JCM900 fx loop treble loss

[lowell]

2 of these 4100 heads came in. Customer says they didn't do this previously. I have however heard at least one other guitarist I know say he used that to his advantage to warm up these brittle and harsh sounding amps. IMO.

First thing I did was confirm the issue. The amps indeed have a slight low-pass effect when using a delay in the loop.

I ended up adding a bright cap on the fx loop mix pot to fix it. I thought it sounded fixed.... they're bringing them back cause now they "sound worse." Not a big deal and I'm happy to work with them to make it sound good in their opinion.

It's opamp-buffered so I don't see any reason that the send can't provide enough current drive.

My question: anyone else run into this issue? Any suggestions on how you'd approach this?

Finally... what's with the 100ohm on the send ground? Ground loop mitigation?


http://drtube.com/schematics/marshal...-iss7-0349.pdf

[alexradium]

bright cap shouldn't be needed with such low impedances.
Maybe the opamp has degrade over time?
I'd change it with a NE5532 and also put a new 10UF in front,even better bypassed with a film cap,470nF.

[Enzo]

Try several unrelated FX units to see if it is universal. Could be unit-specific.

But mainly, look at the circuit. This is not just a pair of cutout contacts. There are resistors doing things. You tried a delay in the loop. Does a straight cord across the loop change the sound? In the empty jack mode, the return has a grounding contact on the 47k resistor forming a 47k/220k voltage divider. The send does have the VR1/R1 across the line.

In the FX loop mode, there is that 100 ohm ground lift. I don't know just what it might cure, but it could be involved in any difference. But each end now has a 10k plus 3.3k resistance to ground. That appears to be a fairly different impedance from the straight through, which might be involved.

WHich leads me to wonder if it makes a difference whether the loop control is set higher or lower.

You tried a brightness cap on one or the other pots, but do you have a 31-bander around? You might run music through the system and have the EQ in the loop to see just which freq bands correct the problem. Might tell us something.

[Enzo]

Try several unrelated FX units to see if it is universal. Could be unit-specific.

But mainly, look at the circuit. This is not just a pair of cutout contacts. There are resistors doing things. You tried a delay in the loop. Does a straight cord across the loop change the sound? In the empty jack mode, the return has a grounding contact on the 47k resistor forming a 47k/220k voltage divider. The send does have the VR1/R1 across the line.

In the FX loop mode, there is that 100 ohm ground lift. I don't know just what it might cure, but it could be involved in any difference. But each end now has a 10k plus 3.3k resistance to ground. That appears to be a fairly different impedance from the straight through, which might be involved.

WHich leads me to wonder if it makes a difference whether the loop control is set higher or lower.

You tried a brightness cap on one or the other pots, but do you have a 31-bander around? You might run music through the system and have the EQ in the loop to see just which freq bands correct the problem. Might tell us something.

[Chuck H]

I'm with Alex. The bright cap certainly shouldn't make things worse, so it seems something may actually be degrading. I vote for the 10uf cap. I might try replacing it with a 2.2uf film cap. Even a 1uf should pass plenty below guitar frequencies. Nothing to lose.

And I like Enzo's inquiry about whether the affect is level dependent. Could be useful info. It could also be operator error. Even among experienced players it's damn rare to find one that isn't baffled by adjustable effects loops. Or worse, thinks they know what they're doing, but don't and are still using it wrong. If, say, the user has the output set too low or too high and is then compensating with extreme and irregular adjustments available on the effects unit. In which case it could be the effects unit causing the problem because, while that extreme adjustment may be available for pinch circumstances, the circuit isn't really very good like that. Enzo's suggestion of trying a patch cord could reveal this.

[lowell]

If I remember correctly a patch cord did not affect the sound in any way. I tried that first to see if it was the loop or not. I definitely think there's a bit of user "hearing more high end" now saying it sounds scooped and worse. I thought it sounded exactly like not using the loop after adding the bright cap. I used my decade box to dial in just the right value.

Adjusting the pot definitely affects it. Even more so with the bright cap. Sounds best with the pot dead center... post installation of bright cap.

I was thinking that the 10uf cap was suspect. I know alot of electrolytics as coupling caps can cause problems overtime. But I didn't think that until the amp left the shop and he called me again.

I'll remove the bright cap and see if the tantalum does it.

I will also try Enzo's advice as well. I have a Furman parametric. Been waiting for it to come in handy for repairs.

[Chuck H]

If your guy is local you could have him bring the whole shebang in and set it up where he believes the phenomenon is at it's worst. This would give you the opportunity to observe how he's using the loop and, hopefully without bruising an ego, give a lesson on unity gain, noise floor and the effect of phase errors on frequency if needed. If the effect seems unreasonably real in the face of circumstances you can put your scope on the output and your signal generator at the input. Do a frequency sweep and record the levels @ frequency with and without the loop in use.

