Hi everyone, I have a question regarding testing channel balance in my stereo tube amp(s) or any amp for that matter.
Over the years I have owned several amps and one of the things that I am always listening for is channel balance/whether one channel is louder than the other.
I own a rebuilt Dynaco ST-70 (with new MultiCaps and new resistors all around) which I recently retubed and re-biased. I couldn't help but notice that even with all this, one channel seemed louder and more dynamic than the other, but I could never tell for sure, as I always doubted my hearing, or assumed it could have been my room acoustics or a problem somewhere else in the signal chain.
I finally acquired a two-channel oscilloscope from a friend and decided to test both channels on my amp through an online signal generator. I'm not an electronics expert and I don't know what half of the controls on this scope mean, but I managed to hook the probes to the output(s) on my amp and fiddle with the knobs till I got two nice looking waves.
Indeed, one wave seemed larger than the other (greater amplitude), regardless of which frequency I fed into the amp.
Assuming that all the caps, resistors, and tubes are functioning as they should (which I have yet to verify), does this indicate a serious imbalance in my output transformers?
I also tested a couple other amps I had around the house and all of them produced slightly different waves on each channel.
Is it simply a reality with these older amps that the balance will not be perfect? I have always kept my balance on my preamp dead center with the assumption that everything is working identically on both channels, but perhaps it exists for precisely this reason? Or am I wrong here.
I hooked up my preamp and adjusted the balance visually to compensate until I had the same sized wave on both sides - the balance knob now rests about 10-15 degrees off-center. I'm hoping this is a temporary solution, but again perhaps this is simply the reality of older equipment (as I said, the two other amps I had also produced slightly different waves on each channel).
I'm hoping someone on here could shed some light. Perhaps my methodology is flawed? Maybe I will take a snapshot of the settings on my scope and maybe someone can point out some obvious problems in my test setup.
Anyways, any advice would be appreciated.... thanks in advance!
Over the years I have owned several amps and one of the things that I am always listening for is channel balance/whether one channel is louder than the other.
I own a rebuilt Dynaco ST-70 (with new MultiCaps and new resistors all around) which I recently retubed and re-biased. I couldn't help but notice that even with all this, one channel seemed louder and more dynamic than the other, but I could never tell for sure, as I always doubted my hearing, or assumed it could have been my room acoustics or a problem somewhere else in the signal chain.
I finally acquired a two-channel oscilloscope from a friend and decided to test both channels on my amp through an online signal generator. I'm not an electronics expert and I don't know what half of the controls on this scope mean, but I managed to hook the probes to the output(s) on my amp and fiddle with the knobs till I got two nice looking waves.
Indeed, one wave seemed larger than the other (greater amplitude), regardless of which frequency I fed into the amp.
Assuming that all the caps, resistors, and tubes are functioning as they should (which I have yet to verify), does this indicate a serious imbalance in my output transformers?
I also tested a couple other amps I had around the house and all of them produced slightly different waves on each channel.
Is it simply a reality with these older amps that the balance will not be perfect? I have always kept my balance on my preamp dead center with the assumption that everything is working identically on both channels, but perhaps it exists for precisely this reason? Or am I wrong here.
I hooked up my preamp and adjusted the balance visually to compensate until I had the same sized wave on both sides - the balance knob now rests about 10-15 degrees off-center. I'm hoping this is a temporary solution, but again perhaps this is simply the reality of older equipment (as I said, the two other amps I had also produced slightly different waves on each channel).
I'm hoping someone on here could shed some light. Perhaps my methodology is flawed? Maybe I will take a snapshot of the settings on my scope and maybe someone can point out some obvious problems in my test setup.
Anyways, any advice would be appreciated.... thanks in advance!
OT mis-match is unlikely. OT ratio goes by turns-count which is usually spot-on.
The old Dynacos did not use high-precision resistors in the NFB network, which will directly skew the gain.
Internal faults causing large loss of gain will show small loss of gain under NFB.
There's too many things could be mis-matched. More testing needed, perhaps beyond your present skill/confidence level. Especially as some test points are High Voltage.
The old Dynacos did not use high-precision resistors in the NFB network, which will directly skew the gain.
Internal faults causing large loss of gain will show small loss of gain under NFB.
