balancing the PI via trimmer?

[12AX7]

I have 2 questions regarding doing this. Note that i do understand that the PI tube will throw it off if i change to another tube, but thats why i wanna do this. Because I will never know if it's in balance and when i change the PI tube i can readjust. I simply want to see how PI balance affects loud volume levels.
1)-is the goal of perfect in the PI actually done there to assure the power tube grids see the same voltage, and if so then i assume i can adjust the trimmer by reading ac voltage at the 6V6 grids with a test signal inserted ?

2)-anything special i should know about the trimmer i use, and also what value do you think is going to have enough range?

[HotBluePlates]

1)-is the goal of perfect in the PI actually done there to assure the power tube grids see the same voltage, and if so then i assume i can adjust the trimmer by reading ac voltage at the 6V6 grids with a test signal inserted ? ...

Yes, comparing the driving voltage to each half of the output stage is a means of checking for balanced driving voltage.

The goal of a balanced driving voltage is maximizing clean output power while minimizing distortion of the output section.  If you're not seeking maximum volume or maximum clean, a perfectly balanced phase inverter is of dubious value.

2)-anything special ... about the trimmer ... what value do you think is going to have enough range?

Depends on what phase inverter circuit you're using.  All of the major types using a tube (paraphase, split-load, long-tail) are either self-balancing/balanced by the design, or have a self-balancing variant.  Only the non-self-balancing paraphase would be very likely to develop mismatched drive signals and a bit more distortion.

Adding a trimmer to any common long-tail or split-load inverter is most likely going to have the effect of creating much more imbalance than there is in the stock circuit.

[12AX7]

It's a typical LTP like a marshall 800 etc. I just find a lot of amps tend to get a much harder attack and lose some of that sag and hair as you turn them up and i thought by getting the PA as clean as possible it might allow the preamp tone to shine thru more rather than being colored to the point of overpowering the preamp distortion i spent so much time perfecting. I could be wrong but i wanted to find out myself how it affects it. I checked the 6V6 grids with a signal injected and with the master about where i have it at stage volume they read about 10Vac apart with no tubes installed. One at 64 the other 74. I noticed the higher the master the further apart they are. they were only a couple volts apart at what i'd call loud home volume.  Tubes are within less than 1mA apart.

[HotBluePlates]

... I checked the 6V6 grids with a signal injected and with the master about where i have it at stage volume they read about 10Vac apart with no tubes installed. One at 64 the other 74. I noticed the higher the master the further apart they are. they were only a couple volts apart at what i'd call loud home volume.  ...

Sounds like something is wrong with the master volume you're using.  Is this a dual-pot master?  The ganged pot could be crap with mismatched sections.

Or you could be seeing the effect of grid current, though that seems less likely.

[SILVERGUN]

Here's some info on how to set it, but also some insight into why:
http://ampgarage.com/forum/viewtopic.php?t=18048

[12AX7]

Theres nothing wrong with the amp or it;s MV. I'm simply saying most tube amps I've played loud tend to sound a bit different than i would hope when cranked up with respect to the very few i've played that sounded like god when cranked. So i wonder if the right balance, be it balanced or OUT of balance would make them all get that god like cranked tone. I wanted to experiment is all.

But heres a question after reading all the info in those links in the last post. When those of you who have tried this and were able to get an amp sounding more glorious when cranked, at the point it sounded it's best loud did you find the best setting to be with the balance IN or OUT of balance and if out, by how much?

[SILVERGUN]

For me, this was just about enhancing the amps ability to produce harmonic feedback...
It did not seem to add or detract from any "god like cranked tone", and in fact the base tone did not change at all afaict.
I'm talking, moving the pot 1/8 of a turn at a time to find a sweetspot where it would be more inclined to feedback.
Once I found that spot, I didn't even take voltage measurements because the objective had been achieved.
I would like to imagine that it was as close to perfectly balanced as possible, but wouldn't really know unless I scoped the output.
You have to take into account the fact that even if you have matching output tubes (at idle), and you are feeding them a perfectly equal signal, that the balance will change as the output tubes put out more power and eventually, naturally drift apart some as the current changes with increased output.
 
