Ever since I saw Borgdog's Nobs table I've had an interest in making my own home brew table. I've authored or helped out on over a dozen Visual Pinball EM tables and a few months ago I got the opportunity to help out on a Cuphead table that I really liked. The table has a great theme, fun game play, a good rule set, interesting graphics and a lively sound track. After we got the table released I asked the designer of the table if I could use his resources to make a home brew version of this table and that started up this project.

My first hurdle was learning how to work with Mission Pinball Framework (mpf). The really nice thing about mpf is that it automatically takes care of all of the low level processes like player and ball management, extra balls, multiball and interfacing with OPP boards. The drawback to working with a framework is that it expects exactly what it wants and if you don't give it exactly what it wants, you have problems.

The mpf site has great documentation and the google groups page has a lot of very nice folks to help noobs like myself make progress. After about a month of trial and error (much error) I ended up with a working mpf set of configs. MPF has a "monitor" section that lets you set up an image of the playfield with all of the switches, coils and lights on it and then click on these to see how the game plays which is critical to troubleshooting the configs. The monitor can also display variable values and other data to help sort out how your configs are running.

Since I had played the table many times in its visual pinball format so I knew that it should be a fun table but I wanted to try out the mpf vpx bridge before I bought a donor table and invested a lot of time and money into a project that either wouldn't work out or wouldn't be fun to play. The mpf vpx bridge is a software interface that lets you run your table in mpf while letting vpx take care of keyboard interface, physics and letting mpf know which switches have been hit. With this interface you can (sort of) white board the table before you start investing in a physical game.

Here is a video of game play in the mpf vpx interface. I could only get one audio track to record so you'll only hear the mpf sounds. You won't hear any vpx sounds like solenoids firing, ball roll or anything else that vpx would play. Also because my video card was driving three monitors and trying to record, there is a bit of lag to the game play (which accounts for some of my poor play). Lastly vpx has a ball control feature which is super helpful in testing hard to hit shots. Unfortunately the SOUL lights are shifted by the flippers and the flippers on my vpin cab also control the left and right ball motion so there's a little funkiness there. Additionally if you jam a vpx ball (with ball control) into a drop target or stand up, you can generate multiple hits which you will see late in the video. This would never occur in normal vpx play but I figured you would't want to watch an hour long video...

Now that I knew the software is working I needed to start working on the hardware side of things. I took a look around at the various options for driving the table and wanted to go with the least expensive stable option which lead me to the Open Pinball Project. I was a little confused at first because it looked like there was a change in the small driver boards because the original boards are now obsolete.

Through more searching I came across a new prototype board that was looking for beta testers called the Cobrapin Board designed by @cobra18t. The details of the board can be found at this site in the mpf docs.

https://docs.missionpinball.org/en/dev/hardware/opp/cobrapin/index.html

The really slick thing about this board is that instead of multiple boards for power, switching, lighting and solenoid control, this is a all in one solution. I ordered one of these configured for use in a 24vDC table. Cobra18t has been great to work with and has been very helpful in answering my questions about this project.

20201103_074540 (resized).jpg

This table was designed by Onevox and A_Bord did quite a bit of blender and other graphical updates. Great to see them following along!

Next step was to find a suitable donor machine. My original thought was to pick up a cheap non-working EM so I contacted the guy that I've bought all of my rehab projects from to see if he had any.

https://vpinball.com/forums/topic/just-bought-my-first-table/
https://vpinball.com/forums/topic/darling-restoration/
https://vpinball.com/forums/topic/pirate-gold-restoration/
https://vpinball.com/forums/topic/grand-tour-bally-1964-restoration/

He said that he had an Allied Leisure Suit Thunderbolt from 1976 that powers up and plays for $150, so I hopped in the van and drove to Northern Illinois to pick it up. The cab is in good shape and well built, once I got it home the game powered up and played flawlessly as advertised. It has a really nice chime unit in it that I'll sell off as well as the MPU, score and credit displays. The downside of ALI is that they use white nylon plastic where others use steel. There's a cheap feel to the components which I'm sure was a big part of their business model. The good news is that most of the components are still functional after 52k plays and 44 years so they exceeded their intended life span.

The table has pretty much everything except for the lockdown bar. This is unfortunate because ALI uses a bar with pins on it like Bally did in the 1960's so that is going to be hard to find. I should be able to use the pop bumpers, flippers, slings, three stand up targets, posts, arch and apron so that's a good start on parts.

ALI's artwork is pretty bad and the game play looks to be so so. This is a good thing because it doesn't make me feel too guilty taking a fully functional table and gutting it out.

20201029_140146 (resized).jpg

20201029_140312 (resized).jpg

My current plan is to take the VP playfield graphic to a print shop and have it printed at full size. I'll glue that down to the plywood and hand cut a whiteboard from that so that I can test out the machine before making the final version. If the white board is good, it will become the final version.

Another option is that I know a guy you has a CNC machine that can accept a full 8' x 4' sheet of plywood. I could make a gcode file with my Vectric software and have him cut it for me but that would probably add a fair bit of cost.

mbwalker that sounds like a good plan because you could do various shapes with it. I was initially thinking Forstner bit but that will be hard to control and manhandling a playfield with a drill press is probably more work than it's worth. BorgDog how did you do your inserts on Nobs?

This looks like it could be pretty useful for boring the counter sunk insert pockets so long at the chuck and shafts don't have much run out to them.

https://www.amazon.com/Milescraft-1318-DrillMate-Drill-Guide/dp/B014A1Z92I/ref=sr_1_2

71esbu541HL._AC_SL1500_ (resized).jpg

Nice weather, November and Wisconsin don't usually go together so I took advantage of today's temps in the low 70's and cleaned up and painted the legs. Since I have a nice set of black leg bolts left over from a previous project, Cuphead will get black hammer painted legs.

20201105_161203 (resized).jpg

Yes the bike stand is like a lot of tools, once you have one you'll find other uses for it.

Can anyone recommend these two options for insert and GI lighting respectively. I need to work with 5v DC.

https://smile.amazon.com/dp/B01DC0J0WS/

https://smile.amazon.com/dp/B01AU6UG70/

71OTLWRlAnL._AC_SL1100_ (resized).jpg

71kuC8MiyZL._AC_SL1500_ (resized).jpg

As I mentioned in my earlier posts, one of the goals of this build is to keep the costs down. That means reusing parts from the machine whenever possible so today's post looks at converting the cheap looking ALI 3" flippers into 2" flippers. I say cheap looking because although they are all plastic, they are still stable and functional after 52k plays so maybe ALI was on to something?

ALI uses lots of nylon plastic in their design, the flippers are no exception. I spent some time looking at them to figure out if there would be a way to modify them. These flippers have a star shaped end that fits into the linkages to drive the flipper. On the plus side, this only gives a few possible orientations, should make a very strong joint and avoids the issue with the set screw making an indent in the shaft that prevents setting the flipper orientation just right. The downside is that I'll have to rotate the mech below the playfield to get the correct orientation. Since this is a home brew and there looks to be lots of room, I don't see this as a big issue.

I had some 2" bats left over from another pin and some 3" metal bases so I marked up the metal bases and cut them to fit the 2" bats. I removed the top pins on the metal bases, drilled a receiver hole and tapped the hole for the screws that hold the bats on. I then marked the base shaft length and cut to a length that would fit into the ALI flipper shaft.

The ALI flipper bat and shaft are an all-in-one unit so I sawed off the bat and center drilled a hole that could receive a metal base. Because of the amount of stress that these will see, I figured there would be no stronger joint than a thread so I cut a 1/4 20 thread into the shaft and drilled and tapped the hole in the plastic shaft to receive this thread.

When these are mounted on the white board i'll make fine adjustments to the bats with thin washers and then epoxy the base and shaft into their final position. This might work great, it may fail spectacularly, that's what the white board is for...

0 (resized).jpg1 (resized).jpg2 (resized).jpg3 (resized).jpg4 (resized).jpg5 (resized).jpg6 (resized).jpg7 (resized).jpg

We'll see how the flippers work out, thanks for the cnc info. I do a fair bit of cnc work but my machine is not large enough to do a playfield, even in multiple passes. I should be able to hand make the pf without too much difficulty, this is my side business http://www.semillerguitars.com/

More clean up, painting and fixing. Pulled the coin door while the good weather lasts and gave it a coat of black hammer paint just like the legs. At the advice of BorgDog I ordered a vibratory tumbler for polishing small parts. This big stuff works out fine with my bench top grinder and a wire wheel. Here's a photo of the feet being polished because no one wants rust marks on their carpeting.

The pop bumpers are made of, wait for it... plastic of course. Here's a shot of one of them assembled. Unfortunately one of them must have had something drop on it because there is a perfect knockout of the center with the crack lines shaped like a blow to the top. Fortunately it pops back together very tightly so the question was "how to stabilize it." Before I popped it back together I was thinking 3D printed splint but once I saw how nice it went together I thought "epoxy pour."

I have some two part West System epoxy left over from the last kayak that I built so I taped the cracks on the top surface and mixed a small cup (a little heavy on the hardener) and did a pour. I now have a very solid hockey puck. I'm thinking of doing the same to the other two bumpers because it's pretty easy to do now while everything is apart.

The plastic part that squeezes the ball away from the bumper needs to be measured to see if it's the same size as the metal ones because if it is, I think I'll switch those out.

1 (resized).jpg

2 (resized).jpg

3 (resized).jpg

4 (resized).jpg

5 (resized).jpg

The Brains of the Operation was the next step. When my son rebuilt his gaming desktop this fall I made sure to keep his motherboard, processor and graphics card for future use. Several years ago these were the original computer components for my virtual pin cab before that got upgraded to its current configuration.

