JC-120 Distortion malfunctioning

[Mr_bibbles]

http://www.tangible-technology.com/s...-120%204th.pdf

Hey y’all,

I have. Jazz Chorus that is stumping me. Everything sounds fine until you turn on the distortion circuit. I know these have a reputation for having lackluster distortion, but something is definitely malfunctioning; if you strum softly it’s not so bad, but if you strum a chord with gusto it pushes the distortion into a blatty mess.

I replaced q5, q6, and q7, and the d4 and d5 diodes with it (sk117 and in4148 respectively), and a couple off spec caps.

Does anyone have a link to an explanation for this type of clipping circuit? Or would someone who is familiar help walk me through the functions? I know my way around tube amps but solid state circuits are still above my pay grade. I don’t have a scope but I have an audio probe.

Thanks

[J M Fahey]

I have. Jazz Chorus that is stumping me. Everything sounds fine until you turn on the distortion circuit. I know these have a reputation for having lackluster distortion, but something is definitely malfunctioning; if you strum softly it’s not so bad, but if you strum a chord with gusto it pushes the distortion into a blatty mess.

That´s the way it is.

I replaced q5, q6, and q7, and the d4 and d5 diodes with it (sk117 and in4148 respectively), and a couple off spec caps.

But you kept the circuit, so .......

Does anyone have a link to an explanation for this type of clipping circuit?

It´s a very poor FET circuit, which gets poorly overdriven, and has not enough gain, ugly clipping, no pre or post EQ, a mess.

Or would someone who is familiar help walk me through the functions?

See above

[g1]

Agree.
Unless you are comparing it to another of the same model and finding it substantially different. The distortion circuit in these is pretty much unusable.

[Mr_bibbles]

Agree.
Unless you are comparing it to another of the same model and finding it substantially different. The distortion circuit in these is pretty much unusable.

I am comparing it to another JC-120, and I’ve worked on a few before. It is definitely broken. It’s not just a ‘bad sounding distortion’, It’s malfunctioning; I think it’s a transistor biasing issue. It sounds blatty and uneven, like it stutters as it’s fed signal. I don’t really know where to start because I’m not sure what function each transistor in the circuit performs, should I just look up SK117 operating parameters and check voltages?

Thanks guys.

[J M Fahey]

I calculated what voltage, signal and waveform yu should have in a properly working preamp.

Had to guesstimate a lot because Roland does not care to show DC voltages (except at the supply itself) nor even *label* where supply voltages come from

I am supposing preamp gets +27V straight from supply regulator (not even a regulator, just a ripple filter), so confirm it first; if different all others will change accordingly.



The very cheesy "Distortion" is just a starved Fet gain stage, Q5, getting about 1/4 of the "27V" so about 7Vdc.
Its Drain "should" be around 5V DC if stage is biased as a "gain stage which gets too much signal and can´t avoid clipping" .

As cheesy as can be, for multiple reasons:

* FETS are more inconstant than a teenager Daughter, parametrs vary all over the place, even same brand and model, same batch, , as much as 5:1
If you replaced it, good luck trying to get one which works there even close to original.

* FETs HATE low voltage supplies; the lower the voltage the crazier more inconstant they become; so you may very well have there a FET so misbiased that it´s slammed groundwise or railwise , slimilar to a "broken amp", and so chops and farts when fed signal.

* to boot such signal has not been preprocessed even minimally, such as cutting bass for clarity (what *every* distortion does) or cutting some treble afterwards (what most Distortions do) to attenuate buzzyness.
So even if working "properly" this Distortion is a mess.

Can´t believe they didn´t add , say, a Tube Screamer or any of their other excellent distortions and can only guess this amp was designed by a Jazz lover who **hated** Distortion, Rock, etc. and only grudgingly added one, the worst he could so nobody uses it.

[DrGonz78]

It's good enough for Jazz... oh wait a moment!! I remember fixing a dead distortion channel on one of those amps. When I returned it to the shop owner I warned him to inform the customer how bad the distortion channel sounds when working properly.

More on the subject of the thread or a question that came to mind. Is there a way to increase the power supply voltage to these FETS? Of course that opens up Pandora's box. Has anyone ever modded these amps to get the distortion channel sounding better?

[J M Fahey]

FETs are wonderful amplification devices, biased like tubes, similar impedance characteristics, clip very similar ... pity rhey are such inconstant b*tches so sooner or later everybody stopped using them.
Japanese loved them (Yamaha, Roland, etc.) but had to heavily depend on large purchases of Factory selected parts, and even so ...
They use Fets preselected, measured one by one, and identified by a coloured dot of paint, that´s why you always see a letter at the end, "Y", "G", etc. meaning Yellow, Green, etc.

