[Brinkman]

Jim,

Have you considered designing a big-brother version of your Bugle Power Supply?

I have noticed a number of your customers going the split chassis route for both the Clarinet and Cornet2 and am considering it myself. However, as an educational exercise, I'd like to improve the power supply to make it more like the Trumpet and Castanet supplies (LCLC, maybe a filament choke). I know you've mentioned on the forum that you'd incorporate such a supply in any future Cornet2 successor.

I was thinking that a stand alone HV + filament power supply module might be a great kit as a me-too for upgrading the Cornet and Clarinet and/or form the basis for an All-Purpose bench-test power supply for folks interested in tube circuit design, such as the supply schematic in my RCA receiving tube manual. Of course, a modern one designed by you would be magnitudes better.

You assuredly have a lot on your plate, so forgive me if I'm stepping on your toes.

[hagtech]

No, I haven't thought of such a separate supply. 

I have, however, designed an LC input supply for the CORNET3.  Ok, it is code-named SAKURA.  The other change is that the B+ is a 100V lower, so I had to drastically re-bias the 12AX7 and re-do the EQ.  Trying it out now.  We'll see.  Could be a total sonic failure, or might end up sweet as heck.  Waiting for the last few parts to come in...

jh

[bluesky]

Hi Jim

Cornet MKIII, now this interesting, tell us more!  I have always wondered about some sort of hybrid Cornet/Trumpet thing.

Bluesky 

[Brinkman]

Just thought I'd compile some quotes I've been contemplating on recently. The oldest one is from 2003. I put some text in boldface...

The Cornet is a very carefully tuned circuit that puts tube bias, plate load, etc. all in harmony for optimal results in that configuration.  Yes, you can do better with plate chokes.  Going to a balanced design will also help (Trumpet).  Interstage transformers may work out better than coupling caps.  Higher B+ helps, too.  There are always compromises in any design.

I think it is the number one reason the TRUMPET sounded so good, not the fact that it was differential.  So imaging a single-ended CORNET style phono with TRUMPET style power supplies. Good power factor means choke input supplies, no diodes pulse charging caps, like every other design out there.  And I mean all of them.  Because I want to do the choke thing on the heaters too, not just B+.  Well, this leads to a lot of iron, a low of magnetic fields, and more cost.  But not that much more. Pretending I used the same chassis, I came up with all new designs for a phono and a monoblock.  The phono has 6.3Vdc heaters from balanced bipolar choked rectification.  Normally this ends up at +/-5Vdc with a pair of 6.3V windings, but I made it lossy in an LCRC topology.  Result is about 20mV ripple.  Then for B+, using the small power transformers (that fit under hood), I can barely squeeze out 210Vdc or 220Vdc.  That's way lower than what a CORNET requires.  Ok, how about using 12AY7 or 6072 tubes for the gain and EQ stage with CCS? I think noise will go down, linearity up, dynamics too.  Heck, I almost don't need the CCS, a 50k plate load might do the trick.  I like this tube.

*

I have, however, designed an LC input supply for the CORNET3.  Ok, it is code-named SAKURA.  The other change is that the B+ is a 100V lower, so I had to drastically re-bias the 12AX7 and re-do the EQ.  Trying it out now.  We'll see.  Could be a total sonic failure, or might end up sweet as heck.  Waiting for the last few parts to come in...

Wow, Jim!

I see you only mention one 12AX7; does thus mean we could expect a 6072 or a 6688 in the new circuit?

Also, could we expect to see any of your thoughts I bolded up above to be implemented? I'm particularly curious about plate chokes and interstage transformers (inductive coupling vs cap coupling), especially since I've been using the K&K Audio driver in my ST-70. Input trannies provide high input impedance and handle phase splitting. Also CCSs for each tube. Sounds fantastic running some KT88s.

Can't wait to see what you end up with!



*quote provided by WGH

[tubesforever]

Hey Brinkman, while a Cornet 3 might be some fun, I need Jim to build a balanced SRPP line stage with absolute polarity.  Let's keep him focused.

