The tube circuit is differential...equivalent of a 300V?

[mike_mccue]

I'm curious about the ART MPA Gold preamp. I had an early version with apprx 40vDC on the plates. I was familiar with the idea that it was starved plate design.

The current spec is this:

"The selectable Tube Plate Voltage function delivers a wide variety of tones. The tube circuit is differential, and provides the equivalent of a 300V power supply on the high voltage setting. Not to mention it has very low distortion and noise. Our normal setting maintains ART’s classic tube sound (warm with smooth overload characteristics)."

Does anyone know what that means?

Why wouldn't they say something like 250vDC?

Is there more to the spec than seems obvious?

thanks,
mike

[mike_mccue]

I went off and learned about the idea of differential preamp circuits.

I guess the idea of using 2 sides of a 12AX7 is a popluar idea in hi-fi preamps.

I'm assuming that the ART MPA Gold is running each plate at apprx +/-150vDC (I've seen a few claims of 180vDC) and they have to keep the plate current low to keep it physically cool.

The only way to get the equivalent gain of a single 12AX7 section with a plate voltage of 300vDc (max rating) is to use a differential design.

I have a real basic question: Does differential mean that the 12AX7 is functioning as a miniature push pull amplifier?

I'm not sure how that relates to linearity or how you could compare that to the transconductance curve you would expect at the familiar 260vDC sweetspot on a single stage of a 12AX7. Can someone help me understand?

Does any of that make sense?

thanks,
mike

[mike_mccue]

I think one thing that is really confusing me is that I'm reading about opamp designs that use + and - voltage supplies, but many of the 12AX7 differential amp schematics I am looking at seem to be powered off the same B+.

There must be something obvious that I'm missing.

thanks,
mike

[Steve Conner]

"The selectable Tube Plate Voltage function delivers a wide variety of tones. The tube circuit is differential, and provides the equivalent of a 300V power supply on the high voltage setting. Not to mention it has very low distortion and noise. Our normal setting maintains ART’s classic tube sound (warm with smooth overload characteristics)."

Does anyone know what that means?

It means they paid some advertising copywriter to make it sound impressive. I wouldn't try to read into it any more than that. If you really want to know how it works, you might be best calling ART or buying one and taking it apart...

Yes, differential might refer to a push-pull amp like the LTPI (Long Tailed Phase Inverter found in many guitar amps) but really once the marketing guys have been at it, it could mean anything. A push-pull amp tends to be more linear than a single tube stage, as it cancels even harmonics. Which is a strange thing to do in a product that's supposed to deliver tube sound...

Also a differential amp doesn't require + and - rails to work, after all, the LTPI runs fine off a single positive rail.

My blind guess is that even the "high voltage" setting is only something like 48v, or they wouldn't have had their ad guys use weasel words like "The equivalent of 300V". I remember when Rocktron brought out the Piranha, their ads screamed that it ran the tubes on a full 250V.

Tubes generate a lot of dirt at these low voltages, but maybe with the differential circuit they were able to cancel a lot of it out.

[mike_mccue]

Thank you Steve (again),

I received a schematic tonight.

There is so much solid state stuff surrounding the tube I'm having trouble following the signal path.

It does appear they have a 150vDC plate voltage but there seems to be some lower voltage that can be switched in for a "low gain" setting.

I'd like to be honest about this. Yesterday, I advised an aquaintance that this preamp was nothing like the tube preamp it is advertised as. I owned a previous version for a week and sold it because it sounded like an industrial mass produced solid state preamp. I opened it up with a tech friend, learned about the starved plate design, and became very disappointed. After explaining this my aquaintance basically said (I'm paraphrasing) "what do you know! I read a great review in SOS".

I don't know a lot about real electronics engineering (I'm a sound recordist, a life long sound gear enthusiast, and a good field *engineer*) but when I went to read the latest specs I found that crazy description.

I'd like to take the opportunity to learn some more myself and maybe help my friend learn a bit as well.

Would you consider looking at the schematic?

I'm also trying to learn why the general statement that a diferential amp, with regards to a 12AX7, is more linear than a single ended class A design.

I know a bit about Transconductance curves so it seems to me after reading about differential amps that they run the circuit at lower voltage along a linear portion of the curve but must be loaded so that they don't flow too much current. It seems to me that the lower current equates to less gain so the dual stage design is basically required to bring the gain up to something relative to a single ended circuit at a higher voltage.
I guess the even order harmonics are canceled out by the reversed polarity of the two triode circuits?

