Hi guys I have a JEN SX-1000 which currently makes no sound what so ever. Here is the story so far:

Bloke I got it from said he replaced the M110 but still no sound...

I got it home and yep no sound but powers up.

I opened it up to find A LOT of the ribbon cables were plugged in back to front!!
I re-plugged all these the correct way round, powered it up and checked that the PSU is ok with a volt meter. All good

When powered now, no audio output BUT you can faintly hear in the background an oscillator which changes pitch when you move the tuning knob on the front.
This is free running, I can't quite remember if pressing the keys changes pitch too as it's been a while since I last turned it on.

This is about as far as I have got. I have access to an oscilloscope and I also have a volt/ohm meter. My electronics isn't very good any more and can barely remember my Logic as I haven't done anything like this in years.
Some one suggested to start looking at the M110 first with the oscilloscope but I'm not sure what to look out for

Can any one help out?

Thanks

Edited by vinyl_junkie (17/03/12 01:40 PM)

Don't know this particular synth, but would like to help if I can. Do you have a schematic that you can post up somewhere. Given that, I can find my way around everything - well, nearly everything!

--------------------
It wasn't me!
(Well, actually, it probably was)

Hi and thanks.

Yep the schematics are here: Jen Schematics

And the data sheet for the M110 is here:

http://www.milton.arachsys.com/nj71/pdf/m110.pdf

The M110 Is the brain I think as it produces the waveforms, detects the keyboard triggering and also triggers the env's etc

Cheers

Martin

Edited by vinyl_junkie (17/03/12 04:09 PM)

If that single page jpeg is too big here it is broken down in the relevant parts to make it a bit easier.

Master oscillator, voice generator, PSU

Waveform generator, LFO, glide osc.

Filter, envelope generators

Filter (continued), noise generator, VCA

Edited by vinyl_junkie (17/03/12 04:14 PM)

Also from another website I found re the M110

"The main issue is the half-broken M110 master oscillator. I say half-broken, because the keyboard scanning and trigger/gate works, but its the oscillator outputs that seem dead (this might have something to do with the range switch being a make-before-break, so shorting together two outputs momentarily...something else to replace).
Since the oscillator also resets the sawtooth at each cycle the sawtooth output does not work."

Righty Ho {rolls up sleeves}

First of all, is that last comment a general reference to M110 problems, or to the specific synth that you have?

I'm very surprised if the switch is MMB, there is nothing about the circuit that would require it, and BBM (break before make) is far more common. Having said that, the outputs seem to be what is known as open collector, as there is a 47k pull-up resistor (to +12V) in which case shorting them would do no harm anyway.

If you are getting correct keyboard scanning and gate operation that suggests the M110 is operational. First place to check with the scope is on the wiper of that octave selector switch. CMOS chips can sometimes go S/C on their inputs, and there is one hanging on that octave O/P.

If you get nothing there, temporarily tack a 47k resistor between one of the other M110 octave outputs and +12V then switch on and see of you get something on that one (thus ignoring the switch and what follows).

However, if you get a nice square wave on the octave selector, move to the waveform selector, (switching it to square wave) and see if you get anything on the waveform level pot.

Let us know how you get on

P.S.
Set the pot to half way, but measure at the hot end (so any possible later faults don't kill the signal).

--------------------
It wasn't me!
(Well, actually, it probably was)

Edited by Folderol (17/03/12 05:38 PM)

The last comment would be general problems relating to M110 issues.

Right so far I have checked with the scope pins 24-21 of the M110 which appear ok.
Square waves of varying footages as expected and turning the master tune knob affects pitch.

I think the keyboard scanning/triggering works, still need to confirm this.
Will have a gander at the octave select switch in a mo and try what you have said

Edited by vinyl_junkie (17/03/12 06:25 PM)

Ok mate here is what I've done so far.

The Octave select switch appears to be ok, I have checked the input of the 4013 IC which is beeing fed by the wiper of the switch right? (pin 11) and you get a nice healthy square which changes when you move the switch.
I can also confirm the gate works ok; It's high when no keys are pressed and drops down when you press a key, only goes back up when you let go of the key.
The Trig is a bit more tricky but I think it works; again it's high when no keys are pressed and when you press a key you get a momentary dip.

Right where to now

Oh and thank you for your time!

On the other hand I think there is something wrong with the waveform selector switch...

Also I don't get one thing;

You have the octave selector switch and that's ok, the output of that goes into the CD-4013
BUT there is a dotted line taken from the wiper that also goes to what looks like another switch? (I think the switch is duplicated just for the sake of making the diagram easyer as there is only one octave range switch)

So the signal is also being taped off to various resistors feeding the 301 IC's (OP Amps), I think there is a problem around these parts as I don't get any waveforms after that.

