On Tuesday, 4 December 2012 22:42:12 UTC+11, Alex Pogossov wrote:
> "patrick-turner" <
in...@turneraudio.com.au> wrote in message news:c1f25434-b68d-4194...@googlegroups.com... > Hi all r.a.t regulars, and to anyone searching about Jadis SE300B amps. > I have put up a page detailing my work on reforming a pair of Jadis last > month at
http://www.turneraudio.com.au/Jadis-300B-reformed.html > > Anyone wanting to make a decent sounding/measuring 16 Watt SET amp are > invited to have a look. > > Patrick Turner I found using timber circuit boards funny. dielectric absorbtion and nonlinear leakage conduction due to the residual moisture in timber will cause sound coloration. Definitely such insulator will give a "wooden" sound -- shallow, flaccid, limp bass.
Wooden boards for circuits have been used for last 150 years. They were routinely used in early radio circuits where circuit impedances were often high in hundreds of k-ohms.
I have measured all use of timber and found the only place one has to worry about the multi megohn resistance between 2 screws 10mm apart is for grid circuits where Z is high and there is a HV connection on the same peice of wood. The wood tends to charge up, and you get a tiny Vdc someplace where you were not to expect it.
But your statement "Definitely such insulator will give a "wooden" sound -- shallow, flaccid, limp bass." - is utter rubbish because the bass quality of the reformed Jadis is nothing short of spectacular, and sounds like a good 50W amp.
Real audiophools with determinations always use teflon -- no absorbtion, no leakage, easily machined, cut, drilled.
Audiophules with determination like to believe they know best by using teflon that almost nobody else uses because of cost and difficulty in its use and the inability to use glue to fix it. Audiopules are notoriously pig-ignorant, knowing almost zero about real world electronic or metal engineering or about electronic properties of material and how they affect distortions and noise and sound. What they DO KNOW is mostly a jumble of bullshit notions not based on any known facts, but based on "what others have said" on the Net, and on slaes BS and on what Stereophile authors have poked down their necks. They then profess to be able to tell if any special capacitor brand has been used or special wore of special any fukken thing. Fact is, when I have yested their perceptive abilities, they rarely can pick whether I have used Wima caps or Auricaps, teflon or polyester, or any difference between either solid state amps I have made, or tube amps. Turns out the audio enthusiasts who are not complete raving loonies like whatever I make.
You can dril blind holes and drive screws and pins into them. Will not carbonise, burn or change properties when overheated during soldering or when a component is burning.
I understand all that, and would love to used 10mm x 10mm bars of teflon to make up circuit terminal strips. Its fiendishly expensive stuff. Probably, like so many modern plastics, its production causes an environmental nightmare somewhere.
The amps I build rarely ever need to be worked on later because anything to be done by anyone other than myself would make things worse, because nobody understands his own product better than me.
So, having wired up the Jadis with a few hardwood plywood boards and hardwood terminal strips using brass plated steel cupboard hinge screws for terminals, its more than likely they'll never ever be changed. If someone were to acquire what I have done and replace terrible horrid wood with teflon, they'd be telling a whopping lie if they insisted it sounded better. One could say music should sound organically wondrous, and be supported on natural substances where possible, including wood, and iron, copper, silicon because all are natural, ie, found on our planet, without much monstering and adulteration by industry. Wood Knot Teflon sound clinical, dry, souless, empty,and unacceptable? Bloody audiopule arguments can be reversed back onto the stupid bastards. Fact is Jadis SE300W amps were a bloody horror story, and now at least just TWO samples sound wonderful. The Jadis site says they are discontinued, and I guess, and I am only guessing, maybe all samples of that model had OPTs without any air gap, plus the whole pile of other circuit mistakes and design mistakes which made +1,000 dB more worsening of sound than any timber board for CCS transistor and protection board, and test terminals for monitoring Iadc and Iac of each 300B without needing to move it.
