1103 replies to this topic

[mjr]

 

 

[weird output glitching when approaching the machine - does it happen when the machine is already on or just when booting?]
 
Yes machine already on. I love in Wyoming we get wicked static electricity due to how dry it is here.

 
Okay, that seems like something must be wrong at a connection level.  Even if the static levels are pretty high it really shouldn't do that. 
 
You can deal with the static issues to some extent by making sure that all of the external metal is wired together with a grounding strap to Earth ground - plunger, side rails, lock bar, legs, etc.  Given your staticy environment that would probably be a good idea in general, and of course it's good for human safety reasons anyway.
 
But the glitching makes me think there might be some other issue, like a bad ground or power connection somewhere on the boards.  (Not necessarily at the power plug level - I'm just talking about connections between parts, like ground and Vcc pins on the ICs and so forth.)

So how do I fix this? And my machine worked great for 2 months with all these toys before a problem started

 

I'm really not sure what the next thing to try - the symptoms are hard to interpret.  The thing that makes it hard is that it sounds like it's working most of the time, and whatever's wrong only manifests under stressful conditions that are hard to reproduce.  Well, not hard to reproduce exactly, but hard to keep going in steady state so you can study what's going on.  When something's just flat-out not working you at least can trace through all of the connections until you find what's wrong, but I don't know how you trace something like this that only happens for a brief instant and makes everything blow up.

 

Maybe do a careful inspection of the boards and see if you can find any bad solder joints - maybe the thing that changed was that a weak solder joint finally broke completely from vibration or something.  I'd also install some grounding straps around the metal parts and see if that helps with the glitching.  Is winter any worse than summer in terms of the static issues?  If so that might be what suddenly changed.

 

You've got the power supply grounding right, you've got the diodes in place, so all of the basic protective stuff is there.  

 

How big is your 24V supply, wattage-wise?

Edited by mjr, 17 February 2018 - 08:13 PM.

[yagesz]

24v 15A 360w

 

Alright Ill pull boards and have a look. This is just very aggravating. I am going to pull the PSU for now and try controlling everything with the 12v supply and use a step up board.

[mjr]

24v 15A 360w

 

That seems like plenty of power for the knocker.  

 

 

This is just very aggravating. I am going to pull the PSU for now and try controlling everything with the 12v supply and use a step up board.

 

Might be worth a try.  I guess there's an outside chance the 24V supply is the problem.  Taking it out of the picture for a while would at least say one way or the other about that.

[inventagain]

Mjr,

Do you have any bare PCBs you can sell me? I have several KL25Zs and want to try your MAIN and POWER boards. If you have them what are is the pricing for each and for the set with CHIME.

If you do, what are the payment methods?

I live in St. Peters, MO 63376

Thanks,

Grant

[mjr]

Do you have any bare PCBs you can sell me? I have several KL25Zs and want to try your MAIN and POWER boards. If you have them what are is the pricing for each and for the set with CHIME.
Thanks,
Grant

 

Grant - looks like we connected via email.  For others interested, I've been keeping a supply of the (unassembled) boards on hand, so I have them available for anyone who needs them within the US.

[yagesz]

Hey MJR::

 

I am still struggling with my system crashing. I removed the 24v PSU I had. I swapped for a 12v Step UP Booster board running off my PC ATX PSU. Everything seemed to be better. The shaker was working. Knocker wasn't crashing the system thought I had finally fixed it....

 

Then I plunged a ball and this burned out a PSU. Any idea what is going wrong.  I can't enjoy my machine when my plunger is a problem too. 

I am a nose hair shy of rewiring my entire cabinet from scratch just so I can test one piece at a time to make sure it doesn't cause me a problem. 

 

Also removing the 24v PSU seems to have diminished the surging I was seeing before.

Edited by yagesz, 04 March 2018 - 05:03 AM.

[mjr]

Then I plunged a ball and this burned out a PSU. Any idea what is going wrong.  I can't enjoy my machine when my plunger is a problem too. 

 

Hmm.... it's really hard to imagine how plunging the ball and a PSU failure could be connected, at least at any sort of direct electrical level.  I don't know what you're using for a plunger, but whatever it is, that's all logic-level stuff that really shouldn't be able to cause a PSU overload even if it tried.  So, I'm thinking this had to be either purest coincidence, or was connected to something only incidentally related to the plunging action, like a 24V feedback toy firing at the same moment, perhaps.

 

The only other way the two could be connected goes back to something outside of the DC side of things, like static discharge off one of the metal surfaces of the plunger.  Did you ever run around the cab and connect everything to a grounding strap?  The plunger would be a good thing to include in that if you haven't already, since that's a bunch of exposed external metal. 

