Looks great. This will be a must-have mod.

The sooner the better so we can all test them for JJP for the next 6 months in the real world and Eric can get his open/closing chest on the LE re-run once it's proven durable and JJP makes a deal with Pinballj for solving their engineering puzzle.

I think some people don't have access to 3D printers or have soldering skills, etc. When I did the spike power supply quiet fan mod thread (here: https://pinside.com/pinball/forum/topic/replacing-stern-spike-noisy-ps-fans) , I never intended to be selling them, but the overwhelming majority of people in that thread preferred a plug and play kit even though I posted part numbers and step by step instructions to do it yourself. I've since sold hundreds of the plug n play fan kit. I doubt even 10% of that have done it on their own from scratch following my initial instructions. So I totally understand what's going on. Best case if he doesn't want to go pure retail is to do something like the spike fan thread. Provide instructions/part numbers/files for those that want to do DIY and team up with mod couple or whoever to sell premade kits to those that don't want DIY and just want something they can install and run.

That coil is the one that releases the balls. It looks like there's just a linkage that goes down under the playfield, pulling the lid open when the forks go up and releasing the lid to close when the forks go down.

I have no idea, haven't taken it apart.

There was a reliability issue in the field. What part was unreliable, I don't know. BUT it seems like the servo method would be less jarring for the whole assembly. Given how thin those mounts are for the chest, I can see them breaking if it's being banged open and closed with the force of the forks.

It's not so much violence on the bracket as violence transferred to the chest lid and mounts. Given that they're already shown to be breaking just from the ball whipping into the chest, the amount of violence of this action is very likely to break stuff and is probably why it was abandoned. The servo is likely the best long-term solution given that it's much gentler on the chest and mounts from the video.

Honestly that doesn't look THAT violent, and the simplicity is something to be appreciated. I'd be willing to test this out on location to see how it holds up. Worst case, if it doesn't, there's a servo version...

Except the brackets holding the chest on. There are definite downsides to both. But the fact that we have two potential ways to solve the issue is great!

More solutions to choose from, the better. I think the electronic will be best long-term, but my mind's open.

Fantastic work. Can't wait to get this installed!

Thanks so much! Really looking forward to this!

Do you recommend PLA, ABS, or something like Nylon for the bracket material? Is there a lot of stress on it?

I don't know where you live, but lots of libraries have 3D printers now and you can print one there if your library has them. Also check your town for services that offer 3D prints.

The progressive ones have realized that recreational book reading isn't coming back on a large scale, and google killed the need for research repositories in each city, so they've branched out. Even our backwoods town library has 3D printers (they're not GREAT but they're good enough).

The heart is the only thing that makes sense as it's part of the story and the chest "beats." Honestly, though if it opens and closes and the LED is brighter (green then red), that's probably enough.

A simple black cover back there might be best. If everything is embellished, nothing stands out.

I should have the parts to start on this mod later today. I'll let y'all know how it goes. FYI, that bracket print with 60% fill medium speed and .2 layer height with a .4 nozzle is about 7 hours. The required supports really slow it down, but what are you gonna do?

First tip:
Lowes has the hinges noted already in black/dark brown:
https://www.lowes.com/pd/Gatehouse-2-Pack-1-1-2-in-X-1-1-4-in-Oil-Rubbed-Bronze-Surface-Broad-Cabinet-Hinge/1000141901

Second tip:
The recommended Arduino boards on Amazon are not actual Nanos, but crappy Chinese knockoffs (which is why you're getting 3 for $12). Two things you need to know - first, these require a different Serial to USB driver than the real Arduinos. They want you to get it from a russian site, but I wasn't cool with that so I found it elsewhere. Second, you have to select the processor with "(Old Bootloader)" next to it. If you just select the processor without the OLD version, it just won't work. You want to upload simply, NOT upload with programmer.

