Into the rabbit hole...36 Watt EL34 Superlite TMB

Double-Bubble! Place for discussing the 36W version...

Moderators: CurtissRobin, colossal, zaphod_phil, Daviedawg, Graydon

User avatar
JMPGuitars
Super Duper Admin
Super Duper Admin
Posts: 3954
Joined: Tue 09/18/12 8:00 pm
Location: South Central, MA
Contact:

Re: Into the rabbit hole...36 Watt EL34 Superlite TMB

Post by JMPGuitars »

Billy_Goat wrote:
Mon 03/08/21 2:54 pm
With all due respect, I'm not a fan of wiring components without providing solid mechanical attachment for the same which your photo appears to indicate, so I built a little turret board where the two caps connect directly to the jacks and the two 'in series' resistors from the caps are attached to turrets. Then shielded wire from there to the main turret board and the last 68K resistor placed on the main turret board. I also added some foil shielding underneath the FX turret board and grounded that. The new turret is also now 1.5 inches above the plate wires lying on the chassis. All components and wiring is now more or less 90 degrees away from the plate wiring lying against the chassis. Nothing parallel.

It seems better upon first check, but admittedly I only had a few mins to play and analyze. So, the jury is still out atm.

If it's determine that this rebuild is still not working properly, I'll reluctantly try wiring as your photos suggests.

Cheers
No worries, I understand that not everybody likes to do things PTP, and I too prefer mechanical support for things. BUT, consider this: if you make good mechanical connections on the solder joints, and twist the leads of the components before soldering, you still have mechanical connections and support. The FX loop is physically light, and shouldn't generate any heat, so risk is low. Another option is to use your tag board to hold a zip tie locking the loop physically in place. There's plenty of ways to deal with it, find what works best for you. That's assuming you haven't already sorted the problem.
0 x
'I installed a skylight in my apartment yesterday... The people who live above me are furious.' - Steven Wright

Modern Ground Schemes
Soldering Technique
B+ Voltage Reduction
Amplifier Tools & Parts Info


Web Design: DolceVittoria.com
Guitars / Amps / Effects: JMPGuitars.com
(anti)Social: Facebook · Instagram
Items for Sale

User avatar
TriodeLuvr
Frequent poster
Frequent poster
Posts: 616
Joined: Tue 10/16/18 11:19 am

Re: Into the rabbit hole...36 Watt EL34 Superlite TMB

Post by TriodeLuvr »

Billy_Goat wrote:
Mon 03/08/21 2:54 pm
TriodeLuvr wrote:
Thu 03/04/21 10:11 pm
Is it wired like this? What's to keep the loop return signal from mixing (potentially out of phase) with the dry signal from the 68K resistor?
Hi TriodeLuvr,

Yes, 68K resistor was in place as required. Thanks!
My point was that the 68K might allow a return signal to mix out-of-phase with the dry signal and create the comb effect you're describing.

Jack
0 x

Billy_Goat
Newbie
Newbie
Posts: 44
Joined: Thu 11/19/20 9:07 pm

Re: Into the rabbit hole...36 Watt EL34 Superlite TMB

Post by Billy_Goat »

TriodeLuvr wrote:
Mon 03/08/21 3:33 pm
Billy_Goat wrote:
Mon 03/08/21 2:54 pm
TriodeLuvr wrote:
Thu 03/04/21 10:11 pm
Is it wired like this? What's to keep the loop return signal from mixing (potentially out of phase) with the dry signal from the 68K resistor?
Hi TriodeLuvr,

Yes, 68K resistor was in place as required. Thanks!
My point was that the 68K might allow a return signal to mix out-of-phase with the dry signal and create the comb effect you're describing.

Jack
Hi Jack,

Sorry, I thought you were asking if I had installed the 68K resistor rather than asking if the 68K resistor might be causing a mixing issue.

With the above in mind, I don't know, but I'm guessing that could be in play based on how I believe the circuit is likely working. Either way, I followed the build as documented. Are you suggesting removing or altering something here?

Cheers
0 x

User avatar
TriodeLuvr
Frequent poster
Frequent poster
Posts: 616
Joined: Tue 10/16/18 11:19 am

Re: Into the rabbit hole...36 Watt EL34 Superlite TMB

Post by TriodeLuvr »

Billy_Goat wrote:
Mon 03/15/21 8:45 am
TriodeLuvr wrote:
Mon 03/08/21 3:33 pm
Billy_Goat wrote:
Mon 03/08/21 2:54 pm

Hi TriodeLuvr,

Yes, 68K resistor was in place as required. Thanks!
My point was that the 68K might allow a return signal to mix out-of-phase with the dry signal and create the comb effect you're describing.