There's too many things could be mis-matched. More testing needed, perhaps beyond your present skill/confidence level. Especially as some test points are High Voltage.
Thanks for the tips! As long as something isn't catastrophically wrong with my amp (ie dying transformer), I'm not afraid of getting my hands dirty to fix the problem.
I was under the understanding that transformers could "fade" with age because of something I read ages ago, but your reasoning seems sound to me.
I guess my next logical step would be to replace said NFB network resistors with higher quality, more precise resistors? Or would the circuit itself need to be modified? Replacing a resistor or cap or two wouldn't be out of my scope of abilities, but modifying the circuit itself might be. Also, could you point me to specifically which resistors you are referring to? I understand very generally what each part of a tube amp does and can identify most parts, but when it comes to the specifics ("NFB network") I will admit I am pretty clueless.
I was under the understanding that transformers could "fade" with age because of something I read ages ago, but your reasoning seems sound to me.
I guess my next logical step would be to replace said NFB network resistors with higher quality, more precise resistors? Or would the circuit itself need to be modified? Replacing a resistor or cap or two wouldn't be out of my scope of abilities, but modifying the circuit itself might be. Also, could you point me to specifically which resistors you are referring to? I understand very generally what each part of a tube amp does and can identify most parts, but when it comes to the specifics ("NFB network") I will admit I am pretty clueless.
You should not assume all components function as they should. Nor should you replace any components randomly. Instead measure the signal levels at the different points of the amplifier and compare the results of both channels.
The test points could be:
- grid1 of voltage amplifier
- anode of voltage amplifier
- anode and cathode of phase splitter/or grid1 of output tubes
- anodes of output tubes
- speaker output
When comparing both channels, you will see where the gain difference takes place.
When a 're-XXXed' unit misbehaves (where XXX can be caps, resistors etc.) the most likely cause is a mistake in the 're-XXX' process. This is why such processes are often a bad idea. They are a poor way of fixing faults (for that you need fault tracing) and often introduce new faults.
Check the feedback resistors. If OK, disconnect feedback and check gain as described by artosalo.
Check the feedback resistors. If OK, disconnect feedback and check gain as described by artosalo.
On a quick read of your post, I have not seen any indication that your 'scope is accurate!
Never believe what a test instrument is saying unless you verify it. There could well be a noticeable difference between the channels.
A simple way to check this is connect both probes to the same signal. The traces should then be the same. Many scopes have an Invert switch on the channel inputs and an Add mode for the inputs. This displays the difference (Add + Invert one channel) between them. If both signals are the same amplitude you should see a straight line.
Also search for "probe compensation" to ensure that both probes have similar frequency responses.
Never believe what a test instrument is saying unless you verify it. There could well be a noticeable difference between the channels.
A simple way to check this is connect both probes to the same signal. The traces should then be the same. Many scopes have an Invert switch on the channel inputs and an Add mode for the inputs. This displays the difference (Add + Invert one channel) between them. If both signals are the same amplitude you should see a straight line.
Also search for "probe compensation" to ensure that both probes have similar frequency responses.
Thanks for all the replies, everyone.
I did try simply using one channel/probe on my scope and moving it from one channel of my amp to the other, and the amplitude difference was still apparent... and from what I could tell with my eyes it was the same "difference" observed when using separate probes on each channel... so I have a feeling it's not a scope calibration issue but I will try what you suggested and report back later today!
I did try simply using one channel/probe on my scope and moving it from one channel of my amp to the other, and the amplitude difference was still apparent... and from what I could tell with my eyes it was the same "difference" observed when using separate probes on each channel... so I have a feeling it's not a scope calibration issue but I will try what you suggested and report back later today!
This kind of fault can be located with single channel/probe.
Just analyze both channels separately but with the same input level.
Just analyze both channels separately but with the same input level.
The Dynaco ST70 60s vintage PCB is not that robust to begin with. I would isolate where the imbalance is before start replacing stuff. For example, you can check the NFB signal voltage with your scope at the 1K NFB resistors to see if they matches (front & back). You can check the four 0.1uf phase splitter output capacitors at the front end PCB. You can try swap the two 7199 if they are not balance. Swap the EL34s and etc.
- Status
- Not open for further replies.