I'm sure I'm saying all kinds of not quite accurate statements.

If you can take the time to correct anything I'm not getting right... I would very much appreciate the help.

thanks very much,
mike

[Enzo]

Well we can't look at a schematic you didn't post or link to...

When i see a mic pre calling itself differential, I don't think of op amp type circuits. To me that conjurs up images of balanced circuits. I could be wrong of course. Instead of taking a balanced mic signal and turning it into a single ended signal through the amp, then turning it back into balanced at the output, some circuits just use two identical signal paths through the amp. One for each polarity at the balanced input. So the + input and the - input each has its own circuit through to the output. The two sides don't really interact.

Look here:
http://www.waltzingbear.com/Schemati...Altec_459A.htm

Ignoring for a moment the transformers to match impedance at the ends, the signal path through this amp is separate for the two polarities.

I look forward to your schematic.

[mike_mccue]

Thanks for taking a look at this thread Enzo,

Here's a link to the schematic:


ART-MPA-Gold-schematic.pdf

best regards,
mike


edit to add: The Altec schematic is something that I can reasonably interpret... that's for providing an example.

[Enzo]

Oh, looks like the tube stage is a differential amp after all.

All the solid state stuff does is interface the tube voltages with the rest of the circuits low voltages - as far as I can see.

Looking at the lower channel strip:

The shared cathode see R376 as the cathode resistor, the rest of the stuff down there just controls current.

The plates run at the 150v, yes, but you can't run 150v to the op amps. The engineer types around here can correct me, but that pile of transistors in the plate circuits looks to me like they are just current mirrors. That all provides a suitable level current signal from between Q311 and Q312 to feed op amp U303B, which sends it along. U303A is just an inverter for the phase switch that follows.

After the phase switch is a pretty conventional output stage. Output level control and a balanced output driver stage.

[mike_mccue]

Is it a LTPI?

Is my question about canceling out even ordered harmonics because it's "push-pull"?

Thanks,
mike

[Steve Conner]

Well, I'd say only a marketeer could call that a tube mic pre. I'd call it a solid-state mic pre with "Hint of Tube".

The actual gain is provided by a standard solid-state mic pre circuit, then it gets fed through the tubes to add distortion.

Enzo is correct: that bunch of transistors is a high-voltage current mirror that basically translates the tube signal back to op-amp levels, converting the differential signal back to single-ended as it does so. (It's doing the same thing as the output transformer in a push-pull amp. So yes, it is push-pull, and this is where the even harmonics would get cancelled.)

The output from the mirror stage is in the form of a current, and U301A (U303B if you're looking at the right channel, as I guess Enzo was) is used as a transimpedance amplifier to convert that back to a voltage.

I think they even wrapped a negative feedback loop round the whole mess of tube, current mirrors and op-amp. Why, I have no idea, maybe they thought it still sounded too tubey. Overall I guess I would class it as essentially the same circuit as the LTPI. It's biased by a transistor current source in the cathode, but that's not too far different from just being a really "long" tail resistor.

A differential amp is more linear than a single-ended one made with the same active device because, and only because, the differential circuit cancels the even-order harmonics generated by the active device. Less harmonic distortion is the same thing as better linearity. I don't know why they chose to remove the even harmonics since the main advantage of tubes is the euphonic mixture of even and odd harmonics you get from a tube in a single-ended circuit.

I'm kind of the opposite, I am a professional electronic engineer, but I design lab instruments and gizmos for industry. I don't have anything to do with the audio industry, I just play with amps and recording gear for fun.

[teemuk]

Kind of what I thought as well. I really wonder what is the point of using tubes in that circuit if not for marketing purposes. There’s about 20 dB of negative feedback wrapped around that differential stage and – the way I see it - that pretty much corrects all distortions introduced by the tubes. It also causes the whole stage to clip very abruptly (think a generic OpAmp since this is basically one anyway), which – I guess - is why they had to add that transistor shunt to decrease the tail current at higher signal amplitudes. Technically this is a very interesting circuit but its actual performance….

[mike_mccue]

FWIW,

You two are describing what I hear when I've listened to these units.

I'm going to read your posts in detail and follow the schematic... it may take me a while to digest.

Thanks very much for taking a look and offering insights.

best regards,
mike