5e3 Deluxe w/ Weak 6g16 Tremolo

[cjonesplay]

I've built a 5e3 Deluxe from scratch and added the 6g16 Vibroverb Tremolo (per Sluckey's schem). I've used a yellow LED and lowered the caps/resistors, per his suggested notes in the schem, and I'm getting very close to his indicated voltages, but trem is still weak.

For the amp I used Rob R's "Improved Layout" with a master vol and FB switch, but I didn't add the low power switch. Of course, I did add the tremolo and tied it to the two power tube 220K resistors and removed their connection to ground. I also created an artificial center tap, as my PT didn't have one. Pic of the insides can be provided if needed.

I'm using 370 volts DC off the rectifier to power the trem, which is just 18 volts less than Sluckey's schem calls for (388v), but I don't think that's the problem. My bias on the JJ 6V6S's is running at 95% dissipation with a 250r cathode, but I did try increasing that resistance to 470r to cool the bias, with no improvement.

I'm hoping someone can help me understand what's going on. Too bad Sluckey's not on here anymore:( Here's a voltage comparison to the attached schem:

Trem Input Voltage: 370 vs. 388v schem
Pin 1/7 Voltage:  191 vs. 196v schem
Pin 3:  1.87 vs. 1.96v schem
Pin 8:  191 vs. 204v schem
Osc Signal:  45 vs. 50vAC schem
Freq Range:  2.5Hz to 8Hz vs. 2.7Hz to 10Hz schem

[Willabe]

A red LED will give a stronger trem then the yellow LED.

[cjonesplay]

Yep, I tried a red LED but it wasn't significantly stronger. Found them both to be about the same, actually.

[roarshock]

Have you tried applying the output of your trem circuit on the cathode of v1 or on the phase inverter like the 5E9A? That seems like a worthy thing to jumper in for a try. It should be easier to wiggle the bias of the smaller preamp tubes.

[ac427v]

I get good trem with Sluckeys trem mods on a cathode biased amp. But it does not have the negative feedback switch or the Master Volume.
Could those be the culprit?

[ac427v]

JPEG version

[bmccowan]

I also had success with the LED mod on a similar cathode biased amp - a GA-16T. But when I first tried it, I had tried an LED pilot light assembly I had on hand. The assembly included a current limiting resistor. That was dumb, I know, but in my defense Worked great after I got rid of that resistor.

[cjonesplay]

No luck moving the insertion point to the PI and preamp tubes . I tried lowering the resistor connected to the Intensity pot to around 100k, but that introduces an enormous amount of thump that increases with depth and is present even when the volume is all the way down. I put a 1n4007 diode on the depth pot to help deal with this, but it's still there.

I got rid of the master volume and NFB, but it made no difference. Not sure what else there is to do here, other than live with the thump.

[cjonesplay]

I should also state I tried several different resistor values on the intensity pot, 100k was the most intense but the thump is there even with a 220k. I've checked my heater wires and other wiring, nothing I move seems to make the thump less/more intense.

[Willabe]

The trem oscillator output signal is not weak if your getting a loud thump.

[cjonesplay]

It's a tradeoff - I can make the signal stronger with the resistor off pin 8 to the Intensity pot, but the thump becomes worse as I do this. When I turn the amp off and the signal fades, the trem sounds much stronger for a moment. Not sure what that means, other than I have an issue with bias

[bmccowan]

I have no concrete suggestions here, just some ideas from past projects - mine and others.
With similar tremolo circuits, many have found the selection of the oscillator tube to be critical, swapping multiple 12ax7s before finding the right one.
Some found that upping the value of the grid stoppers on the output tubes helped the trem.
I built a 6G2 - stock circuit except I added adjustable bias. I had noted that the stock circuit is biased pretty cool. When adjusting the bias I found that there was a very sensitive trade-off. Too hot and thumping became way too noticeable. The sweet spot was not easy to find. I know your amp is cathode biased, but maybe you are right to suspect a bias issue.
There are many threads here on trem issues, and there are many solutions. So don't give up too soon.

[ac427v]

"I got rid of the master volume and NFB, but it made no difference."



I was hoping you would do that while troubleshooting the trem.  Did you then reconnect to the power tube insertion point to see if the original plan would work?
 It seems that using a known good circuit as in my Sumatra amp should work. Or use the Fender Excelsior as a model of Bias modulated cathode biased 6V6 power amp.