I'll post a pic of this switch and board to help out

Edited by vinyl_junkie (17/03/12 07:41 PM)

I don't want to step on Folderol's toeseroles as he clearly knows his casseroles, but when you see two switches on a diagram with a dotted line connecting them, it generally means they are mounted in the same physical switch and will move together. The switch will have two wipers, so make sure you're checking the right one. They are not connected in any other way, and the signals are not shared or split etc. - they're two separate switches operated by the same knob/toggle.

Following the signal flow through the flipflops and through the waveshaper, where exactly do you lose the signal? I take it you've traced the signal on the wiper of the waveform switch and there's nothing there? First place I'd check would be the square wave output (the middle wave-form switch position on the diagram) because that's fed right from the output of the first flip-flop.

Edited by nathanscribe (17/03/12 07:56 PM)

Quote nathanscribe:

---

I don't want to step on Folderol's toeseroles as he clearly knows his casseroles, but when you see two switches on a diagram with a dotted line connecting them, it generally means they are mounted in the same physical switch and will move together. The switch will have two wipers, so make sure you're checking the right one. They are not connected in any other way, and the signals are not shared or split etc. - they're two separate switches operated by the same knob/toggle.

Following the signal flow through the flipflops and through the waveshaper, where exactly do you lose the signal? I take it you've traced the signal on the wiper of the waveform switch and there's nothing there? First place I'd check would be the square wave output (the middle wave-form switch position on the diagram) because that's fed right from the output of the first flip-flop.

---

Thank you! That makes total sense

I'm not getting any output on the second wiper (The one that's dotted on the diagram)
And also no input on that hence the no output! So nothing is feeding the waveshaper etc
But the other half of the switch works ok and it's feeding the CD-4013

Right, well, you won't be getting anything on the other wiper (I'll refer to the one higher up on the diagram as the 'upper') as the wiper would be tied to ground. Make sure you're getting 0V there. If I were you I'd not worry about that part of the switch right now.

Are you getting a good square wave on the square wave output - that is, by the switch marked "waveform", before the filter? What do you get from the sawtooth output? EDIT: check the terminals (switch inputs), not the wiper (switch output).

Edited by nathanscribe (17/03/12 08:07 PM)

Also I spot what looks like a VERY scorched resistor on the board!

I'm removing the board so I can have a closer look at the circuit traces on the other side and further investigate this.

Here is a pic of the net of what the OSC board looks like (The M110 is under the keyboard by the way)

To the left of the connector is the octave switch, next to it is the C4013 (mine is CD4013)
Under the C4013 is the resistor that looks burned on mine.

That photo's too small. Do you have a bigger version?

What's the resistor labelled?

Well that was rather quick, and good news too.

You need to check that you get something out of that chip and onto the waveform selector.

Go to the hot end of the level pot (fed by the waveform switch). Might as well try all three positions on the switch. If only one position works, then use that for the rest of the tests for the time being. If none work then suspect that 4013 chip.

If you have a good signal, then step along the the filter sections IC1/2. Pin 9 then pin 8. The waveshape should gradually change as you work your way along. Also, pressing keys should make it cycle through the filter envelope.

This will eventually bring you to the LM3080 acting as a voltage controlled amplifier - I hope that isn't the fault as it is obsolete, and would require some board mods to get a replacement to fit.

--------------------
It wasn't me!
(Well, actually, it probably was)

Quote nathanscribe:

---

Right, well, you won't be getting anything on the other wiper (I'll refer to the one higher up on the diagram as the 'upper') as the wiper would be tied to ground. Make sure you're getting 0V there. If I were you I'd not worry about that part of the switch right now.

Are you getting a good square wave on the square wave output - that is, by the switch marked "waveform", before the filter? What do you get from the sawtooth output? EDIT: check the terminals (switch inputs), not the wiper (switch output).

---

Not getting any other waveforms mate as the waveform shaper is not being fed anything from the second wiper.
I get output from the first wiper but nothing from the second.

The lower half of the knob are the inputs from the M110 (all ok) then the big long leg from the wiper (right hand side) is ok, outputs into the 4013 IC
BUT the secondary inputs on the selector switch (top left) I get nothing! hence nothing on the wiper

Quote nathanscribe:

---

That photo's too small. Do you have a bigger version?

What's the resistor labelled?