The biggest problem with this step was that the Windows 10 OS was on a hdd and I wanted to transfer it to a solid state drive for Cuphead. I had a small ssd lying around but it didn't have an OS so I spent the better part of a day trying to transfer or clone the hdd. In the process I screwed up the boot partition on the hdd and I ended up formatting the ssd and did a fresh install of Windows 10 using an key from a computer that is no longer in use.

Here are photos of the computer mounted on a piece of plywood running mpf Cuphead on a TV.

20201109_183213 (resized).jpg

20201109_183224 (resized).jpg

mbwalker thanks for the rust removal tip!

Power, that's the next thing on my list.

There are a lot of ways of powering a cab and when I first looked at this aspect of the project I really didn't know which way to go. I started off by measuring the amperage draw of the various coils in the cab while TB was still functional. The results that I got were jaw dropping. The primary coils on the flippers drew a massive 8.5A while the hold coils drew 1.5A. The drop target reset coils drew 4A while each of the other coils (pop bumpers & kickers) drew 3A each.

If there was a 2 ball multiball a maximum of up to 23A or more (both flipper hit at the same time while 2 other coils go off) could be drawn at 35v for a whopping 805W. Coils can draw even more than their stable current draw when they start up so power consumption could be massive. I knew that the fuse ratings for the solenoids was a fraction of this current draw so I reached out to BorgDog and cobra18t for advice.

Borgdog told me that his TNA uses a 48v 600W power supply (12.5A) and it runs perfectly. This reassured me that since professional pinball designers think that 600W is enough power then my calculations don't apply in the "real world." Cobra18t told me that mpf uses clever software devices to try to space out the firing of coils and it also limits the length of time that a coil fires to reduce power needs. His Cobrapin board also has some beefy caps on the power side of things that can discharge to help out when current draws are large.

Armed with all of this info I knew that I could either buy switching power supplies or reuse what I already have. Since the machine has a great transformer in it that is already driving the coils at an unregulated 35v DC I decided that its 880W (110v with 8A main fuse) would be more than enough for the cab's power needs. In following with a main theme of this project, this was also the cheapest option.

First off I checked out the big electrolytic caps in the machine and two of them had ESR's of over 6 Ohms so they had seen better days but the one that had a 2W bleed resistor across its terminals has a sprightly ESR of .14 Ohms so it still reads that it is in great shape. I reused that cap on the 35v power supply and bought a $12 20A 300W buck converter for the 5v output. I didn't put a smoothing cap on this circuit because the buck converter can take care of smoothing the output.

I pulled all of the other power supply feeds from the "power board" since I didn't need 180v, 17.5v, 6.5v or any other outputs. I then moved components around on the board to make better use of the space and soldered a lead onto the 6.5v AC tap on the transformer. This tap goes to a fused bridge rectifier and the full wave rectified output goes to the buck converter.

Here's a photo of the power supply with a meter showing the 5v DC output. I should also mention that I replaced the old fuse holders since it is just a matter of time before they start breaking off wings. Also there is a single point grounding terminal just above the big cap.

20201110_183638 (resized).jpg

Thanks @jabdoa, can you link this post to my mpf build link?

Thanks, I'm getting closer to actually being able to beta test for you.

The cab is not the most exciting part of any project but it's got to get done. ALI uses lots of different materials at least in this cab. There's good old fashioned plywood as well as particle board and the new modern space age material (at least in 1976) MDF. I've rehabbed cabs from Bally, Williams and Chicago Coin and one thing that's different about this cab is that the paint has an orange peel texture to it. Using this plus the dark grey splatter pattern really did a great job at hiding the imperfections in the plywood of the cab. I didn't notice many of the defects until I got a few coats of white on the cab and head.

One thing that I did a little different this time is that I used my flat bed hoist to hold the cab. This is working out great because it lets me move the cab around the workshop which is really nice when you have a shop that is as small as mine. Much better than the saw horses that I've used in the past. I did some bondo repairs to chipped sections as well as remove the cab's partial serial number.

I also got around to disassembling the other two intact pop bumpers and did epoxy end pours on them. Today I picked up a set of three metal "ring and posts" on Ebay to replace the plastic ones. From my crude measurements from my Williams Tropic Fun, the metal posts may end up being a little too long so I may have to shorten them and cut additional threads.

I'm planning on making custom "Goblett" flippers so I wet sanded off the red hot press "FLIPPER" graphics from each of the 2" bats. I'll use my Cricut to cut a vinyl stencil for these.

20201111_200201 (resized).jpg

20201111_210111 (resized).jpg

20201111_210127 (resized).jpg

The flippers will be a stencil so I can match the paint colors with Createx. Currently I'm working with A_Bord on the cab art. My plan is to try to mimic the art style of pins from the 60's to 70's so three color (red, white, blue), splatter background and stencils with small stand offs to get a little overspray around the edges. These are what I'm going with for the cab and head, the front will get two stacks of chips on either side of the coin door.

Cuphead-Cab-2 (resized).png

chips (resized).png

As I'm getting closer to actually building the cab I'm working on tying up a lot of loose ends. One of these it the attract mode light show. MPF has a really cool tool that was made by Pinlandia called "MPF SHowmaker." This tool gives you the ability to make complex light shows with just a few set up tasks.

If you populate all of your non-gi lights in MPF Monitor, MPF Showmaker can know the x/y coordinates of each of these lights. The tool then lets you choose a shape and place it's start position, start color and size and then set up an ending position, ending color and size and transition between the two in various ways.

I think the best way to show is this by demonstration so here is a quick screen cast of the tool in action.

Here's a video of the cuphead playfield in MPF monitor showing a couple of different shows running to demonstrate how this will look on the actual playfield.

Today was splatter day in the workshop. I mixed up some grey Createx paint using 30 drops of white to 5 black and one blue and used a toothbrush to splatter the paint onto the cab. One nice thing about Createx is that it cures with heat so you can work really fast with it. Here's a section of the main cab showing the pattern.

I totally screwed up on my apron and shooter guide painting. I usually use high build auto primer to coat over the old paint on the apron and then shoot white onto that. This time I picked up so Rustoleum all in one Paint and Primer, figuring that would build pretty quickly. It did to a good job of covering up the original ALI paint but unfortunately the original art telegraphed through the white paint when you held it one an angle to the light because of slight height differences in the ALI lettering. I ended up taking the apron and shooter guide outside and used a wire wheel on my bench grinder to take them back down to bare metal. The original ALI graphics paint was very hard to remove. I think you could use it as tank armor.

20201114_133749 (resized).jpg

Painting the artwork is next up. I've done this a few different ways in the past and the way I like best will only work for the "CUPHEAD" logo because the lines separating the red and blue on the characters are too fine for this. To do this technique I project the artwork onto a wall with a piece of poster board on it and trace the letters. These are then cut out with an x-acto knife and used as a mask for airbrushing on the colors.

The nice thing about this technique is that you can get some overspray around the edges that looks just like the old EM's since this was (sort of) the way they were painted. Unfortunately the Createx paint that I use will curl the posterboard (water soluable) so the fine lines between colors on the characters will most likely lead to disaster.

For the characters I fed the graphics into Circut Design Space and cut them from vinyl. The blue sections were removed and the stencils were then put onto the cab and painted by brush. The red sections were then uncovered and painted and all the paint was set with heat (Createx will air dry but will not set until hit with heat).

The final result turned out great! Now for the other side, head and front...

20201114_180621 (resized).jpg

20201115_130152 (resized).jpg

20201115_130856 (resized).jpg

20201115_132428 (resized).jpg

20201115_144603 (resized).jpg

20201115_145509 (resized).jpg

The painting saga continues... Stencils were made for the sides of the head and front of the cab with vinyl on the Circut because the lines are way too fine for cardboard. I had some minor problems with the vinyl pulling up my white paint where the paint went over the original yellow and orange of the cab but otherwise the job went smoothly.

20201116_133608 (resized).jpg

20201116_133618 (resized).jpg

20201116_133629 (resized).jpg

20201116_133633 (resized).jpg

20201116_141314 (resized).jpg

Thanks!

I don't know why but I'm always excited when a machine can stand on its own four feet. Here are photos of the cab, head, coin door and rear cooling vent all back together again. You can now see how the artwork from the cab blends with the head. I've got a Williams lockdown bar sitting on top for reference.

20201117_161904 (resized).jpg

20201117_161923 (resized).jpg

20201117_162136 (resized).jpg

20201117_162200 (resized).jpg

Speaking of lockdown bars, I've got the lockdown bar blues...

The cab was missing its lockdown bar but does have the receiver. I contacted a couple of pinball parts guys that I've bought from in the past and neither has one, has seen one or knows someone who might have one. ALI used a lockdown bar with posts and slots, sort of like Bally did in the early 60's but the spacing of the posts is different (I tried a 64 Grand Tour lockdown bar just to be sure).

I was afraid that this could be a problem so I pulled an "easy to find" Williams receiver and bar from another pin but unfortunately there are issues with interference of the receiver with the plunger. In addition to this ALI aprons are 9" deep while Williams are 8 1/2" deep. The difference has to do with the depth of the lockdown bars (ALI's hangs over the front edge while Williams is all inside the cab). Here's a photo showing that I can't get the old PF to fit into the cab because of the 1/2" difference in sizes. I have to be mindful of plunger reach because I've got to get the playfield and shooter guide into the right position to meet the plunger correctly.

I've put a wanted ad on Mr. Pinball but who knows what good that will do. I think my best bet is to make my own "receiver" and use a Williams lockdown bar. I took a look around and an old buddy of mine (Hauntfreaks) in the VP world came up with this idea for holding down what looks to me to be a Williams bar. I should be able to drill into the metal housing that is spot welded onto the casing of the bar and attach clasps that the hasps can clamp to pull the bar down. This is very similar to how arcade machine control panels are held in place so it should work out fine once I get the geometry figured out.

