(Topic ID: 253732)

MPU 100 Don't boot on first try

By oldschoolbob

4 years ago


Topic Heartbeat

Topic Stats

You

Linked Games

No games have been linked to this topic.

    Topic Gallery

    View topic image gallery

    Oscilloscope_FrontPanelSwitches_3a.jpg
    IMG_3978 (resized).JPG
    IMG_3969a (resized).jpg
    S20191111_0011 (resized).jpg
    S20191111_0010 (resized).jpg
    S20191111_0005 (resized).jpg
    pasted_image (resized).png
    S20191110_0011 (resized).jpg
    S20191110_0002 (resized).jpg
    S20191110_0010 (resized).jpg
    S20191110_0009 (resized).jpg
    S20191110_0008 (resized).jpg
    S20191109_0004 (resized).jpg
    S20191107_0003 (resized).jpg
    S20191107_0002 (resized).jpg
    IMG_4003 (resized).JPG
    There are 252 posts in this topic. You are on page 1 of 6.
    #1 4 years ago

    I just finished rebuilding a Bally MPU. I thought I'd give it another test before I put it the parts drawer. I connected my power supply and got a solid on LED. I turned it off then on again and it booted fine. Tried it again a few times and it booted every time. Then I tried a Stern MPU and it booted first time. I went back to the Bally and it got solid on , on the first try then booted. It does this every time I let it set for a few minutes. I got 5.2 V and 13 V with solid on and when booted. I then tried my old power supply and got the same results. The only game I have available at this time is a working Meteor (MPU 200). So I connected only the J4 to test in the game. Same results - solid on the fist try. Turn it off the back on and it boots.

    I've tried two power supplies and in a game. It just won't boot the first time.

    This is going to be a tough one. Any suggestions where to look?

    Thanks

    Bob

    IMG_3874 (resized).JPGIMG_3874 (resized).JPG
    #2 4 years ago

    Check the voltage across the zener diode at VR1. If it's lower than 7.8 volts, replace it.

    #3 4 years ago

    Thanks Quench,

    I just checked VR1 - upper side reads same as TP2 , 12.9 V. Lower side reads 5.14 V.

    Just to check I tested the Stern board. It reads 4.61 on the lower side. But it boots fine every time.

    Tomorrow I'll check to see if I have a new VR1.

    Thanks

    Bob

    #4 4 years ago

    Before coming upstairs I checked my parts drawer. I have "ONE" VR1 - 1N4738A Diode, Zener, 8.2V, 1W.

    Didn't you say something about having problems with VR1's a while back? Didn't you say you usually replace them?

    Thanks

    Bob

    #5 4 years ago
    Quoted from oldschoolbob:

    Didn't you say something about having problems with VR1's a while back? Didn't you say you usually replace them?

    Yes, if I measure 7.8V or less across that zener diode I replace it. The voltage across that zener is kind of used to determine if power is good at startup.
    Measure the voltage directly across the zener diode (one meter lead on each zener leg).

    #6 4 years ago
    Quoted from Quench:

    Measure the voltage directly across the zener diode (one meter lead on each zener leg).

    With power on or off?

    #7 4 years ago
    Quoted from oldschoolbob:

    With power on or off?

    On.
    Your measurements said there's 7.76V across it (12.9 - 5.14).
    I was just saying to measure directly across the zener to see how much voltage is across it.

    #8 4 years ago

    Sorry - I guess that was a dumb question (must be late).

    Putting one lead on one side and another on the other side of the zener will tell me how much voltage is going through the zener. I didn't know that. I love learning stuff.

    Now I know what a zener is for.

    I'll get on it tomorrow.

    Thanks

    Bob

    #9 4 years ago
    Quoted from Quench:

    Your measurements said there's 7.76V across it (12.9 - 5.14).

    If you subtract the difference from one side to the other side and get 7.76 V going through it - wouldn't a "1N4738A Zener, 8.2V" depend on the input voltage? Let's say I have 12 volts on the input and get 8.2 V across - A 5 V or 25 V input wouldn't give me the same 8.2 V would it? Why is it called a 8.2 V zener?

