Un-Balance the P.I.

General discussion area for tube amps.

Moderators: pompeiisneaks, Colossal

Post Reply
JammyDodger
Posts: 120
Joined: Thu Aug 09, 2007 7:44 pm
Location: Hangtown, CA

Un-Balance the P.I.

Post by JammyDodger »

Hey All,

It seems most amp designs attempt to provide a balance signal to the Phase Invertor. I have read somewhere recently about intentionally un-balancing the P.I by using different gains. I would think that this would produce an asyametric waveform. Not necessarily the best sound.

Has anyone tried this? Any comments?

Cheers, Mike
The Jammy Dodger
User avatar
stoo
Posts: 156
Joined: Thu May 25, 2006 3:46 pm

Re: Un-Balance the P.I.

Post by stoo »

I think all attempts to totally balance the output (P.I. and output tubes) is more the realm of "HiFi" and not as important in Guitar amps. Unless you going for ultra clean tones.
stew
User avatar
LOUDthud
Posts: 464
Joined: Tue Sep 25, 2007 9:27 pm
Location: Texas

Re: Un-Balance the P.I.

Post by LOUDthud »

It actually sounds better because it's imbalanced. A perfectly balanced square wave sounds nasal.
User avatar
David Root
Posts: 3540
Joined: Fri Aug 04, 2006 3:00 pm
Location: Chilliwack BC

Re: Un-Balance the P.I.

Post by David Root »

For MI amps, less than 10% imbalance is pretty clean. Voltage imbalance for an LTP is measured by V Pin 1/V Pin 6 (for a 12A*7 tube), IIRC.

Hi Fi LTPs are usually constant current sourced (CCS) by a BJT or a tube triode with a smallish cathode resistor to ground, and usually run a lot more plate current because of that.
User avatar
odourboy
Posts: 796
Joined: Fri Apr 27, 2007 4:07 pm
Contact:

Re: Un-Balance the P.I.

Post by odourboy »

We recently had this discussion over in the Dumble forum and the expert concensus was to perfectly balance at the OT output (using an accurate scope into a dummy load) gave the best sounding setup.

I tried this on my amp and using a spectrum analyzer, noted that this point closely corresponded where I was also getting the highest production of 2nd harmonics. I'm now a believer.
"Let's face it, the non HRMs are easier to play, there, I've said it." - Gil Ayan... AND HE"S IN GOOD COMPANY!

Black chassis' availble: http://cepedals.com/Dumble-Style-Chassis.html
drz400
Posts: 509
Joined: Mon Nov 06, 2006 4:53 pm

Re: Un-Balance the P.I.

Post by drz400 »

odourboy wrote:We recently had this discussion over in the Dumble forum and the expert concensus was to perfectly balance at the OT output (using an accurate scope into a dummy load) gave the best sounding setup.

I tried this on my amp and using a spectrum analyzer, noted that this point closely corresponded where I was also getting the highest production of 2nd harmonics. I'm now a believer.
I agree, I have tried both and balanced sounds better to me, fatter and punchier. I just looked for symetrical clipping
User avatar
VacuumVoodoo
Posts: 924
Joined: Fri Feb 17, 2006 6:27 pm
Location: Goteborg, Sweden
Contact:

Re: Un-Balance the P.I.

Post by VacuumVoodoo »

David Root wrote: Hi Fi LTPs are usually constant current sourced (CCS) by a BJT or a tube triode with a smallish cathode resistor to ground, and usually run a lot more plate current because of that.
Not neccessarily, you set the CCS for whatever plate current you want. The plate current you need is dictated by the PI tube you choose: 12AX7 will be fine with less than 1mA per plate while 12AT7 will want up to 4mA.

To perfectly balance a LTPI you need two trimming points, one for DC balance to set identical voltage on each plate and one AC balance trim to set for equal signal amplitudes on each PI output. CCS will give you almost perfect DC balance with equal plate resistors so youwill only need an AC balance trim.

Given a balanced PI you have to balance the PP output stage. Here we have tube matching and OT symmetry to deal with. Tubes matched only for idle current and not for transconductance will require the PI to be AC unbalanced in such a way as to compensate for differences in power tubes transconductance i.e. to produce symmetrical onset of clipping.

This creates a dynamically varying distortion spectrum: due to intentional PI AC imbalance you will be getting even harmonics in the power stage when it's not clipping and a lot of odd harmonics once it clips. That's where that swirly compression (or whatever you want to call it) comes from when you go in and out of clipping. Some like it, some hate it.
Aleksander Niemand
------------------------
Life's a party but you get invited only once...
affiliation:TUBEWONDER AMPS
Zagray!-review
User avatar
David Root
Posts: 3540
Joined: Fri Aug 04, 2006 3:00 pm
Location: Chilliwack BC

CCS PI

Post by David Root »

I stand corrected, it does depend largely on the tube type. You wouldn't try to put 4 mA thru a 12AX7, for example.
Post Reply