15708766403_0b45b2bb03_c (resized).jpg

20201117_125406 (resized).jpg

I haven't played Cuphead either but I love the "rubberhose" animation style of the late 20's - 30's and was drawn in by the VP table because of that.

Today was a good day for a "road trip"! I had spent the better part of an hour trying to improve my beat up spare Williams lockdown bar and was realizing that it was probably a loosing cause. I then received an PM from animesuperj saying that he had a lockdown bar that I could use so I drove about 40 miles to pick it up. I've met a fair number of folks through pinball and the vast majority are great people. Thanks animesuperj, after this Covid thing is done you're welcome to come by and play Cuphead first hand, that is assuming I get it built and it's worth playing...

Well this is no white washed series of posts, you get the success along with the failures...

Painting the apron, shooter guide and the flippers should be a walk in the park, I've done it several times before without a problem. All you have to do is lay out the graphics in Photoshop, cut vinyl with the Circut, mix Createx colors, paint and then pull the masking off and bask in the glory of beautiful pinball excellence. Yeah, its worked out that way several times in the past but NOT TODAY.

I did all of the fiddley stuff mentioned above and got all of the masking in place, mixed the paints, painted the graphics and then pulled off the masking. That's where things started to go South. You may recall that I mentioned that I didn't do my usual high build primer under the color coat because I picked up a paint and primer all in one product? Well some of that beautiful white paint lifted up with the vinyl so I was left with some bare metal spots. This wasn't the usual crappy paint under primer that fails but paint and primer in one failing on clean bare metal. This is bad.

I then proceeded to mask off the colors, filled the pits with Createx white and then shot the whole thing with more Rustoleum white. Why did I use the crappy paint? I'll answer that with another question, "what's the hardest color to mix?" At least for me it's white because there are so many varieties (just check out byou local paint department if you don't believe me) and if you are off by just a little bit in terms of hue and saturation, it looks horrible.

The end result was pretty good and I'll seal the whole job in with clear coat once it sits and cures for a week or two. Cue the Darth Vader Soundtrack "Rustoleum Paint and Primer, you have failed me for the last time!"
20201118_184130 (resized).jpg20201119_095950 (resized).jpg20201119_111247 (resized).jpg20201119_125045 (resized).jpg20201119_150838 (resized).jpg20201119_150910 (resized).jpg

It's kind of fun to see your virtual artwork in real life!

Time to take a break from artwork and get back to building stuff. As I mentioned earlier the game did not come with a lockdown bar but I was able to get a GamePlan (not ALI) bar from nimesuperj so I needed to come up with a solution for a receiver. This is a harder problem to solve than it might seem at first because there is a lot of geometry to take into account.

First off, will the receiver be narrow enough to let the 42" playfield drop into position? ALI receivers hang out over the front of the cab towards the player and the retaining pins that lock into the receiver actually slide into holes bored in the top of the cab front. This meant that a standard Williams or Game Plan receiver will not fit. In addition to this the playfield has to sit at just the right location so that the plunger can reach a ball resting against the shooter guide. Also the bars that run on the back side of the lockdown bar need to be able to clear the top of the apron.

Yeah, lots to consider! I ended up using some adjustable locking hasps from Amazon to achieve all of these factors. The nice thing about the GamePlan lockdown bar is that the locking tabs are further towards the edges than the Williams bar so I was able to use those as anchor points for the hasp grabs. I needed to grind down the stabilizing bar that sits closest to the playfield to get it to clear the top of the apron.
20201122_191417 (resized).jpg

Next up I needed to get the computer, hard drive, power supply and table power feed mounts figured out. The game board from Thunderbolt was not wide enough to fit the CPU and CPU PS so I ended up adding 1/2 inch plywood extenders under these to mount them to the board. I 3D printed a bracket to stabilize the GPU as well as a bracket for mounting the HD.

To feed power to the beast I bought a 15 foot three conductor extension cord and wired it to a junction box with double outlets in it. I ran a wire from the incoming hot to the game's on/off switch and shortened the original TB power cord. In addition I removed the TB power feed access to the on/off switch and thus set the power to be always on. This way when the table's power switch is flipped, it energizes the four outlets which turns on the pin.

I've set up the BIOS of the CPU to boot when it sees power so switching on the table automatically starts the CPU. The extra outlets will be for the backbox TV and possibly a power supply for the amp needed to drive the speakers. My current plan is to feed the amp 12v DC from the CPU PS but if this dirty power causes audio problems I'll have the option of adding a dedicated PS for the amp.

20201122_191356 (resized).jpg20201122_191523 (resized).jpg20201122_191558 (resized).jpg20201122_191607 (resized).jpg

Finding a TV for the machine was a bit of a problem. The goal was to make the TV as wide as possible but because of the width of the machine most 40" TVs are too wide by about 1/4" with their bezels on. One common fix for this in the world of VP is to uncase the TV to remove the bezel but this is fraught with problems since the thin film display will fracture or loose pixels if you look at it funny. I personally have destroyed one TV in the making of my Multiball Vpin and I had no intention of repeating that stunt here.

I drove over to Lake Geneva's Best Buy to take a look at a 39" TV from Vizio but when I got there there were none on display to see how the TV looked. The TV was also HD which I've learned now means 720p while "Full HD" means 1080p. I was not about to buy a TV without seeing if the colors were washed out or if the pixelation would drive me crazy so I went about measuring 40" sets to see if one might have the magical width needed to fit into the cab.

It turned out that a Full HD 40 set by Hisense would squeak in by about 1/16" so I brought it home and by the grace of God it fit.
20201124_152234 (resized).jpg

Next up it was time to take the training wheels off and see if I could get MPF to talk to the CobraPin board. When MPF is run in virtual hardware mode it doesn't need to see the exact switch and coil numbers that a particular hardware system like OPP requires. This let me get by to this point in time with rather generic numbers for switches and coils like "1".

I took my photoshop image of the playfield and laid out imaginary wire runs to see how best to run the wires for the game's switches. Because CobraPin does not have enough direct switch inputs for Cuphead I had it set up to use a switch matrix in addition to 22 direct inputs. I used the direct inputs for high priority switches like flippers, pop bumpers and slings as well as remote switches on the cab that I didn't want to send multiple lines from a matrix to like start and the various tilt/slam switches. The matrix is currently set up as 4x6 just because I figured it would have fewer wires than a 3x8 matrix. If this is not the case I hope someone with experience wiring switch matrixes will let me know.

I applied USB data and power to the CobraPin and had all kinds of mpf errors that I had to sort through. These were all "switch and coil number don't exist" errors that I was able to make go away with changes to my switch and coil numbering system. Once I had eliminated the errors it was time to apply 5v DC from the power supply. The little red LED for the supplemental 5v LED lighting went on so this was a good sign. I then hooked up a 50 bulb string of addressable LEDs and let mpf drive them to display the attract light show that I made earlier. SUCCESS!

Backbox, it will hang down into the back of the cab and display scores, videos and other artwork during game play.

Today was the day to make big cuts into the head and cab to make the TV fit. I started out by removing some of the MDF trim on the base of the head's front panel and then drilled holes through the bottom of the head and top of the cab back to see how things lined up. The back plate on the cab is the limiting factor as to how far forward the TV can sit in the head. The further forward the better because it will give me room to mount speakers and baffles inside the head.

I drew the TV cut out profile on the bottom of the head and cut it out with a saber saw. I then put the head onto the machine and traced that outline onto the top of the cab. There were reinforcing blocks behind the back plate in the cab so those had to be cut and pried off the give access to the entire span of the cab so the TV could fit in.

Lastly the original backglass did not sit perpendicular to the floor and neither will the TV. I put the TV on top of supports in the bottom of the cab to line everything up and see how much fill needed to be placed in front of the TV at the bottom front of the head. I picked up some .080 acrylic that will sit in front of the TV and will be painted on its back side to fill the side spaces around the TV. I needed to add an allowance for this .080 at the top of the TV and also cut a groove in the back of the fill pieces at the bottom front of the TV. Lastly I cut a spline to fill the old groove for the backglass.

After all the pieces were fitted I cut a piece of wood to span the back of the head and drilled holes to accept M6 screws and mounted the TV to this plate. Once I had the TV in position and square to the top and sides I used the old mounting brackets for the head's light/credit/score display plate and fixed the TV mount to the head.

Lots of fiddly work but it's very important to the overall look of the machine.

20201125_095501 (resized).jpg

20201125_112728 (resized).jpg

20201125_200154 (resized).jpg

20201125_200244 (resized).jpg

I'm not sure about how hard it will be in comparison since I've never played the video game. I do like the animation and theme though.

Vader77 I really wish that I had done this on my Vpin. When I built it I thought that the BG and DMD had to go on their own screens so I built mine that way (shows how much I knew...). Were you the first one to set up a Vpin this way because I thought I was copying Borgdog's idea but I see yours is from way back in 2013!

BorgDog the space to the sides of the TV will have speakers or speaker baffles mounted on it. I'm going to paint the plexiglass on the back side to make the paint look reflective and deep. The side spaces will have plywood cut for them that will be hidden behind the painted plexiglas. If I can get speakers that are narrow enough I'll mount them to the plywood. If not I'll 3d print speaker grills and baffles that I can use to direct the sound from speakers mounted on plywood behind the TV to the output holes beside the TV. I'll also put a car subwoofer in the cab and drive all of the speakers with a Lepai amp powered from the CPU PS unless it is too dirty in which case I'll add a dedicated 12v PS.

Here's a photo with the plexiglass mounted and all of the fill trim painted, splattered and sealed.

20201127_211939 (resized).jpg

Thanks @onevox! How many ohms are they and are they full range speakers minus the bass (ie not just tweeters)? The 10" sub woofer speaker that I have is 4 ohms.

