MiSTer Reboots When Packaged in Ultimate Case?

[ElMariachi003]

Hi everyone,

UPDATE: I resolved my issue, only to find that it came back as soon as I wrapped it back up in my Ultimate MiSTer case... Don't know if it's a possible overheating issue? (The FPGA has a heatsink and the fan is operating normally)

I just got my MiSTer set up today, and have been having issues trying to get it up and running. It seems to reboot after only a minute or two, and after a few attempts, the screen will just stay black with the Red/Yellow slow flashing in a alternating sequence.

I have tried re-imaging my SDcard (with Mr. Fusion 2.5), and 2 different ones at that. Mr. Fusion does its thing on boot up, and I get the MiSTer Dashboard after the reinstall. This is where my problems begin. By the time I get to the Script menu to perform an update all, the screen will go black, and it either goes back to the initial Dashboard, or stays with the screen black and the LED's flashing in the sequence described above. I don't even have to do anything; the unit will just reboot on its own after a minute or two at most. When it does boot multiple times, it seems like the amount of time before it reboots again gets shorter. NOt sure if that's a sign of overheating, but I do have a heatsink on the Cyclone in addition to the fan from the IO board.

The SD's were SANDisk 128 and 256 GB cards and I am using a 4A rated PSU. I even tried reimaging the SD card a few times. One more thing that I tried was to insert the card that came with the DE-10; it will boot to the Linux desktop, and I have left it running for 10 minutes with no problem.

After so much time waiting to get a DE-10 from Terasic, this is... frustrating to say the least. Has anyone experienced this?

Thanks in advance.

[AngelicLiver]

Have you checked the 6 DIP switches are in the correct position?

[Stinky]

Do you have an inline power switch? Try without.

[ElMariachi003]

I checked the DIP switches, and although they were correct, I decided to toggle them all based on other search results I found about them.

That appears to have done the trick! Thank you both for the suggestions.

EDIT: I guess I spoke too soon - however, after two attempts, I have isolated the problem to my Ultimate MiSTer case, of all things. I initially tested it prior to putting the case on, then after reinstalling the case, ran into issues again. Took it off again, and the MiSTer is working. Very odd. Anyone else experience this?

[AngelicLiver]

Sounds like something could be shorting somewhere. I'm not personally familiar with that case though.

How is the USB hub getting power? Is it one of those metal prong jobbies (I'm really not a fan of those)?

[LamerDeluxe]

Is your fan blowing air out of the case instead of pulling air in?

[ElMariachi003]

Is your fan blowing air out of the case instead of pulling air in?

Hmm, I just happened to stumble on a thread about fan orientation in the Hardware Forum. It is currently label side down. Based on your question and the content of that forum thread, I will try flipping the orientation to see if blowing the air out and up makes a difference. Thanks for the suggestion.

[ElMariachi003]

Sounds like something could be shorting somewhere. I'm not personally familiar with that case though.

How is the USB hub getting power? Is it one of those metal prong jobbies (I'm really not a fan of those)?

I’m using one of those small loop cables. One of the things I checked was to see if it was heating up at all, but it isn’t. Based on my updated resolution process, it seems to be related to operating within the confines of my case, so fan orientation will be my next troubleshooting step. Thanks for the input/suggestion.

[LamerDeluxe]

Is your fan blowing air out of the case instead of pulling air in?

Hmm, I just happened to stumble on a thread about fan orientation in the Hardware Forum. It is currently label side down. Based on your question and the content of that forum thread, I will try flipping the orientation to see if blowing the air out and up makes a difference. Thanks for the suggestion.

The fan in my acrylic case has the label upwards and blows the air out of the case, which, at least with this case, is how it is supposed to be.

[ElMariachi003]

Well, I tried to flip my fan today, however, the Noctua fan I have is designed to accept the panhead of the screws on the fan side; it has a small inset for the head, allowing the screw to peek out a little more on the other side of the PCB. By sitting flush on the face of the label side, my screws are too short to hold the ADC board. I also tried to see if I could swap the wires in the JST connector, but I couldn't push the clips far enough to disengage them. Oh, well - I'll need to get a screw about a mm longer before I try this again. I suppose I could just leave the ADC board plugged into the DE-10's connector as opposed to being held by the IO board with that 4th screw, but I'll just wait. In the meantime, I guess my MiSTer will remain naked for now.

[Stinky]

I doubt it's overheating I run my Mister full-hamburger in a case without a fan and never have any problem. At most you would see some glitches not a full system fault. It is far more likely your issue is power related, inline power switches are known to be a bad thing if you have one.