I'm making a bracket/container for the fake arduino board so you can snap it to the bracket in a vented enclosure instead of having to zip-tie it to the block as it is now. If people hang tight for a day or two I should have it tested and I'll post the change.

Ha! Even roughing it out, I spent way more time on this than I should have. This is a rough print. If you go to .15 or .10 it can look much better.

Here's the Nano enclosure. There are two channels (left and right) inside the enclosure, under the board to run the wiring down to the end of the box and out so it has a clean look. This just has the two switch leads attached for the picture.

nano-box-bottom (resized).jpg

I have no idea what heat these generate, so I vented the top, just in case.
nano-box-top (resized).jpg

Then the Nano enclosure snaps into the modified bigger bracket. Since it takes so long to print a full bracket, I only printed the base of that area to test the fit. It just clips in and out. Pretty nice.
nano-braket-clipped-in (resized).jpg

Here's the top part with cable management for the leads to the switch and the leads coming up from the servo on the side. I'm pretty sure I'll be cutting the connector off the servo and soldering the control line straight to the Nano and the power leads to the 5V adapter.

nano-bracket-top-w-cable-management (resized).jpg

I only have the fake Arduino Nano board from Amazon that the OP suggested for this project on the BoM. No idea how it stacks up to other real or fake Nanos. I'm assuming they tried to fake convincingly, including dimensions, but I have no idea.

Which boards did you get? By pre-soldered do you mean the pin headers are on the board? That might be a clearance problem. Since there's so few wires to connect, I didn't even put the header pins on the boards I got.

With just a handful of wires to connect, it's easier just to push the wires though the holes and solder them directly to the board. I'm hoping to have everything wired up with the revised bracket and nano enclosure later tonight to make sure it all comes together cleanly. Then I can worry about installing it to get the chest working.

ATX adapter on the computer power supply to tap the 5V without affecting anything else.

The existing bracket just has a block and a space to zip-tie the Nano to it, which looks...messy. This cleans everything up and puts it in a clip-on block and has hooks to manage the wires so the install will be much cleaner.

If the block size was increased to allow holes. As it is now, the design is very tight. No wasted space. It's only designed to clip in.

I'll post the revised parts so you can print it if you want. I want mine to be a very clean install, but I'm pickier than most.

Ha! I ALMOST made the corners rounded, but talked myself out of it to concentrate on functionality/accessibility/wire management. Sanding for a utilitarian product is a bridge too far, so I'm not doing that. These improvements are essentially for my own OCD mollification.

The main advantage is isolating all the mods from the JJP system. No additional power draw within their design.

The nano box is done, I'm just fine-tuning the main bracket. Wire-management for the servo is done and great, but the channel for the switch wires is too small. Once I adjust that the main bracket will be ready to go. I also tightened up the servo cutout on the main bracket since the original was too wide. I have a super-OCD nano box with a junction box, but no idea how long that will take (and let's be honest, that's overkill, but I am ill), so I'm just going to give you the revised bracket and nano box STLs as soon as I get the switch wiring channel fixed.

Here's where it's at before adjusting the switch wire channel. Note that this is still a very low-res print so the plastic's a little rough:
bracket_v2 (resized).jpg

The servo leads would be cut and routed into the nano box, or you'd have a connector coming out of the nano box and connect the servo to that. The servo lead can probably be shortened to a custom length by opening the 4 screws on the servo and shortening the wire from that end, but I haven't tried that (yet). I will have to for my super-OCD version, though.

For the record, on the servo the pinouts are:

Brown: Ground
Orange/Reddish: +
Yellowish: Signal/Control

And here's how the board wires up (power for the board comes through the mini-USB):

arduino pinout (resized).jpg

Fixed. Thanks.

Do you have a picture of your base bracket with the switch attached? I think I have this right, but want to make sure.

Thanks for the pic!

I'm screwing the switch from the top down with a thicker platform so it lines up with the main plate it's attached to on the bottom for maximum strength. But the direction of the screws shouldn't matter.