 

But I still think static is a pretty far-fetched explanation in the first place.  If you were frying motherboards or KL25Zs or LEDs, static would be on the list of usual suspects, but I don't think it's very likely to fry power supplies.

 

And overall, I'm still having a hard time figuring out how your power supplies keep getting fried.  ATX PSUs are by supposed to be protected against shorts and overloads - it's part of the spec.  In principle it should be impossible for anything on the DC side to actually damage the PSU.  Even an outright short across the rails shouldn't do more than shut it down temporarily until the thermal overload circuit resets by itself.  That should rule out a lot of possible causes, which you'd think would narrow it down nicely, but what's making it so hard for me is that it narrows it down so much that I can't think of anything that's left to look at.  Except for far-fetched external things like static electricity. 

 

Or maybe house wiring surges?  Do you think there's any chance it's something like that?  Do you have all of the AC inputs connected through a good surge protector?  The single time that I've ever had an ATX PSU fail seemed to be caused by a house wiring surge.  (Hard to be sure, of course, but it was an ordinary desktop PC that had been working in its configuration for a couple of years, and one day every piece of electronics in the thing failed at once - from the PSU to the motherboard to the hard disk.  The only likely explanation seemed to be a power surge on the AC.)

I am a nose hair shy of rewiring my entire cabinet from scratch just so I can test one piece at a time to make sure it doesn't cause me a problem. 

 

I hate to say it, but if this were happening to me, I think that's exactly how I'd proceed at this point.  I'd probably start by just taking out all of the feedback stuff for a while to see if that eliminates the problem, making sure to run it long enough in that configuration to be certain that the random element has had plenty of time to recur if it's going to.  And then gradually add back one thing at a time.  Sporadic problems like this are the very worst things to debug because you never know if "the problem didn't happen in the last X days" means that (a) it's fixed and it's never going to happen again, or (b) it's not fixed and it's just waiting until day X+1 to show up.  All you can do is play the odds and go by the observed time between incidents in the past; if the longest time you ever saw between faults was X days, you gain some confidence when it doesn't appear after X days, more confidence when it doesn't recurs after 2X days, even more after 3X days... there's never a point where you're 100% certain, but if you can go 3 or 4 sigmas out without seeing the fault recur,  you can start to be pretty confident.

[yagesz]

Ok. So I will rewire from scratch. Remove all toys. Play with LEDs only at first. Make sure nothing dies.

Replace my surge protector as I have a good APC laying on a shelf I don’t use since running my whole server with ups

Then add toys 1 at a time.

Bright side is I will have prettier wiring then I do now.


Sent from my iPhone using Tapatalk

[stevied007]

Is there any way to use the PWM output to create a "peak and hold" for flipper solenoids?

 

Eg: Full output to activate solenoid, then lower output to hold the solenoid in position if button is held.

 

Or is a mechanical solution the best option?

[mjr]

Is there any way to use the PWM output to create a "peak and hold" for flipper solenoids?

 

Eg: Full output to activate solenoid, then lower output to hold the solenoid in position if button is held.

 

The Flipper Logic option is exactly what you're looking for.  You can find it in the Config Tool in the output port configuration section.  Each port has a little flipper icon next to it.  Click that to bring up the options - you can set the initial time interval where the output is held at full power, and the reduced power level to use after the initial interval expires. 

 

You'll need to experiment with your specific devices to find the ideal settings.  You want to find the lowest power setting that keeps the device switched on, since that should give you the least heating of the coil. 

 

Once you find settings that work, I'd test the coil to make sure its temperature does stabilize at the lower setting, because I'm not sure it's a foregone conclusion that all coils are safe to operate indefinitely even at reduced power.  If you can operate the coil for a couple of minutes without it getting uncomfortable hot to the touch, it's probably pretty safe - a couple of minutes should be enough to either reach thermal equilibrium, so if it's not getting too hot by then, it's probably not going to no matter how long you leave it on.

[yagesz]

 

 

 

 

So...

Only thing I can conjure at this point is my board is not soldered correctly somewhere. I unhooked EVERYTHING except LEDs and Plunger. Plunger is a 10k Potentiometer. If in game and game is going nuts on LEDs system will crash on a plunge. If I am in PBX with no lights on at all I can get it to do it as well.  Where on my boards should I focus my attention on soldering points.  I only have the Mainboard and KL25Z hooked up. No Power Board.  I can take video for you if you like to see but I can do it with the computer sitting on the desktop.  I even thought it was maybe the rapidness of the springs on the plugner. I can go slow and trip it to. My only thought at this point is the ground loop between everything is compromised. Or the KL25z is compromised.

