• TalkBass has been independent since 1998. Add your voice.
    Create a free account to reply to discussions, view embedded media, and browse with fewer display ads.
    Join free Log in
    Want zero display ads or expanded classifieds tools? Compare plans.

Biasing mti svt, am I doing this wrong?

poorbassist15

poorbassist15

Dec 29, 2009
1,372
148
Danville, VA
www.facebook.com
#1
I'm back again, just put valve art 6550a's in my newest acquisition and I'm biasing according to procedure listed on the amp. I bias one side to 72 millivolts (reads 72.4) and between k1 and k2 to precisely 0 mv(and it's exact too). My question is the balance control. If I run it down to zero +/- .01 it has bad sounding farting distortion (I think it's what they call crossover distortion, at least compared to the YouTube videos) but if I run it to where either side at idle is near 0 on the meter and peaks during playing at 2.5 millivolts between k1 and k2 it sounds huge.


I tried doing the method on the amp, 25v output into 4 ohms and 40hz. Same thing. Sounds much stronger moving much more air to the points stated above.

Am I an idiot or is this normal? I know a little about working on tube amps, I've done it before, but this seems out of normal range and wrong. If it is, then why does it sound better out of range?
 
#8
poorbassist15 said:
Considering it's mv, it's still probably close to spec. Could this be one of my cathode followers failing? They're ge 12bh7s, and I have no idea how long they've been in there.
Click to expand...


It's not unusual to replace the drivers when you replace the 6550s. Weak drivers would make it impossible to increase the current to the desired level. The harder the driver tube conducts, the more it pulls the bias down, which increases current in the 6550s. So if you have weak drivers, you won't be able to get the current up into the proper range. The first section of the driver tubes does not have any impact on bias though. So if this section is weak the amp may still bias fine, but the first driver section may clip prematurely.

Do you understand what the balance control is doing? It's not really related to bias at all. Basically the balance adjust the two outputs of the Phase Inverter (PI). The idea is you want both sides of your push pull output section to be producing a symmetrical and proportional output.

But what happens if one side of the amp is just stronger than the other side? Perhaps by chance, three of the 6550s are stronger. Also it's not unusual for the output transformer to be slightly out of balance, so the plate voltage will be slightly higher on one side. I believe the O'scope method will be way better than the DVM method.

What I have noticed with my amps is the best Balance setting varies with frequency. For example, if I set the control for the best performance at 40 hz, it will not have the best sound at 80hz, and vice, versa. FYI @cerrem actually recommends setting the Balance control with the amp being fed a 400hz signal.
 
  • Like
Reactions: BassikBrad and bobyoung53
#9
It's slightly more towards the left side, but not by a lot, not hard over or anything. It seems like one of the drivers is stronger, or at least that was my assumption. I was confused, as it set to .2 mV is horrid sounding, it set way over to the other side is not quite as bad but still bad. I've basically set it where I think it sounds best, the multimeter method is telling me I'm doing it wrong but I trust my ears. I will check it with the scope, when I get it. I really need a dummy load, 40 hz at 25v even on 1 cab is rattling plates out of the cupboard.
 
#10
Do you know what you are doing or what you are measuring?

What you are describing doesn't make any sense.

I believe there is a sense resistor that measures the sum of the 3 plate + screen currents from each push-pull side. Measuring from MT1 (or K1) to ground should read ~72mV (adjust VR1) and MT2 (or K2) to ground should read 72mV (adjust VR2), the balance is a separate procedure. This is done with no signal, volume down, the amp should be terminated to either a resistive dummy load or a speaker. This is the direct measurement method, and verifies correct bias current from both sides of the push-pull pair. Ignore the indirect method, that was a shortcut that proved problematic.

For the balance control, I use a THD analyzer, drive the amp to 25VRMS out into 4 ohms resistive load, and set VR3 for the minimum THD.

There is an alternate null method which involves measuring the voltage from MT2 to MT2 (or K1 to K2) and setting this for zero volts AFTER doing the direct bias procedure above. If you have to adjust VR3 much, you will need to go back and repeat this entire procedure to bring the circuit parameters back within range.