The power is coming from the mini-USB. You can get an adapter that splits 5v off the ATX power supply to a USB connector so you're not tapped into anything in the system.

I've finished the redesigned big bracket that can take the nano-box snap-in and has the some wire management. I'm printing both now to make sure there are no unforseen issues, then I'll send you the big bracket and nano box files tomorrow that you can use for anyone that wants them.

I'm going to continue on to finish off a Super-OCD version for me that has a junction box with plugs and some other niceties for a super-clean install.

Yeah, I was initially confused about the 5v on the board. Assumed there would be a solder pad, but they had to make design consessions due to the tiny size compared to an Uno board which has an actual DC power plug and USB by combining the power and data into one Mini-USB.

Micro USB is crappier real-world. It SUPPOSEDLY is more reliable on paper, but connectivity issues happen way faster than mini USB. The main reason everyone went to micro is it's thinner and, well, mobile needs that. Plus, the nano this board is copying came out more than 10 years ago. The shift to micro USB hadn't really happened in earnest in mobile by then - the iPhone had just come out!

Wait, so are you saying connecting the computer USB to the Nano mini-usb for 5v power wasn't reliable so you just sent the Nano 5v directly from the ATX power supply to the mini-USB, OR are you saying you didn't use the mini-USB to supply the 5v at all?

IF you didn't use the mini-USB, where did you put the 5v power input in on the Nano?

So basically put a Y on the 5v output from the ATX power supply and send one lead to the servo and the other to the 5v header on the nano, then?

Okay, here's a link to the revised jjPotC bracket and nano box that snaps into the revised bracket to clean up the look of the install:

http://www.irelandfamily.com/pinside/Base_Arduino-box-and-bracket.zip

I'll leave it up for the forseeable future and you can use it as you like.

Note that the nano box works better getting into and out of the clips if you slide it in from the outside and then shift it over and press down. It clips in VERY snug by design. It's not going anywhere.

Also, the nano itself clips into the nano box very snug as well. If you're having problems getting it to snap in, just use needlenose pliers to compress the angled plastic pieces that go around the mini USB connector a little to give it more room to move freely.

I think it's just shy of 7 hours at 0.15 layer height (which is why on the SUPER OCD version I'm making for me, I'm just printing the top part where most of the changes are for testing - takes it down to less than an hour and half). The need for lots of supports on the big bracket really slows printing down. FYI the wire management Ls on the big bracket don't need supports, but the clips on each side of where the nano box goes do and the giant space between the bottom and top do.

The nano box doesn't need any supports of you make it on its back.

So the chest lids are different sizes inside?

Sounds like he's doing a version with the light and a version without and you choose.

So off in Super-OCD land, I have the connectors and wiring worked out for the one I'm making for the one here. As was previously discussed, I won't be using the USB for power to avoid stability issues. The connector opposite the servo and switch on the top of the nano box will be connected to the ATX power supply's 5v.

super-ocd-chest-bracket (resized).jpg

Thanks for fixing it so quickly. I wasn't sure why it was doing it, but it seemed like a long-term reliability issue. Glad the fix was easy!

I'll be putting it on a route machine. It's ALMOST done and then I can install it. Went a little crazy with the design. (Before you say anything, some of the graphic keys are wrong ATM. The wiring works, though. )

I think it'll hold up fine, but I'll be the first to tell you if it doesn't.
Nano-box (resized).jpg

I assume you're using PLA....

Don't need supports for them on most printers. Should print without them. Only the giant open area on the right of the upright needs supports. On my OCD version after these, I actually added a little wedge to the back side of the wire brackets to give it a little more heft.

DEFINITELY doesn't need supports for the wire brackets on this. I have three of these for printing, so I know they work without supports for the wire brackets.

I am using support on the really big gap, just not for those little wire management pieces. They come out fine. If you don't want to waste time and just check, cut off the bottom of the model and just print the top part where all the action is and see how it prints. You can print that top part at 40% fill or so in less than 2 hours. I'm using Prusa and Hatchbox filament pretty interchangeably so I don't think it matters.