[neogeo81]

maybe does the case push on the reset button?

[ElMariachi003]

Thanks for the suggestions, although until I get a chance to test the fan with airflow pulling up and out of the case, I’m still going with the “trapped heat” theory.

I know that it’s not the case top - the buttons still sit loose on the case, and even with the top screws removed (but the top just lying there connected), it would still reboot.

I don’t think it’s a power issue either, because I was able to run it “naked” for 3+ hours on several occasions already without a hiccup. As an example, I ran it off and on for about 5 hours yesterday as I added/tested ROMs for additional cores yesterday (so pumped that I got Neo-Geo working!)… I’ll revisit these, but for now, I’m gonna just find a slightly longer 3mm screw to test that theory with. I’ll revisit all the other suggestions it turns out to be a dead end as well.

[MostroW]

if you take the MiSTer setup out of the case and run it barebone, does it keep working?

you might have a dead short to the case somewhere, i've had that once with a pc case which my father bought for his pc that kept that machine in a dead boot loop.
once i ran it outside the case it ran perfectly fine, when i replaced the case with a decent one there were no troubles anymore.

not saying it's definitely your issue, but might be worth looking at if you didn't already.

[ElMariachi003]

if you take the MiSTer setup out of the case and run it barebone, does it keep working?

you might have a dead short to the case somewhere, i've had that once with a pc case which my father bought for his pc that kept that machine in a dead boot loop.
once i ran it outside the case it ran perfectly fine, when i replaced the case with a decent one there were no troubles anymore.

not saying it's definitely your issue, but might be worth looking at if you didn't already.

Thanks for the suggestion; at this time though, I don’t suspect the case itself, because the Ultimate MiSTer case I have is made of PCB board material. Other than the small tabs that go over the standoffs to secure the side piecesand the rings around each screw hole, there is nothing conductive on it. If I had one of the metal cases available, I would definitely take that into consideration. Right now, I’m still using the bottom piece, which is loosely secured and has the silicon feet attached so that the standoffs don’t scratch any of my surfaces otherwise.

[C-R-T]

The case may cause flexing on the mister board. Are any of the pieces wonky or are they perfectly flat?

[Malor]

My experience with PCs has usually been that similar symptoms means something is shorted. I'm inclined to think that it's not heat, because you should get fine results for the first few minutes, and then see failures as the air heats up inside.

Because these boards are handmade, one real possibility is a cold solder joint somewhere that gets just enough pressure to go wonky when it's mounted in the case. If that's the cause, then the problem will probably get worse over time as the joint degrades from board flex. It might be worth the time to inspect the peripheral boards very carefully with a magnifying glass, looking for dull and/or bumpy solder joints. They should be shiny and smooth if they've taken properly.

NVidia cards used to have a similar problem, and you could sometimes resurrect them for awhile by putting them in a 400F oven for ten minutes, which melted and reflowed troublesome solder joints. I'm not sure if that's safe with Mister peripherals, though, and I would *definitely* not do that with the DE-10 itself.

If you see questionable solder points, doing manual reflow with a soldering iron would be more difficult, but much safer.... an oven reflow is a shotgun approach that could easily wreck the board.

[annette]

Direction of IO board fan airflow will have no effect at all and is not a cause of the rebooting.

I think you have a faulty board maybe bad solder joint or short out made by force once fitting of the case, contact ultimate mister seller this a hardware fault and not correct operation.

[ElMariachi003]

Well, I have not had any issues with my “naked” MiSTer since my last post - as a matter of fact, I had it on for a little over 2 hours straight today as I finally got a stereo Y adapter for the audio jack to test on my CRT with sound.

I will take the above into account, however, so I will check the IO board thoroughly this week. The boards themselves are flat, as the pin connections are all fully seated on both sides and the standoffs on either side of the IO board are fully tightened.

As someone new to the MiSTer community, I am extremely grateful for all the responses in this post. Here’s hoping I can get this sorted out soon so that I can put the MiSTer back in its case where it belongs, even though I’m enjoying the heck out of it as is.

[ElMariachi003]

I finally had a little time to experiment with mu MiSTer today. Long story short, I was able to use my MiSTer, "fully dressed", for close to two hours today! Of course the "bad news", if you can call it that, is that I honestly don't know why it's working, after two previous attempts to install the case.

