Phase inverting preamps

netaron · « **on:** 14 Mar 2007, 05:14 pm »

Hi all,

I have a tube preamp with only one 6DJ8 tube for gain. From what I understand, this makes it a phase inverting preamp, the manual states to switch polarity at speaker end to accommodate the phase inversion. I have tried it both ways; it seems that the top end opens up slightly when phase is corrected at the speaker inputs. But I have been forgetting about this issue when changing equipment around. Just curious why the preamp designer didn't account for this within the preamp, is it as easy as changing polarity the wiring inside the preamp outputs to correct for this?

Thanks,

Haron

Geoff-AU · « **Reply #1 on:** 15 Mar 2007, 03:10 am »

not really, as the pre-amp output is a signal and a ground. reverse this and the signal travels down the outside of the RCA, removing all shielding benefit and letting more noise in. aside from that it might just flat out not work

the reason you should do it at the speakers is because speakers are isolated and don't care what a ground reference is they just have two terminals.

netaron · « **Reply #2 on:** 15 Mar 2007, 04:57 am »

Makes sence, thanks Geoff.

Haron

AKSA · « **Reply #3 on:** 15 Mar 2007, 06:07 am »

Haron,

I wrote a long response to this earlier today, but lost it when I was interrupted by a call and my wife took over the keyboard!!

Any single gain active device which gives current and voltage amplification, such as a plate loaded triode, will invert phase. This is inevitable. To preserve phase you must use a two terminal hybrid configuration, such as a complementary feedback pair, or two cascade common cathode/emitter gain blocks.

So, unless you use two triode sections, you will invert phase. The cathode follower (input at grid, output at cathode) and the common grid (input at cathode, output at plate) preserve phase, but the former has no gain (in fact voltage gain is less than one, though current gain is very very high) and the latter has punishingly low input impedance, requiring more drive than a big block Chevy to crank it over. This latter circuit is used in RF and Microwave technology because it's very, very fast.

Cheers,

Hugh

netaron · « **Reply #4 on:** 15 Mar 2007, 04:30 pm »

Thanks for the clarification Hugh.

I do have another question for you, when I turn off my preamp, I get a loud thump at the speakers. I have tried switching off the amp first, but since the N55+ holds so much power in reserve for a long time makes this a mute point. So I am guessing there is some DC leakage from the caps? I will proceed with changing out the relay that disconnects the pre outputs, but I don't think this will help. Any idea where I need to start poking around?

Thank you as always.

Haron

RonR · « **Reply #5 on:** 16 Mar 2007, 05:27 pm »

Haron,

Do you have bleeder resistors fitted on your 55N+ PSU?
They drain the stored energy from the smoothing caps at switch-off. I have them fitted, and the LF55 (Same PSU) drains in around 4-5 seconds.

AFAIK there is no impact on sonics.

Cheers,

Ron.

netaron · « **Reply #6 on:** 16 Mar 2007, 09:24 pm »

Hi Ron,

Thank you for your post, it sounds great, only I don't know what value to use and how to wire it to the PS.

Haron

AKSA · « **Reply #7 on:** 16 Mar 2007, 10:44 pm »

Ron,

Great advice! Thanks!

Haron, around 10K one watt is fine. Your stereo amp has four power rails; two for each module, one positive, one negative.

Simply wire a 10K 1W resistor from each rail to star ground on all four rails. You can conveniently mount the resistors on top of the power supply pcb.

That should do the trick!

Cheers,

Hugh

netaron · « **Reply #8 on:** 19 Mar 2007, 02:27 am »

Hi Hugh,

Thanks, it works great, no turn off thump. So should I expect my electricity bill to go up slightly now

?

Haron

NewBuyer · « **Reply #9 on:** 19 Mar 2007, 03:51 am »

Quote from: AKSA on 15 Mar 2007, 06:07 am

...Any single gain active device which gives current and voltage amplification, such as a plate loaded triode, will invert phase. This is inevitable. To preserve phase you must use a two terminal hybrid configuration, such as a complementary feedback pair, or two cascade common cathode/emitter gain blocks.

So, unless you use two triode sections, you will invert phase...

Hello Hugh,

Just wondering about this please: Would this apply then to DACs with internal tube buffers?
For instance, some AudioCircle members are using the Paradisea DAC right now, which internally uses a single 5670 tube in Loftin-White fashion (i.e. coupled directly to the opamp). Does this DAC probably invert phase?

AKSA · « **Reply #10 on:** 19 Mar 2007, 08:02 am »

Hi NB,

Very likely, if it's Loftin White it will, but phase can be re-inverted with suitable wiring on a transformer, so it's possible to have a single gain stage with a transformer loading - it's common enough - which will NOT phase invert.

The Cathode Follower and Grounded Grid won't invert phase either, though the former has no voltage gain and the second is notoriously difficult to drive.

Welcome to our forum, Sire!

Cheers,

Hugh