    Is there a formula to figure what voltage you get with a certain voltage going in? If I had a 14 V source and wanted a 5 volt output, What size zener would I need? If I had a certain size input and a certain size zener could I calculate what the output voltage would be?

    This stuff is fascinating.

    Bob

    #10 4 years ago
    Quoted from oldschoolbob:

    Is there a formula to figure what voltage you get with a certain voltage going in? If I had a 14 V source and wanted a 5 volt output, What size zener would I need? If I had a certain size input and a certain size zener could I calculate what the output voltage would be?

    If you wanted 5v out of a zener, you'd use a 5v zener.

    The purpose of the 8.2 volt zener is to hold the reset low to the cpu chip until the 12 volt line is higher than the voltage rating of the zener. Zener diodes in this case (and on the solenoid driver board in the high voltage section) act as crude voltage references. You aren't getting 8.2v going to the reset pin on the cpu, because of transistors q1 and q5.

    The theory of operation says this is so that the +5 line will be stable with voltages higher than the zener voltage. If you look at the spec for the 323 regulator it will tell you the minimum voltage needed for a stable +5 output and the zener is selected for higher than that.

    #11 4 years ago

    There's plenty of online info about zener diodes. Essentially when they're connected in reverse, at a certain voltage (spec of the specific zener diode) they conduct. This makes them suitable in simple voltage regulator circuits - remember that 13V zener in your bench power-supply you were building at the start of the year?

    The spec of the VR1 zener on the Bally/Stern MPU board is 8.2 volts. But I've seen plenty of them running around 7.8 volts and at that voltage, the boards don't like booting from some bench power-supplies.
    To be honest if you leave that zener on the board it will probably work fine in game due to the rate that the supply rails reach steady levels - different rate to bench power-supplies.

    #12 4 years ago

    First thing I checked was across the zener = 7.86 V. I also checked above and below the zener. 13.03 and 5.16.

    Just for grins I checked the Stern board. 8.25 across the zener. 12.87 above and 4.60 below.

    Next I replaced the zener. I still get solid on , on the first try but now it takes several attempts to get it to boot. But at least the zener now reads 8.02 across the legs.

    It looks like Q1 has been replaced (bad solder job). But why would it work after several tries?

    I knew this would not be easy.

    Thanks

    Bob

    IMG_3924 (resized).JPGIMG_3924 (resized).JPGIMG_3925 (resized).JPGIMG_3925 (resized).JPGIMG_3928 (resized).JPGIMG_3928 (resized).JPG
    #13 4 years ago

    What happens if you pull out the 5101 RAM? Do you get the flicker every power-on?

    When the board doesn't boot, what voltage do you read at the reset pin of the CPU (Pin 40 of U9)? If you then very briefly short pin 40 to pin 39 on the CPU, does the board boot?

    #14 4 years ago
    Quoted from Quench:

    What happens if you pull out the 5101 RAM? Do you get the flicker every power-on?

    Same thing - solid on the first try then it boots on the second try. The socket at U8 has been changed. It's a AUGAT socket - looks like machined pin. I looked over the RAM and it looks pretty good - no corrosion.

    When I get solid on I read 5.48 at U9 pin 40. When I short pin 40 - 39 it goes to boot. After booting I get 5.49 at Pin 40.

    Thanks

    Bob

    #15 4 years ago

    Is that replacement Q1 the right transistor?

    Looks like you'll have to test the resistors/transistors/diodes in that "valid power detection" circuit.
    The schematics have a couple of voltages in that circuit listed.

    Also, temporarily disconnect capacitor C13 and see what happens.

    #16 4 years ago

    Q1 looks right "F2N 4401 EBC"

    Next I lifted one leg of C13. And it booted first try - I tried again and solid on.

    I checked the voltages from the schematic. See photo - my results are in red. I tested while it was solid on.

    I think C13 has also been replaced. Bad solder joints. The reason for the bad solder joints is because of missing solder pads on the top side. Looks like C80 and CR7 have the same problem. This board is not as good as I thought. And it looks like I'm not the only one to lose pads on this board. Do you think we can save it?