Extension? I shortened mine...

Yeah, the wire on the servo is way too long as it comes. I trimmed it down and resoldered it to clean up the look.

You're limited on the close of the solenoid by the speed of the servo (which isn't super-fast at 5v), so it will never "slam" shut, and basically just closes quickly at full speed. Also, the way the code is now, if you're wanting to beat the "heart" as it's still opening and closing, it needs to be rewritten to use the millisecond timer so it can be more like a state machine rather than single task delay loops that halt the code while one thing is happening then letting it move to the next task.

Well, the original design was strictly mechanical and tied to the fork movement mech, so maybe the wear and tear was too great when you didn't have a servo controlling the speed of the lid closing (meaning, it slammed shut over and over until it broke something). We'll never really know.

The servo solution does add another potential point of failure, but it at least controls the speed of the lid somewhat, so I think it will be very reliable long-term. I've got one more thing to get working on my OCD version (green/blue ocean light matrix also attached to the Arduino for dim moving lagoon lighting under the chest when it's open) and then I can finally install this.

I think I used a 8-32 Type F thread-cutting screw (there aren't any threads on the fork assembly holes).

Two tips from my OCD version experiments:

There's an atx connector in the backbox (outside the computer case) where only the 12v (yellow lead) is being used, so if you make a splitter, it's a perfect place to tap the unused 5v (red) on that connector right from the power supply connector with no draw or effect on anything else.

Putting a led matrix below the chest to light the playfield under the chest when the chest shot is open can't be done. The playfield is too reflective and you see all the leds. I'll be moving it behind the chest and reflecting it into that area instead, but it cannot be directly lit.

Nope. Doesn't work. The playfield is just way too shiny.

No, it doesn't have anything to do with what Harry's shipping.

I went off into the imagination desert and made a different "OCD" version with a two-chamber heart light in the chest lid and a 4x4 matrix of LEDs that I programmed to transition in and out of colors that shift from greenish to dark blue to simulate the sea. My *idea* was to put that below the chest (it's very thin), but when I did it, you could see the reflection of the matrix lights in the gloss of the playfield, so I had to re-think it. I'm now making a bracket that will shift the 4x4 matrix behind the chest, aimed down to light up the channel under the chest without the reflection of the LEDs.

This is an earlier version of the Nano box (that's an upgraded version from what I posted for people to print of the basic Nano box enclosure). It has the two LEDs, 5v input from the ATX power supply in the backbox, and the AUX is the connector I'm using to drive the "ocean" matrix.

Nano-box (resized).jpg

Two questions on this video.

Why did you use the optional 4-way connector on the servo instead of the 2-way arm it also comes with?

Also, the Steel kwik-link rods come with an end that has a springloaded clip that can snap into the holes on the servo arms eliminating the need for any other parts to connect the rod to the servo arm. I used the second hole from the end and drilled it out a smidge more to make the rod move freely once snapped into the servo arm holes, but this eliminates one whole other part.

Yeah, that's the way I went, too. It made more sense to me, plus with all the wire bunches down by the servo, the less that's down there, the better.

The Servo on the BOM came with 3 or 4 arms, including the 2 way one.

https://www.amazon.com/gp/product/B01KZKOIZC

And then I just clipped on the kwik-link spring-loaded end onto the second from the last hole on the 2-way arm after drilling it out a very small smidge so the hole was a little bigger and the link moved easier:
https://www.amazon.com/gp/product/B0006O4G8M

The clevis clip is the part that goes on the servo end (it comes with the Kwik link rod). snapping into one of the holes on the arm.

You can just use the Du-Bro 605 E-Z Connector with the set screw only and not use the other parts of it for the top.

Went with two individual LEDs for the heart instead of a strip and made two halves of a "beat" so there's a side for each ventricle, left ventricle "pumps", right ventricle lights up, and they both fade.

