Ensoniq EPS Classic Freezes after 30 minutes

[Hyde]

Title says it all. I reboot when it freezes but, still won't do anything after OS loads. I let it cool down & it works fine again. Anyone know if recapping that power supply will fix this? Should i look for something else? I'd love to be able to rely on this machine a bit more.

[Rasputin]

Rebuilding the power supply to make it a bit more robust probably isn't going to do any harm, unless you're bad with at soldering. Checking the PSU with a multimeter and your 5 senses is always a good place to start, though.

The regulators are more what I'd suspect than the caps, so make sure they have a good path to sink the heat. A bad diode in the rectifier bridge could add some unwanted chaos to the system too.

[Hyde]

Rebuilding the power supply to make it a bit more robust probably isn't going to do any harm, unless you're bad with at soldering. Checking the PSU with a multimeter and your 5 senses is always a good place to start, though.

The regulators are more what I'd suspect than the caps, so make sure they have a good path to sink the heat. A bad diode in the rectifier bridge could add some unwanted chaos to the system too.

OK, Thanks. I'll have a look/smell/hear/feel/esp at it.

[Rasputin]

ESP... good call. Way better choice of the fifth sense than taste. Putting your tongue on the PSU could work out pretty poorly. Excellent point.

[Hyde]

So, I'm wondering... I had put this keyboard aside, for the time being. Was working on an issue with the other EPS(very low output on one side). Reading through the repair manual I came across a paragraph which reads "Units with a serial number higher than 16582 must use OS 2.35 or higher." The keyboards came with 2.20. The keyboard with the freezing issue is way above this serial number. the other unit, well below. I'm wondering if this would have any effect on the issue of shutting down? 2.49 is on its way. Unfortunately no one has a floppy drive anymore. So, I had to buy the OS disk & have it shipped

If you have any suggestions on the low output issue on the older machine, I'd love to hear them, as well. In the headphone out put, I only get sound on one side. With both outputs in use, I only get one side, as well. I've reset connections & chips. There is a very faint signal, Not useable.

[Hyde]

Update:
New software=Same issue

I have two units so, I decided to swap out power supply with one that is known to be good.=Same issue.

So, I know the issue is not power supply. Thinking of exchanging some of the removable chips to see if anything changes.

[Rasputin]

Update:
New software=Same issue

I have two units so, I decided to swap out power supply with one that is known to be good.=Same issue.

So, I know the issue is not power supply. Thinking of exchanging some of the removable chips to see if anything changes.

There aren't that many removable chips that would pertain to just one channel going bad, so I would think about what you're trying to achieve here. Firstly, I would identify which exact channel is faint. Is it left or right?

You also know that the signal is going bad at a common point before the signal gets split to the outputs and the headphone amp. This will help you narrow it down.

I would advise you to look at Transoniq Hacker, issue #39 (September 1988). There is an article by Mr. D. Lord called "Inside the EPS" that will point you in the right direction.

http://www.buchty.net/~buchty/ensoniq/t ... DF/039.pdf

Hint: Page 10, figure #2 will tell you just about everything you need to know to troubleshoot your problem. See those op-amps right before the output jacks? Are they tied into the headphone amp? No? Then take another step back toward the DOC chip and look there, and so on.

[Hyde]

Update:
New software=Same issue

I have two units so, I decided to swap out power supply with one that is known to be good.=Same issue.

So, I know the issue is not power supply. Thinking of exchanging some of the removable chips to see if anything changes.

There aren't that many removable chips that would pertain to just one channel going bad, so I would think about what you're trying to achieve here. Firstly, I would identify which exact channel is faint. Is it left or right?

You also know that the signal is going bad at a common point before the signal gets split to the outputs and the headphone amp. This will help you narrow it down.

I would advise you to look at Transoniq Hacker, issue #39 (September 1988). There is an article by Mr. D. Lord called "Inside the EPS" that will point you in the right direction.

http://www.buchty.net/~buchty/ensoniq/t ... DF/039.pdf

Hint: Page 10, figure #2 will tell you just about everything you need to know to troubleshoot your problem. See those op-amps right before the output jacks? Are they tied into the headphone amp? No? Then take another step back toward the DOC chip and look there, and so on.

Great info! Thank you! Definitely will read.

The post was pointing to the machine that is locking up after 1/2 hour, though. Sorry if that wasn't clear. If I can clear up either of these issues I'll be in a good spot. To clear up both would be really great. Would love to have both machines up & running.

[Rasputin]

I have two units so, I decided to swap out power supply with one that is known to be good.=Same issue.

So, I know the issue is not power supply. Thinking of exchanging some of the removable chips to see if anything changes.

