1103 replies to this topic

[LynnInDenver]

It's an accessibility helper option in Windows called Sticky Keys. It's there because some people have issues holding down CTRL or SHIFT and pressing another key, either because they're missing a hand or they have motor coordination issues.

 

In Ease of Access in the Control Panel, you go to keyboard settings/make keyboard easier to use, and it will be under "make it easier to type". Go ahead and turn it off there, and also go into "Set Up Sticky Keys" and tell it not to turn it on when SHIFT is pressed five times. It's the one major accessibility feature in Windows that actively gets in the way of virtual pinball on a new install.

[ikendrick]

LynnInDenver

 

Hi LynnInDenver,

 

100% correct. DOH!!!!

 

I thought I had turned that stuff off as I have had issues with that before in gaming. 

 

Thank you so much for saving my sanity. Works like a charm now.

 

Ian.

[cerimorgan]

I've built the Main and Power boards and have worked through the testing of the Main board. I've been testing the Main board and had success with all of the button inputs, the small LED outputs, the RGB flasher outputs, and the strobe output. Thanks for the fantastic build guide and testing/debugging instructions.

 

However, the knocker output doesn't work. I've checked the soldering twice on all of the components mentioned on the debug page of the build guide under the main board knocker section, and also checked continuity between those components and their immediate connections (according to the schematics), but still no luck.

 

Possibly connected, but my crimp pins connections from the ATX power supply don't seem to be very sturdily connecting to the board inside the crimp pin housings(?). When I connect the housings, I usually have to push down on the wires to force the crimp pins to stay down and make a solid connection. Could I have the crimp pins installed backwards?

 

Are there any tests I should try with the power on for the knocker? I have verified that pin 4 of OK5 is at 12V when the power is on.

 

Any other suggestions? Thanks!

[mjr]

cerimorgan - sounds like you've already done all of the basic debugging on the knocker circuit that I could suggest.  You said you checked the soldering, but you might still try re-flowing all of the connection points in that block.  I mean heating them up with the soldering iron to get the solder to melt again, so that it flows into any crevices in the contact points that it missed the first time.  Beyond that, I guess you're at the point where I'd suspect a bad component, so you might have to go through the circuit and do a live test to see if you can find any pins with voltages that aren't what they should be.  Some starting points:

 

R37 - should read close to 0V when the knocker output is off in the software, about 3.3V when on

OK5 pin 2 - should read close to 0V when the knocker output is on in the software (indeterminate when off)

OK5 pin 1 - should read close to 0V when the output is off, and around 2V when the output is on, but only for a couple of seconds (then the timer kicks in and sets it back to 0V)

 

I don't like to have to do live testing, since it's difficult - everything's so close together and you have to be really careful about shorting nearby pins. But I'm not sure what else to try, other than just wholesale replacing all of the components in that block.  I don't think the power supply connection is the culprit here, since you probably wouldn't be seeing anything else work if it were.  Although if I were you, I'd want to address that first anyway, make sure it's good and solid, since a flaky power connection is likely to be an ongoing headache.  I'd probably just rebuild the crimp housing from scratch if you have spare pins; something probably just got bent the wrong way the first time, and I've never had much luck fixing those tiny pins once they get bent wrong.  It's usually a better bet to just replace them.

[GunSheep]

MJR,

Checking to see if you have any expansion boards available?

[cerimorgan]

Thanks for the quick reply and the warning about live testing. I was much more careful than I likely would have been.

 

R37 - should read close to 0V when the knocker output is off in the software, about 3.3V when on

OK5 pin 2 - should read close to 0V when the knocker output is on in the software (indeterminate when off)

OK5 pin 1 - should read close to 0V when the output is off, and around 2V when the output is on, but only for a couple of seconds (then the timer kicks in and sets it back to 0V)

 

R37 and OK5 pin 2 behaved as you described. OK5 pin 1 was at about 3.2V when the output is on (and went to 0 after a couple of seconds). Does that imply an incorrect resistor somewhere?

 

I'm going to try and reflow all the solder joints in the knocker circuit.

 

I did also redo the crimp pins for the power and it came out much better with my second attempt. It all seems very secure now.

 

Do you advice on how to terminate the unused yellow and black wires coming from the ATX connector that connects with 2 pins to JP7? I'm currently just using a screw cap for each wire and added some electrical tape to try and secure it (though the tape is not really sticking well to the wire).

MJR,

Checking to see if you have any expansion boards available?

 

GunSheep, I have some extra Main and Power boards (no Chime boards though) if you're interested. I had to order 5 of each from Elecrow and they ended up shipping me 10 of each. Feel free to PM me your address if you're interested.

[GunSheep]

MJR,

Checking to see if you have any expansion boards available?