---

Yea sorry, didn't realise it re-sizes pics this forum

Jen OSC Board

Quote Folderol:

---

Well that was rather quick, and good news too.

You need to check that you get something out of that chip and onto the waveform selector.

Go to the hot end of the level pot (fed by the waveform switch). Might as well try all three positions on the switch. If only one position works, then use that for the rest of the tests for the time being. If none work then suspect that 4013 chip.

If you have a good signal, then step along the the filter sections IC1/2. Pin 9 then pin 8. The waveshape should gradually change as you work your way along. Also, pressing keys should make it cycle through the filter envelope.

This will eventually bring you to the LM3080 acting as a voltage controlled amplifier - I hope that isn't the fault as it is obsolete, and would require some board mods to get a replacement to fit.

---

Ah yes of course! I'll check the out of that IC. silly me!

Thanks

Ok I have Square wave input and output on the waveform selector (no other waveforms)
The square imputed into the selector pot is quieter (this is probs due to the resistor before the selector pot right?)

IC1/2 on here is not a LM13600 like in the schematic but a LM13700N

Just checking the pins as instructed now.

That burned out resistor sill also bothers me, I'll snap a pic in a bit

Quote vinyl_junkie:

---

Not getting any other waveforms mate as the waveform shaper is not being fed anything from the second wiper.
I get output from the first wiper but nothing from the second.

The lower half of the knob are the inputs from the M110 (all ok) then the big long leg from the wiper (right hand side) is ok, outputs into the 4013 IC
BUT the secondary inputs on the selector switch (top left) I get nothing! hence nothing on the wiper

---

I think you misunderstand. The upper part of that double switch does not have inputs and outputs exactly - the wiper is tied to ground, and the other terminals provide a variable resistance to ground for the 301 op-amp. The input to that op-amp is actually the other pin (pin3, with + next to it, the lower of the two inputs on tat first 301). Note that that comes from a transistor which is fed by the second flip-flop.

This means that you won't get a waveform reading on the terminals of that upper switch - its' not in the signal path.

If you're getting a good square wave on the WAVEFORM switch, that's something. If the other terminals at the WAVEFORM switch are not showing anything, then we've isolated a section that's at fault.

Edited by nathanscribe (17/03/12 08:55 PM)

I don't quite get the filter configuration but I do get a signal on pin 8 and 9 of that chip.

There are two LM13700's first one I get a signal on both pin 8 and 9 second 13700 I only get a signal on pin 8.
The filter is set fully open with resonance at 0

Also pressing the keys doesn't change pitch ever

Quote vinyl_junkie:

---

Also pressing the keys doesn't change pitch ever

---

If your M110 is working, I'd be looking at the two BC309 transistors and associated resistors/caps that are marked above the flipflops on the diagram. The 110 provides a current that varies with frequency, that will cause a frequency-dependant ramping voltage to be present on the input of the 301 via those transistors, and which is reset by the flip-flop triggering the BC239 slightly to the right. What that means is that this part is generating a sawtooth wave, which is buffered and amplified by the 301 op-amp.

That said, if you're still getting no audio output from the synth when using the one good known waveform, follow Folderol's directions and trace the post-filter signal.

If you need a replacement 3080, I have a bunch of spares and can send you one or two to try out.

You should also make sure that pressing a key triggers the envelopes. Set the filter to self-resonating and set it up in way that there's some envelope modulation of the cutoff, with a medium cutoff frequency. Play a few keys and see if the output of the filter sweeps as you'd expect.

Edited by nathanscribe (17/03/12 09:13 PM)

Very interesting.

IC1 & 2 are what are known as transconductance amplifiers and a discussion of their behavior on here would probably count as a cruel and unusual punishment!

BTW the LM13700 is a slightly more modern replacement for the 13600.

It's interesting that you get nothing on the last stage. It suggests that chip is at fault. However, before accusing it can you check that there is -12V supply on the bottom of that 10k resistor (the other end of which is on pin 9).

Also it's not too clear from the drawing which pin is which. I'm assuming that pin 9 is the one that goes to the LM3080.

In any case, it's clear that you have more than one fault (possibly as a result of those plugs having been in the wrong way round) so it might take a while to get everything sorted.

@nathanscribe
I deliberately didn't mention the waveshaping section as I wanted to keep to tracing the most direct line until things were a bit clearer.

P.S.
If you've got some 3080s that could be very helpful

P.P.S.
Hmmm, I'm getting more and more suspicious of that -12V generally - could answer a lot of questions.