The post was pointing to the machine that is locking up after 1/2 hour, though. Sorry if that wasn't clear. If I can clear up either of these issues I'll be in a good spot. To clear up both would be really great. Would love to have both machines up & running.

Ah, I see. I guess that's the danger in mixing multiple issues in a single thread. My thought is this though: Look at the symptoms of what you're trying to fix. You've stated the symptoms are not only related to thermal changes, but overheating specifically and it is reproducible and predicable behavior within a known time window, yes?

(Keep in mind that I don't have access to an original EPS, so I can't troubleshoot with any authority...)

While the PSU proper may not be the issue, it still may be power related as I believe the EPS has power regulators on the mainboard. This is the EPS keyboard and not the EPSm rack, right? Because the keyboard module itself also has a separate regulator, based upon what I know about other similar Ensoniq gear.

I believe there is an unregulated line coming from the PSU that makes its way to the keyboard. If the keyboard processor voltage regulator is shut down then the KPC will fail and also kill any serial traffic -- leaving the EPS powered on and booted, but in a semi-frozen state.

Can the EPS receive MIDI traffic while in this frozen state? If it will play back MIDI in this frozen state then you know it's probably the KPC.

If it can play back MIDI successfully then we can validate our assumption: The display board may seem locked if the KPC is down but that's actually a misleading symptom. The way to test the display/keypad being at fault is to get the EPS to the "frozen" state and then (in a timely manner) turn off the EPS, disconnect the display board from the mainboard, and turn it back on. If the display is blank but responds to pressing the keypad by displaying characters then you've got a strong indicator of a locked KPC.

Edit: Although this test is no longer applicable to this thread, it needs to be clarified that just pulling the display/keypad header isn't what I meant. There is a method in the service manual which outlines how to put the display into self-test mode which essentially cuts off communication between the display and mainboard yet still allows the display/keypad to power on.

[Hyde]

I have two units so, I decided to swap out power supply with one that is known to be good.=Same issue.

So, I know the issue is not power supply. Thinking of exchanging some of the removable chips to see if anything changes.

The post was pointing to the machine that is locking up after 1/2 hour, though. Sorry if that wasn't clear. If I can clear up either of these issues I'll be in a good spot. To clear up both would be really great. Would love to have both machines up & running.

Ah, I see. I guess that's the danger in mixing multiple issues in a single thread. My thought is this though: Look at the symptoms of what you're trying to fix. You've stated the symptoms are not only related to thermal changes, but overheating specifically and it is reproducible and predicable behavior within a known time window, yes?

(Keep in mind that I don't have access to an original EPS, so I can't troubleshoot with any authority...)

While the PSU proper may not be the issue, it still may be power related as I believe the EPS has power regulators on the mainboard. This is the EPS keyboard and not the EPSm rack, right? Because the keyboard module itself also has a separate regulator, based upon what I know about other similar Ensoniq gear.

I believe there is an unregulated line coming from the PSU that makes its way to the keyboard. If the keyboard processor voltage regulator is shut down then the KPC will fail and also kill any serial traffic -- leaving the EPS powered on and booted, but in a semi-frozen state.

Can the EPS receive MIDI traffic while in this frozen state? If it will play back MIDI in this frozen state then you know it's probably the KPC.

If it can play back MIDI successfully then we can validate our assumption: The display board may seem locked if the KPC is down but that's actually a misleading symptom. The way to test the display/keypad being at fault is to get the EPS to the "frozen" state and then (in a timely manner) turn off the EPS, disconnect the display board from the mainboard, and turn it back on. If the display is blank but responds to pressing the keypad by displaying characters then you've got a strong indicator of a locked KPC.

I decided to take a different approach. Yes, both units are keyboards. I'll stick to the "Freezing" issue, for the moment.
SO, I swapped out the keyboards, last night, cable & all. Oddly enough, after about the same amount of time I got a couple error msgs. "ERROR 131 Reboot" & "Bad Fat Signature". No freezing now, though. The reboot happens automatically & it happens just about every time I try to perform a disk function. If I shut down the unit, no improvement. Thusly, I realized that all freezing, previously, also happened when I attempted to do a disk function. Of course, these symptoms all begin after about 30 minutes of operation. This morning, I had the machine up & running for a little over an hour with no issue. Not sure what to make of that? Intermittent issue, I presume? Maybe it is just a different reaction to whatever the problem is.

The other curious result is that the other keyboard (now with the keyboard assembly from the problem unit) is functioning normally. I did reseat the chips on the keyboard assembly. Maybe that helped?

