More SEPs

[Paul Barker]

Doesn’t the ultra path cap have to be bigger? Than the cathode resistor cap was?

There was also another connection that Lynn doesn’t mention, where you still keep the cathode cap,and another cap from same node at cathode to b+ and I think it was 1/3rd the value of cathode resistor bypass cap, I think, if my memory is correct, which injected power supply noise, adjust cap,size on test choosing cap size by best noise cancellation from the cap in b+ to cathode. Cant remember what it was called. Anyone else remember ? Might not be as good as ultrapath due to electrolytic in place, but I think I used polyprop or pio on cathode aswell. It sounded good.

It was first talked about on the Triodeel email group which was closed by Ned in about 1998. Ive lost the email asddress so have nothing left of it. Steve Bench was a big contributor.

[Cressy Snr]

Dunno about the cap size Paul.

I’ll be perfectly honest: I looked at Pete Millett’s E-Linear amp and noted that he used a 40uF cap for his Ultrapath connection. He associated the cap size with the internal resistance of the KT88 in UL mode and argued that a 300B would need a at least a 100uF Ultrapath cap due to its lower internal resistance.
I saw from the datasheets that a 6L6 in pentode mode has about twice the internal resistance of the KT88, so the cap value would be lower than 40uF. I simply guessed what the value might be: no calculations. 30uF was what I had and I figured (by Millett’s logic) that it would do.

It seems to do the job. Total silence from the power supply, great bass and a big sound like the best bits of a Denon DL103.

[Nick]

think it was 1/3rd the value of cathode resistor bypass cap

If I remember correctly (which I may not) the ratio of the two caps was related to the gm of the valve.

[Paul Barker]

Or was it the mu+1 ?

Might just have to experiment.

[Paul Barker]

Steve you’re right, higher resistance is smaller cap. The vtv article was about low resistance triode.

[Nick]

https://www.tempoelectric.com/Ultrapath ... _Olson.pdf

[Cressy Snr]

Must be very good indeed then Steve as it was very good before

Thanks Simon.
Yes the performance of the amp does seem to have gone up another notch.

[IslandPink]

Thanks for the link Nick - excellent.

[Thermionic Idler]

That'll be an interesting article to read, not seen it before. Being based on a Lynn Olson design, my 300B's use Ultrapath but in push-pull obviously:

EDIT - having read the article, it sounds like it's pretty close to what Western Electric were doing in the WE 86. Lynn calls circuit F "Western Electric PP".

[Cressy Snr]

Well blow me down,
I'm trying the SEP as a headphone amp at the moment on a pair of our Ant's Audio Technica AT706s, which plug into the speaker outlets via a box of tricks. Very impressed with the sound over headphones.
What is even more impressive is the complete absence of any background hiss or hum from the phones when there is no music playing, and when I infer silence, I mean complete silence. It's uncanny. So (though I say so myself) I appear to have done a cracking job on the power supply.

[simon]

What's in the box of tricks Steve (Ant)? I've been musing on a headphone amp but don't really know anything about them.

[Paul Barker]

A few of thos early circuits Nick found recently had a choke before the pp node of the pp opt and b+. On the various audiassylum forums sometimes mentioned a choke feeding a centre tapped phase spliting choke. Might have been Dave Slagles suggestion.

Probably cant then use the cap though. Either or. Dont want to make a tank circuit.

Going off memory their was one of the earliest gm70 article in Glass Audio which on listening tests they liked the sound of a loading choke in series with a predominately ccs cathode load for the driver.

[Cressy Snr]

What's in the box of tricks Steve (Ant)? I've been musing on a headphone amp but don't really know anything about them.

There’s probably some kind of step-down arrangement that causes the amp to think it is driving speakers. The stepped down signal then drives the electret transducers in the earpieces. The ‘phones plug into the front of the box via a 5-pin DIN plug. This probably dates these headphones to the 1970s.

EDIT
I’ve found a picture of the insides of the box:

IMG_1037.jpeg (39.97 KiB) Viewed 616 times

[simon]

Thanks Steve. I figured it must be something like that, would be interesting to see a circuit diagram. Or just build something more tailored I s'pose.

[Ant]

That box is just a pair of step up transformers to run the headphones, with a bypass switch on it to run speakers its not an amp. It has 2 different secondaries giving 2 gain levels for the headphones off the same switch