Awesome, I asked my brother who is a sound expert and he said the impedance of the speakers doesn't matter since the amp has separate control for the sub and speakers so that can be used to match them. I'll definitely take them, I'll PM you my address and see if I can send you something to cover postage.

Thanks Boatshoe my brother said the same thing but from my point of view it is progressing at a normal pace for a project. Once I get to making the playfield it will slow down in terms of things to post about.

I put the car sub into the cab and the enclosure is massive so I'll pull out the speaker and ditch the case, I guess the pin cab will be the enclosure! On my Vpin I built a 1.5 cubic foot enclosure for the 12" sub per the subs spec sheet and I wish I didn't. It takes up a lot of valuable real estate and I doubt that it effects the sound very much since it is a sub and already is in an enclosure (the cab).

I held the sub speaker up in a bunch of different spots and the best one looks like the very back of the cab facing the wall behind the machine. It looks to be a good spot so long as the grill doesn't impede standing the cab on end. One other thing about my Vpin that I wish I hadn't done was put a grill over the sub's cone that projects out from the cab. It prevents little kids from sticking pointy objects into the cone but it also makes the bottom of the cab not flat so I can't just put my Harbor Freight lift under and pick the pin up. I have to span the sides with 2x4's so the grill isn't crushed when the cab is lifted. Live and learn...

Subwoofer mounting and custom grill. I moved the 10" sub speaker around the cab and found that the back wall of the cab was the best place to put it. I made a CNC pattern for a 1/2 thick ring to act as a standoff between the speaker and the cab and used it's internal circle to scribe a line that I used to cut out the opening with a saber saw. My saw blade was a little dull and the old ply was pretty tough because the process put almost as much smoke in the air as was at the late 70's Foghat concert that I went to at the Milwaukee Arena.

I needed a grill for the speaker that was thinner than the standoff feet on the back of the cab so I pulled out some scrap acrylic and made a CNC pattern for that and CNC'ed out the grill. I wanted to experiment and see what hammer paint would look like under acrylic so I sprayed it with the same paint used for the legs and coin door. It turns out that hammer paint works by clumping so that there are thin areas that are nearly transparent and other high build areas. I thought it looked best with the paint side out so that's how I mounted it.

To add a little jazz to it I made a Cuphead graphic out of vinyl and put it in the center. No one will see it but I'll know it's there...

20201129_111324 (resized).jpg

20201130_091026 (resized).jpg

20201130_091036 (resized).jpg

20201130_094009 (resized).jpg

20201130_094017 (resized).jpg

Hey BorgDog what sized router bit did you use to make the tapered route for the shooter lane? If I remember right you made a sled and ran the router on the incline of the sled to make this cut. Very nice idea!

Did you release the PUP? I'd love to see it.

I've not posted for a couple of days because I've been slogging my way through how to make the backglass in mpf. I don't know why but I have such a hard time figuring out how mpf works but I do. The documentation is all there but I find that I'll miss critical points and then stumble upon the answer that had me banging my head against the wall right there in the docs. At any rate I now know how to divide the "window" up into "displays" that I can send slides to so that the backglass can consist of a "topper", "main viewing area" and "scoring section." All of this is working on my rotated screen which only uses a portion of the area available (which complicated things at first).

I guess if I had a time machine I'd go back a couple of days and tell myself:
1) MPF always tries to center things
2) MPF's 0,0 coordinate is in the Lower Left corner not the upper left
Those two pieces of information would have saved me hours of wasted time.

Thankfully I have a graphical genius (@a_bord) helping me out with the artwork so I'm pretty sure that the end result will look great. It's getting there that is the hard part.

The past week or so has been spent working on getting the animated backglass designed and coded. Thankfully I've gotten help from A_Bord with his graphical genius and Anthony from mpf with his crazy good computer science skills.

A_Bord came up with some really nice artwork which included a set of animated scoring reels that he made in blender and then generated a series of PNG animations that I can use in mpf for the graphics. The problem is that there is no way in mpf to implement this. MPF does have the ability to drive physical coils of score reels but nothing is there for something more virtual like these. Anthony came to rescue and has developed mpf code to make this idea a reality. It is not fully implemented but the movie output that he sent out shows that it will look stunning once in place.

https://www.dropbox.com/s/n5hth12a8rn6lwd/virtual_score_reel.mov?dl=0

I was able to learn how to make a "window" in mpf which is really the area of the TV that graphics can be displayed. Onto this window I was able to make "displays" for the topper, main section and scoring section. This took me a crazy long time because my mpf skills are pretty pedestrian and because my TV is spun 90 degrees so locating the "window" was a bit of a challenge. In the end I got it working, learned how to make animations in mpf and also used ffmpeg to make 10 Boss Videos that will play on the main screen while a ball is in play.

Here's a video of the current set up minus the score reels. I've got A_Bord's initial score section graphic in place. If you look closely you can see my spinning star animation behind Cuphead and Mugman on the topper. Subtle but cool!

Slow progress, here is the Attract Screen minus borders between the sections.

The top section plays a widget that moves "PRESS START" across a static background. The middle section plays a 3 second animation I made in Photoshop of Cuphead and Mugman thanks to artwork from A_Bord followed by 20 second silent clips from each of the 10 bosses featured in the pin. The bottom section has a place holder score screen that will be replaced by virtual reels. I'm planning on changing this to a 2 player game because I think it fits the theme better. Player 1 can be Cuphead and Player 2 Mugman.

Today I installed the speakers and Lepy amp. I took a 12v DC lead off of the computer power supply to feed the Lepy but unfortunately I got a lot of computer hash noise in the amp. I then hooked up a bechtop power supply to the Lepy and it drove clean and strong. I pushed the Lepy past what I would consider a realistic amount of output and the benchtop PS showed that it was putting out less than an amp. This confirmed that I was feeding the Lepy dirty power from the PC so I then found a 12v wall wart and used that to power the amp. My brother is designing a power filter and I'll probably build that to clean up the cpu PS.

Thanks again Onevox for the sound bar speakers they work extremely well and fit the space perfectly.

More work on the gameplay graphics. I now have animations that will pop up on the screen for the CUP, MUG drop targets, SOUL top lanes and both dash saucers.

20201211_194825 (resized).jpg

20201211_194833 (resized).jpg

Eight days since my last post and the backglass is finally getting close to being done. Working with bleeding edge mpf has been slow and frustrating and I am very much looking forward to getting back to making saw dust and solder smoke. I have literally spent days getting to this point.

Currently the backglass has 20 second cut scenes of each of the 10 bosses with a 3 second animation between them. The display is divided into a static topper, middle video section and lower scoring section. I've gone through multiple versions before deciding on the current one. In game play the full length videos of the 10 bosses are played at random with the theme song for each boss matched up to the video.

There are many animations and scenes that play during the game including, CUP/MUG drop targets and the 10 soul contracts that are collected for clearing the banks, SOUL top lanes, Dash saucers, Knockout, Perdition entry x 3 and more.

The latest addition was that as of yesterday the dev version of mpf now supports "digital_score_reels" which are zip files with a png animation sequence of a score reel with transitions between digits. At present the system will advance a "nines rollover" as soon as the points are scored so that looks a little funny (score 50 points and it goes 00180, 00290, 00200, 00210, 00220 not 00180, 00190, 00200, 00210, 00220) but all in all getting this to work has taken a ton of time and looks pretty awesome.

I've got all of the code written that I think I need at this point in time so now I'm just waiting on the last few parts to arrive before starting making the playfield. One aspect of game play that is very important to the design of the game is the "Dash" shot. There are two "Dash" saucers that are expected to kick the ball up towards the top of the table.

This is one of the problems with bringing a VP game to reality is that in VP you can make playfield devices do things that are not what they are designed to do. A saucer kicker (or hole kicker) is just meant to capture a ball and kick it back into play. There is no expectation that the ball will travel far from the hole.

I picked up three matched Williams kickers from E-bay with the hopes that all will kick the ball with the same strength. Today I built a test bed to see just how far a hole kicker will kick the ball. I set the angle of incline to 4 degrees and powered the coil with the same 35v DC that will be used in the actual game. With a 1/2 piece of plywood and no bevel cut into the hole edge the ball would only go as far as the first line in the video, with a bevel cut into the hole edge the ball will make it to the second line. Both of these distances are only about 4 inches and are way short of what the game design calls for.

I then got to looking at the kicker mechanism and I figured that if the throw of the slug for the kicker arm were increased, maybe the ball will kick further out of the hole. I added a couple of 1/4 washers under the bracket that holds the coil and tried again. The results can be seen in this YouTube video.

Does anyone know why the designers of these kickers put the spring loaded coupling between the part that actually kicks the ball and the part that is connected to the coil slug? It seems to me that this would dampen the kicking strength of the unit but I'm sure that there is a very good reason why they used a much more complex mechanism. I just have no idea what that reason is?

At any rate it is reassuring to know that the Dash kickers can be modified to get more distance from a saucer kicker!
20201227_191606 (resized).jpg20201227_191630 (resized).jpg

Still waiting for the playfield inserts to arrive so time to design parts for holding the LED lights to the playfield. I used F360 to make a clip to hold the ws2811 lights through holes in the playfield so that the light bulb section sits proud to the surface of the playfield.

I bought a bunch of ws2812b light chips with the hopes of custom wiring them up so that the wiring harness exactly fits the playfield. The chips have two sets of three solder pads for +5v, gnd and data lines and these look tight to each other and it will be hard to hold the wires in place while they are being soldered. I designed these two holder parts in F360 to hopefully help out with soldering and installation. The chip holder section has two sets of three channels in it to index, space and hold the wires for soldering. This chip holder's sides have a bevel to them that mates up with a matching set of bevels in the mounting bracket so that the two slide together and lock. The mounting bracket has two holes for screws to attach it to the bottom of the playfield where they will project through translucent white inserts. The white inserts will let me change the colors of the inserts during light shows.