[Rasputin]

I swapped out the keyboards, last night, cable & all. Oddly enough, after about the same amount of time I got a couple error msgs. "ERROR 131 Reboot" & "Bad Fat Signature". No freezing now, though. The reboot happens automatically & it happens just about every time I try to perform a disk function. If I shut down the unit, no improvement. Thusly, I realized that all freezing, previously, also happened when I attempted to do a disk function. Of course, these symptoms all begin after about 30 minutes of operation. This morning, I had the machine up & running for a little over an hour with no issue. Not sure what to make of that? Intermittent issue, I presume? Maybe it is just a different reaction to whatever the problem is.

The other curious result is that the other keyboard (now with the keyboard assembly from the problem unit) is functioning normally. I did reseat the chips on the keyboard assembly. Maybe that helped?

So one is totally fixed now, and the other only exhibits issues upon accessing the disk drive?

I'd check for continuity on the keyboard ribbon cable, and also give the header pins a good polish. When checking continuity I would move the ribbon cable in a couple different positions (bent, unbent, etc.) and double-check if ALL lines are conducting in EVERY position of the cable.

A remote possibility is the positioning of the ribbon cables. Occasionally, if you overlay a ribbon cable over a chip with high EMI emission or sensitivity then you'll get weird errors. Sometimes making the ribbon cables take a different path over the mainboard can make a difference.

"Bad FAT signature" does seem to indicate a problem with disk access, but there is so much interplay between each section in Ensoniq gear that error messages are often misleading.

Of course, swapping the floppy cable, floppy drive, and even different disks is a logical thing to do. Just be careful with removing ribbon cables all the time. Make sure to grab by the plug and not by the ribbon. I know that seems really obvious, but ribbon cables absolutely do matter and can easily go bad from being manipulated incorrectly! I've seen a lot of flaky ribbon cables in Ensoniq stuff, so it absolutely happens.

[Hyde]

Just focusing on one keyboard right now. The unit with the freezing up issue. It displayed the errors last night, after keyboard swap & seemed ok, this morning. The disk drive, cable, or maybe even control chip for disk drive, were on my radar. This is obviously a laymen's theory... & is why I'm asking for a more learned opinion. (Yes, I'm pulling from the plugs. I have a chip puller & its grabs nicely on those plugs I can't pull by hand.

The other keyboard still seems to be ok, even with the swapped in keyboard from the freezing unit. It has the output issue on the left side. but, I'm not concerned with that one at the moment. I'm focusing on the unit with the freezing problem first.

Back to the unit that was freezing. I will run it for a bit to see if ERRORs & Bad Fat Signature promp come up again. Do you think the disk control chip may have any part in my issues? Or should I focus on the disk drive & cable first?

Thank you for your help, Hyde

[Rasputin]

The other keyboard still seems to be ok, even with the swapped in keyboard from the freezing unit. It has the output issue on the left side. but, I'm not concerned with that one at the moment. I'm focusing on the unit with the freezing problem first.

Back to the unit that was freezing. I will run it for a bit to see if ERRORs & Bad Fat Signature promp come up again. Do you think the disk control chip may have any part in my issues? Or should I focus on the disk drive & cable first

Ah, I thought you meant that somehow swapping the keyboard fixed the output issue which didn't make any sense to me.

I would suspect the drive and cable before I would suspect the controller. If disk loading works even intermittently then the FDC is probably fine. Those would usually fail in a permanent way where a pin would be stuck or floating -- it wouldn't come and go.

Easy test is to take the drive and cable from the bad output EPS and (using the exact same disks) see if you can replicate the problem. If the problem is still there then look at more esoteric problems.

[Hyde]

The other keyboard still seems to be ok, even with the swapped in keyboard from the freezing unit. It has the output issue on the left side. but, I'm not concerned with that one at the moment. I'm focusing on the unit with the freezing problem first.

Back to the unit that was freezing. I will run it for a bit to see if ERRORs & Bad Fat Signature promp come up again. Do you think the disk control chip may have any part in my issues? Or should I focus on the disk drive & cable first

Ah, I thought you meant that somehow swapping the keyboard fixed the output issue which didn't make any sense to me.

I would suspect the drive and cable before I would suspect the controller. If disk loading works even intermittently then the FDC is probably fine. Those would usually fail in a permanent way where a pin would be stuck or floating -- it wouldn't come and go.

Easy test is to take the drive and cable from the bad output EPS and (using the exact same disks) see if you can replicate the problem. If the problem is still there then look at more esoteric problems.

I'll give that a shot next. Again, thanks!

[Hyde]

I've run the machine a few times now, with success. No freezing. I pulled the disk drive out, reset any connectors inside the drive which I could access. The drive looked clean so, I didn't bother trying to clean it further. I reset the cable connections to & from the drive. I'm always a little apprehensive to give it the complete thumbs up. So I'll check back in when I get some serious time with the machine. Need to find a 4x expander & a SCSI card