Jack
Hi Jack,

Sorry, I thought you were asking if I had installed the 68K resistor rather than asking if the 68K resistor might be causing a mixing issue.

With the above in mind, I don't know, but I'm guessing that could be in play based on how I believe the circuit is likely working. Either way, I followed the build as documented. Are you suggesting removing or altering something here?

Cheers
You might want to disconnect one end of the 68K and try it again, just as a test. That would eliminate the possibility of an out-of-phase mix from your effects device.

EDIT: If the problem goes away when you do this, it might indicate the need to revise the effects loop. Again, this is just a test. The root cause might be something completely different.

Jack
0 x

User avatar
JMPGuitars
Super Duper Admin
Super Duper Admin
Posts: 3954
Joined: Tue 09/18/12 8:00 pm
Location: South Central, MA
Contact:

Re: Into the rabbit hole...36 Watt EL34 Superlite TMB

Post by JMPGuitars »

TriodeLuvr wrote:
Mon 03/08/21 3:33 pm
My point was that the 68K might allow a return signal to mix out-of-phase with the dry signal and create the comb effect you're describing.

Jack
Assuming it's wired correctly, this method has been around for a long time, and is well tested. The only issues I've seen are PO if it's placed poorly, or the fact that it's passive doesn't work for all effects.
0 x
'I installed a skylight in my apartment yesterday... The people who live above me are furious.' - Steven Wright

Modern Ground Schemes
Soldering Technique
B+ Voltage Reduction
Amplifier Tools & Parts Info


Web Design: DolceVittoria.com
Guitars / Amps / Effects: JMPGuitars.com
(anti)Social: Facebook · Instagram
Items for Sale

Billy_Goat
Newbie
Newbie
Posts: 44
Joined: Thu 11/19/20 9:07 pm

Re: Into the rabbit hole...36 Watt EL34 Superlite TMB

Post by Billy_Goat »

JMPGuitars wrote:
Mon 03/15/21 1:52 pm
TriodeLuvr wrote:
Mon 03/08/21 3:33 pm
My point was that the 68K might allow a return signal to mix out-of-phase with the dry signal and create the comb effect you're describing.

Jack
Assuming it's wired correctly, this method has been around for a long time, and is well tested. The only issues I've seen are PO if it's placed poorly, or the fact that it's passive doesn't work for all effects.
Josh,

I'm wondering if I still have incorrect wiring on the jacks?

The image you pointed to for your build example isn't high enough res for me to zoom in closely to clearly analyze, but are you using CLIFF jacks for your FX jacks? If so, it would appear the BLUE the wire I have connecting my two FX jacks are NOT connected as you have your jacks connected together? Assuming your jacks are CLIFF and all CLIFF jacks are mechanically the same, am I mis-reading this?

Cheers
You do not have the required permissions to view the files attached to this post.
0 x

User avatar
JMPGuitars
Super Duper Admin
Super Duper Admin
Posts: 3954
Joined: Tue 09/18/12 8:00 pm
Location: South Central, MA
Contact:

Re: Into the rabbit hole...36 Watt EL34 Superlite TMB

Post by JMPGuitars »

Billy_Goat wrote:
Mon 03/15/21 3:36 pm
Josh,

I'm wondering if I still have incorrect wiring on the jacks?

The image you pointed to for your build example isn't high enough res for me to zoom in closely to clearly analyze, but are you using CLIFF jacks for your FX jacks? If so, it would appear the BLUE the wire I have connecting my two FX jacks are NOT connected as you have your jacks connected together? Assuming your jacks are CLIFF and all CLIFF jacks are mechanically the same, am I mis-reading this?

Cheers
They are cliff jacks, and they're wired the same as my diagram. I don't know that all Cliff jacks are exactly the same (but most probably are). Though I suspect yours are wired the same as mine anyway, but you switched the jumper wire to the outside instead of inside. It shouldn't matter since one side of the jumper should be lifted when both jacks are inserted.

Take a close look at the diagram here: files/JMPGuitars_18_Watt_Tremolo_TMB_Reverb_Layout.pdf

Both of our FX loops have the capacitors on the right side of the jack. Confirm that the right side of your jack is the latching side (which I'm guessing it is, or you wouldn't get a signal).

Thanks,
Josh
0 x
'I installed a skylight in my apartment yesterday... The people who live above me are furious.' - Steven Wright

Modern Ground Schemes
Soldering Technique
B+ Voltage Reduction
Amplifier Tools & Parts Info


Web Design: DolceVittoria.com
Guitars / Amps / Effects: JMPGuitars.com
(anti)Social: Facebook · Instagram
Items for Sale

User avatar
TriodeLuvr
Frequent poster
Frequent poster
Posts: 616
Joined: Tue 10/16/18 11:19 am

Re: Into the rabbit hole...36 Watt EL34 Superlite TMB

Post by TriodeLuvr »

Does the loop exhibit this issue if you plug in a patch cable?

