Ensoniq EPS-16+ Repair (stuck in test mode)

Pulling out your hair? Don't know what to do or where to go? Ask in here.
Forum rules
READ: VSE Board-Wide Rules and Guidelines

If your Help request has been solved, please edit your first post in order to select the Image Topic Icon to let others know your topic has been solved.
Shabazz
Newbie
Newbie
Posts: 11
Joined: Wed Apr 21, 2021 9:20 pm

Re: Ensoniq EPS-16+ Repair (stuck in test mode)

Post by Shabazz » Thu Apr 22, 2021 3:45 pm

belzrebuth wrote:
Thu Apr 22, 2021 2:36 pm
Also, check the 34pin ribbon for the floppy drive as they are not very reliable.
I've seen two of them fail in Ensoniq machines.
I just checked the fdd ribbon cable and there is no continuity in lines 1 and 33 but it might not be a problem since they are not only ones GND lines.

belzrebuth
Newbie
Newbie
Posts: 47
Joined: Tue Jul 15, 2014 12:43 pm

Re: Ensoniq EPS-16+ Repair (stuck in test mode)

Post by belzrebuth » Thu Apr 22, 2021 4:45 pm

Sorry for iterating what I've said above, I just saw you've put a new ribbon for the floppy.
But, could it be that the keyboard ribbon is bad?
Maybe replace that too as well, just in case?
The rack has a dummy KPC board installed too, so it may be necessary for the operation of the unit even just to pass thru serial data.
I would disconnect the keyboard assembly and test again just to rule out any kbd related issues.

belzrebuth
Newbie
Newbie
Posts: 47
Joined: Tue Jul 15, 2014 12:43 pm

Re: Ensoniq EPS-16+ Repair (stuck in test mode)

Post by belzrebuth » Thu Apr 22, 2021 4:48 pm

Shabazz wrote:
Thu Apr 22, 2021 3:45 pm

I just checked the fdd ribbon cable and there is no continuity in lines 1 and 33 but it might not be a problem since they are not only ones GND lines.
Replace it just in case or at least check that GND properly reaches the HxC.
Check all pins with the ribbon connected and test voltages too.
A bad ribbon can cause all sorts of problems...

A good thing about a regular floppy drive is that you can actually hear when it's doing what it's supposed to.
With the HxC you can't hear when/if a ready signal or a read signal reaches it.
So I would totally replace that cable especially since you found 2 wires that are already bad.
Consider replacing both of the ribbon cables (kbd and fdd) before moving on with the troubleshooting.

Shabazz
Newbie
Newbie
Posts: 11
Joined: Wed Apr 21, 2021 9:20 pm

Re: Ensoniq EPS-16+ Repair (stuck in test mode)

Post by Shabazz » Thu Apr 22, 2021 6:04 pm

belzrebuth wrote:
Thu Apr 22, 2021 4:45 pm
Sorry for iterating what I've said above, I just saw you've put a new ribbon for the floppy.
But, could it be that the keyboard ribbon is bad?
Maybe replace that too as well, just in case?
The rack has a dummy KPC board installed too, so it may be necessary for the operation of the unit even just to pass thru serial data.
I would disconnect the keyboard assembly and test again just to rule out any kbd related issues.
I put a new kpc 20 pin ribbon cable first so it's new. Sure the fdd ribbon cable needs to be replaced too. The kbd tuning process is always smooth and I don't think it will pass without the kbd attached to the kpc.
It would be nice to know how a good waveforms on the MOSI and MISO lines are look like. Unfortunately I don't have working unit around.

belzrebuth
Newbie
Newbie
Posts: 47
Joined: Tue Jul 15, 2014 12:43 pm

Re: Ensoniq EPS-16+ Repair (stuck in test mode)

Post by belzrebuth » Thu Apr 22, 2021 7:47 pm

Unfortunately mine is racked up at the moment and it's a bit of a chore to screw it out and open it up but if things get really weird I wouldn't mind a lot to do that.
If I remember correctly from the unit I repaired thru this thread the MISO and MOSI pins didn't have continuous activity unless something was happening (a keypress for example)
There's a guide floating around this forum written by Rasputin which describes what you should be getting out of a working UART but it may vary from model to model.
edit:
Here it is.
See about halfway of the thread for UART pin description and normal behavior.
viewtopic.php?t=96381

A keyless unit may also have different data floating around when idle but I'm not sure about that.
My guess is that there's a communication error so it may very well be a cable or a connector.
If these are okay you could try probing the immediate logic ICs involved, hex inverters,decoders, multiplexers etc between peripherals such as the floppy and maybe the keyboard itself.