--------------------
It wasn't me!
(Well, actually, it probably was)

Edited by Folderol (17/03/12 09:21 PM)

Thanks for the help, even though she's not yet fixed you all deserve beers and then some!

I'm calling it a night now and I'll have a better look tomorrow and try what you have both said.
I took the main OSC board out (M110 not on it as it's under the keyboard)

Notice how none of the components on the circuit board are marked on the board, must be the Italian way..... Make everything damn confusing lol I'm not sure which resistor the burned one is or what value it is...

Burn Baby Burn

Board Top Side

Underside

Oxydised pins on IC's

Edited by vinyl_junkie (17/03/12 09:43 PM)

Quote Folderol:

---

@nathanscribe
I deliberately didn't mention the waveshaping section as I wanted to keep to tracing the most direct line until things were a bit clearer.

P.S.
If you've got some 3080s that could be very helpful

---

Yes, I didn't want to confuse things, but probably have. Sorry!

My 3080s are the 8-pin DIPs, which I'm guessing are the ones in the Jen - rather than the cans. I'd be happy to send a couple out if VJ's look to be at fault.

Quote nathanscribe:

---

Quote Folderol:

---

@nathanscribe
I deliberately didn't mention the waveshaping section as I wanted to keep to tracing the most direct line until things were a bit clearer.

P.S.
If you've got some 3080s that could be very helpful

---

Yes, I didn't want to confuse things, but probably have. Sorry!

My 3080s are the 8-pin DIPs, which I'm guessing are the ones in the Jen - rather than the cans. I'd be happy to send a couple out if VJ's look to be at fault.

---

Many thanks Nathan! I'll have a better prod about in the synth tomorrow and re-read all your posts again so it sinks in better.
Quiet enjoy this to be honest, I'm slowly getting "it" and how this stuff works

OK, I'm off to bed shortly, and will probably not be around till quite late tomorrow. However here's a couple more thoughts.

That burned resistor is almost certainly the 33ohm one that feeds pin 17 of the M110. It needs replacing but has probably more-or-less survived, otherwise I don't think the chip would be working - although it's not unknown for these things to sort of work by drawing power through their input pins! At the end of the resistor the first colour band is still readable and orange = 3

The 74LS221 is also probably OK. These chips can tarnish strangely sometimes. If it had overheated I would expect it to have also partly melted the chip holder - It's nice to see liberal use of chip holders. It makes repair much easier.

Night all

--------------------
It wasn't me!
(Well, actually, it probably was)

This post is fab. I love it that electronic things can still be fixed by people who know how, and can be bothered. I'm hooked on this post.

--------------------
The only excuse we have for making music in the first place is to make it differently..vis-a-vis our own difference (Glenn Gould)

Ok I get how and what that selector switch does now, silly me not to see that the wiper is tied to ground hence I wouldn't see a signal duuhh.
I checked the voltage to make sure it's 0v but it was 30mv (Crappy grounding?)

I un-soldered the burnt resistor and to my amazement it measured on my ohm meter at have a guess.... Yep 33 Ohms lol Not bad I don't think considering it's burnt out, carbon composite most likely, old and 5% tolerance.
I put it back for the moment as I don't think it's the issue and don't have a spare at hand, I checked the voltages with it in place and they tie up with what the schematics say.

I have re-checked the M110, it seems that the keyboard inputs are ok, the gate and trigger seem fine etc but I don't understand why when I press any key there is no change in pitch (I checked this at the octave range knob output) only the master tune knob can change the pitch.

Quote vinyl_junkie:

---

I have re-checked the M110, it seems that the keyboard inputs are ok, the gate and trigger seem fine etc but I don't understand why when I press any key there is no change in pitch (I checked this at the octave range knob output) only the master tune knob can change the pitch.

---

If, when you press a key, the gate and trigger work, I imagine the M110 is basically fine. I also imagine the diode matrix etc. is OK. I'm hazarding here, but asI said in a previous post, it may be that the transistors and caps/resistors that provide the ramping voltage to the wave-shaper op-amp are at fault.

Just tell us again, what do you get on the Sawtooth output of the Waveform selector? If the answer is 'not a sawtooth' then there's something wrong with the waveshaper section.

There may be other faults beyond that, which Folderol has spoken about before. But testing that one place will tell you if the waveshaper is duff.

Hello again.

Before going much further, I am very suspicious of the -12V supply. If that is seriously low, or not connected everywhere it would account for almost all the problems.

Check the voltages on the regulator 79?12. You should get about -15 to -20 on one side, and almost exactly -12V on the other side (connected to the test point 'blob').