I chose to make mine more accessible and mounted a 10KL bias pot on the rear panel

[sluckey]

What kind of MV do you have?

[12AX7]

I've tried several types but just went back to the typical pre PI because when the PI get's slammed with a unattenuated signal right from the TS i don't like it. I would never need to have the pre PI master that high.

[sluckey]

OK. The reason I asked was because if you were using a dual pot post PIMV the perfectly balanced PI outputs would just get unbalanced every time you turned the MV, due to inaccurate pot elements and tracking.

Here's how I would set up for a balance adjustment... Use a 100K for one PI plate resistor. Use a 75K and series connected 50K pot for the other plate resistor. This will allow you to vary this plate resistance from 75K up to 125K. Hopefully that's enough range to perfectly balance the two outputs.

Feed a test signal sine wave into the amp and adjust the pot for equal amplitude signals at the grids of the output tubes. Recheck the balance for several different amplitudes of test signal. It's possible that the balance may change depending on the test signal level.

I would not use a small trim pot for this. I don't even feel good about putting plate voltage on an Alpha pot. I would trust a high quality 2 watt mil spec pot such as PEC.

Let us know what you think about having a perfectly balanced PI in a Marshall style amp.

[12AX7]

Yeah, like i said when i tested the 6V6 grids they were off by about 2v at  lower volume and with the master where i use it at the highest they were 10v off. If i happen to have a small enough trimmer i'll try it today. If not it'll have to wait till the next time i get to the elec surplus locally.

[SILVERGUN]

When I think "god-like" tone in a high gain build I don't think 6V6 and...
 

I've tried several types but just went back to the typical pre PI because when the PI get's slammed with a unattenuated signal right from the TS i don't like it. I would never need to have the pre PI master that high.

If you haven't already tried it, I would suggest subbing in a 12AT7 for the PI (last 2 high gain builds I did wound up with AT7)
It will take a bigger input signal before clipping your already clipped signal coming from your high gain preamp.
This should allow you to open up the master a little more and get closer to god.
And besides clipping less, it will also have lower gain itself, which should allow you to open the amp up more....with 6V6's you want less grid V anyway. (which could be the cause of grid current that HBP eluded to earlier..)
I truly believe that this would get you closer to heavenly tone than the balance control.
I would raise the value of the shared cathode resistor on the AT7 to 820R or 1K and give it a listen.

[sluckey]

A 47K resistor and 100K pot would work fine also.

[12AX7]

I tried it but to be honest i didn't hear much. I used a 56k and 50k pot. I can't crank it up here but for a few seconds here and there, but heres what i found. I got balance with a total of 64k on the input side where the 82k resided initially. Once i had balance at the 6V6 grids with the master well up, i removed the 1k source and plugged my guitar in. No difference. Not saying the amp isn't good sounding, but there was no better tone that i could detect. I measured the pot and resistor in total and found the areas on the pot (yes, used a pot, had no 50k trimmer)  that were 64k where it was balanced and 82k where the amp initially was. I couldn't keep the high volume testing going but i turned it up as high as i can here and tried nudging the pot thru it's range from to low of 56k to the ful range of the pot so 106k. I heard sight differences in the highs and smoothness but all pretty subtle and at a certain point, i think the lowest, it sounded a bit thin and weak. But at 82k it was about as good as any.

So back to the 82k resistor, as i think theres no real difference to be had with this amp at least. As to the notion of using a 12AT7, i have, and at home volumes (where i play the most by far, only occasionally play out)  at least i didn't like what it did tonally. Maybe next time i play out i'll bring one and try it at stage volume and see how that goes. But as far as balance, i don't think it matters in this amp. granted i only cranked it for seconds, but i could tell there was no change.