Bypassing screen grid resistors???

[nickb]

That's an interesting note on that chart in #15 Bypassing screen grid resistors??? (for very low power tube tho) "Values to left of A-A subject to considerable variation", isn't it?

If I assume that the same applies to power tubes then that would explain to a significant extent why power tubes, even of the same make, might audibly vary when driven hard.

[Alan0354]

Alan, the bypass capacitor does not go in parallel with the screen resistor but from screen to ground.

I was referring to Chuck in post #4 when he said putting a cap in parallel with the screen resistor.

I tried varying the screen voltage together with the -grid bias as VVR control, it did not sound good at all, nothing like varying the +B and -grid bias.

[frus]

Isn't that kind of what Mesa did with their DynaWatt "patent"?

United States Patent: 4713624

apparently, what you need to do is "optimize" the screen resistor and bypass cap so it charges/discharges at some "magic" rate that is musically "appealing" to guitar players

Randall Smith, Mesa Boogie Founder Speaks - YouTube

Btw, I've never played the Express or 20:20, so YMMV, hell even MMMV

[Chuck H]

Nope. Different. I had a Mesa Dynawatt amp and the schematic matches the circuit. If you look at one you'll see that there is no bypass of the screen grid resistors, to ground, parallel to the resistor or otherwise. As far as I could tell the "Dynawatt" system involves 2.2M resistors from the control grids to ground and a relatively high screen grid rail resistance and nothing more. Same ol' plain jane 470r screen grid resistors from the second node of the HV rail otherwise. I pulled the 2.2M resistors and reduced rail resistance to the second node. The amp sounded much better

Then I went to work pulling all the peripheral high pass filters and high end bleeders that only serve to counteract each other. Took out an odd input circuit involving a 1M series resistor. Then a couple of local feedback circuits. Basically stripped the circuit down to something more like a vintage amp. The process was more like Muntz-ing than modding. In the end what I found was that Mesa actually uses pretty good transformers and speakers because the amp sounded really good once I defeated the Mesa circuits.

[jazbo8]

You did what?! Mr. Smith is not amused...

[Alan0354]

I thought Mesa amps sound kind of suck. My idol was Carlos Santana, I tried so hard to like the Mesa because Santana uses it. I can't. I tried Dual rect. Triple rect, dual caliber, Mevrick, Blue Angle , MK IV, MK I..................all different ones. I just don't like both clean and dirty sound.

What is dynawatt? I am too lazy to read the patent, any simple way to explain it? I hate reading legal language,they try to confuse you.

[frus]

oh yeah... but wouldn't the screen power supply work similar to what J M described? It would stay high for a while, then drop (with some 200ms discharge time). only it would sag preamp voltages too

[teemuk]

...the amp sounded really good once I defeated the Mesa circuits.

Matter of taste I guess really.

Anyway, do heed that part of the DynaWatt use in modern Mesa amps is to pump up higher than usual peak power ratings from the amps. The DynaWatt circuit deliberately sags the screen voltage and limits current flow and dissipation of the power tubes. Tube can momentarily dissipate more than the recommended peak power and the sagging quickly limits the dissipation to safer levels. (I assume you did properly bias the tubes for the modified circuit). Anyway, marketing department can sell a higher powered amp, (despite the fact that the power rating is for burst power at ridiculously high harmonic distortion levels), and guitarists who like pronouncedly "squishy" amps (or whatever the term is) get one.

That's enough OT for me. The DynaWatt is patented so more detailed technical information is just one google search away.

[Chuck H]

Randall describes it as "a few hundred milliseconds." but I'm sure that's intentionally vague. The patent also indicates a high value series resistance between the bias supply and the HV rail, which is ridiculous if you consider it. What I saw in my amp was a high value resistor from each control grid to ground.

There may be some measureable phenomenon with the Dynawatt "system" but it looked and sounded to me like a higher than average series resistance between the plates and screens on the HV rail and a slightly higher value than average screen filter cap. That's a big fat so the f@#k what. IMHO the best thing about the Dynawatt amps is that they all seem to use a fairly cool fixed bias (and possibly CC) PI of an unusual design that seems to work brilliantly for guitar. I've never measured but I suspect it's balance fluctuates dynamically. This seems to enhance 2nd harmonic and abate crossover distortion. A very good thing for EL84's. Funny, I've never seen a patent on that PI and it's the only thing Mesa I left in the amp.?.

[Chuck H]

Matter of taste I guess really.

I trust humanity very little in general. But I would be surprised if anyone hearing the amp before and after disagreed with me. Just banter without evidence, really, but I have at least that much faith in the reasonable sensibilities of my fellow man

[teemuk]

Funny, I've never seen a patent on that PI and it's the only thing Mesa I left in the amp.?.

There's nothing extraordinaire in a constant current source -loaded phase inverter. I'm inclined to think very first patents of such could probably date to 1940's if not even 1930's. I'm looking at North American Philips Co., Inc.'s patent at the moment that shows a triode LTP phase inverter loaded by a transistor current source. If that was patentable in 1959 then I can imagine that "constant" current source, achieved only with a very stable DC source and a resistor, muct be technology invented 20 - 30 years earlier. Nevertheless, if there are guitar amplifiers from the early 1970's, sophisticated enough to control PI's clipping by varying its CCS load's current, then there's likely very little left for Mesa to patent. Although that has never seemed to prevent them earlier either.

[Jazz P Bass]

Mesa Boogie Dyna Watt

The Dynawatt PI is a bit different as it basically references the phase inverter cathodes to a 3 volt supply which is derived from the bias supply.
And take note of the low screen supply voltage that is used (and the odd sting of 330 ohm resistors)
Here are the schematics.
And the Patent: http://patft.uspto.gov/netacgi/nph-P...&RS=PN/4713624

[Chuck H]

Interesting. Maybe. Do fluctuations in the PI have an influence on the bias? Or do any power tube grid current fluctuations have an influence on the PI? The bias supply uses a silly big reservoir/filter cap. Whatever it's doing, if you set up the HV rail more like a normal amp and remove the 2.2M resistors it sounds good.

I'm aware that a DC biased PI isn't a new idea. I'm also aware that Randall never paid any attention to prior design when seeking a patent.

[Alan0354]

You can be very tricky writing a patent. I did not read the "Claim" of his patent in detail as it can be full of legal BS. One thing from my experience writing a patent is you can make if very specific for one purpose and obtain a patent even the circuit is very close an existing circuit. I read the first few sentences, it said "For Guitar". That limited the field to very narrow scope and you can argue the design is for very limited purpose ( guitar) and can be defined unique and obtain a patent. These are more the trick by the lawyer than really a unique idea.......It's how you write the "Claims".

A lot of times, patent is not about how bright your idea, it's about how bright the lawyer can "put" your idea into legal form and make it unique.

[Jazz P Bass]

Then again, Mesa seems to patent items that are not at all earth shattering.

More 'bells & whistles'.
Yawn.