 

I will even request that if your up for it I could mail it to you to inspect because I am at a loss at this point and have no clue what to pinpoint. Its not the PSU's. I have burned several High Power and lower power at this point.
 

Also. I can make the LEDS GO NUTS. Nothing happens. Only with plunger in conjuntion.

5 FLashers
4 RGB Flipper Buttons

1 Underglow.

 

24V is hooked up via Booster Board. But is not feeding anything. Everything goes to Terminal Blocks and breakouts from there. The only things hooked up are the leds. I removed all other wiring.

Edited by yagesz, 18 March 2018 - 02:20 AM.

[mjr]

Only thing I can conjure at this point is my board is not soldered correctly somewhere. I unhooked EVERYTHING except LEDs and Plunger. Plunger is a 10k Potentiometer. If in game and game is going nuts on LEDs system will crash on a plunge. If I am in PBX with no lights on at all I can get it to do it as well.

 

So, it sounds like you've narrowed it down to the plunger at this point?  Everything works if the plunger isn't connected, and things start going wrong again when you re-connect the plunger?

 

It that's it, I can't really think of anything wrong with the soldering that would cause this.  My only guess at a wiring problem is that the wrong voltage is connected as the input to the pot.  You have it connected to 3.3V and GND, not 5V, right?

 

Assuming that's all right, maybe the KL25Z itself has developed a problem.  Like, maybe its ADC input (the pin connected to the pot) got damaged by static discharge at some point, and now having any voltage attached there is causing problems in the CPU.  That would be one possible reason for the erratic behavior on the outputs, since those are all controlled by the CPU, although I don't see any way to connect it to the power supply damage, so there'd have to be something else going on for that part.

[Fusionwerks]

Hey MJR, I finally have the need to install the IR emitter after having to replace my backglass monitor.
I can't seem to find the pinout for jp4 to determine which header pins I need to use.
I have chosen the 39 ohm resistor for r11, which I had written down was for a single LED and the 27 was for 2. Is that correct?
I may end up going with a direct power on wire, but i haven't looked into the switch yet, and I don't know where they wire on the board either

[mjr]

Hey MJR, I finally have the need to install the IR emitter after having to replace my backglass monitor.
I can't seem to find the pinout for jp4 to determine which header pins I need to use.
I have chosen the 39 ohm resistor for r11, which I had written down was for a single LED and the 27 was for 2. Is that correct?
I may end up going with a direct power on wire, but i haven't looked into the switch yet, and I don't know where they wire on the board either

 

Here's a connection diagram for the IR LED:

 

 

And you're right about the resistors: use 39 ohms for R11 for a single IR LED, and 27 ohms for two LEDs (wire the two in series).  The resistor in either case should be a 1/2 Watt type (a larger wattage is always okay for a resistor electrically, but they tend to get physically larger the higher you go).

 

For the hard-wired switch connection, those go to the next two pairs of pins on the same header.  Use the "column" marked TV1 on the diagram for the first TV, and use the column marked TV2 for the second.  These are just connections to the relay terminals, so the order of the wires doesn't matter.  Just connect the first TV switch wires to the TV1 column pins in either order, and likewise the second TV for the TV2 column pins.

[Fusionwerks]

It's clear as day when you can see the board, but i have a header soldered on it and couldn't see it. I am going to try the switched power first. I feel its more reliable than trying to make the IR work.

[mjr]

It's clear as day when you can see the board, but i have a header soldered on it and couldn't see it. 

 

Yeah, unfortunately the connectors cover up most of the printing.  I would have liked to have the markings more out in the open, but there wasn't enough room.

 

 

I am going to try the switched power first. I feel its more reliable than trying to make the IR work.

 

If you can physically get to the switching connectors, that should indeed make a nice reliable setup.  I'm actually more inclined to recommend the IR, though, since it's non-invasive.  Getting to the switch connectors can be difficult and dangerous (in terms of damage to the TV) since you have to open up the case.  On my own cab, I've also had one incident of a solder connection to the tiny terminals on the TV switch breaking spontaneously, probably from vibration - so it might be something that requires maintenance work once in a while.  (It's only happened to me once over two years, so it's not like it's a frequent problem, but even so.)

[Fusionwerks]

I may end up going to the IR after all. When I opened up the monitor, I relocated both the power button and IR receiver, so it's easy to try either way. So far, with the switch its works, but not quite right. It fires the relay at boot and the monitor comes on, but the monitor doesn't show anything after it finishes booting. I have tried a few different timings with no real change. Maybe I need a longer delay, so it comes on after boot is complete, but I'm afraid it won't be recognized and default my monitor numbers out of order. I don't know I will have to tinker. Problem is my pc boots funny sometimes. 1 time it will boot in 30 seconds, then the next time, it hangs and takes minutes to finish. I think it's a video card problem. It doesn't like the order that my monitors are set up I think. Something about it wanting to default to the first display port before hdmi...
Maybe you have some suggestions on what to try for the delay? Or at least what to look for to find a starting point?

