Tremelo problem

[hasserl]

I've got a problem with a Tremelo circuit I just installed onto a homebrew amp. The amp is a project that combines a 5F6A circuit and a non-trem/reverb side of an AB763 into a Deluxe Reverb size Fender style chassis. I built this for a guy that had a loaded Hoffman 5F6A turret board and a Bandmaster output trans and a Deluxe Reverb cabinet, and wanted to combine them into an amp. It all turned out really nice, a great sounding amp on both channels. But then he asked me to add tremelo. Considering what I had to work for I decided to go with a Princeton Reverb type circuit. So I incorporated that circuit into the amp. It wasn't easy, it's wired up mostly point to point as there was no room on the4 board for it. But now that all the components are installed I get no tremelo action at all.

It's hard for me to troubleshoot the system, since I really don't understand how it works. I found an older thread where Bruce Collins wrote this about the circuit:

the intensity pot is only coupling very low frequency AC from the LFO.
It is an AC voltage divider not a DC voltage divider.
Yes the intensity pot is connected to a DC supply but the pot is not grounded on the wiper end nor the the oscillator end, so the "voltage divider" action normally seen with a pot can not send the DC anywhere, just AC.
It is still just has a static negative DC voltage sitting on it.

Sliding the wiper of the intensity pot closer to the LFO couples more low freq AC to the grids of the power tubes. Lots of vibrato.
Sliding it closer to the bias voltage supply couples less AC voltage.
The AC now has to go through the enitre 250K resistance and any residual oscillator AC is shunted to ground through the bias filter cap, thus having virtually very little or no effect on the average bias voltage. No vibrato.

So, with the wiper over at the LFO end, the coupled AC creates the vibrato effect by varying the average bias voltage with the AC.
A great tubey sounding vibrato as the power tubes go from cold class AB (maybe if deep enough class B) to normal class AB and into very high idle current class AB.
By the way, since this low freq AC is sub-audible and coupled right to the grids of the power tubes, if the oscillator went any faster then the RC time constant set by Fender (higher frequency), it would become and audio signal!

Well, that helps a little, but I really don't understand what makes the circuit oscillate, and I'm struggling understanding how to diagnose why the system doesn't operate. I've gone over the installation many times and it all seems to be correct, per the schematic. I'm looking for help with understanding this system, and some tips on troubleshooting it.

Here's the schematic of the amp:

[Bruce / Mission Amps]

Does the oscillator run at all?
Can you see AC moving?
Is the .02 cap (feeding the .1uF cap to the level control) grounded?
Compare your PR schem to a tweed Vibrolux 5F11.
I just finished another one this weekend (my 6th in three weeks) and all 6 of of them fired off the first time power was applied with full/rich vibrato...
They are very similar and the oscillator modulates the bias supply in the same way.

[hasserl]

Thanks for the help Bruce!

"Can you see AC moving?"
I connected a true RMS meter to the .1uf cap going to the Intensity pot and could see AC voltage that varied with the intensity pot level. Same with the scope. But there was no tremelo in the sound.

"Is the .02 cap (feeding the .1uF cap to the level control) grounded?"
I believe so, but I will dbl check that to make sure.

I will compare to the 5F11 schem, hope they have on at the Fender Amp Field Guide.

A couple of thoughts I had that may be key here:
1. I'm using 6L6 tubes, not 6V6. I know the bias voltage needs are different for these tubes, and the PR circuit is intended for 6v6 tubes. Might the 6L6's not work with this exact circuit?

2. I'm using a mini toggle switch to toggle the circuit On/Off instead of a footswitch. I don't know why that would make a diference, but if need be I'll swap that out with an RCA jack. Still, don't know why that would make a difference. Switching the circuit to ground open or closed is the same, no matter what is used for a switch, right?

[hasserl]

Okay, comparing the two circuits, the Vibrolux and the PR, the main difference is the Vibrolux does not have the .02 cap to ground that you mentioned, or the 1M resistor between the plate and the .1 cap to the Depth pot. (The plate load and cathode resistor values are different also) I can fairly easily modify my circuit to the Vibrolux circuit and give that a shot.

