Ensoniq ESQ-1 debugging help

[dnigrin]

Another friend mentioned to me that having access to a digital storage oscilloscope would help me in the debugging process. If I got one, what would be the right next steps to continue the debugging?

[dnigrin]

OK, long time since last update. I *did* secure a digital storage scope, and so now revisiting things, since i can get a much better idea of what's going on. I first revisited the DUART activity, on Pin 11, Serial TXB. Here's what I see on power on. Does this seem like expected output?

[dnigrin]

What's also unusual - I mentioned very early in the thread that I actually have *two* nonfunctional ESQ-1's. I checked pin11 on the DUART on the other board (which does NOT go into self-test mode - I get no output on the LCD with button pushes), and its output looks like this - which one seems more appropriate?

[dnigrin]

OK, another update for those following along. @Rasputin, hope you're still out there?

I was able to get my hands on a working ESQ-1:

* I swapped its ROM chips into the non-working ESQ: no change.
* I swapped its CPU into the non-working ESQ: no change.
* I put its entire main board into the non-working ESQ: ESQ works perfectly (ruling out any power supply or display board issues).
* Since the DUART in the working ESQ is not socketed, I have been limited to only swapping the DUART from my *other* non-working ESQ into the one I'm working on: no change. Although theoretically, both DUARTs could be bad, I'm thinking that that's not likely.

So.... where to look next? Is it those 74F139 chips? I do now have my digital storage scope so I can be a lot more effective with my debugging.... Quoting something said on page 1 of this thread, how do I look into this?

If the DUART Chip Select looks suspect in comparison to the other Chip Selects (such as OSLOW, etc.) then that could be a clue. A couple 74F139 chips interpret the address bus and select/deselect which chip is allowed to talk. If there's a problem in that area then the DUART won't ever be allowed to get the init string from the CPU to send to the display.

Thanks for any/all help on this!

[Rasputin]

OK, long time since last update. I *did* secure a digital storage scope, and so now revisiting things, since i can get a much better idea of what's going on. I first revisited the DUART activity, on Pin 11, Serial TXB. Here's what I see on power on. Does this seem like expected output?

The first picture which looks really erratic is problematic. The second one which has two distinct and straight lines at only two different levels is what should be proper.

[dnigrin]

Thanks, will re-look at this, including moving the DUART from the other non-functioning board (with the correct output on the scope), into the first one, and see what happens with its output when in that synth.

[dnigrin]

Confused - see below. All of these are immediately after power-on:

Non-functioning ESQ-1, with original DUART, pin 11 (TXB)


Non-functioning ESQ-1, with DUART from other non-functioning synth, pin 11 (TXB)


Working ESQ-1, DUART pin 11 (TXB)


Not sure why I'm not getting that "clean" two levels, high/low kind of image.... Perhaps I'm not using the scope right or something? I'm definitely still learning it....

[dnigrin]

I think that I'm to blame for the last few posts, primarily because I'm still learning how to use my digital scope - I apologize for those following along. I realize that I was zooming in too far on the waveforms - the scale on some of the previous readings was at the microsecond level (50-100 microseconds per scope division), and I was interpreting transient voltage changes right at power on. Zooming further back out, I now see what seems to be more appropriate readings (at the 100 millisecond per scope division level).

First, from the working ESQ-1. At power on, again from the DUART, pin 11 (TXB). You can clearly see the point at which the pin starts sending data out to the display board:


Now, from the non-working ESQ-1, DUART, pin 11 (TXB). In this image, the overview is at the top, and at the bottom I've zoomed in to an area where I saw some activity - but clearly not enough to be considered a legitimate "pulse" I think, and definitely not the same as on the working ESQ-1 image above:


This is all to be expected, as if I got the exact same output from that pin, I would have expected the display board to go out of self test.