Jack
0 x

Billy_Goat
Newbie
Newbie
Posts: 44
Joined: Thu 11/19/20 9:07 pm

Re: Into the rabbit hole...36 Watt EL34 Superlite TMB

Post by Billy_Goat »

TriodeLuvr wrote:
Tue 03/16/21 11:22 am
Does the loop exhibit this issue if you plug in a patch cable?

Jack
Jack,

I finally had some time to spend troubleshooting a little deeper. It appears that the amp is not really exhibiting any issues with or without the FX loop in play - at least that I can determine measurement-wise.

I hooked the amp up to my Two-Notes Torpedo Reload box in my studio and ran both pink and white noise into the amp input using a spectrum analyzer feeding into the amp input and then monitoring the output with a second instantiation of the spectrum analyzer .

I then patched a short 6" instrument cable into the FX loop ( no pedals ) and compared the no FX output against the patched FX output. While the noise fluctuates a bit which makes a direct comparison a little dicey, but I was unable to see ( large differences ) or hear any obvious differences between the signals with or without the FX loop patched in. I understand this test may not be 100% scientific, but it's all I can think of to try and eliminate any human bias.

Anyway, I'm guessing that the combination of my pedals is not playing nice with the FX loop unfortunately for whatever reasons. With my pedal board in front of the amp, all is well. I wasn't able to bring along my pedal board to the studio this trip, so I was unable to test that to make any deeper confirmations.

Top screen grab below is the signal without the FX patched in. The bottom with the FX loop patched. The frequency graph on the left of both screen grabs is the input signal being fed into the guitar amp input as a common. The right hand frequency graph on both screen grabs is the output signal.

Cheers!
You do not have the required permissions to view the files attached to this post.
0 x

User avatar
TriodeLuvr
Frequent poster
Frequent poster
Posts: 616
Joined: Tue 10/16/18 11:19 am

Re: Into the rabbit hole...36 Watt EL34 Superlite TMB

Post by TriodeLuvr »

That probably rules out oscillation, which is why I asked the question. This result still doesn't eliminate the possibility of shifted-phase signals mixing in that resistor, though. The only way to know is to plug in the effects that are creating the problem and temporarily lift the resistor out of the circuit.

Jack
0 x

User avatar
JMPGuitars
Super Duper Admin
Super Duper Admin
Posts: 3954
Joined: Tue 09/18/12 8:00 pm
Location: South Central, MA
Contact:

Re: Into the rabbit hole...36 Watt EL34 Superlite TMB

Post by JMPGuitars »

Billy_Goat wrote:
Sat 03/27/21 11:04 pm
Anyway, I'm guessing that the combination of my pedals is not playing nice with the FX loop unfortunately for whatever reasons. With my pedal board in front of the amp, all is well. I wasn't able to bring along my pedal board to the studio this trip, so I was unable to test that to make any deeper confirmations.
Ah, I see. Not all pedals work in a passive effects loop, and not all pedals should be removed from the front of the amp, even with an active effects loop.

What kind of effects were you having issues with? Try some simple effects like reverb, tremolo, or delay, with nothing else attached to them and see how it sounds.

Thanks,
Josh
0 x
'I installed a skylight in my apartment yesterday... The people who live above me are furious.' - Steven Wright

Modern Ground Schemes
Soldering Technique
B+ Voltage Reduction
Amplifier Tools & Parts Info


Web Design: DolceVittoria.com
Guitars / Amps / Effects: JMPGuitars.com
(anti)Social: Facebook · Instagram
Items for Sale

Billy_Goat
Newbie
Newbie
Posts: 44
Joined: Thu 11/19/20 9:07 pm

Re: Into the rabbit hole...36 Watt EL34 Superlite TMB

Post by Billy_Goat »

TriodeLuvr wrote:
Sat 03/27/21 11:27 pm
That probably rules out oscillation, which is why I asked the question. This result still doesn't eliminate the possibility of shifted-phase signals mixing in that resistor, though. The only way to know is to plug in the effects that are creating the problem and temporarily lift the resistor out of the circuit.

