Biasing question for the guru's -- using OT value meas.

[donzoid]

I am a total noob...so please forgive my stupidity in advance. (see my introduction).

I "am" however a pretty quick study, so want to ask all the tech guru's for input here 'cause moneys tight and I know I can suss this out with some help.

Last gig we did, my JMP MkII 50 watt dropped in volume (a third to half) all at once. Tone wasn't "that much different" but at that point she was pretty toasty so I flipped over to my backup and put her on the bench.

What I found was that the center tap for the OT coming off the diode bridge was a "completely" cold solder joint. Re soldered and now want to check my bias. I found this method on one of the other forums and want to know if its correct and accurate AND if so, why am I getting the readings I am getting.

So...here is the method (quoted):
"With the amp off measure the resistance between the center tap of the output transformer and either side of the OT primary. Write these measurements down, making sure to identify which is which. Now fire the amp up and allow it to warm up until it stabilizes. Now measure the voltage between the OT center tap and either side of the OT primary. Now divide the voltage by the resistance measured earlier, voilla! You have the idle current dissipation of the tubes. "

And of concern, using this method, here are the values for my 6550 tube pair:
(Actual Plate V = 470, have a speaker load attached)
OT center to white (pin 3 Valve4): 44 Ohm (amp off) / 1.05v (amp on)
OT center to red (pin 3 Valve 5): 44 Ohm (amp off) / .91v (amp on)

Using this division as instructed yields abt .0238 = 24 and 0.0206 = 21. Can that even run if the bias is that low?
Also the difference between 1.05v and .91v seems excessive. But both tubes test on my Senco Mighty Mite (yeah I know...I hear you chuckling) as being well on the good side and no shorts.
By my calculation these numbers should be:

35 watts (tube rating for 6550) divided by 470 plate volts equals .0744 or 74 milliamps. So .074 X .7 (the typical % dissipation as I have observed) = .05212 or 52mA.

Is this method accurate and am I really "that far" off? I would like input before I go making bias pot adjustments, really don't want to do any red-plating experiments. Oddly enough this amp has been running like this for (probably years) and had little trouble, no red plating, not even one blown fuse.

Thanks in advance for your input!
Don

[Leo_Gnardo]

"With the amp off measure the resistance between the center tap of the output transformer and either side of the OT primary. Write these measurements down, making sure to identify which is which. Now fire the amp up and allow it to warm up until it stabilizes. Now measure the voltage between the OT center tap and either side of the OT primary. Now divide the voltage by the resistance measured earlier, voilla! You have the idle current dissipation of the tubes. "

And of concern, using this method, here are the values for my 6550 tube pair:
(Actual Plate V = 470, have a speaker load attached)
OT center to white (pin 3 Valve4): 44 Ohm (amp off) / 1.05v (amp on)
OT center to red (pin 3 Valve 5): 44 Ohm (amp off) / .91v (amp on)

Using this division as instructed yields abt .0238 = 24 and 0.0206 = 21. Can that even run if the bias is that low?
Also the difference between 1.05v and .91v seems excessive. But both tubes test on my Senco Mighty Mite (yeah I know...I hear you chuckling) as being well on the good side and no shorts.
By my calculation these numbers should be:

35 watts (tube rating for 6550) divided by 470 plate volts equals .0744 or 74 milliamps. So .074 X .7 (the typical % dissipation as I have observed) = .05212 or 52mA.

Is this method accurate and am I really "that far" off? I would like input before I go making bias pot adjustments, really don't want to do any red-plating experiments. Oddly enough this amp has been running like this for (probably years) and had little trouble, no red plating, not even one blown fuse.

Thanks in advance for your input!
Don

Welcome aboard Don, and yes that's one method for reading bias current. 21 mA sure is low. If you put a sine wave thru your amp & looked on a scope, you'd see obvious crossover notch distortion. No chance of dialing up a clean tone. I'd recommend @ 40 mA for 6550's. Why there was a sudden shift in your amp's gain while on the gig, I can't be certain. Possibly those output tubes are on their last legs, but there's a dozen other reasons for an unexpected drop in power/gain.

FWIW I use the "shunt method" for reading bias current.

