Dennis Had Inspire Amps

[Musica Amantem]

Comzee: Let us know how you like the 6L6 vs KT66 in that same configuration. I love my 6P3S-E (6L6 variation) with the same Sophia mesh plate rectifier (274B). Thanks!

[Comzee]

Originally Posted by Musica Amantem

Comzee: Let us know how you like the 6L6 vs KT66 in that same configuration. I love my 6P3S-E (6L6 variation) with the same Sophia mesh plate rectifier (274B). Thanks!

kt-66 is one of the few power tube pairs I don't have

[Musica Amantem]

Originally Posted by FloridaBoy

I got a headache reading that. Is it a legal brief?

Now that I read it again, you're right! I just meant, IMO the output tubes are to be picked from a limited family gravitating around the KT66 for easier achievement of an optimal setting (6V6-6L6-KT66-KT88).

All tubes in this amp matter, but the impact of choosing output tubes outside the proposed range makes it more difficult as the increasing power of larger Beam Tetrodes start trading transparency and detailing for sheer power and dynamics, in my limited experience. Sorry for the blabber!

[Musica Amantem]

Originally Posted by Comzee

kt-66 is one of the few power tube pairs I don't have

Neither do I, that's why I was asking for your impressions. I believe the best output tubes must be very close to the KT-66 spec provided we use the right rectifiers and 6SN7's. I'll soon get me a pair to test my hypothesis.

[Musica Amantem]

Originally Posted by Rosco65

The variations in tubes - input/driver, rectifier, and output - are all equally important in the character of an amplifier's sound, especially one as simple and well-developed as the PSE. We should remember that the power supply - including rectifier tubes - is all-important in tube amps.

Lets establish that Dennis new power supply is brand new, in fact it has been changed almost immediately following AA's listening session report to allow for directly-heated rectifiers. AA spoke at how much better the amp with the Infinite Impedance Power Supply handled overload. But subsequently, both AA and Dennis have indicated there is lower distortion at all power levels, not just in overload.

Dennis PS design essentially isolates the input and screen supply from the output tube plate voltage. This would imply that variations - however small - in plate voltage would be less likely to be transmitted to the screen, resulting in a more stable screen voltage.

The excellent All About Circuits online textbook states that "With a constant DC screen voltage, electron flow from cathode to plate became almost exclusively dependent upon grid voltage, meaning the plate voltage could vary over a wide range with little effect on plate current. This made for more stable gains in amplifier circuits, and better linearity for more accurate reproduction of the input signal waveform."

Along similar lines, Dennis told me "The screen grids play such an important role with these beam power tubes. The DC power source feeding the screen voltage potential must be in the purest form without ripple and of course regulated." He was specifically referring to GL KT-77's in this revised PSE circuit, which he called "spectacular", and went so far and to say that "The longer I listen with the KT77’s I get the feeling this just may be a CAD805 beater."

I'm about to send my PSE back to Dennis to be retrofitted. I can't guarantee that I will be able to report accurate A/B comparisons since I haven't owned it that long and it will be at least a couple of weeks between listening sessions. I will keep everyone in the loop.

Let's assume for a moment we never go overboard with power demands and therefore never push the tubes to their clipping zone. Unless unexpected music source dynamics require a sudden peak, one can control the nominal SPL to ensure the amp is far from its limits. Also, nominal headroom can be extended beforehand by choosing more powerful output tubes within the trannies' capabilities.

In the described scenario, would the new power supply design improve sound tangibly?

[pstrisik]

Originally Posted by Musica Amantem

Let's assume for a moment we never go overboard with power demands and therefore never push the tubes to their clipping zone. Unless unexpected music source dynamics require a sudden peak, one can control the nominal SPL to ensure the amp is far from its limits. Also, nominal headroom can be extended beforehand by choosing more powerful output tubes within the trannies' capabilities.

In the described scenario, would the new power supply design improve sound tangibly?

Good question!