If the machine does not start loading an OS disk I'd certainly start with that very issue.
Replace that ribbon asap and try probing the main reset line to see if it changes state after power on and whatever signals are required for the floppy immediately after the machine starts up.
You most probably have a working CPU, EPROM and UART so I would suspect a communication error which causes a halt of some sort.
I've never heard/seen/read about RAM failing on Ensoniq stuff.

Rasputin
Senior Member
Senior Member
Posts: 821
Joined: Thu Jul 21, 2016 4:52 am

Re: Ensoniq EPS-16+ Repair (stuck in test mode)

Post by Rasputin » Fri Apr 23, 2021 12:35 am

Shabazz wrote:
Thu Apr 22, 2021 3:07 pm
The problem might be in communication between the display board and the KPC board. I could be wrong, but i think they talk to each other using MISO and MOSI lines. I cant see anything meaningful on MISO line, just constantly repeating two blips. Nothing changes when i pressing buttons. The MOSI line is dead silent. Could it be the NEC 7538 chip on the Display Board? But as mention before the DB passed the self test.
I didn't have time to thoroughly read the post, but your thinking is mostly correct--except that the display board self-test only determines if the display itself works and that the keyscan matrix is working. It will not tell you if the display processor is sending serial information anywhere. The display comes out of self-test mode when it receives serial traffic from the DUART, so you know the display board can receive serial input, and you know the key matrix works, but we can't tell if the button presses are being relayed.

Display board upd7538 (U3) pin 39 is the serial output. If no signal is being generated at this pin when you push a panel button then that is almost definitely your issue--unfortunately, that is probably the case. If there is traffic on that pin when you press panel buttons then check 74LS14 (U5) pins 11 and 10--they should both show activity when panel buttons are pressed. If pin 11 shows activity, but not pin 10 then replace U5.

If U5 pin 10 shows activity when front panel buttons are pressed then your issue lies on the KPC, most likely.

Shabazz
Newbie
Newbie
Posts: 11
Joined: Wed Apr 21, 2021 9:20 pm

Re: Ensoniq EPS-16+ Repair (stuck in test mode)

Post by Shabazz » Fri Apr 23, 2021 11:37 am

Rasputin wrote:
Fri Apr 23, 2021 12:35 am

Display board upd7538 (U3) pin 39 is the serial output. If no signal is being generated at this pin when you push a panel button then that is almost definitely your issue--unfortunately, that is probably the case. If there is traffic on that pin when you press panel buttons then check 74LS14 (U5) pins 11 and 10--they should both show activity when panel buttons are pressed. If pin 11 shows activity, but not pin 10 then replace U5.

If U5 pin 10 shows activity when front panel buttons are pressed then your issue lies on the KPC, most likely.

Thank you for your insight!

I just probed U3 and U5. That`s what i get:

U3 pin 39 shows some traffic when nothing being pressed and it changes when pressing panel buttons.
U5 pin 11 shows activity, but pin 10 is silent. I am gonna replace U5 and come back with update.

Once again thank you!

Shabazz
Newbie
Newbie
Posts: 11
Joined: Wed Apr 21, 2021 9:20 pm

Re: Ensoniq EPS-16+ Repair (stuck in test mode)

Post by Shabazz » Wed Apr 28, 2021 9:03 am

Hi all!

I came back with an update.

Just replaced U5 and signal on pin 10 is still low (about 100mV). Then i cut MOSI track to pin2 on J2 connector to see if there is something pulling the level on U5 pin 10 low. I was right. Without connection with the KPC board i can see traffic both on the U5s pin 11 and 10. BTW there is traffic even if no keys being pressed. I don`t know if is it ok.

So there is something pulling pin10 voltage low and i think it`s the KPC related problem. Unfortunately i don't have keyboard version KPC schematic, only for the rack version. And they are quite different.

Probably i should start to seek fo spare KPC board.

Rasputin
Senior Member
Senior Member
Posts: 821
Joined: Thu Jul 21, 2016 4:52 am

Re: Ensoniq EPS-16+ Repair (stuck in test mode)

Post by Rasputin » Thu Apr 29, 2021 12:35 am

Shabazz wrote:
Wed Apr 28, 2021 9:03 am
So there is something pulling pin10 voltage low and i think it`s the KPC related problem. Unfortunately i don't have keyboard version KPC schematic, only for the rack version. And they are quite different.
See if this is helpful: http://www.tauntek.com/ASR10highcoilsch.pdf

Just like you looked at and replaced the U5 on the display board, I think there's another similar chip on the KPC which is also used as a schmitt trigger for the serial comms.