Or you can always wire a 1 or 10 ohm 1% "current sensing" resistor between cathode & ground, then you don't have to stick probes into hi voltage. 1 or 10 ohms are commonly used values because it makes the math easy. For instance with 1 ohm, 1 millivolt means 1 milliamp and even a cheap $10 Harbor Freight DVM will read your millivolts. Those that want another digit of accuracy might want to choose 10 ohms instead, but I've never found that necessary. Bias current read by this method will include plate current + screen grid current. Close enough for rock n roll...

[donzoid]

i am convinced that after heating the amp up for three sets of music, that cold solder joint just started letting the OT center tap (cold) solder joint flop around. That of course is now fixed. I just wanted to see if this method was correct BEFORE making an upward adjustment on the bias. I plan to install new tubes shortly anyway, just didn't want to endanger the thing. Its been a real road warrior for me and she is "close to my heart" for a lot of reasons. Thanks!

Oh, and doesn't the shunt method you suggest require lifting grounds to the chassis (replacing with the shunt resistor)? I think if it came to that, I might just invest the 50 bucks in an amp-head bias tester (which from what I've read does the same thing inside the pass-through sockets) just because I have many amps where that would have to be done and it might just save time/effort.

[xtian]

I like the OT measurements to check bias. Easy, accurate, and does not require any soldering.

[Leo_Gnardo]

doesn't the shunt method you suggest require lifting grounds to the chassis (replacing with the shunt resistor)? I think if it came to that, I might just invest the 50 bucks in an amp-head bias tester (which from what I've read does the same thing inside the pass-through sockets) just because I have many amps where that would have to be done and it might just save time/effort.

No lifting required for shunt method. Set your meter for milliamps & whatever scale will show @ 40 mA clearly. One probe goes on the OT center tap, and the other on the plate, pin 3. Then the other plate. For safety, clip probes on before firing up your amp. There's a little built-in inaccuracy because your meter isn't zero ohms through the ammeter circuit, but as long as the meter's resistance is a small fraction of the OT's winding, MOST of the current will find its way thru the meter.

Most of the cheaper bias testers I've seen insert a 1 or 10 ohm resistor on pin 8, then you read mV with your own DVM and convert to mA. Some (with built-in meters) put an "old fashioned" milliammeter into the pin 8 connexions. These last tend to cost @ $120-150. Cheaper to wire the resistors into your amp. (I use Dale 3W 1% 1 ohm wirewound resistors, about a buck apiece from Mouser.) Tradeoff is, you have to unbox your amp to measure. You could put test points (pin sockets) on the amp to overcome this.

If you have a lot of amps, or you're doing service work, the bias testers can be a safe, no-scissors no-glue method to measure bias.

[donzoid]

Thanks xtian, that's exactly why I was interested in trying this method (keeping in mind the voltages I am reading are low, but with respect to ground, can be "heart stopping" of course). And I don't mean beautiful!

[donzoid]

Thanks Leo for clarifying...the shunt method I was asking about, I have read about and watched on youtube. You are talking about the same method I went back and looked and they shunt from pin 1 (in these old Marshalls pins 1 and 8 are already jumpered). While I have no plans (ever) to get rid of this amp, I am still hesitant to punch more holes in the chassis but I have seen those external bias points in several amps and that is an excellent idea. In fact when I eventually build my TW express clone, I plan to have exactly that! Heck I guess I could put the bias pot right there too, and adjust on the fly for tube swaps between 6v6 and other types.

Thanks! I am going to run the bias up this weekend and will follow up on the thread. NO WONDER I have been dialing the gain steadily back on the pre side of the amp. As you said it was nigh impossible to clean it up and I am a volume-knob guy

[JazzGuitarGimp]

Hi donzoid, and welcome!

I've used the OT primary DC resistance, and can attest to its accuracy. One thing they don't tell you, is that the copper wire the OT is wound with, has a temperature coefficient; the warmer it gets, the higher its DC resistance. No need to get into formulae et al, just be aware of this phoenomina and make sure you test the resistance within a few minutes of measuring the voltage drop across said resistance.

Also, you mentioned you will be replacing the 6550's soon. But if the amp has been running cold since they were put into service, I'd say there is a good chance that they still have plenty of life left in them. So it may not be necessary to replace them at this point, unless you are specifically looking for the tone of Brand X tubes.