    Thanks

    Bob

    Capture a (resized).jpgCapture a (resized).jpg
    #17 4 years ago

    I just noticed I missed a voltage - Q1 collector to R138 and R140. Schematic shows 0.3 VDC. I got 9.6 mV. I got the same results with solid on and booted.

    Then I checked the Stern and got 37.1 mV.

    Thanks

    Bob

    #18 4 years ago
    Quoted from oldschoolbob:

    I think C13 has also been replaced.

    Yes, it's a 50nf capacitor. The original value is 10nf.

    Quoted from oldschoolbob:

    Next I lifted one leg of C13. And it booted first try - I tried again and solid on.

    Did it seem to make much difference?

    Quoted from oldschoolbob:

    I just noticed I missed a voltage - Q1 collector to R138 and R140. Schematic shows 0.3 VDC. I got 9.6 mV. I got the same results with solid on and booted.

    Then I checked the Stern and got 37.1 mV.

    Have you got a 2N3904 to replace Q1 with?
    Does R138 and R140 measure good?
    How about the diodes at CR5 and CR7?
    Is the big R11 resistor good?

    #19 4 years ago
    Quoted from Quench:

    Yes, it's a 50nf capacitor. The original value is 10nf.

    Should I change it if I have a 10nF?

    Quoted from Quench:

    Did it seem to make much difference?

    Didn't seem to make a difference - but it's still disconnected. Would that make a difference on reading 9.6 mV at Q1?

    Quoted from Quench:

    Have you got a 2N3904 to replace Q1 with?

    I'm pretty sure I have a replacement from GPE - either 2N3904 or 2N4401. I'll check tomorrow.

    Quoted from Quench:

    Does R138 and R140 measure good?
    How about the diodes at CR5 and CR7?
    Is the big R11 resistor good?

    Can these be checked in circuit?

    Shop is closed for tonight - wife went to bed.

    I'll get on this tomorrow.

    Thanks

    Bob

    #20 4 years ago
    Quoted from oldschoolbob:

    Can these be checked in circuit?

    Check them in circuit. If they measure out of spec then investigate further.

    Change that C13 capacitor to original spec. Its larger value isn't helping because it's delaying the 5V reference state used for the reset.

    If you feel like delving in, pull out the oscilloscope and enable dual channel mode.
    Connect one of the scope input leads to the banded side of zener diode VR1
    Connect the other scope input lead to the banded side of diode CR5
    Take a snapshot of the signals the moment you power up. Lets see the relationship between the rise time of the 12V line and whether the 5V reference state used for reset has stabilised.

    Compare the result to your working Stern board.

    #21 4 years ago

    Tomorrows "to do" list:

    1 - change C13 - I should have one.

    2 - replace Q1 - I should have one.

    3 - check R11, R138, R140, CR5 and CR7.

    4 - pull out the oscilloscope . That will take a while to set up - I've never used the second channel. My scope has a run/stop switch. If I hit that switch before I power up I think it should capture the moment it powers up and no more - is that what it's for?

    Quench, you know you're lucky I'm not your neighbor - I'd be there all day asking questions. I can just see me with a schematic running along side of you while you're cutting grass. you'd never get anything done.

    Thank you for teaching me so much.

    Bob

    #22 4 years ago
    oscilloscope a (resized).JPGoscilloscope a (resized).JPG
    #23 4 years ago

    I have no idea what I'm doing but I'm having fun learning. I didn't follow my to do list. I started with the oscilloscope first.

    Blue probe is connected to CR5 - Yellow is VR1. First scan is Stern. Second is Bally.

    I'm having problems with my scope. The vertical and horizontal controls don't seem to be doing anything. I'm sure I got something messed up in the menus somewhere. I'll mess with that later.

