Strange oscillations in speaker... Tremolo at fault or bad tube?

[Britt Rossman]

Hey folks,

Recently put the finishing touches on a 6G2-style clone and I'm having a little trouble in our burn-in vetting... for some reason when the tremolo is turned up (depth), it is accompanied by the speakers moving back and forth to the pace of the tremolo. Now I've heard hum do this as the bias is wiggled, but to actually be causing large physical movements of the speakers? My 15" WGS speaker barely moves, while the pair of JBL K110's we chose for the cab are very sensitive and move a lot.

I've been scratching my head a lot on this and havent been able to find anything definitive, but I'll share what I can... The tubes check out fine. The bias is fixed by tapping the HV of the secondary, then is modulated (from that fixed mid-point). We are using 6L6GT's instead of 6V6 that are standard to the original 6G2 design. I'm pretty sure the speakers are okay too...

Today I heard a variation of the symptom - instead of the speakers/hum making a rhythmic "wum, wum, wum", there seemed to be two overriding pulses at the same speed but off set, making a "wu-dum, wu-dum, wu-dum"...

Totally stumped here. Any leads as to where to start looking beyond this would be much appreciated. Thanks in advance!

- Britt

PS - Here's a schematic, too. The only difference with mine other than 6L6 subs are a different tone stack and a more robust PSU (used a choke, etc.)...

http://www.thevintagesound.com/ffg/s..._6g2_schem.gif

[Jazz P Bass]

Try disconnecting the feedback loop.

What bias voltage do you have?

-35Vdc is a tad low for a 6L6 pair.

[Tom Phillips]

Check the power amp section balance too. A badly unbalanced output section can cause that problem with the 6G2 tremolo circuit. Looking for something badly out of balance not just slight variances.

[SoulFetish]

I would defer to the guys with more experience with bias modulating tremelos, however, I've seen this type of oscillation can be caused by a faulty decoupling capacitor leaking DC to grids of the following stages. Since it seem that it's dependent on the tremolo's modulating signal, I would check this capacitor for leakage/DC resistance/etc.:



Jazz & Tom, what do you guys think? I'm sure a leaky cap here would cause problems, but have you guys seen a faulty cap leak DC into an LFO signal?

[Britt Rossman]

I'll try lifting the FB, but I don't think that's it...

Bias is set around -40 last I checked, and ~445v on the plates. Tubes are well matched, but OPT yields a difference of 1-2mA of plate current from left to right. Current across 10R resistors to GND read 38.5mA to give ~17W per 6L6, which is reasonably low, but the original OPT used was for a PP pair of 5881s...

[Jazz P Bass]

Is that 38.5ma's per tube?

That's more in line with an EL34, no?

[Tom Phillips]

Britt,
I should have mentioned this first.
With an output tube bias modulating tremolo circuit like the one you are building it is necessary to set the bias such that the largest tremolo modulation amplitude (depth) does not force the output tubes into cutoff or saturation. Yes, it is a compromise, but you cannot simultaneously have a hot bias and real deep tremolo with the 6G2 circuit. I suggest that you try a cooler bias setting like ~25ma per tube and determine if that reduces the problem. Give it a try and let us know the results.
Tom

[Britt Rossman]

Thanks for the clarification Tom! That makes a lot of sense... Though I was under the impression that the bias is fairly nominal (around 60%)...

so I can either get a deep trem with cool bias or moderate bias with shallow trem? I'd prefer to get as much power as possible, and if the trem wind up being softer I think that is a more than reasonable compromise... as it stands, I thought it odd that the trem is so strong for this design...

So say we go with reducing trem strength and leave bias where it is (maybe cool it a bit...) - in order to reduce trem strength, would I just limit the maximum depth with a series resistor? Maybe just try it by ear until the problem is mostly gone? I can't think of another way to deal with it...

[g1]

You mentioned "prefer to get as much power as possible" with reference to warmer bias. Cooler bias (within reason) should not reduce power. Lots of amps are biased around 50% or even less idle dissipation.

[Jazz P Bass]

If you stop & think about it, wouldn't a 'hotter' bias lower the overall available B+ voltage?

So, technically, a hotter bias would have less voltage to work with.
Hense, a lower output.

[J M Fahey]

Hotter bias *lowers* power big time.
The problem is not somewhat reduced +B voltage, although that is a factor too, but reduced current swing capability.

Highest power and highest efficiency comes from Class B (0 idle current) or *very* small idle current .

Biasing to a fixed dissipation point is nonsense, is throwing the baby with the washing water.

NO amp design Engineer will do that, unless heīs designing straight Class A, which is not the point here, or relies on crude cathode resistor self biasing which puts severe constraints on what he can choose.

Time and again you check an amp with original Factory settings and invariably itīs "cold" ... by Internet Forum and Guru standards that is

Loudest most powerful tube amp commonly available for a given tube complement is MusicMan ... which runs as cold as can be, only a few mA idle current.
No one else in the World can touch them or even come close as far as highest RMS power at clipping using 2 or 4 6L6 or other tubes they have also used.

Musicians donīt measure power, donīt look at waveforms, most a few can do is read mA or mV , multiply by plate voltage, get a dissipation number and compare it to datasheet specs, and thereīs where the trip ends.

[Britt Rossman]

Here is my thought. If I bias to a high resistance, I let very low negative voltage to the grids, thus being most positive and closest to flowing grid current (which is bad!) This is also bad because the voltage is low which drive current high - previous bias arrangement had voltage set by two parallel resistors, and when one opened for some reason, grid current temporarily went to >100mA...

So by lowering my bias point (increasing negative voltage), I am operating at a safer point for the tube... at least this was the thought in that the tubes are literally not as hot...

[SoulFetish]

So, I don't know if I missed it somewhere, but was this problem identified and solved? It's been great to follow along and listen to some great points on biasing this circuit. The bias point ended up being the problem?

[loudthud]

The Concertina phase inverter causes the LFO signal to be imbalanced at the power tube grids because the PI's output impedance is different between plate and cathode. Easy to see with an oscilloscope. Smaller coupling caps to the power tubes might make a small improvement. Try 0.022uF. An oversize OT exacerbates the problem. There will always be some imbalance. Live with it.

[Britt Rossman]

Loudthud - I'm not sure I was clear in the OP.... I'm not worried about a imbalance or a little hum (my favorite pickups are unpotted p90s )...

I fear the problem is that on these sensitive old JBLs, the problem has been exacerbated to the point where the speakers may be in danger of overexcursion, and subsequent voice coil damage.

Soul - I'm sorry but not yet... 2nd year of grad school started last week, so I'm trying to stay on top of those things... I'll definitely post when I can get back into her.

In the mean time, I'll post some pics!
EDIT::
or just one....