[Merlin]

Bias-wiggler trem relies on tube nonlinearity to work, but most audio tubes were designed to be as linear as possible, so it can be difficult to get deep trem out of them this way. Biasing your power tubes cooler should help, because it pushes the operating point further into the more nonlinear region. It may also help to use a much larger cathode bypass cap on the power tubes, otherwise the shared bias resistor tries to fight your attemps to vary the gain.

[cjonesplay]

Now is a good time to thank everyone for their helpful recommendations:) This situation is one where the collective knowledge is leading to quantifiable solutions... cheers:)

It turns out that cooling the bias does have a subtle effect on the intensity, but not enough to completely solve the problem. Upping the 250r cathode resistor to 330 is a slight improvement, but not revolutionary. 470r does make a difference, but the bias is colder than I'd like at 65%...

Per Merlin's suggestion I increased the cathode bypass cap from 25u all the way to 330u, which increases the intensity dramatically (and thump) slightly, but then I get spitty blocking distortion with the original 250r cathode resistor installed. Upping that resistor decreases the distortion and cools the bias.

Next I tried upping the 1.5k Grid Stoppers, per bmcmcowan's suggestion, up to 56k. Viola! The blocking distortion is gone, so I can keep my 250r cathode resistor (although I might switch to a 330r once I can order a 10w one), and the trem is very choppy.

I settled on a 150k resistor connected to Pin 8 of the tremolo tube. It's got some thump on higher settings, but gives the max possible range without being ridiculous. More changes might yield better results, but I hope this helps someone else with their trem problems, as my searches didn't produce any of these suggestions/results.

[bmccowan]

You are exhibiting great patience - I wish I had as much!

Next I tried upping the 1.5k Grid Stoppers, per bmcmcowan's suggestion, up to 56k. Viola! The blocking distortion is gone, so I can keep my 250r cathode resistor (although I might switch to a 330r once I can order a 10w one), and the trem is very choppy.

That suggestion originally came from PDF, I believe, when he was helping another with a trem thump issue.
Keep us posted. Trem issues are tough. I have an early GA-40 with a very odd trem circuit. It works only when it's raining.

[cjonesplay]

I’m still not 100% rid of the thump, so I’m probably going to continue working on this, but for now it’s manageable. I just hope someone else can benefit from this eventually

[shooter]

It works only when it's raining.

My Pinto only ran when is wasn't raining

[Merlin]

I’m still not 100% rid of the thump, so I’m probably going to continue working on this, but for now it’s manageable. I just hope someone else can benefit from this eventually

Some of the thump is probably caused by the output tubes not being perfectly identical, which erodes CMRR. One way to improve this would be to use individual bias resistors for the output tubes, which should help equalise their idle currents somewhat.

[Willabe]

It works only when it's raining.

My Pinto only ran when is wasn't raining.

Distributor cap probably had fine cracks in it, had that happen a few times.

You could spray it with WD-40 and that would dispel the moisture and it would start, until you got a new cap.   

[shooter]

was a cracked voltage regulator, the previous owner replaced everything else, I was living in CA at the time and it only rained like twice a year, even SNOWED once, for 3 minutes 

[bmccowan]

I once had an old Avocado Green Fiat. It only ran on Wednesdays.

[cjonesplay]

Merlin, your last comment got me thinking and I remembered this feature of the 5e3 as described by Rob R:

The standard 5E3 Phase Inverter is slightly imbalanced due to where the cathode output was tapped. The plate resistor is 56k but the cathode has a 56k and 1.5k resistors which bring its load to 57.5k. This imbalance will push one power tube more than the other and create more 2nd order harmonic distortion (which can be a good thing for the overdrive tone).

I’m wondering if that 1.5K could be causing that imbalance. Would be a shame to have to eliminate it, but I should try it just to see if that makes any difference. I’ll report back if there’s anything to report:)

[SEL49]

I’m wondering if that 1.5K could be causing that imbalance. Would be a shame to have to eliminate it, but I should try it just to see if that makes any difference. I’ll report back if there’s anything to report:)

That 1.5k is necessary to bias the tube. Don't remove it. Instead, connect the coupling cap to the junction of the 1.5k and the 56k. I don't know if this will cure the trem thump.

[mresistor]

I once had an old Avocado Green Fiat. It only ran on Wednesdays.

  a hump day Fiat