    Thanks

    Bob

    stern 1 (resized).jpgstern 1 (resized).jpgbally 1 (resized).jpgbally 1 (resized).jpg
    #24 4 years ago

    I just checked the resistors and diodes. Both Bally and Stern:

    R11 - Bally = 88.2 - Stern = 96.5
    R138 - B = 48.1k - S = 45.9k
    R140 - B = 19.71k - S = 20.45k
    CR7 - B = 0.561 V - S = 0.563 V
    CR5 - B = 0.593 V - S = 0.553 V

    Next I'll check my parts drawer for Q1 and C13.

    Thanks

    Bob

    #25 4 years ago

    I just discovered something real strange. I replaced C13 and was going to buzz my connections. The schematic shows one side to ground and the other to R12. I looked for R12 and it looks like a diode there. See first photo in post 12 above. I checked it with my ohm meter and I read like 5.2 m ohms and open in the other direction.

    Can this be right????

    Am I missing something?

    I checked the Stern board and it's a resistor there.

    WHAT???

    Bob

    #26 4 years ago

    Adjust the vertical position of the traces so that zero volts is on the first grid line at the bottom of the screen. Since you're measuring positive DC voltages the trace shouldn't go downwards so this gives you the full screen for better resolution.
    When you're measuring AC voltages, then have the zero volt traces running in the centre of the grid.
    See how you go about getting the vertical controls working again.

    Quoted from oldschoolbob:

    I looked for R12 and it looks like a diode there.

    Someone has changed R12 to a blocking diode for an external battery. Actually I do the same thing but use a 1N5817 diode which has very low voltage drop. Since there's no battery currently on your board, the other end of that diode is going to an open circuit.

    #27 4 years ago

    I just got back on line after studying the schematic. I noticed R12 is going to the batteries. Someone put in a diode to block the batteries from being charged. That's pretty clever.

    I change C13 and buzzed the connections. All good.

    I then connected the power supply and tried to boot. It took 5 tries before it booted.

    I haven't changed Q1 yet.

    First I need to figure out how to get the scope to adjust. Too many buttons and menus. I guess I'll get out the manual.

    I'll be back when I get this figured out.

    Thanks

    Bob

    #28 4 years ago

    I think my oscilloscope is real sick. I can't get any knobs to function. No vertical (channel 1 or 2). No horizontal. No trigger level. I just can't figure it out.

    Bob

    #29 4 years ago

    Do you have a "default setting" button on your oscilloscope? I just got a Siglent oscilloscope and I've already just about worn out the "default setup" button cause I keep messing things up. Saves me every time! I haven't used an oscilloscope for my pinball machines yet but it is absolutely essential for tuning up a Vectrex.

    #30 4 years ago

    It does have a default setting. But the strange thing is if you change your settings after a few seconds that becomes the new default setting. I wish it had an "original" default setting but I can't find it. The "auto set" works fine but I just can't change any settings. I don't know if it's some setting I screwed up or if it's just broke. I've read most of the manual and can't find anything to reset the settings.

    Thanks

    Bob

    #31 4 years ago

    I just replaced Q1 and checked its continuity. Still don't boot first time.

    Could U9 be causing this problem? Maybe I'll switch U9's tomorrow. I'm running out of ideas.

    Thanks

    Bob

    #32 4 years ago
    Quoted from oldschoolbob:

    Could U9 be causing this problem? Maybe I'll switch U9's tomorrow.

    Yes, give it a quick try tomorrow.

    #33 4 years ago
    Quoted from oldschoolbob:

    But the strange thing is if you change your settings after a few seconds that becomes the new default setting.

    It doesn't have a store settings button that's stuck?
    Is there anything in the menus about restoring factory defaults?

    #34 4 years ago
    Quoted from Quench:

    Is there anything in the menus about restoring factory defaults?

    Not that I have found yet. Don't know where to look - it has a zillion menus. And I can't find anything in the manual about factory defaults. I think I can download the latest software and put it in the scope. I don't know if that would help?

    Thanks

    Bob

    #35 4 years ago
    Quoted from oldschoolbob:

    I think I can download the latest software and put it in the scope.

    If you press the Utility button and go to the System Info menu, what firmware version do you have loaded?

    The latest Hantek firmware appears to be dated 25-Jan-2019

    #36 4 years ago

    I don't think I ever changed the firmware. I got the scope in 2018 so 25-Jan-2019 would be a later version.

