OK... I'm stuck.

[digi2t]

Still working on the USS-1 clone board. Everything is working....

Except the @#$%ing envelope function.

Just to backtrack a bit, the 4013 is working fine in place to the MFC4040 for the sub-harmonic, so many thanks to those that steered me right on that quest. TBH, none of the 4040's I have seem to work in this, but just as well that the 4013 is doing the job, it's way easier to find.

Next...

The power supply section we're using is +/-12v. The trace that on the net shows positive voltage as high as +14.5v, and the original unit I have here is topping out around +/-13.5v, so I figure there was some wobble room there. The fuzz works, the phaser works, sample/hold is a go, and sub-harmonic is old news. I can't seem to get the envelope to work though.

The envelope section actually serves a few functions; In envelope mode, it gives you the attack/decay swell of the note (I'm guessing that's what the 3080 marked "10" is doing?). With the filter mode on, it adds an autowah function. I think it also affects the phaser, acting as a attack delay, delaying the onset of the phaser after initial attack. Anyway... right now, it ain't doin' jack shit.

I read up on the 4001, and I saw that it comes in buffered and unbuffered flavors. I've tried both, and nada. As it stands now, there is no envelope effect on the filter (or phaser, if there's supposed to be), and in envelope mode, I only get a few farts when I pluck hard. Sensetivity is at max, with attack and decay set fast and slow respectively.

I've even audio probed both the original unit and the clone board, and up until the 4001, I'm getting the same signal. When I get to the 4001 though, voltages don't seem to be fluctuating much at all.

Here's a copy of the schemo, with my voltage notes in red. There are also some resistor value discrepancies between it and my unit, so I'm running with my values (in red).

[ElectricDruid]

One thing that strikes me about that at first look (and that's complicated enough to need a lot more than one look) is that it seems to be a envelope detector followed by an envelope generator, rather than the typical envelope follower that we usually see. E.g. it triggers a circuit which then generates an entirely new envelope.

So there's two things that could be wrong: The triggering might not be working, or the envelope generation might not be working.

Do you get any trigger pulse at the point where it says "jumps to 11V at leading edge of transient"? That looks to be the output from the triggering circuit and the input to the env gen (I think?).

Is it possible to trigger the env gen manually? The voltages in the envelope generator part look terrible, tbh, so I think I'd start by trying to separate that part out, add a button to pulse the input and see if I couldn't get it generating some AD curves.

[digi2t]

Do you get any trigger pulse at the point where it says "jumps to 11V at leading edge of transient"? That looks to be the output from the triggering circuit and the input to the env gen (I think?).

I hooked up my DMM, and used the MIN/MAX feature. I get a max of about 1.7v. Nowhere near the 11v stated on the schemo. I had actually checked that, but neglected to add it to the schemo notes.

Triggering it manually is a great idea. Going to try that tonight.

[Gus]

Do you know if the 4001 is a 4001?

Is it NOS or something new? If new is it from digikey, mouser, element 14, allied etc.?

[Kevin Mitchell]

Are you using the TTL or the CMOS versions?
What are the exact part numbers you are working with? Looks like the original uses TTL according the the schematic. So if you're using a CD4001 in place of a MC14001CP it will not trigger the same doing tasks outside of the digital realm - as in it's not being used as a standard logic gate so subbing one for the other without further work will be problematic.

[digi2t]

Are you using the TTL or the CMOS versions?
What are the exact part numbers you are working with? Looks like the original uses TTL according the the schematic. So if you're using a CD4001 in place of a MC14001CP it will not trigger the same doing tasks outside of the digital realm - as in it's not being used as a standard logic gate so subbing one for the other without further work will be problematic.

Not that I knew about that, but I did do the test with the original unit. It uses an MC14001CP. I carefully removed it, soldered in a socket and tried a CD4001BE. Worked fine. I can try it again tonight to confirm, but like I said, I didn't hear any difference.

Right now, like I said earlier, the audio is basically dead, apart from some farts when I pluck or strum hard.

Do you know if the 4001 is a 4001?

Is it NOS or something new? If new is it from digikey, mouser, element 14, allied etc.?

The 4001BE's and the 4001UBE's were all sourced from reliable vendors (Digikey/Mouser/Newark).

[digi2t]

Alright.... progress!

As per Tom's idea, I used a 9v battery, and sent +9v to the pin 6 junction. The envelope opened up, and I get the audio. I took it a step further, and threw a pot on the +9v to modulate the voltage. A B1K pot was allowing to use a bit less than half the sweep, say 400 ohms to cut in and out, kind of simulating a modulating signal.

I'm getting the volume modulation, like I would in envelope mode (also tells me the 3080 is working), and with the filter mode on, some filter frequency shifting. Whatever is supposed to be flipping and flopping on the 4001 is doing so.

So would that point to the trigger being the issue? If it's supposed to be getting up to +11v at pin 6, it ain't. The farting is because I'm just scratching the minimum for a split second.

