6G9-B Tremolux (Brownface) voltage check - running too hot

[Bruce / Mission Amps]

There should be no resistors on pins three.
The only thing connected there should be output transformer's blue or brown primary windings.
The resistor that sets the overall bias voltage for your amp is the 33K one over by your pilot light that has a 25uF@50v electrolytic capacitor across it.
It is the bias filter cap that I mentioned that should be replaced.
Use at least a 47uF@100v here.

[Bruce / Mission Amps]

Ha ha.. I never heard of using a triode mode tube like that to reverse engineer an idle current!
Guess that proves I don't know everything!! ha ha

[mlannoo]

There should be no resistors on pins three.
The only thing connected there should be output transformer's blue or brown primary windings.
The resistor that sets the overall bias voltage for your amp is the 33K one over by your pilot light that has a 25uF@50v electrolytic capacitor across it.
It is the bias filter cap that I mentioned that should be replaced.
Use at least a 47uF@100v here.

First off, thank you for your time! I have to remember to stay off forums when I'm at work recalling discoveries from the night before! sorry ... that was pin 5 with the resistor. anyway ...

The BiasRite measured:
V6 = ~58-62ma
V5 = ~47-52ma

Shouldn't these values be closer to each other? Does that indicate a failing tube?

Using the BiasRite was easy but I wanted to manually take the readings so I followed your steps and got stuck calculating the idle current -

DC resistances of OT center tap to pin 3 on each power tube:
V6 = 70.6
V5 = 78.3

Actual DCV from OT center tap to pin 3 on each power tube:
V6 = 425vdc
V5 = 425vdc

(here's where things don't seem to make sense)

Voltage divided by Resistance = current
425v / 70.6 = 6.019 (doesn't seem correct to me)

As for the bias supply filter cap ... it has already been replaced with a 100uf/100v cap parallel with a 27Kohm resistor.

Thanks again ... I'm learning a lot.
-ml

[tubeboy]

Ha ha.. I never heard of using a triode mode tube like that to reverse engineer an idle current!
Guess that proves I don't know everything!! ha ha

Bruce, I recall you mentioning this is a friendly forum for information exchange. I never claimed to know more than anyone else and trust they would put me back on track if I'm in error. To speak to an earlier suggestion that the 100K resistor would cause a lesser % of change has me puzzled. This circuit is simple voltage divider with two rc time constants. The discharge constant is 1/3 the charge constant. If AC input were doubled the output will double. As I understand this is no accident. Any increase in line voltage will increase the bias voltage and protect the tube from excessive plate power. A typical 6L6 in operating class AB will have a B+ to bias voltage ratio near 10 to 1. No calculations here, just an observation.

Chris

[Enzo]

I don't speak for Bruce certainly, but I got the impression he was patting you on the back for the insight, and deprecating himself in the process.

mlannoo
DO you really get 425 VDC from OT CT to the plates? Or do you get 425v to chassis at each point? If you actually get 425v between CT and each end, that would mean the transformer was open. Or that the plates were grounded. Your number indicate 6 amps of 400+ volts flowing, and I highly doubt your power transformer can provide that, after all it would be somethign like 2500 watts. You measure the voltage from CT to plates with your meter probes in exactly the same places as when measuring the resistance of the windings.

Given the resistances of the windings, and using your bias-rite currents, I would expect about 4 volts across each half-winding. As in 425 at the plates and 429 at the CT, or something similar.

In this case, the actual B+ voltage doesn't enter into it. yes it is an important itme and will determine wattage, but for current measurememtns all you need is the resistance of the winding and the voltage from one end to the CT., NOT to ground.

Your mis-match could be a lot of things. Are the tubes a matched pair to begin with? I don't mean matched as in both 6L6, nor do I mean same brand, I mean an actual "matched pair." Selected for their characteristics.

The tubes could be different, not necessarily failing. The different resistances of your windings enter into it. Are the bias and screen voltages at the tube sockets identical or not?

[Bruce / Mission Amps]

No, it can't be 425v.
You probably didn't really measure the voltage across the center tap of the OT to lugs 3.
We looking for the voltage dropped across the DC resistance of the OT.
Your voltmeter....Red lead on the center tap and Black lead on lug 3.
That is the number you divide by the 70 ohms.

Your bias rite must be way out of calibration or your meter is acting up.
No, maybe you don't have the tremolo shut off and that is why you see the variation in each tube's idle current.


You can solve this backwards (ignoring any other currents)...
I (current) x R = volts
say 58ma in one of those power tubes,
.058 x 71 ohms = around 4.1 volts.
And that is more like what I'd expect to see... less then 10v across either leg of the OT primary.
Reread the other reply I made... I gave you a hint of what to expect.

Anyhow.... if you are averaging around 58ma for the pair of tubes at 425 volts.
That would be +25 watts each.
Not enough to blow the glass off the tube but still too high for what I'd like to see.
Now change that 27K to 33K to 36K resistor and see how that works.

[Bruce / Mission Amps]

Huh?
Let me rephrase that....
I've been working with vacuum tubes since I was 12 years old and I never heard of considering a power tube in triode mode like that to reverse engineer an idle current... and that proves I don't know everything.
That's all.