Jack
Jack,

Yeah, I understand that there may still be a phase issue that the test above wouldn't not uncover. I was looking for low hanging fruit here. Without understanding exactly how the circuit is mixing signals, I'm having a hard time wrapping my head around how to determine a phase issue outside of thinking I'd have to connect my oscilloscope somewhere before and after the 'mixing' of the signals? Even if I could prove that, I'd be hard pressed to know how to resolve.
JMPGuitars wrote:
Sun 03/28/21 6:14 am
Billy_Goat wrote:
Sat 03/27/21 11:04 pm
Anyway, I'm guessing that the combination of my pedals is not playing nice with the FX loop unfortunately for whatever reasons. With my pedal board in front of the amp, all is well. I wasn't able to bring along my pedal board to the studio this trip, so I was unable to test that to make any deeper confirmations.
Ah, I see. Not all pedals work in a passive effects loop, and not all pedals should be removed from the front of the amp, even with an active effects loop.

What kind of effects were you having issues with? Try some simple effects like reverb, tremolo, or delay, with nothing else attached to them and see how it sounds.

Thanks,
Josh
Hi Josh,

For the record, I've tried 4 separate and different wiring schemas for the FX loop so far. The last being wiring as you indicate in your build picture, Josh. Meaning, the components wired directly to each other and then shielded and shrink wrapped.

Well, as I'm sure you are aware, pedal boards are a fluid thing and mine is constantly changing... Currently my board consists of in it's signal order... trying to keep all of the time based FX's in the FX loop.

In a perfect world...

Guitar --> JHS Little Black Buffer --> Origin Effects Cali76 Stacked --> AnalogMan Sun Lion --> Chellee Ponyboy Overdrive --> Fire Custom Shop Carpe Diem Boost --> Boss RC-3 Lopper --> Amp Input

FX Send --> Source Audio EQ2 --> Neo Instruments Ventilator II --> Ten Effects Barracuda --> MXR Carbon Clone Mini Delay --> EHX Oceans 11 Reverb --> FX Receive

Since the above doesn't work so great atm, all pedals in front of the amp like this works as expected...

Guitar --> JHS Little Black Buffer --> Origin Effects Cali76 Stacked --> AnalogMan Sun Lion --> Chellee Ponyboy Overdrive --> Fire Custom Shop Carpe Diem Boost --> Boss RC-3 Lopper --> Source Audio EQ2 --> Neo Instruments Ventilator II --> Ten Effects Barracuda --> MXR Carbon Clone Mini Delay --> EHX Oceans 11 Reverb --> Amp Input.

Question for Josh or Jack - if you don't mind... for troubleshooting purposes, what parts of the FX loop can be removed without affecting the rest of the amp? Can I disconnect the two connections to and from the FX loops jacks at the terminal board and / or do I also need to remove the 68K mixing resistor?

Thanks again in advance!
0 x

User avatar
JMPGuitars
Super Duper Admin
Super Duper Admin
Posts: 3954
Joined: Tue 09/18/12 8:00 pm
Location: South Central, MA
Contact:

Re: Into the rabbit hole...36 Watt EL34 Superlite TMB

Post by JMPGuitars »

Billy_Goat wrote:
Mon 03/29/21 7:22 pm
Since the above doesn't work so great atm, all pedals in front of the amp like this works as expected...

Question for Josh or Jack - if you don't mind... for troubleshooting purposes, what parts of the FX loop can be removed without affecting the rest of the amp? Can I disconnect the two connections to and from the FX loops jacks at the terminal board and / or do I also need to remove the 68K mixing resistor?
If you remove the FX loop, and don't want to remove the 68K, you can jumper it.

You might need to research the pedals you're trying to put in the FX loop. Some pedals don't like passive loops.

Thanks,
Josh
0 x
'I installed a skylight in my apartment yesterday... The people who live above me are furious.' - Steven Wright

Modern Ground Schemes
Soldering Technique
B+ Voltage Reduction
Amplifier Tools & Parts Info


Web Design: DolceVittoria.com
Guitars / Amps / Effects: JMPGuitars.com
(anti)Social: Facebook · Instagram
Items for Sale

User avatar
TriodeLuvr
Frequent poster
Frequent poster
Posts: 616
Joined: Tue 10/16/18 11:19 am

Re: Into the rabbit hole...36 Watt EL34 Superlite TMB

Post by TriodeLuvr »

JMPGuitars wrote:
Mon 03/29/21 7:33 pm
Some pedals don't like passive loops.

Thanks,
Josh
Is it because the loop is passive, or because this particular passive loop is creating a problem unnecessarily? In addition to the possibility of mixing signals that are out of phase, this loop connects the pedal's output back to its input through the three resistors. If the pedal has a high input impedance and a lot of gain, that's a potential problem.

I'm not seeing the purpose of the 68K that's straddling the loop. The input impedance of the LTP is very high. Why can't the 68K simply be permanently removed?

