Clarinet Break Through

tubesforever · « **on:** 30 Dec 2009, 01:53 am »

While Jim Hagerman was busy perfecting the Cornet 3, I took liberty to ask him if he could plug in some values to hit the 12AU7 tubes lighter on the Clarinet.

My goal was to see if the reduced voltages could still keep the circuit happy and quiet down the whistling and squeeling some of us hear with our Clarinets.

So in a moment of extreme weakness, probably due to the exhaustion and exhilaration he achieved with the success of his C3 improvements, Jim graced me with some values to try on our Clarinet line stages.

Now for some history. The SRPP (series resistance push pull) circuit Jim employs in the Clarinet is a bit of magic. Here's why. Jim utilizes two single ended triode gain sections running a series resistor push pull configuration to deliver the sonics of a single ended triode gain stage with the linearity of a solid state line stage.

As an added benefit to his design, you can throw out the negative feedback. You don't want it or need it. In fact you will never miss it. I did not know what negative feedback sounded like until a friend with a Cary line stage showed me what his system sounded like with zero to 15db of negative feedback. It sealed my desire to find cleaner zero feedback designs.

You get to have your cake and eat it too with a Clarinet. I love the circuit even with any flaws.

With Jim's original circuit my Cornet 2 sounded delicious, but CD, SACD and DVD-A could sound really ragged if I tried to juice up the volume beyond a natural listening level. The sound would distort and the magic was gone.

Worst still, the Clarinet could begin to whistle or squeel if they were not measuring "new" on the ol tube meter after 3-5 hours of use.

Jim Hagerman to the rescue! The fix is just 4 silly little resistors.

Remove the four 680 ohm resistors at R 308 and R310.

Replace the R 308 resistors with 1.2k ohm resistors and the R310 gets 820 ohm resistors.

Be prepared for a shock.

First of all the sound is a little more relaxed and not quite right up there in your face. I like to sit in rows 12 - 15 because the sound is meaty and full. I get more of the prime seat sound now.

Second I can run significantly more volume without the sound ever distorting with my digital sources. I like this a lot.

There is more depth and finesse to the sound now. Its a really nice presentation overall and the tube squeels are gone. What's not to love!

Dang Jim, the Clarinet is now a 9+ out of 10 in my book. It has SET magic, passive preamp clarity and detail, and it just begs you to turn it up just a little more.

The modified Clarinet reminds me of my old Bedini 6677 with its uncanny fashion of playing louder and louder without distorting. The Clarinet line stage is best line stage I have ever owned.

Wagner's Ride of the Valkyries and Stravinsky never sounded better in my room.

FOUR DAMN RESISTORS .....Get to it! Report back.

And by the way Jim, thank you sincerely for being a diy rock star!

david62 · « **Reply #1 on:** 30 Dec 2009, 02:32 am »

Thanks for posting that information.It is especially nice for those of us that have yet to build our Clarinets and can benifit from the start.
Dave

Bill Epstein · « **Reply #2 on:** 30 Dec 2009, 11:59 am »

Krap!

I've got more resistors than Mr. Vishay but not those 2 values, OF COURSE!

david62 · « **Reply #3 on:** 30 Dec 2009, 07:34 pm »

Which tubes do you advise using in the Clarinet,and where is a good source for tubes?
David

tubesforever · « **Reply #4 on:** 30 Dec 2009, 08:16 pm »

I like the JJ Tesla and the Tesla S tube. The Tung Sols would be good. If you can find the Sylvania Mil Spec 12au7 tube these are sweet. I have not been able to do a lot of tube rolling yet. I will get to it. The NOS tubes I own would start to howl or squeel after a half day of run time. I need to see now if that problem is cured with the NOS tubes measuring Good or better.

I have been buying tubes from tubesandmore, tubedepot, tubestore, and off ebay.

Cheers!

Bill, its that damn Murphy guy again. I am going to kick his butt someday. Murphy's law comes up so often in my life that I should count him as one of my reliable "friends".

Happy Holidays!

Bill Epstein · « **Reply #5 on:** 30 Dec 2009, 10:15 pm »

Hey David, and everyone else, too. The cool thing about AES tubesandmore.com is that you search for 12AU7 or CV4003, etc and you get some name brands as well as a cryptic "NOS" and a price. I've gotten Amperex in the past as well as the exploding Mullards, first time that ever happened.