[mlannoo]

I don't speak for Bruce certainly, but I got the impression he was patting you on the back for the insight, and deprecating himself in the process.

mlannoo
DO you really get 425 VDC from OT CT to the plates? Or do you get 425v to chassis at each point? If you actually get 425v between CT and each end, that would mean the transformer was open. Or that the plates were grounded. Your number indicate 6 amps of 400+ volts flowing, and I highly doubt your power transformer can provide that, after all it would be somethign like 2500 watts. You measure the voltage from CT to plates with your meter probes in exactly the same places as when measuring the resistance of the windings.

Given the resistances of the windings, and using your bias-rite currents, I would expect about 4 volts across each half-winding. As in 425 at the plates and 429 at the CT, or something similar.

In this case, the actual B+ voltage doesn't enter into it. yes it is an important itme and will determine wattage, but for current measurememtns all you need is the resistance of the winding and the voltage from one end to the CT., NOT to ground.

Your mis-match could be a lot of things. Are the tubes a matched pair to begin with? I don't mean matched as in both 6L6, nor do I mean same brand, I mean an actual "matched pair." Selected for their characteristics.

The tubes could be different, not necessarily failing. The different resistances of your windings enter into it. Are the bias and screen voltages at the tube sockets identical or not?

It's more than likely me not using the DMM properly. Probably me reading 425v at each point to chassis. I understand now and will try again tonight. I have a matched pair of 6L6's coming in a day or so from thetubestore and will try this all again when they arrive. Not sure why my bias meter was fluctuating so much.

[tubeboy]

Huh?
Let me rephrase that....
I've been working with vacuum tubes since I was 12 years old and I never heard of considering a power tube in triode mode like that to reverse engineer an idle current... and that proves I don't know everything.
That's all.

I guess I read too much into that. Please accept my apology. Now that I read about the 27K in place of the 33K things make more sense. I’ll be curious to hear how the new tubes work out.

[mlannoo]

Here's what I found measuring manually:

DC resistance-
V6 - 70.6
V5 - 78.3
Voltage drop across DC resistance of OT-
V6 - 4.2
V5 - 3.8
-----------
V6 = 59ma (25w)
V5 = 48ma (20.4w)

And here's what my bias meter read ... although I could have let it warm up longer:
V6 = 60ma
V5 = 50.2ma
-----------------------
My new set of 6L6's arrived and I will install them tonight and check the bias. They are a matched pair (with equal plate current of 17).

As for my replaced Mojo PT, here's what they (Mojo) had to say about the old serial number:

I have a Mojo power transformer in my '62 Fender Tremolux (brownface) ... the serial number says:

MVP BR
MOJO
February 1997

Yep that designation stood for Mojo Vibrolux Power Brown - the Vib and Trem were in fact the same and was more currently our mojo757 - which we eventually phased out because the mojo758 is the same but with a bias tap - the browns derived the bias from the HT which can still be done with the 758.
http://www.mojomusicalsupply.com/search.asp?q=MOJO758

So it looks like I have the correct PT ... which is nice.

Thanks Bruce for your help! Nice website btw ... I've got to save a couple bucks up for your 5E3 kit - looks great!

[mlannoo]

Anyhow.... if you are averaging around 58ma for the pair of tubes at 425 volts.
That would be +25 watts each.
Not enough to blow the glass off the tube but still too high for what I'd like to see.
Now change that 27K to 33K to 36K resistor and see how that works.

Installed new power tubes - kept the same bias resistor and cap from before which was a 27K and 100uf/100v. calculated that the tubes were running roughtly at 25 watts each - still a bit too high. So I swapped the 27K resistor to a 33K and now the tube are running at around 14-15 watts. that's probably a little low? I don't have any 30K resistors to try (splitting the difference). Can I parallel two resistors to achieve a bias in between? or is there some other part of the equation i can adjust to find the sweet spot?

Thank you,
Matt

[MWJB]

Well if your maths are good, 14W/425v = 33mA. Your plate voltage has probably risen slightly since you rebiased colder so if we estimate a worst case scenario and you have 15W/435v = 35mA. Perfect.

It's always a good idea to actually measure plate current & plate voltage to take the guess work out.

Yes, if necessary, you could parallel or add resistors in series with the 33K to get the bias current to your desired setting.

[3124+]

I have a '61 Tremolux 6G9-A. It is all original. The B+ measures about 430VDC even though the schematic indicates that it should be 365V. I wonder if some Tremoluxes left the factory with overwound power transformers?

mlannoo, do you find it necessary to set the volume fairly high? My Tremolux seems to have lazy volume knobs. In both channels, I have to turn up past 5 in order to fill the room. Unusual pot taper or something?

[Enzo]

One issue is mains voltage. Old days, mains were 110v, while today they are 120v nominal and often higher. 360v B+ at 110VAC will rise 3 volts for every 1 volt the mains rise. SO if the mains are 13v higher than design, then the B+ will be 39VDC higher than spec.

Also, are you taking these readings with all the tubes installed? if they are out, then the B+ will rise to well above normal.

[3124+]

My readings were taken with all of the tubes installed. I have a variac to lower the mains voltage down to 110VAC, and the plate voltages remain considerably higher than spec. I really do think that some of these Tremolux heads left the factory with out of spec power transformers.