Jack

Image
0 x

User avatar
JMPGuitars
Super Duper Admin
Super Duper Admin
Posts: 3954
Joined: Tue 09/18/12 8:00 pm
Location: South Central, MA
Contact:

Re: Into the rabbit hole...36 Watt EL34 Superlite TMB

Post by JMPGuitars »

TriodeLuvr wrote:
Mon 03/29/21 9:37 pm
Is it because the loop is passive, or because this particular passive loop is creating a problem unnecessarily? In addition to the possibility of mixing signals that are out of phase, this loop connects the pedal's output back to its input through the three resistors. If the pedal has a high input impedance and a lot of gain, that's a potential problem.

I'm not seeing the purpose of the 68K that's straddling the loop. The input impedance of the LTP is very high. Why can't the 68K simply be permanently removed?

Jack
I didn't design the loop, but I, and many many others have used it without issue. There are limitations to every type of fx loop. You're welcome to experiment with the loop circuit and report your findings. ;)

High gain pedals in general should stay in front of the amp. A high gain pedal should push the preamp, but would completely skip the preamp in the loop.

Thanks,
Josh
1 x
'I installed a skylight in my apartment yesterday... The people who live above me are furious.' - Steven Wright

Modern Ground Schemes
Soldering Technique
B+ Voltage Reduction
Amplifier Tools & Parts Info


Web Design: DolceVittoria.com
Guitars / Amps / Effects: JMPGuitars.com
(anti)Social: Facebook · Instagram
Items for Sale

User avatar
TriodeLuvr
Frequent poster
Frequent poster
Posts: 616
Joined: Tue 10/16/18 11:19 am

Re: Into the rabbit hole...36 Watt EL34 Superlite TMB

Post by TriodeLuvr »

Well, I ran this through a simulator last night, and the only impact I see from removing that resistor is about 1/2 dB signal loss in "straight-through" mode. And if the grid resistors were changed to 1M instead of 470K (like the original Marshall PI), that number diminishes to less than 0.3 dB. Maybe the original design is intended to always mix the pedal output with the dry signal, but that's not what I would want. Or maybe the 68K is a failsafe to keep the amp alive if the shunt in the "Send" jack fails. A patch cable would solve that though. I say take it out, nothing to lose, everything to gain. :D

Jack
0 x

User avatar
JMPGuitars
Super Duper Admin
Super Duper Admin
Posts: 3954
Joined: Tue 09/18/12 8:00 pm
Location: South Central, MA
Contact:

Re: Into the rabbit hole...36 Watt EL34 Superlite TMB

Post by JMPGuitars »

TriodeLuvr wrote:
Tue 03/30/21 10:55 am
Well, I ran this through a simulator last night, and the only impact I see from removing that resistor is about 1/2 dB signal loss in "straight-through" mode. And if the grid resistors were changed to 1M instead of 470K (like the original Marshall PI), that number diminishes to less than 0.3 dB. Maybe the original design is intended to always mix the pedal output with the dry signal, but that's not what I would want. Or maybe the 68K is a failsafe to keep the amp alive if the shunt in the "Send" jack fails. A patch cable would solve that though. I say take it out, nothing to lose, everything to gain. :D

Jack
The resistor only belongs there with the FX loop. He as asking about leaving it in out of convenience while the FX loop is removed for, I'm assuming, experimentation.

He should definitely remove the resistor without the FX loop if that's how the amp will be permanently. But if he's just waiting to experiment more, and doesn't want the resistor interfering with his signal chain without the FX loop, then a jumper is fine temporarily.
0 x
'I installed a skylight in my apartment yesterday... The people who live above me are furious.' - Steven Wright

Modern Ground Schemes
Soldering Technique
B+ Voltage Reduction
Amplifier Tools & Parts Info


Web Design: DolceVittoria.com
Guitars / Amps / Effects: JMPGuitars.com
(anti)Social: Facebook · Instagram
Items for Sale

User avatar
TriodeLuvr
Frequent poster
Frequent poster
Posts: 616
Joined: Tue 10/16/18 11:19 am

Re: Into the rabbit hole...36 Watt EL34 Superlite TMB

Post by TriodeLuvr »

JMPGuitars wrote:
Tue 03/30/21 1:36 pm

The resistor only belongs there with the FX loop.
Why? What purpose does it serve, with or without the loop? It's sitting between a 500K pot and a 1meg input at the PI. Unless there's a compelling reason we haven't discussed, it should be removed from the circuit.

Jack

EDIT: Also, the .01uF cap connected to the Send jack serves no purpose.
0 x

Post Reply