Anyone try a regulated PS in an Ultrapath preamp?

[KT]

I was revisiting some old projects and have a question about the Ultrapath preamp.

One of the selling points of the DRD, or Ultrapath, circuit, is that it effectively removes the power supply from the signal path. This allegedly reduces the contribution of power supply components to the sonics, ameliorating the need to use some ultra-pricey parts in the PS for good results.

I'm curious if those of you who have built the preamp found this to be true. I know that the Ultrapath cap is supposed to contribute to the sound (with GE motor run caps sounding very good, IIRC), but how about the power supply parts?

I'm specifically curious about the use of a regulated supply.

I have an old QPSA-1 tube-level regulator from Curcio Sound Lab, and was thinking of using it in the PS. I imagine this could save me from purchasing some potentially expensive chokes.

I wonder if any of you have noticed enough of a difference in the sonic contribution of various power supply components in the Ultrapath circuit to suggest that a regulator may contribute too strongly to the sound (since many regulators are described as having a mechanical or electronic signature).

Thanks!
KT

[leadbelly]

First of all the DRD is not the same as Ultrapath, and in fact is completely irrelevant if discussing preamps since the DRD is a power amp design. The acronym DRD refers to the choke loaded driver scheme.

You make want to read: Ultrapath = Ultra Simple...Except that it's not for some insight.

[KT]

Thanks for the clarification, Leadbelly.

I'll look at the article a bit later.

Best,
KT

[KT]

Sorry, it is the QPSA-7, not 1.

Link was to a great discussion of the Ultrapath circuit. http://www.tubecad.com/2008/08/blog0147.htm

Based on the findings, it seems the circuit suffers from some serious shortcomings.

So the Ultrapath actually is more succeptible to PS artifacts and ripple, not less so as originally thought? That's really interesting.

I did notice that the circuit at Electra-print had been modified from the original one: the PS now is diode rectified and does away with the chokes. There is also a stack of diodes tying the cathode and ultrapath cap to ground. I guess all these changes were made to minimize PS noise.

http://www.electra-print.com/images/ultrapath.jpg

So is anyone using a non-battery powered Ultrapath preamp happily, with no hum or PS noise? Since I have the transformers for the original design, I feel like I'll need to try it for myself and make adjustments where necessary.

I'm just wondering if it would be worth it to use the regulator ahead of the original choke loaded filter. Just thinking out loud, but would that eliminate some of the ripple and noise, or just make things worse?

Thanks!
KT

[dsavitsk]

Originally Posted by KT

Link was to a great discussion of the Ultrapath circuit. Ultrapath = Ultra Simple...Except that it's not

Link was to a bad strawman argument. Really, most of that has little to nothing to do with the UP circuit. The UP requires an OPT, making the first half unrelated. The parafeed example is dumb, and a parafeed cap in a western electric connection is essentially an UP circuit -- no need to add the extra cap to nowhere.

Yes, the UP will have terrible PSRR, but the advantage is that the signal current loops are shortened. Is the UP the best way to do this? Maybe not, but it is a viable way and has some advantages. If PSRR is a concern, then build a good regulator.

I did notice that the circuit at Electra-print had been modified from the original one: the PS now is diode rectified and does away with the chokes. There is also a stack of diodes tying the cathode and ultrapath cap to ground. I guess all these changes were made to minimize PS noise.

http://www.electra-print.com/images/ultrapath.jpg

The mods don't make much sense -- replacing the cathode bias resistor (or diodes here) with a CCS would be the much better option -- really more in the UP spirit I'd think. You want the impedance of the bias mechanism high, not low.

So is anyone using a non-battery powered Ultrapath preamp happily, with no hum or PS noise? Since I have the transformers for the original design, I feel like I'll need to try it for myself and make adjustments where necessary

I'm not, but I have built one before, and so long as the PS is clean it works fine. Here's my version using a glow tube instead of the cap for the signal current return: ecp.cc

If you want to build the original, build a good shunt regulator being fed by a good CCS. Then bias the tube with a CCS and you should end up with a very nice and quiet circuit.

[KT]

Thanks for the input, Crusher. Your experience and thoughts make me feel more reassured about building the circuit.

Nice looking build, by the way. Classy wood base and top plate. Colors of the wood, top plate, and TX paint match up very nicely. The layout looks very proportional, nothing out of place. Very solid. Good discussion, too, which will take some time for me to digest.

EFT presentation link looks like a great read. Again, will look at it perhaps later today.

As for a CCS, what is your opinion of the Bottlehead Anticipation? I have a kit that I bought for an older Foreplay lying around. Perhaps that would do for the CCS?

OK, just checked and they don't make the anticipation any longer, but here is the description:

C4S is a high impedance current limit circuit for preamp and driver tubes, designed by John Camille as an improvement in effective impedance, power supply rejection and stability over previous designs using FETs or tubes. It includes two PNP transistors and two LEDs per tiny PC board (2 boards are provided in the kit for a stereo pair) and an extensive 28 page manual written by "Buddha" Camille himself. This current source may be used as a plate or collector load for voltage differentials up to 300 volts. Use of NPN transistors will allow use if the board as a current sink for cathode loads in cathode follower and differential circuit applications. The circuit current and bias are easily adjusted with just two resistors. The C4S circuit boards are designed for easy series and parallel resistor configurations, for precise matching. What does all this mean? The 5 megohm effective impedance characteristic of these active loads permit near theoretical gain performance from voltage amplifier stages, and can reduce distortion in vacuum tube audio circuits by several hundred percent. Noise generated in this circuit is up to 40 times less than several commonly used circuits. The result is clarity, bass speed and drive that beats resistive loads, choke loads, and other tube and solid state loading schemes like mu followers/stages and SRPP’s...

Sorry I don't have the schematic on hand. I'll have to rummage around my parts bin and post it when I find it.

Or is there another CCS you'd recommend looking at?

Thanks,
KT

[KT]

Hi Crusher,

I just found the discussion of Bottlehead's CCS on your site. Reading it now.

Thanks,
KT