Oh... another thing that I forgot to mark on the schemo, the 470k resistor to +V1 at the base of transistor X (just before the 4001), is not 470k in the original unit. It's 270k. That's what I have in there now.

I've tried 2N2222's, 2N3904's and 2N5088's so far. All give more or less the same result. Is this a biasing issue? 

[Slowpoke101]

Just a quick question regarding the 100K resistor that biases transistor 'W'. Are you certain that it connects to the the collector of transistor 'X' and not to V1?
I quickly bread-boarded that part of the circuit and it performed rather poorly. When the 100K was connected to V1 the circuit worked as expected.

[anotherjim]

It's scary indeed. The 4001 flip-flop doesn't look like it's free with both outputs at the same logic level. Ok, so the first thing I'd do is check those cross-connections between the top & bottom gates 'cause a 1 and a 0 input shouldn't make a 1 output with a NOR gate.
When checking logic just get the input pin voltages and see if the output pin voltage agrees with the truth table (NOR both ins need 0v to get 12v output) but beware CMOS high impedance means if a pin is disconnected, your meter with tell you it's got 0v so continuity check pin to pin links.

[digi2t]

Just a quick question regarding the 100K resistor that biases transistor 'W'. Are you certain that it connects to the the collector of transistor 'X' and not to V1?
I quickly bread-boarded that part of the circuit and it performed rather poorly. When the 100K was connected to V1 the circuit worked as expected.

Yeah, I've checked that on the original unit, and it does. Here is a side by side mirror image of that section...



I'm wondering if the fact that the voltage is a bit lower is throwing this off? I can lift that resistor off the board and jump it over to V+ to test. I'll report back later today.

It's scary indeed. The 4001 flip-flop doesn't look like it's free with both outputs at the same logic level. Ok, so the first thing I'd do is check those cross-connections between the top & bottom gates 'cause a 1 and a 0 input shouldn't make a 1 output with a NOR gate.
When checking logic just get the input pin voltages and see if the output pin voltage agrees with the truth table (NOR both ins need 0v to get 12v output) but beware CMOS high impedance means if a pin is disconnected, your meter with tell you it's got 0v so continuity check pin to pin links.

Will test here as well, but I'd be a bit surprised if there is something funky here. Once the trigger voltage comes up, it seems to work properly. Regardless, no harm in checking to be sure, and I will.

[anotherjim]

If pin1 is 0v and pin 2 is +12v, no way can pin3 be +12v. Logical conclusion is pin 2 must be 0v cause I don't have a reading for that pin. I think the little voltage bump you see is the DC average of a fast glitch due to propagation delays in the gates. It wants to work but can't.

[digi2t]

The good news... the envelope output to the filter is there...

WE HAVE AUTOWAH!!!!

I had switched to a different jfet for testing, and the new ones had a reverse pinout. Of course, this dumbass put the new jfet in this spot in backwards, so the filter section wasn't receiving the modulation signal.

The... so so news....

Tried moving the 100k base resistor from transistor W to V+ directly, and it didn't help. Actually made it worse. No farting. Put it back.

Increasing the 270k base resistor on transistor X to 470k (contrary to what's in the original unit, but what's marked on the schemo) made a slight improvement. I'm now getting up to 2.8v at the collector. In other words, louder farts. Disconnecting it all together bumps the collector voltage up to 5.3v, and the muting is no more. I can hear the audio signal, and the decay pot is acting like a volume control. Still a long way from 11v though. This probably because the collector voltage is high enough to open up the 3080, and since pin 11 goes low, the pot is now varying the voltage to the 3080? I reconnected the resistor.

There are the voltages from the 4001 as it stands now;
Pin 1 = -0.395v
Pin 2 = 0
Pin 3 = 12v
Pin 4 = 0
Pin 5 = 12v
Pin 6 = 73mV
Pin 7 = 0
Pin 8 = 73mV
Pin 9 = 73mV
Pin 10 = 12v
Pin 11 = 12v
Pin 12 = 73mV
Pin 13 = 0
Pin 14 = 12v

BUT!

If I set the attack and decay to long settings, and I give pin 6 a "jump start" with the 9v battery, the envelope attack/decay shaper starts working, and stays working. It'll stay on and maybe a bit more if a nudge towards shorter settings, but if I push the controls more than that, it shuts down. I can nudge it open again with a 9v jolt, but shuts down again.

It's really is as if I'm not getting enough of a voltage jump at transistor X collector to wake up that part of the circuit.

[Slowpoke101]

This probably will not help at all but it does confirm that the circuit does actually work. I constructed the circuit from the input transistor 'T' and then going all the way to the point marked 'C' - Envelope Output. I connected an oscilloscope to the envelope output so I could see the attack-decay envelope shape. It is easy to trigger the circuit and it is quite sensitive without the IC input buffer (I found the circuit diagram later). The envelope shape is as expected and does vary with the controls.

At first it wasn't easy to trigger as I had made a mistake with transistor T's bias. I connected the 150K resistor to V1 instead of 0V, once that was corrected the circuit worked well. Normal strumming or picking actions on the guitar easily trigger the circuit.