    I also think found how to get back to the default settings.

    I'll give it a try tomorrow.

    Thanks

    Bob

    default 1 (resized).JPGdefault 1 (resized).JPGdefault 2 (resized).JPGdefault 2 (resized).JPGdefault 3 (resized).JPGdefault 3 (resized).JPGdefault 4 (resized).JPGdefault 4 (resized).JPG
    #37 4 years ago

    First I tried the oscilloscope. I went to default setup - I pressed cancel to go to the previous setup. I did that several times and it still didn't work. Then I downloaded the latest firmware and installed it. Still don't work. Looks like all the control knobs (potentiometers) are not working. Might be just a bad connector but I think it's still under warranty. I got it through Amazon so I'll contact them next.

    Next I swapped the U9's. (Stern to Bally). I first put the Stern U9 in the Bally and it booted right up. Turned it off then on again and got solid on. Got solid on several times and it finally booted again. I put the Bally U9 in the Stern and it works every time.

    The sockets on this board are AUGAT and the connections look really small. Should I change the U9 socket? I know I'll lose a bunch of solder pads. They just fall off this board.

    Thanks

    Bob

    #38 4 years ago
    Quoted from oldschoolbob:

    The sockets on this board are AUGAT and the connections look really small. Should I change the U9 socket?

    Carefully pry the plastic IC socket cover off. It will expose all the socket pins pins and you'll be able to see what condition they're in. Lever your screw driver off U15. Don't twist screw drivers under the socket against the PCB because it will damage traces. Lever off a close by IC.

    Do the same with the socket at U8 and U11 and inspect their pins for corrosion.

    #39 4 years ago

    Thanks Quench,

    I don't like to lever off of components. With my luck they'll break. I placed a toothpick on the ends of the socket and levered off that - worked pretty good. The pins looked OK but they are really small and some aren't closed completely. I decided to replace the socket. I'm very carefully heating each pin and removing them one at a time. This is going to take some time but I'm not doing anything anyway. I'll let you know how it works out.

    The U8 socket has already been replaced - looks like machined pin socket.

    This has been a crappy day. I didn't get the MPU fixed and I didn't fix the oscilloscope. I could have just stayed in bed. Hope your day was better.

    Bob

    #40 4 years ago
    Quoted from oldschoolbob:

    The U8 socket has already been replaced

    Sorry, I meant U7 where the 6810 RAM is.

    Quoted from oldschoolbob:

    I'm very carefully heating each pin and removing them one at a time. This is going to take some time but I'm not doing anything anyway. I'll let you know how it works out.

    Actually that's how I remove those sockets when I replace them. Pry off the plastic cover then heat each pin one at a time and pull them out with my fingernails
    Pretty much no risk of damaging traces provided you only start pulling the pins when you feel them moving in the molten solder. It takes me about 20 minutes to remove all the sockets on a board like that.

    Good luck with the oscilloscope - I hope the supplier resolves the issue.

    #41 4 years ago

    Hey Bob,

    What model of Hantek do you have?

    The scope indicates you are measuring 128V? By chance do you have the probe setting to 'X1' and not 'X10'?

    pasted_image (resized).pngpasted_image (resized).png
    #42 4 years ago

    Hi Mark,

    It's DSO5072P.

    Good catch - it was set on 10X.

    But the problem is none of the knobs are working. Can't adjust volts/div , sec/div , trigger level , not even the VO menu. It has 8 knobs and not a one will work. If it was one knob out I would suspect a bad pot. But they're all out so it must be power to the pots or something like that. I'm trying to get a schematic.

    I contacted the supplier but nothing yet.

    Thanks

    Bob

    #43 4 years ago
    Quoted from oldschoolbob:

    Hi Mark,
    It's DSO5072P.
    Good catch - it was set on 10X.
    But the problem is none of the knobs are working.
    Bob

    I know, but one thing at a time. If the attenuator is set wrong, you could saturate the AtoD, causing the vertical scale to potentially look like it was not working. Granted, the other functions (i.e. time base) should have still worked.