 

GunSheep, I have some extra Main and Power boards (no Chime boards though) if you're interested. I had to order 5 of each from Elecrow and they ended up shipping me 10 of each. Feel free to PM me your address if you're interested.

 

Thanks Cerimorgan. PM on the way. 

[mjr]

OK5 pin 1 was at about 3.2V when the output is on (and went to 0 after a couple of seconds). Does that imply an incorrect resistor somewhere?

 

Hmm.... I think that might mean that OK5 (the optocoupler chip) is defective.  The 3.2V out of the timer circuit is correct - that's exactly what you should be seeing at IC11 pin 3, which is the output pin of the 555 timer chip that controls the 2-second time limiter for the knocker output.  During the 2-second "on" interval, that pin outputs a "high" voltage of just slightly below the 3.3V supply voltage, so 3.2V is just about right.  And it drops to 0V when the 2-second timer expires.  So the reading you're seeing - 3.2V for a couple of seconds, then 0V - says to me that the whole timer block up to and including the 555 chip is working correctly.

 

But what should be happening when OK5 pin 1 is at 3.2V is that the OK5 LED side should be conducting current and consuming power.  If that were happening, the voltage reading on pin 1 should dip well below the full 3.2V coming out of the timer output - it should drop to something around 2V.  If you're reading the full 3.2V, I think it means that OK5's LED side isn't conducting any current - it's acting like an open circuit.  The other end of the LED is OK5 pin 2, so either the voltage on pin 2 is wrong, or the LED isn't working.  Since you also measured the voltage at pin 2, and that looks to be right, it sounds like the problem is between pins 1 and 2, which means either the chip is broken or you've got a bad solder connection at one of those pins.

 

So the first thing I'd do is try re-flowing the solder on pins 1 and 2 of OK5.  If that doesn't fix it, I think you probably have a dead OK5 chip, in which case the only thing to do would be to replace it.  Which unfortunately fits my own experience - on the first prototype I built of this board, I had *two* bad PC847 chips (well, partially bad: each one had one non-working LED).  Not quite identical to your issue - PC847 vs P817 - but they're basically the same chip.  Maybe they're especially delicate, or there's a quality control problem at the factory.

Edited by mjr, 27 September 2021 - 06:40 PM.

[mjr]

One other quick thought - you raised the possibility that the 3.2V on OK5 pin 2 could mean a wrong resistor value.  It occurs to me that that actually is a possibility.  If R18 were a very large resistance value, like maybe 47K instead of 47 ohms, that could result in similar voltage readings to what you're seeing.  So do a quick test of R18's resistance and see what it reads.  (It should be okay to measure it in-circuit, with the power off.)

[cerimorgan]

I verified that R18 is 47 ohms. I replaced C5, C8, C9, IC11, OK5, R6, R8, R10, R12, R13, R14, and R18. Still have C7, R37, T2, T3, and Q1 to replace, but I saw your message about OK5, so I figured I'd give it a go, but same result. I looked more closely again at the OK5 connections and noticed that the solder isn't sticking to the pad for pin 1. I'm guessing I need to clean off the contact pad(?)

 

One other thing I initially forgot, but noticed as I was finding replacement components... when initially order parts, I couldn't find the 2N4401 NPN transistor anywhere, so I replaced them with 2N3904 PBFREE. From what I could tell from the datasheet, the lead layouts seemed to be the same, but I may have misread that (page 2 shows a diagram where the leads are labeled, but it's not clear to me whether it's labeling them from the flat side or the rounded side). This would be for T3.

 

The replacement OK5 came from the same batch as the original one. I just ordered a few more from DigiKey to try and rule that out (or fix it!)

 

Thanks so much for your help so far! While I wait for the replacement octocoupler, I'm going to move onto connecting and testing the power board.

[mjr]

The 2N3904 looks fine to me for T3, and that seems to be confirmed from your measurements - the 2-second timer wouldn't be working if that transistor weren't doing its thing correctly.

 

If you want to test the PC817 chips out of circuit to see if they're really the problem, you could set up a little breadboard circuit like this:

 

+3.3V  ------- 47 ohm resistor ------- PC817 pin 1

GND ------- PC817 pin 2

GND ------- 1K resistor -------- PC817 pin 3

+3.3V ------- PC817 pin 4

 

If it's working, you should see about 3V on pin 3 in that configuration.  If that reads correctly, try disconnecting pin 1, and pin 3 should go to 0V.  If both of those read correctly, the chip is probably good after all.  Could be worth testing to make sure this isn't a red herring.

 

If it's more convenient, you can use a 5V supply instead of 3.3V.  Just change the 47 ohm resistor to about 200 ohms.  And of course the expected voltage reading on pin 3 will be closer to 5V.

Edited by mjr, 28 September 2021 - 04:07 AM.