We'll see how well all of this works out in reality when the playfield is made...

20201229_112205 (resized).jpg

20201229_112226 (resized).jpg

20201229_112252 (resized).jpg

20201229_112317 (resized).jpg

Started making saw dust today, here's the white board playfield with the side rails, apron and top arch mounted to it.

20201230_161656 (resized).jpg

Thanks! I think it looks pretty sweet too but I'm handling it way more than I would like to at this stage. It still doesn't have any clear coat on it since I was planning on hitting it with Spraymax when I clear coat the playfield.

At any rate it is great to see this start to come together. Even though it's only a blank piece of 1/2" plywood it still looks like progress after over a month of just working on MPF.

I made a Photoshop poster of the playfield at 1 to 1 scale and put it on the blank plywood to work out the details of how everything was going to go together. There are a ton of details that needed to be worked out like how will the playfield mechanisms fit under the playfield and not block each other, switches or lights. My mechs are ALI and have different geometry from what Onevox used for his VPX layout so that had to be taken into consideration as well.20210101_100400 (resized).jpg

After all of that had been worked out, I printed out circles with center holes on transparency film and used that to mark the centers of the 3/4", 1" and 1 3/16" inserts with an awl. Once everything had been marked I pulled back the paper overlay and drilled 3/32" holes for each mark. The areas that could be reached with the drill press had that used otherwise these were hand drilled. 20210101_100446 (resized).jpg20210101_101817 (resized).jpg

For those who haven't used Forstner bits before, they cut an flat bottomed hole but first cut around the edge and then clean up the middle. They follow a center point until the perimeter starts to cut and then the perimeter guides the bit. Because of this you never want to cut all the way through the wood from one side because it will badly splinter out the bottom side unless it is supported very well. For this reason it's common practice to drill part way through from one side and then flip the board and finish the hole from the bottom. You have to leave a center hole for the second side so you have to not bore so far though from the first side that you mess up the center hole.

I bored out the bottom insert profiles with a Forstner bit that was 1/8" smaller than the insert itself. This is because inserts need a shelf under the to restric how deep they sit in the hole. I bored about 2/3 of the way through the playfield and did the bottom side first because that way the center plug will be drilled down onto the bottom hole when the top insert section was bored. This would leave a cleaner finished product than boring it out the other way.20210101_100523 (resized).jpg

The inserts that could be drilled with the drill press had the stop on the drill press set to stop the top hole at exactly the right depth. The holes that could not be reached by the drill press were hand drilled and I needed a way to know when to stop. I designed a part in F360 that could be 3D printed that fit around the Forstner bit, had a depth stop that was adjustable with a set screw and also had clean out sections on the sides to let the wood removed by the bit clear the bit. This worked very well and I was able to get all of the 3/4" and 1" inserts bored and temporarily set. I'm waiting for a 1 3/16" Forstner to arrive to bore the last two inserts.20210101_132758 (resized).jpg

20210101_133051 (resized).jpg

Sure PM me your email address and I'll send you the stl file. It fits my 1" Forstner bit but I used it on my 3/4 bit as well and although it was slightly off center the bit was run at such a slow speed that it wasn't an issue.

Today I finished up most of the playfield cut out work including cutting the shooter lane groove. This one is a complete rip off of a technique from @borgdog. As he had told me in a PM it is very hard to find a 1 1/16" round nose router bit so I went with a 1" bit to cut the groove.

The trick is to make a sled for your router base with a slot in the middle that is wide enough for the 1" bit and set up fences on either side to guide the router down the middle. On the far end you put a slat of wood that is the same thickness as the deepest cut of the groove and then set the router to just touch the table at that end where you want the end of the groove to be. All you then have to do is run the router the full length and presto, you're done. Works like magic!

20210103_194344 (resized).jpg

20210103_194851 (resized).jpg

I had picked up a four ball trough from Terry at Pinball Life a few weeks ago and today I figured out how to mount it. I looked at a few modern Stern machine playfields and saw that they cut a large rectangle for the trough which is what I did as well. I didn't think to buy a scoop for the trough since I wasn't exactly sure how all of the geometry would work out so today I spent time designing scoops in F360 and printed them from ABS. It took a couple of tries to get the geometry right and the scoop is printed is draft resolution with 50% infill for a quicker (2 1/2 hour) print. The actual part will be 100% infill with a finer layer height.

Since I used the original ALI metal trough back guide I could only mount the scoop to one side on the play field. The plan is to drill holes in the metal guide, tap holes in the player side of the scoop and mount it on that side via screws through the metal plate.

Here are s couple of photos of the trough, scoop and playfield and a quick video of the scoop in action (minus the inner shooterlane wall).

20210104_211931 (resized).jpg20210104_212008 (resized).jpg

Sure PM me with your email address.

Well as I said in a much earlier post, I'm going to show the successes and failures of this build.

I started off populating the underside of the playfield by starting with the lighting. I figured that you populate a PCB starting with the lowest components like resistors so why not start with the lowest component on the playfield, lighting. Everything was going along great until I saw that some light brackets were going to interfere with roll over switches so I figured that those switches and other components like kickers, dts, pops, ect should go on first.

I started with the three bank drop targets since they looked like they would cause the biggest problems because of their size, I was right. I bought two really nice old gottlieb 4 bank dts and figured that I could pull one target and use those as 3 dt banks. I spent half a day working on getting those to fit using different orientations, configurations, I even cut one down to be a three dt unit (they are kind of modular) and it still wouldn't fit with the Dash kickers. The reason for this is that the coil that resets the dt's sits out to the side of the unit and increases the footprint of the unit. I needed to find another solution or I would have to change the geometry of the playfield layout. Something that I REALLY don't want to do since I know this layout should be good.

I took a look online and saw that Pinball Life sells a "modern" 3 bank dt where the coil sits within the footprint of the dt bank. I called to get the dimensions of the bank and it looked like it would fit so I drove down to Huntley to pick up two of them. They cost more than my donor machine...

I hooked up the DT to my 35v PS to see how it would reset and it didn't. It could reset 2 dts but not the full bank of three. It was then that I realized that these coils were designed to work at 50v not 35v, so now I was faced with either making a 50v ps for the table, increasing the coil strength or decreasing the return spring strength of the dt mech. More to come on this

On a positive note I have found a way to increase the kicking distance of saucer or hole kickers. If you raise up the coil of the kicker, the kicker's arm can move further along its arc and kick the ball a longer distance, here's a video of that in action. This means that the VPX Dash kicker shot can be reproduced in real life!

This post is intended for mature audiences only, please so not read this and drive as this can cause drowsiness.

I really don't want to add a 50v PS to the cab and finding springs with lower tension for drop the targets will probably be very difficult/impossible so I figured that my best bet would be to try to increase coil strength. I have no use for these coils so if I destroy one, oh well.

Coil magnetic field is determined by three factors, Current, Number of Windings and Length of the coil. Increasing the current or number of windings increases coil strength while increasing the length of the spool will decrease coil strength. One trick that I've used in the past is to cut a coil and pull some windings off as a means of increasing coil strength. You might think, "that's crazy, you're decreasing the number of windings so you're going to decrease the coil strength!" Well that's correct but if you think about the geometry of a coil, the last windings take the longest distance around the core of the coil because they sit on top of all of the other windings. If you pull those off you will have the most favorable decrease in resistance per loss of windings.

A coil's resistance value is determined by the gauge of the wire and the length of the wire. The coil that came with the DT unit is a Williams 26-1200 which has 26 gauge wire and 1200 windings on the spool. The bigger the gauge number, the narrower the wire so 26 ga wire has a diameter of .016 while 20 ga is .032. As you decrease wire cross sectional area, you increase resistance so 26 ga wire has 40 m ohms of resistance per foot while 20 ga has only 10 m ohms per foot. By pulling off windings you decrease the resistance of the coil which increases the current (I = V/R) so if you increase the current faster than you decrease the number of windings, you will increase coil strength. That's why this only works for the first windings because they have the longest path around the core.

I started by pulling off 75 windings and the bank reset the dt's better but would not set all three so I pulled off another 75, better still but not all three. I kept on pulling off groups of 75 to see what would happen knowing that this would probably not work and the targets continued to reset more and more briskly until I got down to about 750 wraps at which time the sparks were getting pretty large and there was no added benefit. Oh well, it was worth a try because I had no idea how much coil strength would be needed and maybe I could get one of these coils to work.

Next I calculated the amount of coil strength in milli Teslas (mT) that 50v would provide with an intact 26-1200. Using ohms law I calculated the current at 4.6A and with a coil length of 35mm and 1200 windings that should be around 198 mT. At 35v I was generating 138 mT with the original coil. By pulling off windings I was able to hit a peak of 145 mT at 825 windings and around a whopping 7A of current. This goes to show that when these manufactures design a coil they are shooting for the best combination of lowest current with the best number of mT's.

Next I took a look online to see if there is a stock coil that will deliver over 200 mT at 35v. Pinball life sells a Stern coil that is 24 gauge with 940 turns. According to an online coil resistance chart this coil has a resistance of 5.72 ohms which at 35v is 6.1A and by my calcs should deliver 205mT. I ordered a couple of these from Pinball Life and we'll see how they work out. I also ordered a 1 pound spool of 24 ga magnet wire from Remington Industries (BTW a great online source for wire of all types) so that I can rewind the coil that I destroyed to put it back in service should the Stern coils not be strong enough.

Sure but I already have a beefy ALI transformer and my CobraPin board is set up for 35v so I'd like to stay at that voltage.

I love it when the math works out! Full three bank reset with 35v DC and new Stern 24-940 coil that should generate over 200 mT of field strength.

AD72 good point. Higher voltage makes interfacing with modern equipment much easier but the downside is that it makes the older equipment run hot. If you're reusing parts or buying old scrapped parts like I am then lower voltage and careful coil selection may be the better way to go.

