PJCCS added to Quickie

[machinehead]

My Quickie now has the PJCCS added to it as of this weekend.

The sound is much improved, more open and lively, less dull?

It also seems to have more gain, less need to turn up the volume knob quite as high.

This is will fresh batteries.  I thought the purpose was to keep current at a  constant while the voltage of the
batteries degraded through their life span.

Seems that it really does more than that.  

Can you guys explain more of the operation theory?

Thanks.

[Paul Joppa]

Current source (and also choke) loads reduce distortion in voltage amplifiers. If you are familiar with plate curves and load lines, you can more easily visualize this effect.

The stock load resistor of 4000 ohms loads down the tube, which has an internal impedance of 4000 ohms. This reduces the output by half, or 6dB. The current source does not load down the output, so the gain is increased by 6dB; however the output impedance is also increased so it will have more trouble driving long cables. (I posted some more details on this here recently.)

[bundee1]

How is the bass since you added the PJCCS?

[machinehead]

bundee1, I need to do more listening and get back to you. I need to put it back into my system.
I didn't notice a lack of bass in my system to begin with. 

[machinehead]

Put the Quickie back in the big rig again tonight.

Plenty of bass.  I can't remember what it sounded like without the constant current supply.

I can tell you that there is a big sonic difference in different brands of tubes, or maybe its just a different pair of tubes.
I did have this guy match these pair, that could be it.  Not sure if the stock tubes are matched on a tube tester or not.
Left and right balance seems better with these new tubes.

[bundee1]

what tubes are you using at the moment?

[machinehead]

a pair of RCA's unknown year, from the box, I would say 50's or 60's.

There is a guy in FL, online I bought them from.

[Dr. Toobz]

Current source (and also choke) loads reduce distortion in voltage amplifiers. If you are familiar with plate curves and load lines, you can more easily visualize this effect.

The stock load resistor of 4000 ohms loads down the tube, which has an internal impedance of 4000 ohms. This reduces the output by half, or 6dB. The current source does not load down the output, so the gain is increased by 6dB; however the output impedance is also increased so it will have more trouble driving long cables. (I posted some more details on this here recently.)

Does this mean that a plate choke with a DCR of about 4k would also "load down" the tube like a resistor? Yet, I thought the chokes also increased gain over a plain old resistor load, so what are they doing differently?

[Paul Joppa]

The AC impedance of a plate choke is very high, and will not load down the tube at audio frequencies. The DC resistance is much lower, and sets the DC operating point which also affects the maximum voltage - but not the audio-frequency gain which is set by the impedance at audio frequencies.

Clear as mud?  :^)

[machinehead]

Would it not make sense to add a fuse to the Quickie, if so, what rating, and where?

[Dr. Toobz]

The AC impedance of a plate choke is very high, and will not load down the tube at audio frequencies. The DC resistance is much lower, and sets the DC operating point which also affects the maximum voltage - but not the audio-frequency gain which is set by the impedance at audio frequencies.

Clear as mud?  :^)

I suppose :-) What's interesting to me is that I've done comparisons of the PJCSS and chokes, and while close, the chokes won. I had one channel hooked up with one, and one with the other, and A/B'ed mono material back and forth. I also had the PJCSS hooked up for a while on its own, and then went back to the chokes (now with LED's, too - which I'm sure add very little resistance, but do drop the voltage somewhat). I'm not sure why I would prefer the chokes - I think they might sound more "natural" and work to tame my digital source (DAC) - and I noticed bass note clipping sooner with the current source as opposed to the chokes. I think this must be because either a) the PJCSS maybe squeezes a little more gain from the tube than the choke, pushing it to the very edge of distortion-free operation, or b) the choke can store current whereas the current source cannot. Would the choke be able to swing a higher voltage than the B+ through a current source, considering a choke's ability to store energy?

Considering the fact that the chokes add hum (only perceptible when the S.E.X. is turned up all the way), the PJCSS is probably a better bet for most people. I think the chokes should be considered, though, if you need to tame a harsh digital front-end or need warmer, "fatter" sounding bass. Both are way better than the plate resistors. It's nice to have some options in order to "customize" the Quickie to a particular system!

[Paul Joppa]

Good point - yes, the current source will clip by limiting the positive excursion of the plate voltage, a bit below the battery voltage. The choke will allow that voltage to rise higher.

The stock resistor load will also limit the positive voltage peak. The limit will however be earlier but softer, as the tube gets forced into the high voltage/low current ares where it becomes less linear - you will get more compression, which while distorted is softer sounding.

Very roughly speaking, driving a high impedance load of 50K or more, and with fresh batteries, the stock resistor load can in theory produce a maximum of 5vRMS. The current source will have basically the same peak output; some experimenting with R1, the current set resistor, might optimize that. The choke load can produce 7vRMS, and if you change the cathode resistor to 2K (from 1K) it should approach 9vRMS but with reduced peak current drive available. These numbers are theoretical predictions - I invite any owner with a 'scope to put them to the test and get some more data up on the forum here!

[Dr. Toobz]

....and if you change the cathode resistor to 2K (from 1K) it should approach 9vRMS but with reduced peak current drive available.

One more question - what effect would changing the cathode resistor to 2k (and getting that 9vRMS maximum) have on sound quality? I've always wondered this, since I've never messed with the bias side of things.