Yes, the 600 Ohm variant of the tri-filar Edcor would be the one to use.
I used an older pair of Punkydawg’s matched 8mA JFets.
The adjustment procedure for P1 and P2 is very much a see-saw affair. I started by opening up both pots a full turn (counterclockwise),then alternated one turn on each of the pots until the bias current started to come up. I monitor the bias current across R2 and the offset across the speaker terminals. Initial adjustment can be done one channel at a time with the lid off. Final adjustments need to be done after the amp has reached a stable temperature with the lid resting on top.
I used an older pair of Punkydawg’s matched 8mA JFets.
The adjustment procedure for P1 and P2 is very much a see-saw affair. I started by opening up both pots a full turn (counterclockwise),then alternated one turn on each of the pots until the bias current started to come up. I monitor the bias current across R2 and the offset across the speaker terminals. Initial adjustment can be done one channel at a time with the lid off. Final adjustments need to be done after the amp has reached a stable temperature with the lid resting on top.
Thanks for that, what bias current did you finally settle on? I seem to remember the biasing is a see saw affair using the 2 pots.
the Jfets are not a problem - have those in NOS Toshiba's.
I also assume you have the IRFP048 mosfets installed?
How low did you manage to get the offset voltage down to?
Just gathering all the relevant info before the build.
the Jfets are not a problem - have those in NOS Toshiba's.
I also assume you have the IRFP048 mosfets installed?
How low did you manage to get the offset voltage down to?
Just gathering all the relevant info before the build.
I‘m running the IRFP048s at 1.6A; was able to get the offset below 10 mV over the entire warm up period. I could increase the bias, but the amp sounds really nice as is. I’ve got the old Keratherm pads left from the previous boards that were installed in this chassis. The new boards have the Mosfets relocated closer to the center of the heatsinks. Warm up happens remarkably quickly.
@TungstenAudio - if you are revising the board layout, can I suggest that you add parallel lands for the R1* value of 0.0909R to have parallel resistors on the pcb rather than piggybacking the second parallel resistor on the first resistor for R1* - a neater solution.
Hi
As @6L6 mentionned "I'd echo having the possibility of other transformers, preferably Edcor and CineMag, (if you have to choose just one, Edcor please) for the exact same reason Nelson's new F6m is being made... the Jensen transformers are getting very expensive"
From a European perspective, i couldn't find a local dealer for those.
Jensen are 50USD
x2
Add shipping from the US to Europe with a reliable carrier +50USD
Add Customs taxes
Add VAT
= we are talking 200€! Argh!
Edcor are more available here...
As @6L6 mentionned "I'd echo having the possibility of other transformers, preferably Edcor and CineMag, (if you have to choose just one, Edcor please) for the exact same reason Nelson's new F6m is being made... the Jensen transformers are getting very expensive"
From a European perspective, i couldn't find a local dealer for those.
Jensen are 50USD
x2
Add shipping from the US to Europe with a reliable carrier +50USD
Add Customs taxes
Add VAT
= we are talking 200€! Argh!
Edcor are more available here...
I've had some time to look at the Edcor tri-filar footprint. It is much larger than the Jensen (more pins) and physically larger to the point of requiring an entirely new layout. Still feasible to fit into the overall dimensions of the PCB, but a lot more work to basically start over.
I would like to present the Gerbers for the JFet input PCBs now so our eager builders (greedy boyz & girlz) may get a head start with the Jensen transformers if they have some already or can manage the cost (less $$ in the U.S.)
I would like to present the Gerbers for the JFet input PCBs now so our eager builders (greedy boyz & girlz) may get a head start with the Jensen transformers if they have some already or can manage the cost (less $$ in the U.S.)
I bought my Jensens from Cable Solutions. Search under products by manufacturer.
https://www.cs1.net/
Yes, I bought four of them to test the new PCBs.
https://www.cs1.net/
Yes, I bought four of them to test the new PCBs.
. 
Nice job, I like that layout better. It looks like it places the output FETs more centrally on the heat sink. I'll have a closer look at the schematica later. it looks really interesting to me.
I installed two pairs of motor run caps (big 70 uF polypropylene in oil) in the F6 test bed yesterday. Each pair is effectively dedicated to one of the channels by proximity and short wires. Technically this is still a shared PSU, but the big film caps give a nice approximation of what a full dual-mono PSU would sound like. I got the general idea from the Wiley Zen Mod.
The addition was immediately noticeable, as the soundstage became detached from the speakers. While the sonic landscape doesn’t occupy the entire rear wall of my listening area, it is no longer bound to the locations of the speakers. Height, width and depth are all improved, plus individual instruments have a more palpable sense of being physically present. (Speakers and preamp need to be up to this.) This is the first time I’ve used motor run caps in a shared PSU this way; I‘m plesed with the results.
Just for grins I checked bias and offset this morning, and took the opportunity to goose the bias slightly to 1.65A. Further listening today has been at least equally pleasant. The F6 is remarkably dynamic and liquid in this configuration.
The addition was immediately noticeable, as the soundstage became detached from the speakers. While the sonic landscape doesn’t occupy the entire rear wall of my listening area, it is no longer bound to the locations of the speakers. Height, width and depth are all improved, plus individual instruments have a more palpable sense of being physically present. (Speakers and preamp need to be up to this.) This is the first time I’ve used motor run caps in a shared PSU this way; I‘m plesed with the results.
Just for grins I checked bias and offset this morning, and took the opportunity to goose the bias slightly to 1.65A. Further listening today has been at least equally pleasant. The F6 is remarkably dynamic and liquid in this configuration.
Attachments