I've got an old presentation slide projector that I hook up to the computer and project onto poster board taped to the wall with that.

More work on lighting. As I mentioned earlier I designed a holder for the small chip 2812b's that spaced out the three wires going in and the three wires going out.

20210110_171912 (resized).jpg

These could be slid into a couple of different sized holders that could be screwed down to the playfield.

20210110_171922 (resized).jpg

Some of the LED's had to be installed under playfield devices so those were inserted into routed out channels

20210110_171929 (resized).jpg

I put together some mpf code to drive the LED's as a test, there are about 12 left to go...

After nearly 300 solder joints, I am now done with light wiring! The GI lights were run and used the little under-playfield clips to hold them in place. Thus far the clips look to work well but we'll have to wait until the game has been played for a while to be sure that they are a good solution. I added in sections of wire to lengthen the GI strand if more than one light would need to be turned off to bridge a gap between holes. If only a single light would not be used to make the span, I just zip tied off the light and turned that position in the strand off in the mpf light configs.

20210113_192408 (resized).jpg 20210113_192423 (resized).jpg

I set the GI lights to 255, 197, 143 but I think that they still need more amber to them. I stressed the system by running 50 GI lights on full white while running the insert lights on an attract pattern. There were occasional flashes in the GI strand so I pulled out the 2200uF cap on the input of the 5v buck converter and put in a 10,000 uF cap and that removed any flashing.

I put a single chip 2812b into each of the pop bumpers and that seems to be doing a nice job with lighting. These are connected by 3 pin molex so the pop bumpers can be removed while the rest of the light strand stays on the playfield. If anyone is looking for a good excuse to buy a 3d printer, check out all of the black ABS parts used in the lighting section of this build. I also printed low profile wire retainers to guide the wires around playfield mechs.

20210113_192024 (resized).jpg 20210113_192055 (resized).jpg 20210113_192104 (resized).jpg

@cobra18t, me too!

Bord the light show that I thought would look so cool is a little underwhelming so I'm going to try to re-do them with strobing the inserts top to bottom and side to side to see if that looks any better.

@Zitt, thanks for the offer! The bumpers are ALI with their plastic base and I'm not sure about the bodies but they look a lot like the ones I've worked with on Williams and Bally EMs. The lights are chip 2812b's and if you ream the channels for the original light wires up by one size from what just fits down the channel (sorry don't remember the bit size) you can still leave a small portion to mate up with the lower section and three 22 gauge wires will fit up the channel.

20210114_153524 (resized).jpg

This is what the modern Stern dt's look like. I'm going to put some super tough vinyl stickers on them for the CUP and MUG.

20210114_153556 (resized).jpg

Today's job is to figure out a way to get leaf switches behind the DT's. The new DT mechs fit perfectly but they leave no room under them for mounting standard EM through playfield leaf switches. I took the switches apart and tried all kinds of different geometries to see I could make them fit including mounting the bracket on top of the playfield but nothing would work. I ended up boring and tapping new holes in the mounting bracket to raise the switch by about 3/16" and them trimmed off about 1/4" from the top of the large front leaf so that it can fit under the plastics.

20210114_154619 (resized).jpg20210114_154700 (resized).jpg

I'm amazed at how well Onevox 's playfield geometry works with actual pinball parts. The cutout for the dts and switch was exactly the right size for the parts. This makes me pretty confident that the geometry of VPX tables and parts is very good.

Sorry I haven't worked on a machine newer than RaB, they have optos for rubbers?

Zitt you're right I was thinking about the Stern coil that I juiced them with, they are these parts from Pinball Life
https://www.pinballlife.com/mm5/merchant.mvc?Screen=PROD&Product_Code=PBL-100-0083-00

Yes but do you know your customer number. All kidding aside, Steve is a font of knowledge.

Thanks Wally, you were a big inspiration to me for undertaking this project! Seeing your shop and playing the games that you've made gave me the bug to try something like this. I've looked all over for Two Brothers Pinball beer here in Janesville but have yet to find any.

For those of you who don't know, Wally is a home builder legend and also happens to be a really nice guy. Over the years he has hand built a bunch of classic pins like Medieval Madness, Cactus Canyon Continued, Scared Stiff and Monster Bash to name a few.

Today a few of the last "problems" to be solved for the table came together. I need a spinner for the left lane and bought a used one (Middle Earth?) that unfortunately is fairly wide and I couldn't figure out how I could mount it and get a hole in the playfield for the drive shaft.

20210116_203359 (resized).jpg

I found a modern style spinner at Pinball Life that just came yesterday. It has a microswitch that sits above the playfield so that should make finding real estate for it easier. I printed heavy vinyl artwork and applied it to both sides.

20210116_203447 (resized).jpg

I also printed out CUP and MUG art for the drop targets.

20210116_203455 (resized).jpg

Lastly, when Bord and I worked on Onevox 's VPX table, we added in a two ball mutiball and needed a way to block off the Perdition lane until one ball drained. Bord came up with a post that could be dropped and raised to take care of this and I looked all over for a mechanism that could fit this bill. I found an "up post" on ebay that was from an old 2" flipper EM that also came yesterday. I stripped and cleaned it and printed stand offs to raise it up enough to fit in the tight space that is available for it under the play field. I had some leftover fiberglass tubing from stunt kite repairs that fit the post like a glove so I cut that to length, mixed up some epoxy that was dyed black and did a pour to fill the tube and glue everything together.

20210116_203544 (resized).jpg

For those who have never done this, you can add a drop or two of dye to epoxy and not effect the cure or strength of it. It is a very common way to do mother of pearl inlays with the black epoxy holding in the MOP and filling any gaps in the inlay routes.

DSC03822 (resized).JPGDSC03834 (resized).JPG

I wasn't exactly sure if those were the right mechs for the job. The really nice thing about the old mech that I got is that it requires no current to keep the post in either the up or down position since it is locked mechanically up or down. The posts on Pinball Life all looked to need to be energized to keep them up and if a player got a really long MB in CH that could be for a while which could generate some heat (around 2.9 A at 35v or 100W). Lastly, the old mech was $4 on ebay ($13 with shipping) and I'm trying to keep costs down as much as possible.

If I couldn't get the post mech to work or fit, my back up plan was to handle entries into the Perdition Lane during MB in the mpf software. Since I wasn't exactly sure how to go about doing that (probably a new mode would have worked) I figured that I would stay with the original VPX design and put in the post to block the lane.

The last two days have been spent wiring up the switches for the table. The Cobrapin board can accept 22 direct input switches and 64 matrix switches. MPF requires switches that are associated with an auto-fire coil to be on the same bank (blue pill board 0 or 1) so these coils influence the switch set up as well.

20210120_175255 (resized).jpg

I ran one direct input bank to the switches that are on the cab like flippers, start, tilt etc. I then set up the pop bumpers and slings on direct switch banks and assigned several more switches based upon location and wire runs to these direct switch banks. The remainder of the switches were distributed to the switch matrix and put onto columns and rows based upon where they were located.

The one big problem that I ran into was the drop targets. The DT unit is set up so that it receives 12v to drive the light diode and then there are four more pins for one column and three rows. The problem is that OPP uses inverse logic to the rest of the pinball world. These drop targets expect positive voltage on the rows and for the column to go to ground when the rows are to be read. OPP strobes by putting the column to 5v and the rows go to ground. For more information on how a switch matrix works check out this site: https://iotexpert.com/pinball-a-switch-matrix/

I worked with Cobra18t for a day trying different combinations of columns and diode locations to see if we could get the DT's to work as a part of the switch matrix but nothing worked. I ended up putting the DT's onto the direct switch banks and reassigned six direct switches to the switch matrix.

20210120_175325 (resized).jpg

Lastly I needed to power up the optos on the trough which required 5v and somewhere between 50 and 100 mA. I put the light diodes in series and used a 33 ohm resistor to give them 50 mA and found that they worked perfectly well at this level of current.

I now can completely simulate a game using the switches on the playfield! It's funny how much heavier the playfield has gotten with all of the mechs and wiring on it. Next up is powering the coils.

Thanks wayinla I looked online last week, found a copy of the Phantom Tollbooth and read it. What a fun book! Sort of like Alice in Wonderland without all of the "trippiness" and it also has a great theme.

Thanks BorgDog for mentioning that the switch matrix pins can be used as direct switch pins. That's great that you don't need to use a switch matrix. It's not that a switch matrix is bad, it's just that there are some confounding issues with it's logic. Cobra18t has been talking with the lead of OPP to see if they can update their firmware to match the rest of the world. Too late to help me but might be there for you. The way that I got around the matrix on the DT bank is that I shorted out the diode on the board's DT column, put a switch ground line to the column pin and then ran three direct switch lines to the three row pins. That works flawlessly. You might want to use an extra external board so that you can use that to run the wires to your cab switches separate from the harnesses that run to your playfield just to make pulling the PF easier.

Well, I screwed up, as I mentioned before I'll tell you about both the good and the bad...

This past weekend I hooked up all of the coil connections and triple checked the diode orientation and continuity down all pin lines to solder joints. Everything was good. I started up Cuphead with no balls and no lights hooked up, hit the start button and nothing happened, good. I dropped in two balls hit the start button and the game started and ball one gracefully ejected into the start lane!

20210122_093640 (resized).jpg

I then dropped the CUP DT's and they all scored and the bank reset on the first pulse of 100 ms and I was surprised that they pulsed a second time but that must be a part of mpf. Same for the MUG DT's, this is going great! I then hit the middle pop bumper's switch and it scored but nothing happened, hit the second and the same thing. I cut power to the machine and there was a little magic smoke smell in the machine so I pulled the PF and couldn't smell what went up so I checked the resistance of the first pop bumper coil and it had minimal resistance. I cut the diode and it checked out OK so oh no, there's a short in the coil! The 8A fuse on the 35v DC had blown so I knew my protection circuitry was working.