[CoGT3]

Originally Posted by Musica Amantem

Let's assume for a moment we never go overboard with power demands and therefore never push the tubes to their clipping zone. Unless unexpected music source dynamics require a sudden peak, one can control the nominal SPL to ensure the amp is far from its limits. Also, nominal headroom can be extended beforehand by choosing more powerful output tubes within the trannies' capabilities. In the described scenario, would the new power supply design improve sound tangibly?

I posed that question earlier to Analog and this was his response:

"do know that one of the beneficiaries of the new PS is the screen grid voltage. The screen grid’s purpose is to reduce capacitance that arises between the control grid and the plate. Such parasitic capacitance can cause the tube’s circuit to become self-resonant at some frequencies, and it will reduce the tube’s achievable gain at higher frequencies. This problem is called the Miller Effect, and the screen grid helps to resolve it. A triode amplifier will typically require some type of “neutralization” circuit outside of the vacuum tube to avoid the detriments of the Miller Effect. ((G6B10) Vacuum Tubes - Ham Radio School.com)
Any variance in the voltage at the screen grid will cause fluctuations in the electron flow, introducing unwanted variance in the signal.

From Dennis' comments...."What I discovered was when an output device, in my case a vacuum tube, became non-linear and produced various spurious distortion products back fed to the power supply. In the case of a basic power supply supplying both the output devices along with the driving front gain sections I was able to detect these modulated parasitic products in the driver stage compounding the problem of output distortion of the final product.

So what he is saying (I think) is that the clipping/distortion produced by a transient dynamic peak is compounded by bleeding into the (assumed) regulated power supply feeding the driver tubes and the output tube screen grids, causing even greater distortion. Call it unintentional negative feedback of the worst type.

So, in a flash of Inspire-ation he came up with "the idea of using the elements of the rectifier tube to isolate the power supply from any distortion or variance produced by the output tubes. Think of it kinda like a isolation or interstage transformer in some ways. There is no direct contact linking the two ends of the circuit. Instead, Dennis is using the vacuum of the rectifier tube to isolate the regulated power supply from the rest of the circuit, hence the "infinite impedence" part of the circuit. The novel part of it is that he is running direct current through the rectifier, and not AC as would normally be applied to a rectifier.

Now with respect to the 45 amp, remember that there's no screen grid in the 45, so you would lose the benefit of this effect. However, any amp will use driver tubes, so they would benefit from this application. My undereducated guess is that the amp would benefit, but not so much as a tetrode amp."

Short answer is yes, with the pentode amps it should be a benefit at less than the clipping point.

[Musica Amantem]

With the Infinite Impedance mod, what takes care of the original tube-rectifying job and what are the tradeoffs involved? Is the rectifier still doing its originally (essential) intended duty on top of the new role?

Are the mod's ramifications the same with the PSE than with the single tube per channel Beam-Tetrode SET?

Thanks!

[CoGT3]

My understanding is that it uses solid state diodes to rectify the AC to DC. The DC then passed through the rectifier tube, then to the regulated power supply sections. The vacuum in the rectifier tube acts to stop backward progression of power distortion from making back into the diode rectifiers.

Hopefully Analog can clarify as needed.

[Analog Addict]

Originally Posted by CoGT3

My understanding is that it uses solid state diodes to rectify the AC to DC. The DC then passed through the rectifier tube, then to the regulated power supply sections.

First part is correct. Soiled state rectifier. The type of rectifier tube now used in the circuit is directly heated, since Dennis grounded the 5 VCT filament supply.

Originally Posted by CoGT3

The vacuum in the rectifier tube acts to stop backward progression of power distortion from making back into the diode rectifiers.

This part is to my knowledge, incorrect. The trick to this circuit is the power supply providing rock solid voltage to the screen grid. As someone else has stated, it's small oscillations in the voltage at the screen grid which then get amplifed as well. If there is noise or hash in that voltage, that gets amplified as well.