SVT Phase Inverter Tube Selection ???

Surfnski · #1 02-19-2011, 02:34 PM

Hello all!

I want to get some opinions on the importance of the phase inverter tube selection for the SVT.

I plan to retube the power section of my recently purchased 87 SVT. I have a nice set of NOS 6550A's and two GE 12BH7A's, the only thing I need now is the 12AX7 for the PI.

What tube is best suited for PI duty? Any particular brand or type, long plate or short plate better for any reason?
I'm thinking I would want to use something mil spec/ruggedized.

Please let me know what you guys think.

mongo2 · #2 02-19-2011, 02:52 PM

If I was to be concerned about anything with a 12AX7 in the PI role it'd be that the two triodes are balanced.

Bassmec · #3 02-19-2011, 04:29 PM

Quote:

Originally Posted by mongo2

If I was to be concerned about anything with a 12AX7 in the PI role it'd be that the two triodes are balanced.

Vital otherwise you have more cone excursion from the center line one way.
I would see if you can find the fabulous Mullard CV4004 Or Telefunken, Ecc83 Then Rca 12AX7 then GE then Amperex and Brimar

.

Jerrold Tiers · #4 02-19-2011, 09:20 PM

Quote:

Originally Posted by Bassmec

Vital otherwise you have more cone excursion from the center line one way.

Well, NO, not really........ the transformer will not pass DC.......

The balance control will take care of gain mismatch within limits, assuming the unit you have has that. HD (the numbered run of 500) and VR do, CL etc do not.

Without it, matched is better.

username1 · #5 02-20-2011, 05:38 AM

I like to use jj long plate tubes for all the 12ax7 positions in my svt.

JOE EDO · #6 02-20-2011, 06:06 AM

a well constructed tube that can handle the voltage pounding it will get. I had my tech go though a few and we found that the chinese Penta brand with matched tiodes worked best as a PI. I have a SVT VR with GE's in the power section, so we really pushed the bias/ PI to it's limits.

lem8r · #7 02-20-2011, 06:40 AM

SVT uses one triode for PI. It is not long tail PI but cathodyne. So balanced tube in PI is useless. Better to use it in next buffer stage.

Jerrold Tiers · #8 02-20-2011, 10:20 AM

Quote:

Originally Posted by lem8r

SVT uses one triode for PI. It is not long tail PI but cathodyne. So balanced tube in PI is useless. Better to use it in next buffer stage.

Almost......

Clarifying................

There are 3 tubes involved...... The true phase inverter, part of V1, which is one section and doesn't need matched.

The next two stages are in one tube, one for each bank of outputs, V2, and V3... It is a gain stage followed by a buffer that drives the grids of the output tubes.

None of these need matching in one tube, but V2 and V3 can use matching BETWEEN tubes for units not having a balance adjustment.

Surfnski · #9 02-20-2011, 01:03 PM

Thanks for the info guys.
I have a nice matched pair for V2 & V3. I think I'm going to use a GE JAN for V1, seems like a waste since it only uses one triode...oh well.

JOE EDO · #10 02-20-2011, 01:26 PM

anyone try using 12BH7's in place of the 12au7's?

coreyfyfe · #11 02-20-2011, 03:42 PM

Quote:

Originally Posted by JOE EDO

anyone try using 12BH7's in place of the 12au7's?

I actually have the inverse going at the moment, I have 12AU7s in the place of 12BH7s in my 70s SVT. The tech who was setting up the amp didn't have a matched pair of 12BH7s but did have a matched pair of NOS 12AU7s to put in place of the BH7s that were in there which were apparently not even close to being a match. I have the 12BH7s along with a few others and I've wanted to play the tube swap game but haven't had a chance to test them to find the best match before playing around.

JOE EDO · #12 02-20-2011, 05:09 PM

Thanks corey...I've heard that the 12bh7 can draw a bit more current than the 12au7 which leads me to want to believe that they would push the power section harder. However, trade off could be the 12bh7's causing componet damage? Let us know how those 12au's hold up and your thoughts on their performance.

okcrum · #13 02-20-2011, 05:58 PM

Quote:

Originally Posted by JOE EDO

anyone try using 12BH7's in place of the 12au7's?

I'm using RCA 12BH7s from my old stock in my CL. Jerrold and I talked about it a bit, and it really boils down to availability and personal preference. There wasn't any difference in the sound, and the circuit is designed to work fine with 12AU7s.

The only real difference I could find (other than 12BH7 uses double the filament current of 12AU7) is that the 12BH7s have a higher pulse voltage rating. Both 12AU7 and 12BH7 were originally designed for vertical oscillator/amplifier use in televisions. This might make 12BH7s a bit more likely to survive certain faults in the amplifier.

coreyfyfe · #14 02-20-2011, 06:18 PM

Quote:

Originally Posted by JOE EDO

Thanks corey...I've heard that the 12bh7 can draw a bit more current than the 12au7 which leads me to want to believe that they would push the power section harder. However, trade off could be the 12bh7's causing componet damage? Let us know how those 12au's hold up and your thoughts on their performance.

He said it probably wouldn't make much difference in terms of sound, maybe a little earlier breakup, but he also said that since the current handling capacity of the AU7s is lower they may go south sooner than a pair of BH7s would but probably not. Like I said, I'd like to try to match some of the 12BH7s I have before I got playing around with them in my amp since it sounds good now, so until that point I probably won't have a chance to A/B them.

Jerrold Tiers · #15 02-20-2011, 08:43 PM

The difference is really NO difference, since the grids of the power tubes are in series with 47,000 ohms each......

The upshot of that is that the maximum possible fault current with the resistors in place is well within even the 12AU7 capabiliity. The difference is fairly small anyway, and with the resistors there is no need for the added capability.

They ARE different tubes, and i wouldn't say they had to SOUND the exact same. But there is not much "room" for differences to *make a difference* with the circuit used.

Quote:

Originally Posted by Surfnski

Thanks for the info guys.
I have a nice matched pair for V1 & V2. I think I'm going to use a GE JAN for V1, seems like a waste since it only uses one triode...oh well.

Both halves of V1 are used....... one is input amp, the second is phase splitter/inverter

Assume you mean V2/V3 are the matched ones, as they are the ones that can use the matching.

IF you have an OLD SVT, an MTI version, an SLM "SVT-HD" or an SLM or Loud "SVT-VR", there is a balance adjustment which much reduces the need for a matched pair of tubes for V2/V3.

And, of course, since the bias is separate for the two groups, that is no reason to match either..... You really only need groups of 3 outputs matched, and drivers need no be matched at all, because between the two bias controls and the balance, it's all covered.

For SVT-II, SVT-CL, SVT-2PRO, there is no balance control, and in those cases matched driver tubes may be of some advantage.

mikepaulhamus · #16 01-27-2013, 08:37 PM

v1,v2,v3 in the svt vr are all split half signals run to the power section tubes such as 12au7,12au7 and then the 12ax7 driver tube that pushes on the 6550's. that tube is important must have high gain doesn't need matched at all. mismatch actually make the amp sound kind of cool.the 2 12au7's in the power sections should be matched with each other, don't have to be but I find the amp to run easy and cooler. with biasing a little easy also. I use 6189 instead of standard 12au7's. there is more headroom and a bigger soundstage.so the svt's only need to have matched power tubes that's it. you will be wasting your time with any matched tubes beside the 6550's. I have been using svt's since the 80's they havn't change at all. the biggest way to change the sound to make it better is the v1 only put a jj 803 gold 12ax7 in v1 and magic. great tube the best ever made, nos tubes can sound muddy at times, but the jj803's rock.