Cheat Heartbeat-min.gif

Still working on the lagoon lighting part. Undesirable reflection everywhere I've tried to place the 4x4 matrix. Going to try a couple individial LEDs instead next.

It's a pain to get the two nuts holding the hinge removed due to small working space, but it's not hard. You need a 5/16", preferably low-profile socket. As long at the power's off, you don't have to worry about the coil lugs. (I didn't watch the video, so didn't know the coil was being removed - I did it without taking the coil off)

The MAIN thing you need to be very careful of is the middle opto receiver (transmitter? I'm not sure what's on the back side of the chest) just below the hinge. If you knock that out (easy to do) it's a PAIN to get it snapped back in, and you can break the snap-in connector legs.

Yeah, this was the WORST part of installing the lid. I was able to get the middle one back on without breaking the legs, but it's a huge pain in the butt, and if you break them, replacing them is the only way to do it right. Fortunately they're pretty cheap. Mouser has them.

It really is an easy process as long as you know what to avoid. I was doing it blind, and so was Harry, so we've mapped out the hard part for you! Even with the irritating nuts you need to remove, the whole process should take about 15 minutes proceeeding slowly so as not to break the IR brackets.

It's not a bracket, it's an all-in one IR receiver or transmitter casing. I need to look at the diagram to see if it's a transmitter or a receiver that's on that side.

The mouser ones need the ends of the wire leads crimped into a connector, but that's all.

Once the servo finishes moving each time, it needs to be disconnected with a command.

I'm using servo detach because if you don't, the servo is alway applying micro-power and it makes annoying noises and unnecessary (in the big picture) micro movements. Pretty sure having the servo keep grinding the position non-stop is worse for the servo than turning it on and off as-needed. It takes quite a bit of force to move the servo, so slippage once software disconnected is not a problem. I'm running it on a route machine (with different overall code I did, but servo detach at the end of each movement) and have had no problems.

Replacement Opto parts were actually Digi-key, not Mouser (I realized when I pulled the bags of them I have). The two I have are:

Part Number 365-1626-ND
https://www.digikey.com/products/en?keywords=365-1626-ND%20%20

...and 365-1627-ND
https://www.digikey.com/products/en?keywords=365-1627-ND%20%20

Pretty spendy at about $3 each.

It's pretty quiet unless it's against something that amplifies the vibration. I could HEAR it on my desk, then holding it, I could feel it. But I can't imagine it's good for the life of the servo.

I don't know of any way, but I'm by no means a Nano master. Pretty sure you just need to plug it into a computer and update that way through the USB.

I actually got one of those $129 Dell Inspiron 11" mini Win10 laptops specifically for this purpose, since when I'm testing it post-install, lugging a full laptop every time I wanted to field test an update I completed bench testing at home was a pain.

At least it's easy to correct, just attach the lid rod to the arm on the other side of the servo arm.

We have it running on route and I used servo disconnect from the get-go. It does not creep in any visible way if disconnected after moving. Don't worry about that.

Oh, I didn't look at the way you did it. I decided that the incremental servo commands in the initial code taken from an ardiono sample were a pretty useless difference over just telling the servo to go to the desired position in one shot and then disconnecting. Simpler, easier, less stress on the unit, not much practical difference visually, so that's the way I went.

Check the servo switch with the forks down and see if the arm is bent out enough on the switch to be triggered. I bet the switch isn't triggering.

Nope. Aside from some small wear and tear on the screw holes on the chest lid it's completely reversible.

But if you force the arm down a little more, does the switch base click? If it still has room to click, it's not registering.

Did you get the lighting? If so, can you see it on inside the chest?

If the lights are always "beating" then that likely means the switch is always on or wired wrong.

So with the playfield raised and power on, if the forks are UP and the lights in the chest are on, will manually closing the switch with your hand so it clicks turn them off? Or even when the switch clicks closed manually with your hands the lights stay on?