Edited by Fusionwerks, 20 March 2018 - 04:21 AM.

[mjr]

I may end up going to the IR after all. When I opened up the monitor, I relocated both the power button and IR receiver, so it's easy to try either way. So far, with the switch its works, but not quite right. It fires the relay at boot and the monitor comes on, but the monitor doesn't show anything after it finishes booting. I have tried a few different timings with no real change. Maybe I need a longer delay, so it comes on after boot is complete, but I'm afraid it won't be recognized and default my monitor numbers out of order. I don't know I will have to tinker. Problem is my pc boots funny sometimes. 1 time it will boot in 30 seconds, then the next time, it hangs and takes minutes to finish. I think it's a video card problem. It doesn't like the order that my monitors are set up I think. Something about it wanting to default to the first display port before hdmi...
Maybe you have some suggestions on what to try for the delay? Or at least what to look for to find a starting point?

 

The Windows video drivers do have some kind of dependency on connector types.  I don't remember the exact rules, and they probably vary by video card anyway, so I think you'll have to experiment.  My own experience was that one of my ports (HDMI, I think) simply can't be the primary monitor no matter what settings you make in Windows - it seemed to be hard-wired in the video drivers with no way to change it.  So you might have some similar issue.  If you can figure out which port Windows insists on being the primary, you could try swapping the TV connectors around to make your fastest-booting monitor the primary.

 

Windows does have some boot logs that you can check to see what's hanging it up on those long boots, by the way.  It takes some serious patience to find anything in there, because you have to sift through 10,000 lines of extraneous garbage to find the one line with the interesting bit you're looking for, but sometimes it's the only way to track down weird issues like this.

Anyway, the timing for the TV ON switch shouldn't have any dependency on the Windows boot time.  If anything, you want the monitors to turn on before Windows finishes booting, so that they'll be visible on the hardware ports by the time Windows scans for monitors during the boot cycle.  You only need to delay the ON signal long enough to give the *TVs* time to become responsive to control inputs.  Most TVs take a few seconds to power on to that stage.  The default timing in the firmware is 7 seconds, I think.  Maybe what you need to do is shorten the time to get your TVs online earlier in the boot process - maybe try reducing it to 5 seconds or 3 seconds.  The only thing that should matter is whether the TV is responsive enough to receive the ON switch signal and power on.

[yagesz]

 

Only thing I can conjure at this point is my board is not soldered correctly somewhere. I unhooked EVERYTHING except LEDs and Plunger. Plunger is a 10k Potentiometer. If in game and game is going nuts on LEDs system will crash on a plunge. If I am in PBX with no lights on at all I can get it to do it as well.

 

So, it sounds like you've narrowed it down to the plunger at this point?  Everything works if the plunger isn't connected, and things start going wrong again when you re-connect the plunger?

 

It that's it, I can't really think of anything wrong with the soldering that would cause this.  My only guess at a wiring problem is that the wrong voltage is connected as the input to the pot.  You have it connected to 3.3V and GND, not 5V, right?

 

Assuming that's all right, maybe the KL25Z itself has developed a problem.  Like, maybe its ADC input (the pin connected to the pot) got damaged by static discharge at some point, and now having any voltage attached there is causing problems in the CPU.  That would be one possible reason for the erratic behavior on the outputs, since those are all controlled by the CPU, although I don't see any way to connect it to the power supply damage, so there'd have to be something else going on for that part.

 

 

 

With plunger removed. I still get crashes on my system with the shaker motor and Knocker (both freshly rewired. SO only LEDS. Motor. System crash. Leds and Knocker. System Crash.  Just LEDS... Plays all day.

[mjr]

With plunger removed. I still get crashes on my system with the shaker motor and Knocker (both freshly rewired. SO only LEDS. Motor. System crash. Leds and Knocker. System Crash.  Just LEDS... Plays all day.

 

Okay, I see.  

 

Have you tried it with just the plunger connected?  No LEDs, motor, or shaker?  I'm thinking the plunger must be a whole separate problem, which might be part of why it's been so difficult to figure out what's going on.  If you can get a crash without anything but the plunger connected, that would tend to confirm that idea.  If not, I'm not sure what to make of the plunger crashes when other things are attached - maybe some kind of interaction I'm not seeing, maybe pure coincidence.