They specially designed *one* FET for very low voltage, the one used inside electret capsules, which works as low as 1.5V DC, but which has terrible amplification.

Similar to tubes, they work better with higher voltages ... but most stand only 25V ... you´ll see those in Japanese amps, and Randalls fed exactly that and no more.

Not worth modding JC distortion; either use an external pedal or embed a small distortion PCB (say, an MXR Dist+ which is the simplest) and use switching Fets as a mini , dedicated FX loop.

But in this case check voltages to be able to repair it at least to Roland standards

[drewl]

Maybe change C21 and C25 to "tune" the distortion a bit?

[Mr_bibbles]

Wow

Thanks JM, this is exactly the answer I was looking for, plus a whole lot more. I’ll get to the bottom of it and record what the failure is in case it can help someone else.

Seriously though thanks for the effort... test voltages???? Amazing!

I calculated what voltage, signal and waveform yu should have in a properly working preamp.

Had to guesstimate a lot because Roland does not care to show DC voltages (except at the supply itself) nor even *label* where supply voltages come from

I am supposing preamp gets +27V straight from supply regulator (not even a regulator, just a ripple filter), so confirm it first; if different all others will change accordingly.

[ATTACH=CONFIG]55000[/ATTACH]

The very cheesy "Distortion" is just a starved Fet gain stage, Q5, getting about 1/4 of the "27V" so about 7Vdc.
Its Drain "should" be around 5V DC if stage is biased as a "gain stage which gets too much signal and can´t avoid clipping" .

As cheesy as can be, for multiple reasons:

* FETS are more inconstant than a teenager Daughter, parametrs vary all over the place, even same brand and model, same batch, , as much as 5:1
If you replaced it, good luck trying to get one which works there even close to original.

* FETs HATE low voltage supplies; the lower the voltage the crazier more inconstant they become; so you may very well have there a FET so misbiased that it´s slammed groundwise or railwise , slimilar to a "broken amp", and so chops and farts when fed signal.

* to boot such signal has not been preprocessed even minimally, such as cutting bass for clarity (what *every* distortion does) or cutting some treble afterwards (what most Distortions do) to attenuate buzzyness.
So even if working "properly" this Distortion is a mess.

Can´t believe they didn´t add , say, a Tube Screamer or any of their other excellent distortions and can only guess this amp was designed by a Jazz lover who **hated** Distortion, Rock, etc. and only grudgingly added one, the worst he could so nobody uses it.

[J M Fahey]

Maybe change C21 and C25 to "tune" the distortion a bit?

C21 definitely, and R44, so they combined (that´s why we talk RC networks) cut some Bass to make distortion less farty.
But let mr bibbles repair it first, maybe post some audio, and then improve it *slightly* ; not much wiggle room available.

FETs all over the landscape is the reason why all those "tube schematics rebuilt with FETs" pedals have one trimmer per FET

Sort of acceptable for hobbyists, impossible for a Factory.

[Mr_bibbles]

Measurements:

So the main rail is coming in at 24.3vdc.

The biggest discrepancy is at Q2/c21 junction. Instead of 160mv it’s a flat 15v. After C21 it settles down to .5

Also, the 160mv supply is off on he Q2 side of C2. Seems like it’s fine on the q6 side but my meter has trouble that low, bounces between 100-300mv while reading.


Other than that the 7v and 5v readings near q5 are 5v and 3v respectively. That could be accounted for by he new calculated rail voltage?


Based on this, my hunch is c2, c21 or q2. What do y’all think?

[g1]

Instead of 160mv it’s a flat 15v. After C21 it settles down to .5

Also, the 160mv supply is off on he Q2 side of C2.

Numbers in boxes are AC voltages with signal level shown at input jacks.
Numbers in ovals are DC.

[HaroldBrooks]

Might I suggest using this ? The ST-2 With an EQ in front of it is my "Amp Model" of choice, and it's amazingly inexpensive, and up to around 12 O'clock sounds a lot like a vintage Marshall Plexi. The trick is to not feed it too much treble on the input, and to not expect it to distort like a modern Metal amp and still sound good. For a vintage Lower distortion, it's a fantastic model, and generally I hate amp modeling, but sometimes it's better than other options (poor distortion built into the amp.) It get's some bad press because it's abused and misused, but as an amp modeler, it sounds better than anything in my Eleven Rack, IMHO. Again, I trim back the real highs and real lows from my guitar, and it's pretty good !

https://www.youtube.com/watch?v=Fg5HLdYBskM

[Mr_bibbles]

Looks like I have 2-3 MVac on all leads of q2, at c11, r41, and c2. I replaced c11 (now .022) and Q2 in case it was faulty, no change

[J M Fahey]

Looks like I have 2-3 MVac on all leads of q2, at c11, r41, and c2.

With what signal input?