[hasserl]

A little more info now. I am getting AC voltage at the wiper of the Intensity pot, but it is in the millivolt range. With the ,02 cap to ground it was at around 60omv at full intensity; with the .02 cap to ground removed (but the 1M resistor still in place) the voltage increased to around 1.50v at full intensity. The voltage drops off to zero, slowly, with the Intensity pot turned all the way down.

But, the toggle switch seems to have no effect at all, other than making the scope wave jump when I flip it. But the wave form goes right back to what it was (which is very assymetrical) before I flip it and settles right back down. The voltage level remains the same no matter which position the switch is in. I did verify that the switch is operational, and the circuit to ground is completed with the switch.

BTW, which way is "On", with the circuit to ground open or closed? As I recall Princeton Reverbs, they need the circuit closed in order to function, correct?

[hasserl]

Well, I set this one aside and laid my old Princeton Reverb up on the bench and I am getting the exact same thing out of it. The AC voltage readings at the pot wiper are about the same as the first amp with the .02 cap to ground installed, ~600mv at full Intensity.

Problem is, the PR does not have any Tremelo action either. With the footswitch connected I can switch it either way and it does not effect the AC voltage readings at the Intensity pot wiper, and I hear no tremelo action out of the speaker.

I've never had a footswitch for this amp, and have never used the Tremelo in the twenty years or so I've owned it. I took the footswitch from my Super Reverb and pugged it into the PR, no Trem.

This is getting frustrating.

[MWJB]

Check/replace the 2x 0.01 an 1x 0.02 caps in the PR tremolo circuit, as well as the cathode bypass cap for that triode.

Bias trems don't seem to like high bias currents so try biasing a little colder than you usually might, are you getting any variation in bias current with the tremolo engaged?

[Enzo]

The triode is an oscillator by feeding the signal at the plate back to the grid, but the pulses have to be delayed since the plate and grid are out of phase. The feedback loop is the three caps - .02, .01, .01. Each cap also has a resistance to ground, so think of the loop as a series of RCs to charge up. The values of the R and C determine how long it takes for the cap to charge. The larger the caps, the longer it takes, the slower the trem. The resistors as well, if they are larger, the RC takes lomger to charge. One stage has the variable resistance, so the trem speed can be adjusted.

I don't know what the point of grounding the second loop node to the cathode instead of ground might be.


The switch here grounds one point in the middle of that loop, thus killing the trem oscillation. In many others though, the switch opens one of the resistances. SO in yours, the switch must be open, but in many other Fenders, the switch has to be closed.

Ince there is a hefty oscillation going, THEN there is a tap off the strong plate signal, through the 1meg and the .1 to the pot.

There should be a strong signal at that plate, many volts. Get that happening before you worry about anything on the righthand end of that 1meg

[Satamax]

Mate, what is that 1 meg to the Two caps leading to the pot? That should be 100Ω Look at the princetown schem, it's in no case one meg!

[hasserl]

Thanks for the replies guys, that all helps. I'll check it out again later this evening. Enzo, the description of the circuit does help. Max, I believe it is 1M, even so I jumpered it last night to mimic the Vibrolux circuit with the .02 cap to ground lifted, that did not help the situation.

[Enzo]

SHows 1meg on my Princeton drawings. if you don't get strong oscillation at the plate there, then that resistor won't matter.

[hasserl]

Well, I got it. It was a combination of two things, the 1M resistor from the cathode to the junction of the two .01 caps; and a weak tube. I just didn't have a good installation of that resistor. Again, this is point to point wired, no board for the trem circuit, and it's not the neatest installation. Oh well. It's working good and strong now. The tube was an old RCA I pulled out of a drawer of tube pulls. I thought it was good, but swapping it out with a Chinese 12AX7 brought it to life.

Thanks for the help guys, I appreciate it.

[Bruce / Mission Amps]

You knucklehead!
And what about the other amp the has no tremolo?

[hasserl]

You knucklehead!
And what about the other amp the has no tremolo?

Haha I deserve that!

The Princeton Reverb with no Tremelo, I moved that aside to finish up the first one. Frustration got the better part of me the other day. I just wanted to take some voltage measurements off of it for comparison. So I still need to go back into it and get it straightened out.

BTW, Enzo was right, there is a good strong AC voltage on the plate of that Tremelo tube when it's working right. I think it varied from around 41vac to around 56vac, depending on the setting of the Speed pot. The Intensity pot didn't appear to change the voltage level.

Thanks again