Phase Inverter comparisons

[Merlinb]

" every amp I have seen this stage is fed from the same B+ supply node as the inverter stage, and 300-330v would be the limiting voltage for this stage

Um, I don't get it. The limits for the previous stage are exactly the same as for the cathodyne, e.g., 550V max supply. Typically the grid voltage of the cathodyne wants to be about 1/4 to 1/3 of the supply voltage, but it can be higher. I can't see any reason why either stage has any special limitations? Even if both were fed from 500V, and the anode of the previous stage were 250V, where's the problem?

[MWJB]

550v suppy to the gain stage plate resistor equates to 374vdc on the 12AX7 plate, with respect to the cathode, assuming middle of the road plate & cathode resistor values (100K/1.5K), looking for higher headroom might require a higher value cathode resistor & more plate voltage. Let's say that power tube plate voltage was 525v, drop a couple of volts for the screens, use a 1k PI dropping resistor (like an AB763) and you're still looking at 500vdc+ at the B+ rail & 340vdc on the 12AX7 gain stage plate (with respect to cathode).

Note that on their high B+ models, HiWatt used a lower voltage B+ supply for the PI & preamp tubes.

[Merlinb]

550v suppy to the gain stage plate resistor equates to 374vdc on the 12AX7 plate, with respect to the cathode, assuming middle of the road plate & cathode resistor values

Well yeah, but who's going to use middle of the road values for non-middle of the road voltages like that? Obviously you can't take a circuit and just arbitrarily raise the PSU voltage indefinitely, that would be plain stupid. But you can still use a 550V supply, you just have to stay within the safe operating area, as you would with any component.

[MWJB]

A proven set up like 100K/1.5K is hardly arbitrary, it works in 90% of all guitar amps at a vast range of voltages. It's the middle ground. The point being that larger plate resistor will drop the voltage but after a point will impair fidelity. If you bring up the cathode value (2K+) to increase headroom again, you push up plate voltage. Use too small a cathode resistor and the triode won't handle big signal voltages, negating any benefit in using a cathodyne in place of an LTP.

Perhaps you'll furnish us with a simple schem, with projected voltages, of your 550v supply on a 12AX7 gain stage feeding a cathodyne PI? If you have the answer, and it stands up to ear testing, then share it. Perhaps a comparison with an LTPI equipped, high voltage, super reverb?

There are undoubted ways to get the plate to cathode voltage in the right range but it's also got to sound good to be of any practical value.

[Merlinb]

A proven set up like 100K/1.5K is hardly arbitrary, it works in 90% of all guitar amps at a vast range of voltages. It's the middle ground. The point being that larger plate resistor will drop the voltage but after a point will impair fidelity.

"works in 90% of all guitar amps"
I think you are confusing what is used in 90% of amps and what works. There are infinite combinations of component values you could use to get whatever results you want. It sounds like you're trying to say that the only values which anyone could ever possibly want to use are 100k/1.5k, no matter what the voltages are?? Such nonsense aside, it doesn't change the fact that you can use a cathodyne at exactly the same voltages as any gain stage, which was the whole point of these ramblings.

"impair fidelity"
Words fail me.

[MWJB]

Frankly Merlin I don't know what your problem is...once again your passion for grandstanding is detracting from what might otherwise be an informative thread. This thread could be used to give people practical solutions to problems, or inspiring them to take a new direction (& prove you right)...but you continually fail to give any tangible instruction and insist in talking in abstract notions and "broad brush strokes".

Your attitude & tone of response stinks..."Um yeah but"...."but, why would anybody".."I don't get it"...you like to give the impression that the solution is easy and any fool can do it, whilst holding back on any practical data.

Come on share some info...or, as my old gran used to say, "if you have nothing useful to say, then say nothing". Dump, or get off the pottee.

At no point did I sat only 100K/1.5K is the only thing that works...it's just a yardstick. You know that.

"You can use a cathodyne at the same voltages as any other stage" example please. I want to see my 12AX7 cathodyne inverter tube with 550vdc B+ supply feeding both plates....and I want to know that it can cut the mustard in a comparison in an A/B against a LTPI. You must have amps about that you can experiment on.

Fidelity, you know what it means. Guitar amps may not be "hi fi" per se, but a degree of fidelity is desirable to stop the sound from turning to mush.

An infinite number of comonent values? Really, do you know what infinite means?

[Merlinb]

I have no idea why you are taking the thread in this direction. My only point was that you said there were special voltage limits on the cathodyne, and I say there aren't. That's it. I never said I was recommending anyone actually run the thing at 550V, people can do what they want, I just objected to your misinformation. The point is, the supply voltage is rarely a reason to favour the LTP over the cathodyne, or vice versa.

Yes I know what an infinite series is, and you could have an infinite number of possible values even for a single resistor, from zero to, well, infinity. Not all would be useful of course, but even between 0 ohms and 1 ohms there is still an infinite number of values. If, when you design an amp, you simply copy Fender values with no thought to what you're actually trying to get out, or what the supply voltages might be, then you're not really designing, you're just throwing stuff at the wall.

As for my attitude, I am loath ever to say that some is simply wrong. With comments like "Um yeah but"...."but, why would anybody".."I don't get it"... I was trying to be humble enough not to look like a complete idiot when you finally laid out a flawless explanation of why I was wrong. Which never materialized. Maybe I should have just said you were wrong?