[cerimorgan]

Thanks for the further suggestions. I haven't had time during the week to do anything up until today. I took another look at OK5 on the main board and the pin 1 connection is definitely not in good shape. The solder is no longer sticking to the bottom of the board at the pin. It looks like there's a bunch of resin(?) on the board, causing the solder to not stick. Any recommendations for cleaning it off?

 

I'll probably try to build the knocker circuit on a breadboard with spare parts tomorrow and ensure that it works. The replacement 817 chips arrived today so I'll try with one of those.

 

I did connect the power board to the main board and power for the first time today and managed to get all of the outputs working. Initially 8 or the 32 didn't work, but I inspected and reflowed a bunch of suspect looking solder connections which fixed 7 of them. I was able to trace through the final JP5-1 pin circuit and re-soldered the pins on Q1A which fixed it).

Edited by cerimorgan, 02 October 2021 - 05:58 PM.

[mjr]

If the gooey stuff is from the solder flux, I think the usual cleaner for that is just regular isopropyl alcohol.  If that doesn't work, you might also try melting it with the soldering iron  and using a solder-sucker to remove it.

 

Sounds like good progress with the power boards!  In the worst case, you could leave the main board knocker output unfixed and use one of the chime board outputs for the knocker instead, since they're exactly the same design (assuming you have a spare output that you weren't already planning to use for something else).

[cerimorgan]

I was able to successfully test the PC817 chips on a breadboard. I tested one from the old batch and one from the new batch and they both worked correctly.

 

I'm still having problems with soldering pin 1 of the chip to my main board. I think I've damaged the board and am hesitant to keep playing around with it in fear of damaging it more. I looked through my electronics parts and had almost everything I needed to build a chime board (I ordered the extra parts just in case because it wasn't that much more expensive), so I just ordered some chime boards from Elecrow (not sure why I ordered the parts for a chime board, but not the actual board!), an additional timer chip from Newark, and the capacitors from Mouser. Is there a sub-section of Murphy's Law that states that one vendor will never have all the parts you need in stock?

 

Thanks again for all of your help so far mjr! The guide has been excellent in getting me this far, even with debugging issues with the board, which was my big fear before getting started. My cabinet flat pack from VirtuaPin should ship out in a week or two. I think that's when things will really start coming together.

[mjr]

Is there a sub-section of Murphy's Law that states that one vendor will never have all the parts you need in stock?

 

Well, if there isn't one normally, that sure is the case right now with the Worldwide Semiconductor Supply Chain Crisis going on.  I've run into exactly the same issue recently - sourcing parts (even run-of-the-mill parts) is a huge pain right now.

It's definitely possible that some copper (pad or trace) on the PCB is the problem.  I've had that happen myself after trying to remove parts; the glue that holds the copper down can fail and make the copper separate.  I've fixed that in prototypes a couple of times using little jumper wires, but it's not a great solution for a real board.  Anyway, the chime board should do the trick, and give you plenty of extra timed outputs for future coil-based devices. 

[cerimorgan]

By the way, mjr, you requested feedback on the availability of a generic lockdown with fire button options. Virtua Pin now sell what I think you're looking for: Geniune Widebody Lockdown Bar - w/Fire Button & Lever Guide Ass.

 

They conveniently added it to their inventory the day before I placed my order. My flatpack is scheduled to arrive today and it looks like the second shipment that includes the lockdown bar just shipped today, so I should be able to give further feedback later this week.

[cerimorgan]

A few weeks later (thought I had all the components needed, but was short two 47 ohm resistors), I've finished the Chime board and was able to successfully test all the outputs. I've never been so happy to see an LED blink on and off!

 

In the meantime, I've assembled the cabinet, mounted the playfield monitor and been able to test a couple of button inputs (Start and Launch Ball).

 

I've wired up the outputs from the power board to the lamps on the buttons and they work in the Pinscape Config Tool, but not in Visual Pinball. I haven't really done much reading/debugging yet, but that'll be my next step.

 

Thanks again for the help and suggestion of using the Chime board.

[oldgreg]

Hello fellow pinballers. After months at the last stage of cleaning up my pinball build after I got working 50v knocker and all I think I blew something on my main board. Now button inputs are still working fine but outputs don't. Button leds just weirdly flicker randomly, same for knocker. Tried new kl25z and same behaviour. Anyone had similar issue?

[mrarcade]

This is another lesson about not buying from GGS. Anyone reading this, please save yourself big trouble by buying instead from Arnoz who is reputable.
 

How do you order from Arnoz?

Is there a website with prices etc?

[ACDNate]

https://shop.arnoz.com/en/

 

 

 

This is another lesson about not buying from GGS. Anyone reading this, please save yourself big trouble by buying instead from Arnoz who is reputable.
 

How do you order from Arnoz?

Is there a website with prices etc?