Don't do that right off. If you can avoid cutting, that's always the best way.

I'd first try using flat headed screws instead of pan head. Then, also try putting a spacer on the lid-side of the chest bracket between the bracket and the lid to raise the chest lid enough to let it close.

Put it on the other side of the star connector on the servo. You'll get opposite action.

Has to do with what arm you have it on. You still have it on a "wrong" arm. Having four to choose from (instead of the 2-way one) makes this error easy to make.

Maybe a bad servo? Does it do this backtracking with no rod attached?

After getting the mechanical version fully working, the concensus seemed to be that the action was too violent and is probably the reason trunk lids were breaking during test and the feature was pulled.

The servo motion at 5v is pretty slow, so full speed servo with one command to go to the open/close position is a really nice speed with no hassle and less wear and tear on the motor since each is one move rather than a bunch of small ones every time. Try it, you'll see what I mean.

Are you giving it more than 5v? The input voltage determines the speed to a large extent. I'm giving it a straight 5v and it's a nice speed open and close. I'll take some video of it opening and closing next time I'm at the location.

Did you measure the power at the connector with a multimeter?

Here's the lid with straight 5v and no stepping commands to get it to/from open close, but using a direct command to the open/close position instead. I don't think it's too fast.
chest-with-beat.gif

He could easily do a combination, but I don't know if he wants the upgrade headache...

That's what I was getting at. The timing is almost the same (yours is just a hair slower), but using direct calls rather than stepping the motor repeatedly in a loop each time it opens and closes is less wear and tear on the servo over time.

Is the head bigger than the holes? It cuts the thread in the metal brack that piece goes over.

No, it threads the metal bracket holes on the jjPotC assembly and holds the plastic piece so it doesn't move.

The optos are black and very hard to see, even in real-life given the cramped conditions. The problem is not that you break the opto itself, but that you pop the opto off (they're held on by little clips) by making contact with the opto body and in trying to snap the opto back on, you break the tiny, fragile clip, which is very easy to do if you don't know what you're doing. Once the clip is broken, the opto won't stay in place and you need a new opto.

BTW, I didn't remove the coil. The access is tight, but it is do-able without removing the coil and just putting the coil cover to the side.

Hot glue doesn't last. Temporary solution at best for an area like that.

Some low island greenery which is the background for the chest for much of the movie where it's out might be good, but it'd have to be low so as not to interfere with the cannon shot...

The only other place where it spends a lot of time is on the sandy beach where it's dug up, but that's no good...

dead-mans-chest (resized).png

Check your switch bend. It's probably not bent out quite far enough to detect fork down.

Alternately, if that doesn't work, with power on, raise the playfield, raise the forks and manually activate/deactivate the switch. If nothing happens you have a bad switch or bad wiring to the switch.

Glad it was an easy adjustment instead of a bad switch!

You're probably just sensitive to the motion being done incrementally (or the servo's bad). Aside from doing it as a straight one command motion there's no way to get rid of the steppiness.

Do you have a short video of the motion?

Angle both surfaces slightly to the outside on each side or it's a natural ball trap for airballs or missed canon shots.

Most likely the switch arm not bent out enough for the fork assembly to trigger it. Put the forks up and manually click the switch and see if the servo works.

Might also be the power connection to the fake nano board. Does the light come on on the circuit board?

Is the light on the circuit board on (should still be able to see it through the shrink)? Because if it isn't, you have no power at the board.

The servo maxes out at 6.5v, if I remember correctly. Giving it max 6.5v power will make it move quicker, but I don't think Arduino Nano (and knockoffs) can regulate down from 12v to a specific voltage.

There's a completely unused ATX-style 5v connector in the head that's perfect for this.

So you're plugging your alternate power into the wall outside the machine?

Also the 5v free plug is not shared with anything else, coming straight from the power supply, so it's regulated. Double plus.