Also I noticed that balancing of the filaments of 300Bs with 33R series resistors is unnecessary. With the DC filament supply, one end of the filament is ALWAYS 5V higher or lower, no matter where the self-bias circuit is connected.
There is few mV of hum across the cathodes. Its negligible. But when I made amps with 845, guess what, even with cathode Vdc and few mV of hum, a fraction of a mV got into output so balancing is GOOD PRACTICE even though you say its BS.
Thus the last 5V of the grid voltage swing is underused, because grid current from the more negative end of the cathode will prevent full emission on the more positive end.
In theory, you are correct, in practice, its BS. The difference in emission along the cathode is utterly negligible and unimportant to operation. Would you care to quantify your argument? Jadis had used a regulated 5Vdc for 300B and hum was less than a mV, but they had two balancing 47r resistors. Trouble weas they used just one 5Vdc supply common to both 300B - big mistake when tubes are not matched, and running too hot anyway. The very slight amount of unbypassed resistance in my TWO reformed Rk+Ck circuits = 38 ohms, and the very slight amount of local current FB does SFA good, bad, or otherwise except stop the two parallel cathode bypass caps ever being over currented with AC from low Z of 300B cathode. Not likely in fact.
Similraly, this 5V skew smears the tube cut-off -- one end might be cut-off while the other is conducting.
This is a hi-fi use of class A 300B and they never go anywhere near cut off.
Your argument does not hold water.
Besides the cathode is wearing unevenly, one end always bearing more emission current, but this should not be a tube life limiting factor.
And it just does not matter!
So this balancing is just a sterotypical thinking.
The balancing was a necessity where AC was routinely used in many amps for heating. Usually a hum nulling pot was used, and even in PP amps the nulling is never perfect and in a pair of Sun amps with 2A3 I repaired 10 years ago the AC heating and pot were retained, and I gave a pair of headphones the owner could use for nulling. That worked, and noise only seldom appeared at speakers. My 55W SE55 with 845 had less than 0.25mV of total noise at output. I know all about how to build amps with low noise, and I'll do it my way, and not yours.
The four 33R resistors can be safely removed.
Indeed, but they are staying put.
Now, depending on whether the self-bias circuit is connected to a positive or a negative end of the filament, one would have less or more fixed bias mixed respectively.
The Reformed Jadis now have individual R&C cathode biasing. Audio Note makes a similar amp with similar fatures to what I have done. It works, and cannot easily be fucked right up by some idiot audio nutter who always manages to get his testicles in a knot when he tries to adjust the "fixed bias" which of course is unfixed, and adjustable, and thus able to cause mahem among the Dopes. My dear customer here has an excellnt range of human abilities, and a fine man he is, but he just has no clue how to use a voltmeter, like 99% of the rest of the population. Leak and Quad used cathode biasing for obvious reasons although I'd say again Quad were idiots to have one lone 180r to bias both KT66.
But maybe Peter Walker owned shares in the MOV tube making Co, so the more KT66 people bought because of bias failure with his crummy biasing meant he got richer. Plus, omitting one R and one C allowed him to buy a Morris Major in 1960instead of a Morris Minor, wow, wonders wood never cease.
BTW, about 10 years ago I totally re-engineered a large stereo VAC amp with 4 x 300B per channel. Jadis ain't alone and producing rubbish that smokes. Anyway, in went timber circuit strips, worked well, very rugged, but the grid cicuits did use teflon insulated wires point to point with no grid potential points to any wood board/strip. Caps were boxed polypropylene Wimas siliconed to chassis so that the grid wiring could be slung from short cap leads acting as terminals.
Certainly no lossy PCB boards were used, with stuff all jammed tight, and boards blocking natural air convection flows. Don't worry Alex, I know what I am doing.
I have done many amps over the last 18 years and not one has had to be returned as a result of my poor tradesmanship. I'm now retired, and may not make too many more amps.
Patrick Turner.