Anyway, in order to determine some kind of process of elimination, I decided to install all four walls, but left the top off. I hooked it back up, fired it up, and was able to operate the MiSTer for about 20 minutes, no problem. I then carefully placed the top plate on, without connecting the JST connector for the mounted menu buttons. My primary motivation for this was that I wanted to see if closing off the top produced some kind of trapped heat situation. Again, the MiSTer continued to run for about another 20 minutes no problem, and I could even feel cool air blowing out of the slots on the front of the case. This is where I turned off the MiSTer, so I could carefully connect the plug for the top's buttons to the IO board, then screw on the top, being carful not to tighten the screws too much - just enough to be snug.

And that's it! so far so good. As I mentioned it above, I got to enjoy t for about 2 hours without any problems. I'll be sure to update if I do. Thanks again to everyone who pitched in with advice.

[ElMariachi003]

Well, it is a sad update, but after a few days with a few hours of play time each time, my MiSTer suddenly started with the reboot issue again.

I did a little troubleshooting, and after simply removing the top plate (and disconnecting The JST plug from the IO board (for the menu buttons soldered to the top plate), it started working again, where I played for a good while more. I suspect that maybe this top plate might have an issue, so I will try to contact Ultimate MiSTer about this.

[LamerDeluxe]

So you still haven't tried running it with the fan blowing air out of the case?

[rickdangerous]

Well, it is a sad update, but after a few days with a few hours of play time each time, my MiSTer suddenly started with the reboot issue again.

I did a little troubleshooting, and after simply removing the top plate (and disconnecting The JST plug from the IO board (for the menu buttons soldered to the top plate), it started working again, where I played for a good while more. I suspect that maybe this top plate might have an issue, so I will try to contact Ultimate MiSTer about this.

Send me PM with your info and I send you a top case replacement to despite the problem.
Try to close the case without connecting it. Let me know.

Kind regards
Ricardo

[Malor]

Well, it is a sad update, but after a few days with a few hours of play time each time, my MiSTer suddenly started with the reboot issue again.

I did a little troubleshooting, and after simply removing the top plate (and disconnecting The JST plug from the IO board (for the menu buttons soldered to the top plate), it started working again, where I played for a good while more. I suspect that maybe this top plate might have an issue, so I will try to contact Ultimate MiSTer about this.

Did you try putting the top plate back on again, without the buttons connected? If the machine keeps crashing, then it's probably a cold solder joint somewhere on one of the expansion boards that's cracking under pressure. If it stops crashing, then the problem is probably localized to the buttons or to the I/O board.

[ElMariachi003]

So you still haven't tried running it with the fan blowing air out of the case?

Actually, I didn't even get to that after trying the test below. So an airflow issue (at least in my case) was eliminated once I discovered that I could run as long as the Top was not connected to its JST connector on the IO board. As a matter of fact, I could feel cool air blowing out of the vents from the front panel of the case, even after over an hour of playing.

Well, it is a sad update, but after a few days with a few hours of play time each time, my MiSTer suddenly started with the reboot issue again.

I did a little troubleshooting, and after simply removing the top plate (and disconnecting The JST plug from the IO board (for the menu buttons soldered to the top plate), it started working again, where I played for a good while more. I suspect that maybe this top plate might have an issue, so I will try to contact Ultimate MiSTer about this.

Send me PM with your info and I send you a top case replacement to despite the problem.
Try to close the case without connecting it. Let me know.

Kind regards
Ricardo

Thank you Rick! I noticed that you had PM'd me, so I have replied. Yes, I did place it on top without connecting the JST, and I didn't have any issues. I've removed it again for now so that I could have access to the buttons on the IO board.

Did you try putting the top plate back on again, without the buttons connected? If the machine keeps crashing, then it's probably a cold solder joint somewhere on one of the expansion boards that's cracking under pressure. If it stops crashing, then the problem is probably localized to the buttons or to the I/O board.

Yes, that's exactly what I did, and as mentioned above, the problem went away. Rick had sent me an DM, which I just discovered and replied to.

Thank you for all the replies.

[ElMariachi003]

Just got my replacement Ultimate MiSTer Top plate in the mail today. Installed it and have had it running continuously for about an hour now. So far so good. I'll probably keep it on for the better part of the evening just to make sure all is good; I'll be sure to update this post with the news. Again, a big shout out/Thank you to RickDangerous for reaching out and sending me the replacement plate!

UPDATE: Was able to put my MiSTer through a 4 hour cycle, going through various cores, and of course playing for a while. No problem whatsoever! Will try to perform another cycle tonight for that long to see.