Tube Amps / Music Electronics
For current discussions, please visit Music Electronics Forum. The sunn still shines online!

ampage archive

Vintage threads from the first ten years

Search for:  Mode:  

Re: 6V6GT troubles

8/26/2000 6:37 PM
Peter S
Re: 6V6GT troubles
I have a good Tektronic scope, it's a little bit of overkill for working on tube amps, but it's certainly sufficient. I've tried biasing amps many different ways over the past 30 years or so, but I have the best way is to measure the plate current. I think it's good to use a scope and a signal generator as long as you use it in conjunction with measuring the idle current. BTW, measuring the cathode current with a probe is fine, because most tubes are only drawing a couple ma of screen current at idle. You can compensate for this if you want to, but it's really not neccessary. Using a scope alone is very unreliable and almost always results in a cold bias. Using a scope to see what's going on is a good thing for begginners to do because it gives you a graphic representation of how the different current settings affect the waveform on the scope.  
Peter S
8/24/2000 2:21 PM

Hi Randy!  
I seen a lot of talk on the subjekt 6V6s her on bbs.  
But it seems that one factor has not been mention in the discussion: many 6V6s ar not really 6V6s.  
It has been suggested that some Russian and Chinese tubes marked as 6V6 are actually a 6EY5 or similar.  
Regardless if Sovteks 6V6s are true 6V6s or not,  
I have been using this tubes with out any problem with voltages up to 440v, BUT with 1K screen resistor.  
6V6 in generall seems to benefit greatly from a increased screen circut resistance.  
Heck Jim Kelly was using 480 volt on his amps with a 2K screen resistor, and it`s still the best "clean/sustain" amp I ever played.  
Owe "OJ" Johnson  
8/25/2000 1:46 AM
Randy Jamz

My Grandpa used to have a saying; "Don't open a can of worms unless you intend to fish for trouble." I knew everyone would have a knee-jerk reaction and support plate current measurement. Some fads take longer to run their course.  
I won't waste too much space here defending my own ideas, but I will say;  
1)I have tracked down many an old time audio man and even a handful of classic guitar amplifier manufacturers during the last 20 years. I was learning then (still am now) and asked many questions.  
2)To a man I was told nobody ever measured plate current in any amplifier they ever built. I doubt Fender or Marshall took production time to do it either.  
3)Does that make current measurement wrong? No. I remember the cheesy Cesear Diaz video with his pearls of wisdom on biasing. Set it at -52VDC and the amplifier is biased 'right'. The moral? Everyone does it different, and everyone's method is right for them.  
4)Audiophiles started the plate current measuring fad in an attempt to balance the output tubes. Since I don't like a Hi-Fi guitar amplifier sound, I don't bias it this way. I bias it the way it was done 100 years ago at Fender/Gibson/Marshall/et. al. because I like the tone of those amplifiers.  
Oh well. Bias it how you like. I should have just shut up and 'live and let live'.
8/25/2000 3:58 AM

Randy, do you have an amp handy that uses negative feedback and has an adjustable bias supply?  
Play with the bias voltage and adjust it via the O'scope cross over notch method but with and without the NFB loop connected and log the idle current results and the clean ouput power levels.  
Then try the same thing, with and without NFB, by using the 50%-60% plate dissipation method ignoring the O'scope and crossover notch.  
Really play the amp after both methods, trying not to be biased (no pun intented) and listen to the amp set up either way.  
See what you think.  
Main thing is to listen to them and adjust idle current accordingly.  
BTW, I've heard from more then one assembly line builder too and have been told they did set the rough bias voltage at a predetermined negative votlage but many companies (maybe all of them) then used the OT shunt method to set the power tubes idle current but a bit lower then the +50% dissipation level we do now.  
But in those days I think it is safe to assume the power tubes were more reliable and closer matched from the factory plus they seemed to be randomly closer in all specs then any of the current crop.  
IOW... not too many tubes that were even 10ma apart randomly picked up out of tray.  
But I bet the real sleeper amps of the day were the ones that were biased closer to 50% of dissipation rating and the power tubes, accidently, were very closely matched.  
For my money, if the power tubes are within  
3ma-5ma, they're a good set of tubes.  
I don't waste a lot of time balancing.  
8/25/2000 8:42 PM
Reid Kneeland

" I should have just shut up "
Try that next time. Or at least post some real reasons why the scope method is better, instead of folklore and condescending remarks about "fads". If you really do it because you think it sounds better, than why not bias by ear, and then check the plate disippation to see if you're going to shorten the tube life? Sound and tube life are the main (if not only) concerns when setting bias. The scope show you neither.  
P.S. Going by bias voltage alone IS idiotic, and very few of us would dispute that.
8/26/2000 1:04 AM
Randy Jamz

I've heard the old NFB loop factor into biasing arguement and don't get it at all. So I did the experiment just about two months ago with a Tweed Deluxe clone that was setup class AB. Having the feedback in or out made no difference as to the final bias voltage when I biased it 'right'. I used a radomly selected NOS GE 6V6 and an unmatched NOS Westinghouse 6V6 as well as similar brand (but not matched into pairs) 6V6s from Mazda, Philips, Marconi, and EH. Of course the totally unmatched pair didn't look right on the 'scope (if you know what to look for) and I'll admit the current draw would have shown this too. But guess what; of course no feedback makes the sine wave clip at the same level that the feedback removed...DUH! But pull back the volume control and here comes the hard part...BIAS THE OUTPUT TUBES PROPERLY and the final C- voltage and plate current (when I measured it) didn't change more the 1 volt or 1 or 2 mA. BFD. I'm sure on paper the NFB theory works great, but on paper the bumble bee isn't supposed to be able to fly.  
If the 'scope isn't showing you sound or tube life (it's a stretch, but I suppose it might be able to show tube life in a roundabout way) when you set your bias, learn how to read a 'scope. End of pissing contest.
8/26/2000 1:18 AM
Ken Gilbert

"I'm sure on paper the NFB theory works great, but on paper the bumble bee isn't supposed to be able to fly."
listen, it's a MATTER OF FACT that NFB around an output stage decreases crossover distortion. it's a matter of fact because crossover distortion is distortion introduced by the circuit itself, and as such will be reduced. it works for tubes, it works for solid state.  
so, that said, if the NFB is messing around with the XOD, how do you know what the bias is doing to it?  
"If the 'scope isn't showing you sound or tube life (it's a stretch, but I suppose it might be able to show tube life in a roundabout way) when you set your bias, learn how to read a 'scope. End of pissing contest."
ooooooo--i guess you told him! well reasoned reply, excellent logic and command of the language. you flamed his ass good!  
...about as good as the flames from the tubes as you attempt to set idle current with an oscilloscope. since you seem to one who learned how to read a scope, why don't you enlighten us a bit as to how you could have ANY idea what the DC current is on the primary side, given that the OPT secondary does not respond to it at all?  

<<First Page<PrevPage 2 of 7 Next> Last Page>>