--------------------
It wasn't me!
(Well, actually, it probably was)

Wave shaper is definitely duff, only waveform I get is square. I have checked the inputs pins on the waveform selector switch and also what is coming from the wiper.

Quote Folderol:

---

Hello again.

Before going much further, I am very suspicious of the -12V supply. If that is seriously low, or not connected everywhere it would account for almost all the problems.

Check the voltages on the regulator 79?12. You should get about -15 to -20 on one side, and almost exactly -12V on the other side (connected to the test point 'blob').

---

OK mate, I get -12 that's ok and on the other side I get -23

I have checked all the power outs the only one that seems more out is the +18v supply which seems to be +23v

Next thing would be to check power is getting to all the IC's right?

OK, that's almost a shame The high +18V isn't a problem.

Yes, check it's getting to all the Amp chips.

If it is, the next move it to try swapping the two LM13700s and seeing if that fault (no O/P on the last stage) 'moves'. If it does, you have a faulty chip. If it doesn't that it's a circuit fault somewhere.

I'm still trying to pursue a direct line here. The waveshaper can be sorted later.


The best way to lift the chips on a nice open PCB like this is as follows:

Work a thin-bladed screwdriver under one end of the chip and twist it to ease the chip up just a bit. Repeat with the other end of the chip. Keep going back and forth just slowly walking the chip out of it's holder. This way you can get it out without bending or breaking the legs.

--------------------
It wasn't me!
(Well, actually, it probably was)

Here are the exact voltages I get at the output connector on the PSU, a loom then plugs into this that distributes power to all the boards. I have so far checked all the pins on the PSU but not got to checking power on each board yet.

+11.90v DC

+23.1v DC

+4.96v DC

-12.09v DC

Checking the power to the ICs might be worth it, it's easy enough to do.

If the waveshaper is down, I would personally 'scope what you get going into pin 3 of the first 301. You may have to pull that 301, I don't have any spares but they're readily available and cheap.

EDIT: I feel I'm confusing things with Folderol's advice being along other lines, so I'll grab a drink instead.

Edited by nathanscribe (18/03/12 08:55 PM)

Aye gave up with the wave shaper for the minute as in theory I should get an output as the square wave works.

I can confirm the op amp's are getting the right voltages on all the boards, haven't checked the other IC's yet.

I'll swap the LM's next time I have time for this, the board needs to come out so I can prize the chip out which is a bit of a pain as all the knob caps and washers need to be unscrewed from the front panel.

I'm calling it a early night tonight, got a horrible headache which means I'm not paying attention.
I had the power switched off on the synth but forgot the fact that it's still plugged in... Why is that an issue? Because the back of the fuse holder is still live and they didn't shrink wrap the terminals so they are exposed ;-) That woke me up a bit

I'll get back to it probs next weekend, don't usually have the time during the week.

Many thanks for all your help again

OK, If you're not feeling 100% it's definitely not the time to be messing about with this stuff - as you discovered! It might be an idea to see if you can cover those exposed mains tags.

Some more musings:

I was rather hoping the -12V rail was the problem, because you just seem to have too many unrelated faults.

What I'm now thinking is that at some time it was powered up with those plugs in the wrong way round, which probably put all sorts of voltages in totally the wrong places. This would account for the cooked resistor, the (possibly) faulty 13700, and the lack of waveshaping or envelopes. The amps all use the + and - supplies and if these got reversed anything could have happened.

If it looks like this is the case, you will need to decide whether for curiosity's sake you want to work through finding each faulty part, or just do a chip 'blitz' and replace the lot. The latter is an attractive option as all the chips are plugged and (apart from the 3080) all are still currently available.

If I was doing this for a customer, the blitz approach is what I'd use (it would actually be cheaper for them as far less time would be involved), but being an ornery type, if I was doing it for myself I'd want to know exactly what had gone wrong

--------------------
It wasn't me!
(Well, actually, it probably was)

Quote Folderol:

---

If I was doing this for a customer, the blitz approach is what I'd use (it would actually be cheaper for them as far less time would be involved), but being an ornery type, if I was doing it for myself I'd want to know exactly what had gone wrong

---

It can take far longer than any sane person would want to take, sometimes, particularly when something that sits there looking innocent is secretly destroying worlds...

I was thinking the same thing, I'll get a list of how many chips I need and total price and then have a think about it.
Can the pots be damaged by passing DC through them? I say this because when I first got it and the cables were plugged in the wrong way round when I powered it up that's what it sounded like when you turned them ....mega unhealthy sounds!

When I opened it up it I was amazed most were plugged in back to front!