    Is anything working on the front panel? Buttons OK?

    BTW, I have the same scope. Wonder if it is just a loose connector.

    I'll see if I can round up a schematic too. is it still under warranty?

    #44 4 years ago
    Quoted from Quench:

    pull them out with my fingernails

    How many times have you burnt you fingers? I got some tweezers with Teflon tips. But that's still slow for me. But remember I'm old and slow.

    Bob

    #45 4 years ago

    Bob,

    There might be some interesting tidbits of info for the DSO5702P here: https://www.eevblog.com/forum/repair/hantek-dso5072p-oscilloscope-channel-1-drift/msg1840220/#msg1840220

    Doesn't specifically address your issue directly, but it shows some voltage test points which might come in handy if you have to fix.

    Also, I believe you can control the scope via the usb. It would be interesting if you can still do that.

    #46 4 years ago
    Quoted from oldschoolbob:

    How many times have you burnt you fingers?

    Heh, sometimes they bite for a split second but it's not that bad. It's not for everyone..

    #47 4 years ago
    Quoted from mbwalker:

    Is anything working on the front panel? Buttons OK?

    The supplier contacted me today and asked the same thing. Yes all the buttons work as they should. The volts/div and seconds/div will change when I use the auto set button. So the scope is working fine but I just can't change anything with the knobs.

    Quoted from mbwalker:

    Wonder if it is just a loose connector.

    That's what it acts like.

    Quoted from mbwalker:

    is it still under warranty?

    It's 15 months old. The supplier said one year warranty. But Hantek says three year.

    Those schematics are nothing like a pinball.

    Hopefully the supplier will come up with something showing how the pots are connected.

    How old is your scope?

    Quoted from mbwalker:

    Also, I believe you can control the scope via the usb. It would be interesting if you can still do that.

    I thought of that. I have a laptop next to the scope but I've never connected the two. Have you?

    Thanks

    Bob

    #48 4 years ago
    Quoted from Quench:

    Heh, sometimes they bite for a split second but it's not that bad. It's not for everyone..

    I get burnt enough accidently - I don't need to stick my fingers in there.

    I did get all the pins removed today - some didn't want to come off easily. Just a little clean up and a close inspection and I can install the new socket.

    Thanks

    Bob

    #49 4 years ago
    Quoted from oldschoolbob:

    How old is your scope?

    I thought of that. I have a laptop next to the scope but I've never connected the two. Have you?
    Thanks
    Bob

    Mines probably two years old. Still impresses me for what you can get nowadays for not a lot of money. There is a simple mod if you ever wanted to expand the bandwidth to 200MHz. I think it's just a resistor removal or add. Doubt if you would ever need that for a pin tho. If you look for the video to mod it, you might get an idea of what's inside and maybe spot a cable that if you took yours apart, could check. Don't do it if you think it's still under warranty tho. Don't want to void it.

    USB: I'm pretty sure I tried it once, or at least watched the video. No real interest in actually using that feature, so didn't give it much more than a passing thought.

    Your AutoSet button changing the scale is an interesting comment. I'll have to think about that. It looks like there's and input board, and I'm thinking the control panel might be a separate board? I'd dig up the 200MHz mod video and check it out.

    #50 4 years ago

    Here's the inside from the 200Mhz mod video:
    pasted_image (resized).pngpasted_image (resized).png

    Looks like a power supply board in the image.

    There are 252 posts in this topic. You are on page 1 of 6.

    Reply

    Wanna join the discussion? Please sign in to reply to this topic.

    Hey there! Welcome to Pinside!

    Donate to Pinside

    Great to see you're enjoying Pinside! Did you know Pinside is able to run without any 3rd-party banners or ads, thanks to the support from our visitors? Please consider a donation to Pinside and get anext to your username to show for it! Or better yet, subscribe to Pinside+!


    This page was printed from https://pinside.com/pinball/forum/topic/mpu-100-don-t-boot-on-first-try/page/1 and we tried optimising it for printing. Some page elements may have been deliberately hidden.

    Scan the QR code on the left to jump to the URL this document was printed from.