I desoldered the coil, replaced the fuse with a 10A and put the game back together and it started up fine. Ran all of the checks as before and everything worked until I got to the right sling at which time I blew the 15A 24v AC fuse with more magic smoke smell. I pulled the PF and then saw my problem. There is a mixture of old AC coils, original ALI coils and new DC coils from Pinball Life.

The old ALI coils all had small signal diodes tucked under some tape that were still connected to the coil terminals that I had seen when I pulled the table apart months ago but were hidden from view on the bottom side of the coil bottoms on the pop bumpers so I never saw them while hooking up those. If by chance the signal diode was oriented in the opposite bias as my diode then I just created a short circuit with the signal diode and pow, there goes the diode and the Cobrapin transistor.

I cut all of the signal diodes, put the pop bumper back into service and when I powered up the machine the pop bumper and right sling both fired upon energizing the table so I shut down. This was because the Cobrapin transistor on these pins were shot so I switched the pins for those (thankfully I still had one open pin on each connector) and the table started up and ran fine. The flippers flip, kickers kick, DT's reset and pop bumpers fire well!

20210122_094714 (resized).jpg

The problem is that when I hooked up the lighting, the GI lights won't work. I tested the insert and GI chains and they are fine so there is another problem with the Cobrapin board or the blue pills that I have no idea what it is. I contacted Cobra18t and he suggested that I send the board back to him so he can troubleshoot it.

I should mention that Cobra18t has a kickstarter going for the Cobrapin board, here is a link to it https://www.kickstarter.com/projects/1862357036/cobrapin-pinball-controller

With the Cobrapin board out for repairs I'm going to have to work on non-electrical aspects of the game for a while. One of these is figuring out how to make all of the many walls needed for the table. I contacted wallybgood and he said that he has used a laser cutting service to cut 18 ga stainless steel for him to make his walls. Since I'm making an original table I'll need some test walls to see exactly what geometry is needed to make this thing play the way I think it should play.

I decided to use F360 to make some ABS test walls and if those work well then I'll make steel walls of the exact same geometry. There are a ton to make so this will take a while. I just wish that I could actually run the game with the Cobrapin so the playfield coils could be used in testing.

20210125_110346 (resized).jpg

20210125_113249 (resized).jpg

All of the 3D printed test walls are now done and installed. With the cobrapin board out for servicing I really can't test the wall geometry so I'll move on to finishing off the acrylic for the head and the speaker grills next.

One concern that I have is that with a full plunge, the ball will bounce off of the rebound rubber, make it back to the shooter lane gate but will not reach the center of the SOUL lanes. Part of the problem could be that I need a stronger plunger spring but part of the problem may be friction from an unfinished playfield. This brings up the question of if I should pull the playfield and give it a clear coat to be sure that the current layout will work well?

A few additional observations are that the left orbit lane probably needs to be moved slightly to the right so that I can get a full length wall for the outside of that lane. It could be that I can just move the light inserts a little on the final playfield, I'm just not sure at this point in time.

The last thing is a general pinball observation. I have a Rocky and Bullwinkle that had two switches that sit next to each other on the same row of the switch matrix go out a couple of days ago. It turned out that a solder connection on a standup target failed and took the switches off of the matrix. While playing the game with the glass off I was reminded at how incredibly violent a pinball game is! The glass muffles the sound so that you don't fully appreciate all of the crashing impacts that take place under it. The fact that 28 years of having its stand up target get hit by direct shots from a flipper finally caused this solder joint to fail really brought this home. Pinball machines lead a very hard life and it's amazing that they hold up as well as they do!

20210130_152853 (resized).jpg

Gamemodder The back story is that I found this vpx table and had no idea what a Cuphead was. I loved the rubber hose graphics, played it a few times and really dug the game play and music tracks so I contacted Bord and asked him if he wanted to do a private mod on the table. He agreed and we made what is now Cuphead Pro but not for general release. I contacted Onevox and showed it to him and he really liked what we had done so the three of us polished it up and Onevox released it as Cuphead Pro to differentiate it from the 3" flipper Cuphead.

This pin has a completely different backglass and some different rules so it is technically not either of the vpx versions.

Zitt Good thought, the rubber rebound is new from PBR but I have experience with Titan's products so that might be an improvement. Looking at Pinball Life I have a standard spring in the game and it looks like Terry sells four different types of springs that are "stronger" than what I have. My hope is that one of these will give the game enough strength to make the plunge shot more challenging.

For the last couple of days I've been designing speaker grills in F360. My first attempts were using a pattern of small 2mm circle openings but the printer nozzle head knocked several out of place so I knew that this was not the way to go.

Next I designed a grill with bars that run crossways and length wise on the grill. This printed well and looked pretty nice so I figured that I'd go with these. I thought that they didn't need a lot of infill so I looked online to see if there is a way to add hard points to a print so that the mounting screws have something to grab onto. I found a video showing how to add shapes to an object that can have different infill patterns and density so I was all set to go. I made a new print using this and saw that the next Youtube video in the queue was a Prussa video on a similar topic.

20210131_163305 (resized).jpg

I watched the Prusa video and was blown away. It turns out that you can use Prusa's infill patterns without top and bottom layers to make grills, this is uber cool! I then made what I always had in mind for the grills but didn't quite know how to pull off in F360. I made a new F360 object with a solid front and lettering on top.

Screen Shot 2021-01-31 at 4.30.43 PM (resized).png

I then set a slab in Prusa Slicer for the grill and set that to 40% honeycomb infill so that the Letters are solid and sit on top of the honey comb grill work. I could also set hard points for the corners for the mounting screws.

20210131_161205 (resized).jpg
20210131_174231 (resized).jpg

And the grill printed super well.

20210131_174111 (resized).jpg

While the Cobrapin is away the mice will play, or at least work on the playfield. I pulled all of the roll over switches, flippers, pegs, rubbers, pins, rails, inserts, GI lights and brackets to sand and finish the playfield surface. I didn't pull the pop bumpers or standups because that would be way too much work. As I mentioned a few days ago, I'm not sure if friction is influencing the way the game plays so I put a sanding sealer coat of shellac on and followed that up with a coat of gloss varnish. I'll put a coat of paste wax over the varnish so that the playfield will be pretty low in friction for further testing. My hope is that this playfield tells me everything I need to know about layout so that I can make a fresh playfield, do all of the graphics, clear coat it and then transfer over the parts all in one step.

I made a new set of side rails and shooter lane rail from some maple and made sure that the apron will fit down the top slots. I've also 3D printed shims for all of the switches as well as thin gasket rings for the pop bumpers. I did this because the TB playfield was exactly 1/2" thick and the plywood that I'm using is a little less than 1/16" under a half inch. These will make the parts sit through the playfield like they were mounted on an actual 1/2" thick playfield.

20210202_181835 (resized).jpg

Lastly Cobra17t checked out my board and the only things wrong with it were the two bad transistors. He thinks that my 5v power supply is the problem so I ordered a Meanwell 5v 50W (10A) power supply and I'll use that in place of my current system. In addition I ordered a 12v power supply for my amp because the 12v CPU power supply is too dirty and causes all sorts of extraneous noise. I've been using a 12v wall wart to power the amp but this has no where near the amount of current to be able to drive the amp to what it is capable of doing.

Thanks @j-rail!

I got the playfield put back together so I did some experimentation with the location of the rebound rubber. The decrease in friction helped some with ball travel from the rebound rubber so that was a success.

As I watched the ball path I noticed that it was not hugging the top arch but was making its way more or less straight across the the table from the rebound rubber to the shooter lane gate. I had my suspicions that the geometry of the arch and rebound rubber were playing a part in my problem so I moved the rebound rubber away from the arch so that the ball would be pinched between the inner rim of the rebound rubber and the arch. The results were a dramatic improvement in how far the ball would rebound from the rubber and now the ball hugs the arch on its way back to the shooter lane gate. The reason why I moved the rubber closer to the arch was because shots up the left orbit need to clear the rebound rubber so I reconfigured the walls in this area to accommodate this. Here's a video of the current plunge shot.

Cobra18t's current thought as to why my lights did not work is that my 5v power supply is not up to the job. I ordered a 10A 5v ps as well as a 12v ps for the amp. Today I moved the amp from the back of the machine to just inside the coin door so that I've got easy access to the volume and other controls. I lengthened the speaker wires and put the new 12v power supply up front by the amp. I now have enough power to really drive the speakers and I have to say that I'm surprised at how much sound these little guys put out.

20210206_170753 (resized).jpg

I put the new 5v power supply back where the Cobrapin board will be remounted once I get it back. I also removed my old 5v power supply with its buck converter and large cap. The little switching power supply looks a little lonely back there all by itself...
20210206_170834 (resized).jpg

Today I worked on the backbox acrylic face cover. I mounted it on the head and marked the locations of the speakers as well as the edges of the TV image area. I then cut out the speaker openings and drilled the holes for the mounting screws and test fitted the speaker grills.

Next I masked off the area where the TV will project on the back side of the acrylic. You can really tell that I'm from Wisconsin since where else can you buy your sweetheart a time slot for when the truck will fall through the ice just in time for valentines day (see lower left article).

20210207_141002 (resized).jpg

Folks who have built arcade machines might recognize this trick. I mixed up some Createx black and airbrushed it onto the back side of the acrylic. You can use any type of paint finish from flat to gloss just as long as it covers the acrylic so nothing can show through from the back side. I used Createx because it drys quickly and doesn't stink which is important since it is single digits outside.

20210207_145258 (resized).jpg

When you are done and pull off the masking, you get a nice even and high gloss finish because the finish sheen is really the acrylic front.

