does speaker impedance effect bias setting?

[5thumbs]

I am biasing a friends old blackface Fender Super Reverb. I only have the chassis, not the cabinet. I don't have a 2 ohm dummy load only a 4 ohm. If I set bias with the 4 ohm dummy load, will that be correct or will having the higher impedance effect the bias setting?

[Tom Phillips]

For practical purposes the bias setting made with a 4 Ohm load will not change when the user connects to a 2 Ohm load.

[Leo_Gnardo]

What Tom said ^^^. If you're watching on a 'scope for crossover distortion you may reach a point where you've eliminated it for a 4 ohm load but it will show up with 2 ohms. FWIW I find Super Reverbs lose their crossover notch into 2 ohms at about 35 milliamps, so if you're at that point or a bit beyond, that's where you need to be.

[5thumbs]

Hey guys thanks for the response. My bias setting was 44.7mA? 6l6gc =30 watts so 30*.7/470 (plate voltage) = 44.7mA. Am I getting the math wrong somewhere? should I be closer to 35?

[Tom Phillips]

Side note - I'm surprised that, for an old blackface SR, the plate voltage would be that high (470V) with an idle current of 44.7 mA. Is the line voltage unusually high?

[Leo_Gnardo]

Your math is fine if you believe that 70% "rule." You'll find many of us MEFsters don't subscribe to it. Of course you can listen to the amp at both - or any other - bias setting, find if you can tell the difference. Most sensible folks want to keep the heat down, get some more life out of the output tubes. I can only guess the 70% crowd have no scopes and don't trust their ears, they choose to run the bias unnecessarily high just to make sure they're out of the crossover distortion zone.

[Justin Thomas]

Hey Leo,

Did I hear wrong, or does part of the determination of bias point have to do with where you run the amp at? As in, one bias setting might work great if you keepnthe volume on 3, but if you legitimately play on 10 all the time, the best operation point may change? I don't remember which way does what, but, I just thought I remembered that.

Justin

[eschertron]

Hey Leo,

Did I hear wrong, or does part of the determination of bias point have to do with where you run the amp at? As in, one bias setting might work great if you keepnthe volume on 3, but if you legitimately play on 10 all the time, the best operation point may change? I don't remember which way does what, but, I just thought I remembered that.

Justin

Just as a thought experiment, I'd bias a little hotter if I was always playing at 3, to keep it as much in class A as possible. If I'm grinding like crazy, then crossover won't be as obvious amidst all the clipping and I can keep the idle current lower. Now I'll have to get my bias-rite (tm) out and check my hypothesis

[Leo_Gnardo]

Hey Leo, Did I hear wrong, or does part of the determination of bias point have to do with where you run the amp at? As in, one bias setting might work great if you keepnthe volume on 3, but if you legitimately play on 10 all the time, the best operation point may change? I don't remember which way does what, but, I just thought I remembered that. Justin

Can't say I ever heard anything along that line. Sounds like one of those urban legends. Maybe this is the suburban variety. FWIW in the 80's, NYC amp gooroo wiz Harry Kolbe had an annoying habit of setting super low bias current on all the amps he worked on: "You rockers just want distortion all the time, here it is." Those who wanted a good clean tone couldn't find it with the output tubes barely out of the freezer. I s'pose Peavey and some other manufacturers follow the same idea with their 5150/JSX/XXX and cold-biased Mesas. It wouldn't matter much where you had the volume dialed, you get gritty sound all the time. OK for those who want it I guess.

Volume on 10 all the time still doesn't tell me much. Some players would just honk away full blast, the better ones use their volume controls & fingers to get dynamic variety. Unless instructed differently I set the bias just past the point the crossover notch disappears, you can see it on the scope and hear it too with a sine test wave. Add maybe another 10% for good measure and that's enough. With some amp/tube combinations surprise it's only 50% dissipation - sometimes a little less. Other amp/tube combinations I have to go 90%, AC30 anybody?