[MWJB]

"My only point was that you said there were special voltage limits on the cathodyne, and I say there aren't." Well as far as I can see it IS a factor, maybe not a showstopper but it merits consideration.

"I just objected to your misinformation" Misinformation?! There's that attitude again. You cheeky SOB! I stated that cathodyne PI's gain stages were "rarely run much above 200vdc on the plate"...it's not misinformation, it's an accurate observation, with plenty of references.

You did say that you see no problem with running a 12AX7 with 500v on the plate...well I do see a problem with it regarding the pre PI gain stage. You are aware that others read these threads, many without your knowledge of tube circuits, to suggest that a 12AX7 will happily take 500v on the plate, in a typical gain stage, is somewhat irresponsible (you can't expect them to "fill in the blanks" unless you give them the tools do so) - unless you are going to provide a demonstration as to how this is achieved (you still have not).

Fender or not, it makes no odds (take half a dozen Fenders of any model and they'll all sound different anyway), the point is, that if you are going to the trouble of designing something it has to have something to offer, something that at least compares favourably with a benchmark, preferably offers an improvement...whether that be in tone desired, headroom... whatever. "Anything" is possible if you just expect the pilot light to come on and some noise to come out of the speaker.

If you're not going to contribute any practical/useful information, then I'm through with this thread.

[Merlinb]

Misinformation?! There's that attitude again. You cheeky SOB! I stated that cathodyne PI's gain stages were "rarely run much above 200vdc on the plate"...it's not misinformation, it's an accurate observation, with plenty of references.

Yes, but that wasn't the misinformation now was it. We've already established what you were wrong about- it was the supply voltage thing.

You did say that you see no problem with running a 12AX7 with 500v on the plate...

I didn't say 500V on the plate, I said from a 500V supply, or running it under 500V which means the same thing [EDIT: 550V to satisfy any pedantic readers]. When you run a valve "from x volts" it means the supply voltage; you actually have to specify plate if you mean plate voltage, because that is not a constant.

If you're not going to contribute any practical/useful information, then I'm through with this thread.

I think my first post was the most useful in the whole thread, since it actually answered Lowell's question.

[MWJB]

This is actually what you said, "Although an LTP can be used under higher voltages, a cathodyne can still be used up to the max value given on the datasheet, which is about 500V for a 12AX7." You did not specify "plate", or "supply", you did not link the data sheet that you refer to...you just said 500v.

"I think my first post was the most useful in the whole thread, since it actually answered Lowell's question." I'm sure you think so, you would though wouldn't you. It was one answer to Lowell's question among several...that were made in good faith, without contradiction, or derision.

[Zoe_N_Iain]

I didn't say 500V on the plate, I say from a 500V supply, or running it under 500V which means the same thing. When you run a valve "from x volts" it means the supply voltage; you actually have to specify plate if you mean plate voltage, because that is not a constant.

He's right, he didn't say 500v on the plate at any point.

[MWJB]

He said there was no limit to the supply voltage, now he is saying "from a 500v supply" (it's come down from 550v), meaning less than 500v. It doesn't strike you as contradictory? There are plenty of amps where the supply might exceed 500v at the PI node.

[raiken]

He's right, he didn't say 500v on the plate at any point.

To clarify, the ECC83 datasheet does say "Vao (max) 550V", Va (max) 300V", so you can indeed run them from a 550V supply if you want to, if you are careful about choosing component values to avoid exceeding the ratings.

While we're on the subject, other max ratings of importance are:

Anode dissipation - 1W max
Cathode current - 8mA max
Grid voltage - -50V max
Grid resistor - 2Meg max (cathode biased)
Cathode-to-heater voltage - 180V max

And this rather interesting one:

Cathode-to-heater circuit resistance in phase splitting circuits: 150K ohms (max)

[Zoe_N_Iain]

This is actually what you said, "Although an LTP can be used under higher voltages, a cathodyne can still be used up to the max value given on the datasheet, which is about 500V for a 12AX7." You did not specify "plate", or "supply", you did not link the data sheet that you refer to...you just said 500v.

I actually thought it was pretty obvious that Merlin was referring to the supply voltage, myself.

And since you asked, here's a datasheet that specifies Vao (Maximum Anode Voltage);

http://www.mif.pg.gda.pl/homepages/f...10/e/ECC83.pdf

Since it's possible to swing the signal upto very nearly that value, which would put 550v on the anode for a brief instant, the maximum supply voltage is thus 550v also. Practically, the supply voltage could be much larger, if you had a reason for it to be, provided the transient plate voltage never exceeded 550v (can't imagine why you'd want to, though). Be a bit touchy to do that in a cathodyne, of course, since to set the old-faithful 1/4 B+ grid voltage, you'd be riding close to the limits of the heater-cathode insulation rating, even with a lot of elevation applied.

Whatever the case, I'm afraid I have to agree with Merlin that cathodynes can be run from a higher supply voltage than you may have commonly seen and work well doing it. I've a couple of my amps being regularly gigged (by other people, my playing sucks!) with a cathodyne running at 450v supply and with a 340k plate load on the driver triode, as it happens.

[MWJB]

"I've a couple of my amps being regularly gigged (by other people, my playing sucks!) with a cathodyne running at 450v supply and with a 340k plate load on the driver triode, as it happens." I'm not going disagree anything there, that's perfectly reasonable. You don't even "need" a plate resistor that big (obviously what sounds best is best on that front). I haven't atany point suggested that might this not be a good idea. My only observation is that 450v supply at the PI node is not particularly high.