The Cornet 2 is already so quiet its quieter than my amp.

A balanced line stage and a Trumpet kit.  That's what I want for (add Holiday of your choice here). 

Cheers!

[Bill Epstein]

Hey Tubes, I built the Clarinet more because I had the parts than any real desire to replace the Aikido or the modified ARC LS-1 but now that it's breaking in, and I switched the polarity at the speakers, it's starting to have some of that single-ended Zen I heard from the CAT SL-1.

I never expected that.

[Brinkman]

I need Jim to build a balanced SRPP line stage with absolute polarity.

I'm pretty sure an input transformer, with its primary wired in reverse (in place of say, C302 in the Clarinet) would accomplish what you're after here. The Lundahl LL1690 is a perfect example, and even has dual primaries and secondaries for balanced applications.

[tubesforever]

Brinkman, you probably have a better grasp on this than I do, but I want a pure differential balanced circuit. 

From what I understand a transformer creates separate channel grounding but does not provide common mode rejection.   I want the discrete grounding and the common mode rejection.

Happy Holidays! 

We have got to get together next time I travel to Portland Oregon!

Cheers!

[hagtech]

I see you only mention one 12AX7; does thus mean we could expect a 6072 or a 6688 in the new circuit?

I looked at the 6072 for a number of uses, but it never seems to stick.  My new guitar amp had them for awhile, but they got replaced with something else.  The C3 might be able to use them.  I had a hard time maintaining gain with the low B+.  Switching to a 6072 could make it even worse.  I have the C3 proto built, just need some spare time to fire it up.

I did a paper design of a 6688 based phono.  I like it.  Have not built it yet.  The problem is that most customers was a more mainstream tube - usually something they have on hand.  Regardless, I find the 12AX7 a really suitable tube for a phonostage front end.

I'm particularly curious about plate chokes

No, I don't think I would do this for a phono.  Only a power amp (or the CASTANET).  I'm talking about chokes in the power supply.  Huge difference bewteen rectifiers feeding a choke versus rectifiers feeding a capacitor.

jh

[Brinkman]

Brinkman, you probably have a better grasp on this than I do, but I want a pure differential balanced circuit. 

From what I understand a transformer creates separate channel grounding but does not provide common mode rejection.   I want the discrete grounding and the common mode rejection.

Hey Tubes,
I was trying more to suggest a way to invert the polarity of the Clarinet without adding an extra stage; a single transformer vs. active and multiple passive components in the signal chain. I didn't want to try and take a stab at differential.

I'm not 100% positive about the technicalities of input/interstage transformers, but for common mode rejection, as long as the anti-phases are symmetrical and with equal impedances, I don't see why transformers couldn't be used. Look here, about 3/4 of the way down.

Historically, transformer coupling was used in telephone repeaters and I can't think of a situation more in need of a high CMRR. Common mode voltage can be passed via capacitive coupling between primaries and secondaries, and that is why I recommended Lundahl transformers, which shield their transformers well and have low Primary-Secondary Capacitive coupling.

---------------------------------

As for my original post about a bench power supply kit, it was more inspired by my slow acquisition of test equipment for home circuit construction. There's an interesting school of thought amongst some audio experimenters that an amplifier can be conceptually approached as a modulated power supply. If that's true, then an amplifier can never be better than it's power supply. Of course, the truth is probably somewhere in the middle (or rather, the truth is in balance), but I like the way the concept sells the importance of what is often considered an afterthought, much in the way circuit grounds were discovered to be the true source of electron flow ("B+" as electron sink). In the end, which model is "more correct" is only a human preoccupation.

So that was my suggestion. If a reliable regulated bench power supply was available with highly variable voltage and current, and designed with the same audiophile considerations found in high-end components such as choke-input and separate choked filament supplies, then it could be an indispensable tool for the home-constructor who wants to be sure his power supply is not his weakest link.

There are other tube-based entries out there, but I'd rather spend a bit more for build quality, added flexibility and audio concessions.

Getting off my soapbox...