20210207_151536 (resized).jpg
20210207_191136 (resized).jpg

It's been seven days since my last post. The Cobrapin board came back yesterday and I still cannot get my lights to work. Cobra18t was kind enough to check out the board and replace the two bad transistors. The board drove his Linux based machine just fine in terms of lights but will not drive my Windows based machine. Something happened when I blew those fuses but we just haven't figured it out yet. I tried driving the game with a laptop to see if there was some sort of USB communication problem on the PC in the cab but no lights would drive.

All of the obvious causes like blown fuse, no 5V to the strings, bad USB cables, use different USB ports. route the wire runs away from interference, drive a short chain of lights not in the cab have not turned up a cause. I'll keep trying to come up with a solution...

On a positive note the switch banks and coils all work fine so today I spent some time tuning the coils and setting up some game logic. There are still a few problems with my code but for the most part I can play Cuphead. Here's a video of gameplay as it stands right now.

All lighting is dead and there are two strands of addressable LEDs involved. One is for the GI lights and the other is for the inserts. Cobra18t thinks that the problem is being caused by the "blue pill" driver boards that sit on the Cobrapin board. He is sending me new ones that should get here in a few days so at that time I can see if that is the cause.

Cuphead has been stalled for the last few weeks because of problems with lighting. The game plays great but the lighting doesn't quite work but hopefully that will be fixed in a few days.

The big news from today is that the digital score reels are finally finished! Anthony from the mpf user group made these great digital score reels that look and work great but unfortunately when a new player or next ball starts, all of the reels spin from 00000 to the new score. Not horrible but not what I was looking for. Anthony made a fix for this back on January 24th but it never worked. I kept on asking on the forum for help but Anthony seems not to be around anymore and I think Jan was getting tired of my requests.

My only option was to try to figure this out myself so I spent a couple of days sifting through the mpf code. I figured out that although mpf had been updated with Anthony's code, mpf-mc (media center) had not so I looked at Anthony's push request and edited the code on my machine. The result was that it still didn't work. I looked through Anthony's statements and in every one he used "frame_persist" except for one so I changed that one to "frame_persist" and presto-magic, it worked, sort of...

The problem was that the player who was not up had his score slowly turn backwards to the next lower digit. For example in a one player game on ball 1 player 2's score was 00000 but on ball 2 it was 00000 rolled part way back to 99990 (last 0 is fixed). Eventually on ball 5 player 2 had 88880 as a score. I looked through the code some more and saw that there was a minus 1 applied to the frames because of a difference between the index that starts at 0 and frames which started at 1. I put in a +1 in a section of Anthony's code and this fixed the roll back issue. SUCCESS!!!!

Here's a short video showing the score reels and how the behave with the new code.

As best I can tell there are a few ways of doing the artwork, either overlay, silk screen or by hand. Since I am making the playfield by hand and the locations of holes for inserts and saucers might not be as precise as a CNC machine, my plan is to paint by hand. Large areas will get air brushed and any lettering or small keylines will be clear decal. Long keylines might be painted by hand and any characters will be white decal. I plan on coating everything with Spraymax as a clear coat so before I start I'll make some test decals and shoot those with Spraymax when I clear coat the apron, shooter guide and top arch. If the decals don't tolerate the Spraymax then I'll have to try one of the other methods.

Before I can do any of that I need to finalize the playfield and get these stupid lights working...

Cuphead Lives! Cobra18T sent me some new stm-32 blue pill boards that are programmed to work with the Cobrapin and now I have functioning lights, switches and coils.

Now that everything works I can see the many coding issues that need to be resolved so I am working on those. For example mpf would not set the score reels to 00000 upon a new game so I wrote some custom python code to fix that. I also discovered that if both dt banks were dropped at the same time, they would both try to reset and because of the large current draw of two big coils firing at the same time, neither would reset. I wrote some code to check if they were successfully reset and if not the two coils were fired with a 100ms gap of time between them.

I also found that the perdition entry lights were not lighting at the start of a new ball so that was fixed and I changed from the base end of ball bonus mode to a custom one that doesn't have as long a delay. A 2x bonus on the last ball was added and I'm still trying to make the code for perdition fail work correctly. Once that is done, I think I'll have the vast majority of bugs fixed in the software.

There are a few stylistic changes that I think I need to make. For example when a game ends the attract mode removes the score reels and runs the "Press Start" animation. I think I'll leave the score reels up so that the players can see what score they got as well as making those scores available to passers by. I also will add in some additional attract graphics like a high scores display, thank-you display, "powered by Cobrapin" display and more. Once those are done I'll post a video of the game with all of those features.

In the meantime here is a video of Cuphead driving each insert light in sequence and then each GI light in my LED test mode.

Good one j-rail, sadly it seems that this project is cursed.

I ordered a 48v power supply because the current set up could not raise the CUP dts if the flippers were held. When the power supply was delivered I could hear UPS drop it onto the concrete porch through the walls and then ring the door bell. There was no packing cushion around the power supply in its cardboard box and when plugged in on the benchtop it would put out 48v with no load on it but something broke with the violent impact to the porch because it would not even put out enough power to charge the caps on Cobrapin before the safety shut down would kick in. The power supply was returned and another one ordered.

I've been working on lighting and game logic and I'm pleased to say that all of the logic works like I want it to and the attract mode lighting is now complete.

Well the new power supply just arrived without the drop from several feet onto the concrete porch. Unfortunately it behaves similarly to the one that I sent back and Cobra18T thinks that there is a safety feature of the power supply that is cutting the output current because the big power capacitors on the Cobrapin board look like a dead short to the power supply. It does however only need 5 power cycles compared to the more than a dozen by the first power supply to get the caps charged and power on so I do think there was a problem with the first power supply.

The gory details are that the Cobrapin has a 1.3 ohm thermistor on it that acts as a load to the power supply so that it doesn't think that the caps are a dead short while they charge up. At 48V DC with the 1.3 ohms from the thermistor, the power supply is putting out 48/1.3 or a little more than 30A for the 100 ms that it takes for the caps charge. The power supply is designed to cut voltage output if the current is greater than 1.5x its max rating (9.4 x 1.5 = 14A) so it is going into safety shut down. By design, once the thermistor heats up, its resistance drops down so that the system can see the full 48v. The solution is to replace the 1.3 ohm thermistor with a 5 ohm thermistor so that the initial current draw while the caps charge is 48/5 or 9.6A which the power supply should not mind. I've ordered that part so some time next week I can be back up and running with the 48v power supply. Please stop me if you've heard this before...

Cobra18T is changing the design of Cobrapin to incorporate a MOSFET to switch a 5 ohm thermistor out of circuit once the caps are charged to improve performance on 24v power supplies. It's my job as a beta tester (test pilot) is to find these problems (find interesting ways to crash the plane) so that those of you buying the kickstarter boards (with the improved features) don't run into these issues (have to hit the ejector seat). Kickstarters, you're welcome!

I've finally got the game up and running to the point where I can shoot videos of the table. As I mentioned earlier I was having power issues getting the CUP drop targets to reset while the flipper hold coils were activated. I bought a 48v Mean Well power supply but because the Cobrapin's capacitors looked like a dead short and the 1.3 ohm thermistor did not give enough resistance to the system the Mean Well power supply would shut down after power up.

One solution was to replace the 1.3 ohm thermistor with a 5 ohms thermistor (48v = I * 1.3 ohms where I = 37A [power supply shuts down]or 48v = I * 5 ohms where I = about 10A [power supply rides through to charge the caps]).

Another solution is to add additional windings from the transformer in series and phase to increase the AC output of the transformer. I used the 24v AC windings and put them in series with the heavy 6v AC lamp windings to get a total of 30v AC which should be somewhere between 30v DC with a heavy load and 46v DC with no load. This increased voltage is now enough to raise the DTs of the CUP bank while the flippers are engaged!

I had to go through MPF and reset all of the coil settings to decrease their power with the increased voltage to both protect the pin's mechs as well as tune the game play to make it play more like an EM. There is still some work to be done here as you can see in the videos but I'm closing in on what I'm looking for.

Here is a walk through video that details some design features of Cuphead as well as the game's rule set. You'll have to use your imagination for the artwork but it will look a lot like this.

Cuphead Playfield Layout font test (resized).jpg

Here is a video of some game play.

This playfield is Onevox's work except for the dice lane.

I took a break from working on Cuphead because I really didn't like how the dash saucers were kicking and also the Perdition entry was too hard to get into and I needed some time away from the project. Instead collaborated with Bord and worked on some visual pinball tables. I had forgotten how much fun that was and how nice it was to work on vpx where I actually know what I'm doing!

This past week I've gotten back to Cuphead and made several changes that I think are improving game play. These include moving the mini-posts from just above the in/out lane wire to above the outlane wall, re-aiming the Dash kickers and reconfiguring the Perdition Lane entrance. The miniposts protect against "death lane" losses and also open up the in/out lane opening. I reworked the left Dash Kicker assemblies so that it is aimed more towards the top of the playfield. This now allows the ball to get up to the top kicker saucer on occasion.

20210420_113738 (resized).jpg

The Perdition Lane was much more involved. I had to move some lights, move the blocking post mechanism, move the location of the lower peg (to enlarge the entrance) and print new walls that "armor" the inner aspects of the opening to reduce the elasticity of it to direct more balls into Perdition. All of this required plugging holes, drilling new holes, moving mechs and lights but in the end I think the effort was worth while.

20210420_113742 (resized).jpg

Here is a video of game play to show some of the new table behavior. Note that I don't have a wall up by the top right standup because I'm still trying to figure out what to do with that portion of the game. Balls can get stuck there as you can see in the video.

If anyone can tell me why the ball occasionally leaves the playfield from the left sling that would be appreciated. I think it happens once in gamplay and I don't know if the ball is getting pinched by the rubber or what it is but it makes for glass hits that I really don't like.