Cheers,
Lou

[Leo_Gnardo]

I am still hesitant to punch more holes in the chassis but I have seen those external bias points in several amps and that is an excellent idea. In fact when I eventually build my TW express clone, I plan to have exactly that! Heck I guess I could put the bias pot right there too, and adjust on the fly for tube swaps between 6v6 and other types.

Now you're thinkin' !!!
Drill baby drill, to borrow a phrase from "the Quitter."

I am going to run the bias up this weekend and will follow up on the thread. NO WONDER I have been dialing the gain steadily back on the pre side of the amp. As you said it was nigh impossible to clean it up and I am a volume-knob guy

Still makes me think those 6550's are "last legs" and slipping fast.
I always recommend everybody bring a spare amp. You never know, and the show must go on.

[Structo]

Thanks Leo for clarifying...the shunt method I was asking about, I have read about and watched on youtube. You are talking about the same method I went back and looked and they shunt from pin 1 (in these old Marshalls pins 1 and 8 are already jumpered). While I have no plans (ever) to get rid of this amp, I am still hesitant to punch more holes in the chassis but I have seen those external bias points in several amps and that is an excellent idea. In fact when I eventually build my TW express clone, I plan to have exactly that! Heck I guess I could put the bias pot right there too, and adjust on the fly for tube swaps between 6v6 and other types.

Thanks! I am going to run the bias up this weekend and will follow up on the thread. NO WONDER I have been dialing the gain steadily back on the pre side of the amp. As you said it was nigh impossible to clean it up and I am a volume-knob guy

Hi Don,
When using the 1 ohm resistors, you don't really need to drill any holes.
Just unsolder the ground connection from pins 1 & 8, then solder in a 1 ohm 1% resistor from pins 1 & 8 to a ground.

Be sure to leave the connection between 1 & 8 intact.

Now to check bias, simply set your meter (set to read mV or millivolts) and connect the red probe to the pins 1 & 8 side and the black probe to ground.

The 1 ohm resistor converts the current to voltage.

Ohm's law

E = R x I (Volts = Resistance x Current)

So R is 1, so any current there equals volts (millivolts in this case)

Of course this method requires the chassis to be pulled but you have to measure plate voltage anyway and it is good to measure it because the more current the power tubes pull, the lower the plate voltage will go ( not by much though).

[donzoid]

All I can say right now is...WOW. There I'll say it again...WOWEEEEEE!

I didn't get my chance to re-bias the amp until after the Friday night gig. Instead I used my 100watt and all night, was tweaking with mix just to get it dialed. We opened for Gary Small and the Coyote Brothers on Sunday, so Saturday I was "determined" to get this thing going again. Re-checked the center taps to either side of primary (just to make sure) they were the same so fired up the amp and cranked up the bias pot until the voltage across it was 1.9 (which calculates to abt. 42 mA bias value). Then of course never being able to "leave well enough alone" I noticed that the last re-tube done included a complete set of Ruby 12ax7's.

I know now, as all you experienced people do, that these are just re-brands. Who KNOWS what they are I thought. So...back out to the shop...rifle through boxes...find my box of "throwaway" 12ax7's (tounge planted firmly in cheek here and you'll see why). Ahhhh here we go. So, in V1 I put in a Blackburn Mullard (1963), V2 got a Fender special 7025 (re-branded GE old stock) and since I knew the specs were tight I put a Mesa 7025 spt (one of those mil-spec 12ax7's, my guess is its a JAN) in the PI.

Absolute Heaven. Now I actually don't have "as much gain as I'm used to having". But what I DO have, is the most amazing clarity, punch, and even harmonic balance. Our gig Sunday was in 100 degree heat, mid-afternoon. I figured if she was going to "cook" on me, it would be right there. No chance. Probably the best she has ever sounded since the MV install (25 years ago). Next steps are replace the output tubes with JJ KT88's, and start reverse-engineering the existing PPI MV to see what in the heck is actually in there and possibly change to LarMar. Can't thank you all enough. I am hoping to get some films of Sundays gig's to review how the thing sounded if so, will post up a link.