The 100K resistor for transistor W bias does go to transistor X collector. The connections shown for the 4001 do work even though they seem odd but they change state depending on the voltage level of the envelope output.

As I said - not very helpful, but it does confirm the circuit works, unless I've really done something weird - I'll check for that tomorrow.

[digi2t]

Well no, it does indeed help. At the very least, it motivates me to keep digging.

I took a look at the Maestro Envelope Modifier, and I see that much of the trigger section is almost identical, except for some cap and one resistor value (if I'm not mistaken). It's running at 9v.

Some questions on your setup though;
- What voltage are you testing at?
- What transistors are you using?

I'm kind of tempted to bump up the voltage to 15v. It also seems to work better with lower gain transistors (2N2222a vs. 2n5088).

EDIT:

Swapping to +/-15v did nothing. The filter autowah still works, but in envelope mode, just farting. No audio attack/decay to be had  . The X collector voltage only gets to about 3.5v when I hit the strings, and that's with sensitivity on max.

I'm beginning to wonder if I have a bad 150n cap? Kinda at my wits end here, so I'm hunting.

[Slowpoke101]

I've checked my bread-board layout for errors and found that I have got it right. So far so good.
My test voltage is +/- 12.8VDC. I've tested at 10V and 15VDC as well and noted no difference other than a slight improvement in sensitivity at the higher voltage. Remember that I am not using the IC input amp buffer (why? I am lazy - I ran out of room on the bread-board).

The transistors that I am using a standard jellybean types found 'Down Under' - BC549. Gain averages out at about 350. Nothing special but much higher gain than gain MPS5172 or 2N5172 usually offered.

Something to note - I think that you said that you have the sample/hold system working. That uses the envelope trigger to control the S&H transistor's noise gate so surely the envelop trigger area circuit should be working.

I have noted something of concern that you must check and it will be causing a big problem with your attack/decay shaper. There is something wrong with the way your 4001 is connected. With the circuit un-triggered (just sitting there) the pin voltages should read as follows;

1. -0.2            8.   0.07
2.  12.8           9.   0.07
3.  0               10.  12.8
4.  12.8           11.  0.07
5.  0               12.  0.07
6.  0.07           13.  12.8
7.  0               14.  12.8

Your previous post shows pin 11 as being 12V. It should only hit that voltage during the attack phase. If it stays there the envelope shaper may not work. I haven't really investigated this so this is only hypothetical. But do check out your connections.

Having looked more closely at the full circuit I am impressed that you are willing to take this project on. I hope that you have a win with this. It will be great project. It's as bad as some of the Oberheim analog synths I repair.

[digi2t]

OK, thanks Ian.

I've replaced the initial 2N3904's with lower gain PN2222's (plastic 2222's) for now. Typically, transistors used during this era tend to be a bit lower gain, so I figured I would roll with that.

Yes, my pin 11 is sitting at 12v (15v now since the regulators swap). Looking at the truth table, it seems that it should be sitting at 0v. That didn't catch my attention before. I'll double check with the original unit tonight, see what pin 11 is doing there. If it should be 0v, perhaps I have a short in the board, or the socket? Been through those rodeos before. Continuity tests might be on the menu tonight.

And before I forget, thanks a bunch for your help here. Been staring at this tree for so long, I've lost sight of the forest.

Will report back promptly...

EDIT:

Pin 11 doesn't seem to be the only issue. Pins 2, 3, 4, 5, and 13 also have their voltages flipped as well.

Just checked the original unit, and the pin voltages correspond to yours. Something really whack going on here.

[Slowpoke101]

Pin 1 is the key. It resets the latch (the flip-flop part) after the initial attack phase. That allows the voltage on pin 11 to return to zero which starts the decay phase of the envelop shape. Check its connections with the emitter of transistor S and the collector of transistor R.

Another silly thing that just occurred to me is if pin 11 is high (12V or 15V), how is pin 1 low? Transistor S would have to be turned off but it should have more than 8V on it if pin 11 is high.

[digi2t]

I FOUND IT!!!!     I FOUND IT!!!!      I FOUND IT!!!!      I FOUND IT!!!!     I FOUND IT!!!!
Error on the board. @#$% my life.

Schemo that was sent out to the fab shop...



For some reason, on the PCB there's a trace between pin 6 and anode of D23. We're looking into the source of the error. I've only been staring at it now for 3 weeks. Yes.... I'm slow.

Cut the trace, and everything works as it should.

Mighty thanks to all for their input and bench work (especially Ian). I guess R.G. was right all along.... it's not a failure until you quit.

I need a nap now.

[Govmnt_Lacky]

Mmmmmmm... eagerly awaiting it's arrival. I NEED A PROJECT to get back into the groove! 

[digi2t]

Mmmmmmm... eagerly awaiting it's arrival. I NEED A PROJECT to get back into the groove! 

Where the hell were you the last 3 weeks of my life? 

It's coming buddy. Phil's travelling right now, but I'm sending out the last corrections. Should have it in a couple of weeks or so.