[5thumbs]

Jeez Leo, when did you "MEFsters" become such amp guru snobs, Ha!
Most sensible folks want to keep the heat down, get some more life out of the output tubes. I can only guess the 70% crowd have no scopes and don't trust their ears, they choose to run the bias unnecessarily high just to make sure they're out of the crossover distortion zone.
70% is not a "rule" as we all know when dealing with all the amp variables, just used it for the math. Notch distortion is probably the least accurate method to bias a tube amplifier circuit (not going into it here), I've tried. 45 years of working with TV and radio circuits, I started with an EICO 425 scope, then on to Tektronix analog and now digital scopes.
I was wanting to make sure OT secondary impedance mismatch didn't effect the bias point as this would present the tube plates with a different reflected primary impedance. As I said in my post, I only have the chassis, no cabinet, no speakers. I'm sorry if I offended you with my question. Wish you all the best, Tim

BTW Tom, looking on Leo's original schematic. He shows 460v at the plates, so I would assume with modern wall voltages the +10v difference would be about right.

[Dave H]

I am biasing a friends old blackface Fender Super Reverb. I only have the chassis, not the cabinet. I don't have a 2 ohm dummy load only a 4 ohm. If I set bias with the 4 ohm dummy load, will that be correct or will having the higher impedance effect the bias setting?

It depends on which bias method you are using. If it's the 70% "rule" you don't even need a dummy load, a shorted jack in the speaker outlet will be fine. If using the cross over notch method, from my (very limited) experience using a higher impedance dummy load than the amp is set for will reduce the cross over distortion visible on the scope but the amp doesn't see a fixed impedance at a single frequency. It sees the graph below so how accurate can the cross over notch method be?

[Leo_Gnardo]

If using the cross over notch method, from my (very limited) experience using a higher impedance dummy load than the amp is set for will reduce the cross over distortion visible on the scope but the amp doesn't see a fixed impedance at a single frequency. It sees the graph below so how accurate can the cross over notch method be?

Good point. And not only will the amp be seeing this curve, in real life use the amp will be delivering a complex waveform built out of a collection of frequencies, so it's looking a multiple points on the speaker impedance curve, simultaneously. Oh my what's a mother to do ? ? You could take the lowest point on that curve, use that as your impedance standard, load the amp with that & adjust bias there. Any impedance upward from there, less current is delivered, not likely to put amp into the crossover distortion zone. You've done the best you can. Typical speakers, below the resonance point there's not much clean power available from a guitar/bass tube amp anyway, you're in the low frequency "slop" zone.

70% is not a "rule" as we all know when dealing with all the amp variables, just used it for the math.

On this, we can agree. Thanks for explaining. Every now & then at MEF we meet some bright spark who learned this "70% rule" from some how-to book or fanboy website. I was not aware of the depth of your experience, which is certainly respectable.

Notch distortion is probably the least accurate method to bias a tube amplifier circuit

Please expand! Lacking a distortion meter, those of us fortunate enough to have a scope & oscillator in the workshop are pretty well stuck. I s'pose a dual trace scope could be used to subtract one trace from the other, making a rough sort of distortion indicator. In any case, I'm sure not the only one here interested in what you have to offer on this.

Offended, heck no, I got better things to get offended about, mostly not on MEF.

[5thumbs]

Sorry for the late response, I got busy! This is how Randall Aiken explains the main issues I have encountered - The Last Word On Biasing
Biasing by the crossover distortion method as indicated in the Pittman book and other literature is inaccurate because the point at which crossover distortion appears is very hard to detect and is subject to changes with load impedance, amount of negative feedback, and, in particular, with grid drive if the phase inverter is AC coupled to the output tube grids (as it is in almost all guitar amps).
Also, If you have two class AB amplifiers, one with a plate voltage of 350V, the other with a plate voltage of 550V, and set them both using this method, the amplifier with the 350V plate voltage usually will be biased too cold, while the amp with the 550V plate voltage will usually be biased too hot.

In TV/Radio amp circuits biasing is set for tube life, customers didn't want to continually replace expensive tubes. But guitarist are willing to sacrifice tube longevity for better tone. The datasheet books were all geared towards the more conservative TV/Radio parameters. The first time inside a Fender Deluxe Reverb, I was convinced the 6V6's were ready to explode at any moment as most 6v6 circuits I'd encountered to that point never had more than 350v max on their plates ;-)

[g1]

By this arguement, most Fender amps of the 'classic' era were biased to degrade the tone. I don't think this is true. 70% is very hot compared to how they came out of the factory. Same for many amps still being manufactured today.

[Jazz P Bass]

Then again, there are twp totally different 'crossover notches'.

I set bias to remove the small signal, just out of Class B distortion.

Then (I do not go here) there is the 'just before max power' kink.