So now I think I'm ready to start to march my way backwards to debug this methodically, and with the right tools to do the job. As a reminder, I've ruled out the CPU and the EPROMs as the source of the problem, as swapping them with the working ESQ components yielded no change. So I think I've narrowed the problem down to:

• DUART

• 7665 chip (U24) - sends RESET to CPU and DUART

• 74F139 chip (U19) - sends Chip Enable to DUART

Are these correct conclusions? Anything else I should consider? I should also add that I've re-confirmed that clock and power are getting to both the CPU and the DUART as expected.

Thanks - I really appreciate the assistance as I continue to learn!

[Rasputin]

Have you scoped the working DUART chip select pin and compared it to the non-working?

[dnigrin]

Yes. Seem similar to my (untrained) eye?

Working ESQ-1 DUART chip select (pin 35):


Non-working ESQ-1 DUART chip select (pin 35):

[Rasputin]

I would personally socket the DUART and replace it. The part isn't expensive, and there's a pretty good chance that's the issue, in my opinion.

If your desoldering skills aren't a liability, you could swap U19 (74F19) and U13 (74F19) to test them, but I don't think that's the issue.

[dnigrin]

Thanks - I will likely try both of those things, in the order you outlined as well. For my understanding though - is there something you saw in the scope tracings I posted that has you suspicious of the DUART?

[papashark]

Any update on your ESQ-1 debug? I got one non-working for cheap and is in the same state, display self test mode only. Will follow your steps so far in debugging mine, thanks for documenting all this! Seems these mainboards are more prone to issues than many other vintage keyboards I've worked on...

[dnigrin]

Any update on your ESQ-1 debug?

Unfortunately no - it's still sitting on my bench, waiting for me to take a fresh look at it all again. I decided to work on some other projects for a bit and return to this one with fresh eyes and energy in a bit of time....

FWIW, I did try replacing the DUART as Rasputin suggested above - no change. And I also tried the 2nd step he suggested - again no change.

Good luck with yours!

[dnigrin]

Old thread I originally started, that I'm returning to after a break and with renewed energy. Also importantly, I now know my digital scope far better than I did before!

To recap - I have a plastic case non-functioning Ensoniq ESQ-1, along with a fully working one to compare with. The bad unit goes into self-test mode at startup.

• Swapping the entire main board from the good one to the bad one restores full functionality to the bad one.

• Swapping just the CPU or just the EPROMs or just the DOC from the good to the bad does nothing.

• Replaced the DUART in the bad one yields no change.

I've decided to compare multiple pins at startup on several components on both good and bad devices to see the differences. See below. I'm hoping that @Rasputin or @Meatballfulton you're still out there, or that someone new could help point me to where the problem might be! Clearly, something doesn't look right with OSLO, but I'm suspicious of problems further upstream in the startup sequence, as I see differences there too (for example, even the 7665 that generates the RESETs has differences).

For ease of reference, here's the schematic I'm using: https://defectiverecords.com/temp/esq_1 ... d_RevH.jpg

CPU (U10)
Ch 1. Voltage (Vcc, pin 7)
Ch 2. Reset (/RESET, pin 37, active low)
Ch 3. Clock (Q, pin 35)
Ch 4. D0 (pin 31)

Good


Bad


===================================
EPROM (U20, OSLO)
Ch 1. Voltage (Vcc, pin 28)
Ch 2. Chip Select (/CE, pin 20, active low)
Ch 3. D0 (pin 11)
Ch 4. D1 (pin 12)

Good


Bad


===================================
DUART (U6)
Ch 1. Voltage (Vcc, pin 40)
Ch 2. Reset (pin 34)
Ch 3. Chip Select (/CS, pin 35, active low)
Ch 4. Serial TxB (pin 11)

Good


Bad


===================================
7665 (U24, sends Reset to CPU and DUART)
Ch 1. Reset (/RESET, pin 7, active low)

Good


Bad


===================================
74F139 (U19, sends Chip Select to DUART)
Ch 1. Chip Select (/DUART, pin 5, active low)

Good


Bad


===================================
74F139 (U15, sends Chip Select to OSLO)
Ch 1. Chip Select (/OSLO, pin 5, active low)

Good


Bad