You need to be VERY careful, there is dangerous high voltage throughout the circuitry (enough to kill you, or make you wish you were dead) and also if you happen to have an intermittent trim pot it's possible to cause expensive damage as well.
 
  • Like
Reactions: MaxSpinrun, DrMole, bobyoung53 and 1 other person
#11
I did everything exactly as what you've just described, as that's the procedure silk screened on the front of the power amp chassis. I know the Phase Inverter balance adjusts each side of the power section, and I also know it's push pull and that minimum THD is obviously the goal however at this moment I neither have an oscilloscope nor a way to measure THD. I'm questioning why when I adjust the PI balance pot to where it tells me that it should be at 0v +/- .01v between K1 and K2 it sounds exceptionally bad. I measure 2.5 millivolts at ~40hz when using this method, playing a bass, which google informs me is less than the .01 specified(because I forgot how many spaces to move the decimal), but when I drive it to 25 volts output, I'm getting to the .01 or close to it. I want to know if that there's an issue with my procedure or if I have a weak 12BH7, or even if this is normal and acceptable.
 
#12
poorbassist15 said:
I did everything exactly as what you've just described, as that's the procedure silk screened on the front of the power amp chassis. I know the Phase Inverter balance adjusts each side of the power section, and I also know it's push pull and that minimum THD is obviously the goal however at this moment I neither have an oscilloscope nor a way to measure THD. I'm questioning why when I adjust the PI balance pot to where it tells me that it should be at 0v +/- .01v between K1 and K2 it sounds exceptionally bad. I measure 2.5 millivolts at ~40hz when using this method, playing a bass, which google informs me is less than the .01 specified(because I forgot how many spaces to move the decimal), but when I drive it to 25 volts output, I'm getting to the .01 or close to it. I want to know if that there's an issue with my procedure or if I have a weak 12BH7, or even if this is normal and acceptable.
Click to expand...

2.5 mV is 0.0025V.

It's impossible to say where the distortion is occurring. You could try a new set of driver tubes to see if it corrects the issue. I would just set the amp for the best sound and not worry about it, especially since the amp sounds great and you are within the 10mV tolerance.

Also, I believe if you used a THD meter you would find the lowest distortion would not necessarily correspond with the 0V reading between the test points. The voltage reading is really sort of a hack way of doing it.

Doing the measurement with a signal from your bass is probably more likely to produce this sort of result than using a pure tone from a signal generator; since a bass does not produce a pure tone. As I said previous, the optimum Balance adjustment for one frequency is not necessarily going to be ideal for other frequencies. Although you would hope it would be, it has not been in my experience. Basically each side of the amp (I.E. push VS pull) will have different headroom, gain, and frequency response characteristics. So it's not really possible IMHO for the Balance control to be optimally tuned across the amp's full band width. So you basically tune it to the lowest frequency that amp was expected to reproduce; which is probably also the frequency where the amp would need to produce the most power, given the Bass control provide 12dB of boost at 40hz.

I like this online tone generator: Online Tone Generator - Free, Simple and Easy to Use.
Obviously it's not as good as using actual test equipment, but probably better than thumping on your E string.
 
  • Like
Reactions: AndyPanda and BassikBrad
#13
poorbassist15 said:
I did everything exactly as what you've just described, as that's the procedure silk screened on the front of the power amp chassis. I know the Phase Inverter balance adjusts each side of the power section, and I also know it's push pull and that minimum THD is obviously the goal however at this moment I neither have an oscilloscope nor a way to measure THD. I'm questioning why when I adjust the PI balance pot to where it tells me that it should be at 0v +/- .01v between K1 and K2 it sounds exceptionally bad. I measure 2.5 millivolts at ~40hz when using this method, playing a bass, which google informs me is less than the .01 specified(because I forgot how many spaces to move the decimal), but when I drive it to 25 volts output, I'm getting to the .01 or close to it. I want to know if that there's an issue with my procedure or if I have a weak 12BH7, or even if this is normal and acceptable.
Click to expand...
That's not what you posted earlier. You need to verify through direct current measurement of each side that both halves are in fact carrying its correct load. If not, you could be unbalancing it when adjusting P2, then really unbalancing it when trying to adjust balance.