Shabazz
Newbie
Newbie
Posts: 11
Joined: Wed Apr 21, 2021 9:20 pm

Re: Ensoniq EPS-16+ Repair (stuck in test mode)

Post by Shabazz » Thu Apr 29, 2021 10:58 am

Thanks for your reply

Unfortunately that schematic doesn't really help. There is no schmitts on the KPC board and the MOSI line goes straight from the U5 pin 10 to the pin 45 of MC68HC11 micro on the KPC board. Should i try to replace that micro? Don`t even know is it possible to find.

Rasputin
Senior Member
Senior Member
Posts: 821
Joined: Thu Jul 21, 2016 4:52 am

Re: Ensoniq EPS-16+ Repair (stuck in test mode)

Post by Rasputin » Thu Apr 29, 2021 9:11 pm

Shabazz wrote:
Wed Apr 28, 2021 9:03 am
Then i cut MOSI track to pin2 on J2 connector to see if there is something pulling the level on U5 pin 10 low. I was right.
The KPC IC (MC68HC11) which recieves MOSI is socketed. What happens if you just pull that chip? Is MOSI no longer pulled down?

What I'm getting at: Is there an internal short within that chip? Because that would obviously necessitate replacement (which is going to be hard to find), but if a short lies elsewhere--such as the socket itself--then that would be nicer.

Locating the short is the first order of business.

Shabazz
Newbie
Newbie
Posts: 11
Joined: Wed Apr 21, 2021 9:20 pm

Re: Ensoniq EPS-16+ Repair (stuck in test mode)

Post by Shabazz » Mon May 03, 2021 11:51 am

Rasputin wrote:
Thu Apr 29, 2021 9:11 pm

The KPC IC (MC68HC11) which recieves MOSI is socketed. What happens if you just pull that chip? Is MOSI no longer pulled down?

What I'm getting at: Is there an internal short within that chip? Because that would obviously necessitate replacement (which is going to be hard to find), but if a short lies elsewhere--such as the socket itself--then that would be nicer.

Locating the short is the first order of business.
Hi all!

A brief update.

I pulled The KPC IC (MC68HC11) and the MOSI is still low. Than i started to trace the MOSI line from the 20pin connector and as it turns out there is 1N914B diode which was failed in the open position. I have no replacement diode and the stores are all closed today. So i`ll get this diode tomorrow and hope that helps. Fingers crossed!

belzrebuth
Newbie
Newbie
Posts: 47
Joined: Tue Jul 15, 2014 12:43 pm

Re: Ensoniq EPS-16+ Repair (stuck in test mode)

Post by belzrebuth » Mon May 03, 2021 8:28 pm

Really hoping this fixes it. :D
You deserve a working machine after all the trouble you've been through.
Good luck!

Shabazz
Newbie
Newbie
Posts: 11
Joined: Wed Apr 21, 2021 9:20 pm

Re: Ensoniq EPS-16+ Repair (stuck in test mode)

Post by Shabazz » Tue May 04, 2021 5:57 pm

belzrebuth wrote:
Mon May 03, 2021 8:28 pm
Really hoping this fixes it. :D
You deserve a working machine after all the trouble you've been through.
Good luck!
Thanks!

Unfortunately the last fix didn`t help much.. All the signal levels are fine now but i don`t really like its waveform shape. You can find oscillograms in my dropbox folder https://www.dropbox.com/sh/gm0dt1rv9pxw ... B_28a?dl=0

First i thought it was my cheap oscillograph that needs calibration, but BCK signal from the U5 looks just fine. I guess that'd be the reason for communication breakdown between the display board and the KPC.

What do you think?

belzrebuth
Newbie
Newbie
Posts: 47
Joined: Tue Jul 15, 2014 12:43 pm

Re: Ensoniq EPS-16+ Repair (stuck in test mode)

Post by belzrebuth » Tue May 04, 2021 6:08 pm

I think that either the a capacitive load of some sort has been "introduced" to the schmitt by a short maybe
or there's some slightly excessive current draw/someone else's input resistance has lowered somehow.
The thing is that you have a good rising edge but the rest of the pulse decays at the same rate so since you replaced that chip I would check what's immediately after.
If it's an IC just cut its leg to a convenient spot and see if that restores the signal.
Looks like the schmitt trigger is somewhat struggling in there except if it's a rare case that this signal is indeed correct.
It's not so common for diodes to fail open so I would probably check what's immediately after the schmitt first.
I really need to have a look at the schematic as I talk out of my head at the moment...

Post Reply