Brinkman · « **Reply #6 on:** 30 Dec 2009, 11:16 pm »

In a prior thread, somebody asked about using 12BH7s in place of the 12AU7s. It would require an H+ circuit change, but little else. Since my Clarinet will remain unbuilt for the time being, it will be a while before I'll tackle it. Though I am fond of 12BH7s.

This new info is surely helpful; it should be nice for comparing an 12AU7 to 12BH7. Thanks, Jim. And Jim.

tubesforever · « **Reply #7 on:** 1 Jan 2010, 05:55 pm »

I don't think there are as many 12BH7 tubes to select from if you enjoy tube rolling. If you want to drop these into a Clarinet you will need a different transformer or an additional 12v transformer to give you the correct H+. This is easy but its not a simple tweak to the current configuration.

The reason to attempt a 12BH7 in my line of thinking is if you require more gain. These are basically two 12Ax tubes stacked one on the other. These tubes require twice the heater voltages and a beefy H+ power supply.

Brinkman, from my seat, I would love for Jim to build us an Octal Clarinet 2 and pitch the 12AU7's...but heck I know 1/1,000,000 of what Jim Hagerman has forgotten about tube circuit design.

Here is a follow up from my original post.

Last year I bought about a dozen JAN 6189 tubes which are the joint Army Navy mil spec tube equivalents to a 12AU7. After 3-4 hours I always got squeeling from these tubes.

So yesterday I put in a pair fired it up at 9 am in the morning and ran it until 12 o-clock New Years eve.

Not a single whistle or squeel--Just blissful silence.

I think these new voltages are going to do the trick! I still have other tubes to roll of course but the Clarinet is now happily performing its task in a way only an SRPP can accomplish.

Jim Hagerman, you are a DIY rock star! Thanks for the recommendation! I could not be happier unless of course you can pencil out an octal version of the Clarinet and make this your next product. Jim, don't make me beg.

Bill Epstein · « **Reply #8 on:** 3 Jan 2010, 03:14 pm »

Knowing just enuf to be a danger to myself and others, it doesn't seem like it would be all that difficult to do 6SN7s with the current board.

You'e need to use P-to-P tube sockets wired to the board and change a few resistors. The voltages and the rectifier are OK the way they are, I used the same transformer in my all 6SN7 Aikido.

Anything else?

GRD · « **Reply #9 on:** 3 Jan 2010, 06:36 pm »

A couple of thoughts.

Heater current on the 6SN7 is twice that of the 12AU7, so you'll need to see if the transformer can deal with 1.2 amps versus 0.6 amps. Since the heater is DC, the rule of thumb I see quoted is the heater winding should be rated at 2x the AC current capacity. So a 2 amp AC heater winding can supply 1 amp of DC. 1.2 amps may be OK with a 2 amp winding, depending on the transformer.

Older 6SN7's have a heater/cathode rating of 90 volts. The later 6SN7GT versions have a heater/cathode rating of 200 volts - like the 12AU7's. So that is another consideration if you've got older 6SN7's that you are considering.

Otherwise, the 6SN7 should work (substitute) for the 12AU7.

Bill Epstein · « **Reply #10 on:** 5 Jan 2010, 04:36 pm »

Back to those resistors...

All I can find are 910 and 1.2k. Probably close enough?

fatty · « **Reply #11 on:** 5 Jan 2010, 08:05 pm »

Hi GRD,

I hate to argue with you but the filament current of a 12AU7 and a 6SN7 are exactly the same .300 ma at 6.3 and .150 ma at 12 volts. However, you will need to change the operating points for a 6SN7 because they are different from a 12AU7. This will require some resistor changes. I am quite sure that JH could provide those values.

GRD · « **Reply #12 on:** 5 Jan 2010, 10:53 pm »

Thinking I made a big "oops", I went back and rechecked my recollection of the heater voltage/current requirements:

The 6SN7 has a 6.3 V 600 mA heater/filament. The 6J5 (one half of a 6SN7) is 300ma.
The 12AU7 has a center tapped filament so it can be used in either 6.3V 300mA or 12.6V 150mA heater circuits.