I will say that a loss of top end sounded plausible. Scooped mids absolutely does not so there's some user perception there. I mean, part of the point to using effects is that it changes the tone. If there are any phase/time effects in use that could explain a lot. The perceived mid scoop may be real and it may be because it's supposed to happen. Even with other similar complaints about this loop I wouldn't rule out user error or misconceptions about tone and time based effects on the part of the customer or even all of whom have made the complaint. There's a very good reason I never use adjustable, series/parallel or even fully bypassed effects loops in any of my designs. Going beyond "plug this in here" and "plug that in there" confuses the hell out many guitar players.

[Enzo]

Oh I loved doing that, invite a guy to bring in his rig and show me the problem.

[Chuck H]

Oh I loved doing that, invite a guy to bring in his rig and show me the problem.

Not sure if that's sarcastic or if you actually enjoyed witnessing and correcting ignorance. What I know about you tells me it could be either.

[Enzo]

I genuinely liked when I could have a guy bring in his rig and we'd set it up in the warehouse. He could demonstrate the problem, and I could assess whether it was an amp problem or otherwise.

[lowell]

Completely agree. And he infact just brought his whole rig in. Guitar, pedalboard, cab etc.. The problem is specifically with his MXR Carbon Copy delay. The problem happens with other effects as well, but with that delay it's very extreme. It was like a blanket was put over the speaker cab. I could dial in more highs with the fx loop pot (which has a bypass cap) but I do have to admit it did sound a bit scooped. Like the cap is too small.

I plugged his guitar into the amp input...disconnected everything and just put the carbon copy in the loop. Terrible low pass. I then put the carbon copy into the front end... no issue... sounded great. So there seems to be an extreme interaction with that pedal.

I still have the amp. He took his board as he has some gigs. But I will try my delay with the amp again and see if it's that bad.

I need to grab that pedal from home so will have more to report tomorrow. Gonna start with removing the bypass cap on the pot and putting the circuit back to stock. THEN will try to dial in a good delay sound. If the tone sucks my first approach is to add a tantalum or similar cap in parallel with the 10uf send coupling cap. If that works I'll replace that cap.

I'll reread thru this thread and see what else I can report for solid data.

[The Dude]

I don't have time to look up all the specs on this gear, but will add this. Many effects loops are designed for line level effects units. Many pedals are designed for instrument level ins and outs. There can also be impedance issues because of the different expected levels and matching. One needs to read specs when deciding the best option for connection of these various devices. Of course, if you have a major level or impedance mismatch, you can't expect things to work as intended. There's no hard and fast rule that says pedals go in front of the amp and rack effects go in the loop. You have to actually read the specs. Different devices are designed differently.

[Chuck H]

Completely agree. And he infact just brought his whole rig in. Guitar, pedalboard, cab etc.. The problem is specifically with his MXR Carbon Copy delay. The problem happens with other effects as well, but with that delay it's very extreme. It was like a blanket was put over the speaker cab. I could dial in more highs with the fx loop pot (which has a bypass cap) but I do have to admit it did sound a bit scooped. Like the cap is too small.

I plugged his guitar into the amp input...disconnected everything and just put the carbon copy in the loop. Terrible low pass. I then put the carbon copy into the front end... no issue... sounded great. So there seems to be an extreme interaction with that pedal.

I still have the amp. He took his board as he has some gigs. But I will try my delay with the amp again and see if it's that bad.

I need to grab that pedal from home so will have more to report tomorrow. Gonna start with removing the bypass cap on the pot and putting the circuit back to stock. THEN will try to dial in a good delay sound. If the tone sucks my first approach is to add a tantalum or similar cap in parallel with the 10uf send coupling cap. If that works I'll replace that cap.

I'll reread thru this thread and see what else I can report for solid data.

Can we assume that you confirmed proper signal levels and unity gain?

Re evaluating the schematic I can see that the 10u cap is in the circuit full time. So it's passing signal whether the loop is in use or not. Same with the buffer amp. I don't think it's the cap or the buffer. The loop ONLY adds the effects unit/s plugged into it and the pot/load circuits. The impedance of that circuit is low enough that the problem shouldn't be from any additional cable capacitance. Besides that you've already plugged a plain cable in with no problem. The only thing I can guess is that it's the delay unit causing it. Maybe he just doesn't notice it with other amps? Seriously, after evaluating things I don't see anything that can be done in the way of a repair. You might mess with that circuit till doomsday without improvement because there may be nothing to fix.

[lowell]

I did not confirm unity gain levels. Are you referring to send level and return level being unity? And thanks for that tip.

Customer owns 3 of these. Yes 3. Says this does not happen with other 2 amps. Same pedalboard and cables. Also says it didn't USE TO do this. It has changed over time. The other one that he brought in was doing it too. But that was someone else's. Not one if his OTHER 2.

I know.. this seems crazy. And I need to have him bring the "other" ones in to confirm that they don't have the same issue.

[Chuck H]

And I need to have him bring the "other" ones in to confirm that they don't have the same issue.

When set up the exact same way!