I urge you to learn more about this before going further. You need to understand what you are measuring and why.

By any chance when measuring for balance that you were using the D.C. Scale on your meter. Iirc you need to use the AC scale.
 
  • Like
Reactions: MAXSPINRUN, Sparky Mark and IamGroot
#14
agedhorse said:
That's not what you posted earlier. You need to verify through direct current measurement of each side that both halves are in fact carrying its correct load. If not, you could be unbalancing it when adjusting P2, then really unbalancing it when trying to adjust balance.

I urge you to learn more about this before going further. You need to understand what you are measuring and why.

By any chance when measuring for balance that you were using the D.C. Scale on your meter. Iirc you need to use the AC scale.
Click to expand...

Damnit, I was on DC. I'm an idiot lol. I didn't even think about this. It's so obvious and I should have known better.
 
  • Like
Reactions: bobyoung53 and agedhorse
#16
Wasnex said:
2.5 mV is 0.0025V.

It's impossible to say where the distortion is occurring. You could try a new set of driver tubes to see if it corrects the issue. I would just set the amp for the best sound and not worry about it, especially since the amp sounds great and you are within the 10mV tolerance.

Also, I believe if you used a THD meter you would find the lowest distortion would not necessarily correspond with the 0V reading between the test points. The voltage reading is really sort of a hack way of doing it.

Doing the measurement with a signal from your bass is probably more likely to produce this sort of result than using a pure tone from a signal generator; since a bass does not produce a pure tone. As I said previous, the optimum Balance adjustment for one frequency is not necessarily going to be ideal for other frequencies. Although you would hope it would be, it has not been in my experience. Basically each side of the amp (I.E. push VS pull) will have different headroom, gain, and frequency response characteristics. So it's not really possible IMHO for the Balance control to be optimally tuned across the amp's full band width. So you basically tune it to the lowest frequency that amp was expected to reproduce; which is probably also the frequency where the amp would need to produce the most power, given the Bass control provide 12dB of boost at 40hz.

I like this online tone generator: Online Tone Generator - Free, Simple and Easy to Use.
Obviously it's not as good as using actual test equipment, but probably better than thumping on your E string.
Click to expand...

I was using a 40z tone off a generator app on my phone. Probably not ideal, but I felt it would at least get me close. agedhorse pointed out I should have been on AC and I was totally on DC. This is probably my whole problem right here.
 
  • Like
Reactions: Wasnex and IamGroot
#18
agedhorse said:
That's not what you posted earlier. You need to verify through direct current measurement of each side that both halves are in fact carrying its correct load. If not, you could be unbalancing it when adjusting P2, then really unbalancing it when trying to adjust balance.

I urge you to learn more about this before going further. You need to understand what you are measuring and why.

By any chance when measuring for balance that you were using the D.C. Scale on your meter. Iirc you need to use the AC scale.
Click to expand...

??? As far as I can tell, he followed the correct procedure. See page 6 of the attached service manual. Basically you measure between ground and K1 for 72mV then you move the probe from ground to K2 and adjust for 0V. I think it's done this way to simplify the procedure because the voltage at K1 will shift slightly as you adjust for the reading at K2. So essentially you're not adjusting for 72mV, you are adjusting for a 0V differential. This puts both K1 and K2 in the ballpark of 72mV, and at exactly the same voltage.

I don't believe the tag on the amp is specific regarding whether the Balance adjustment measurement is AC or DC, but per the service manual, the meter should be set to DC. See paragraph 3B on page 6 of the attached.
 

Attachments

Last edited:
  • Like
Reactions: AndyPanda

Latest posts

Top Bottom
Back