Looking at the two tube curves, I stand corrected. Although the 12AU7 and 6SN7 are similar, they aren't substitutes and the resistor values will need to be adjusted. Thanks for the correction.

Brinkman · « **Reply #13 on:** 6 Jan 2010, 04:52 pm »

Quote from: tubesforever on 1 Jan 2010, 05:55 pm

If you want to drop these into a Clarinet you will need a different transformer or an additional 12v transformer to give you the correct H+. This is easy but its not a simple tweak to the current configuration.

The reason to attempt a 12BH7 in my line of thinking is if you require more gain. These are basically two 12Ax tubes stacked one on the other. These tubes require twice the heater voltages and a beefy H+ power supply.

I gotta leave for work, so here's the short version:
I think some folks are confusing current and voltage; voltage drops across series elements and current splits among parallel branches. So if two heaters are in series, the same current would travel through them (say 150mA), but the voltage across them needs to increase (12V). If they are in parallel, the current would need to be doubled (300mA), but the voltage node formed would only have to be 6.3V. So the question with the Clarinet is whether there's enough current in the heater supply for a tube with twice the heater current requirement of a 12AU7 (such as a 12BH7).

To address why one would want to use a 12BH7 is a matter of preference. I know some tube nuts (don't we all) who swear the higher filament current tubes sound better. And if you look at the curves, the 12AU7 and 12BH7 have similar operating points, so resistor changes in the signal circuit are not needed. Thus, if you can meet the heater requirements, the 12BH7 is a 12AU7 variant as far as tube rolling is concerned.

The mu of a 12BH7 is 17.7, whereas a 12AU7 is 17.5, so a gain difference that small is not noticeable. I personally have performed this substitution in another circuit and did indeed prefer the 12BH7. YMMV.

Happy new year.

Bill Epstein · « **Reply #14 on:** 7 Jan 2010, 06:16 pm »

Meanwhile, back at the breakthru

I was getting a faint high-pitched 'flourescent lamp' type squeal and then a louder growl at higher volumes.

This AM before work I replaced R308 with 1k and R310 with 910 ohms, what I had on hand. While I was inthere I replaced the Z-cap output caps with Obbligatos. For first listen, I'll leave the Teflons disconnected.

Stay tuned.

tubesforever · « **Reply #15 on:** 7 Jan 2010, 08:44 pm »

Hey Brinkman or JH,

I am not sure I want to just plug away a B12H7 into my Clarinet or the Cornet 2 knowing my current configured units sound delicious right now.

However, from your post it sounds as if you have already tried the B12H7 in your other tube projects as a direct roll for a 12AU7.

Looking over the Hammond transformer and the capacitance supply in the H+_can this be pulled off with the Clarinet? If its safe to roll I will be happy to buy some and give them a try.

From what other designers have told me, the circuit must be balanced for a 12BH7. The 12AU7 requires less on the current side of the equasion.

So is it safe or is it risky?

hagtech · « **Reply #16 on:** 9 Jan 2010, 01:57 am »

The 12BH7 should plug in and play. Not sure if optimized. Heater draw is a bit more, so the dc will go down a bit. This can be compensated for with that 5W resistor.

jh

flocchini · « **Reply #17 on:** 9 Jan 2010, 04:05 pm »

I have been following this discussion with great interest. The "tube squeal" issue has held me back from the Clarinet. Given that I have been very happy with my hybrid CHIME and CORNET2 I have to ask - Has anybody done a octal Clarinet or is one in the works? My Russian teflons need a home. aa

Thanks

Bob

Bill Epstein · « **Reply #18 on:** 10 Jan 2010, 12:03 am »

Don't hold back.

I had some rude noises from mine recently but when I swapped output caps I discovered that it was a loose connection to one teflon by-pass. Now it's back to oh-so-quiet.

How do you make a Clarinet squeal? I dunno. Has anyone but Tubes had the issue? He says it's on the CD tap of his unit. Maybe the selector switch?

analog97 · « **Reply #19 on:** 10 Jan 2010, 12:12 am »

Quote

How do you make a Clarinet squeal? I dunno. Has anyone but Tubes had the issue?

Bill,

I had reported the squeel as well, but it was MY ERROR!!! There have been an awful lot of posts that suggest a problem with either the C2 or Clarinet, but my gut tells me 99% are user errors.