If we put the output transistors too close the heat won't spread as evenly on the heatsink.
I see that Peter just ran out of his PS boards, but if he has more made would there be any reason not to just use those? They accept 50mm caps and are modular, so if you want to use more caps you can just add more boards.
http://www.diyaudio.com/forums/audio-sector/149672-universal-power-supply-pcb.html
http://www.diyaudio.com/forums/audio-sector/149672-universal-power-supply-pcb.html
Hi there!
I was thinking about running F5 Turbo with 75V rails and 8 2SK1530/2SJ201 pairs per channel, each output device with 100mA bias. So I ask whether the lower bias is adequate for the design and also whether the first stage board will allow paralleling two output boards. I already own the transistors and I'm looking for a design in which I can use them. Any suggestions are highly appreciated.
Thanks in advance
Daniel
I was thinking about running F5 Turbo with 75V rails and 8 2SK1530/2SJ201 pairs per channel, each output device with 100mA bias. So I ask whether the lower bias is adequate for the design and also whether the first stage board will allow paralleling two output boards. I already own the transistors and I'm looking for a design in which I can use them. Any suggestions are highly appreciated.
Thanks in advance
Daniel
How much power do you really need? 75v rails seems quite a lot! Nelson's recommendation is 32 for v2 and higher for v3 but I would say maybe 45max. As far as I understand, it is better to have a lower voltage and higher bias current to go deep into class A.
A1837/C4793 are good for 230V, with the 25V limit of the J74 good for +75V rails as is now.
Alter the voltage divider of the cascode devices for higher rails, and he could go to 125V, if not for the 100V (90V safer) limit of the output MOSFETs.
Bare in mind that F5 is common source : http://www.diyaudio.com/forums/pass-labs/148865-fqa12p20-fqa19n20c-discontinuation.html#post1900943
With 8 pairs of those, logical choice is to use parallel input JFETs (me would pick Violets with aircon)
....................... 
Sorry, what do you mean?
I guess he means that if one wants 100 watts or so level heat dissipation in that configuation the bias needs to be ~100ma.
I was just making some quick calculation on impact of rail voltage increase in V3 and realized that I was underestimating heat sink requirement.
For example, if the bias is 1A per mosfet in Daniel's configuration, heat dissipation per channel is 1kw. My F5 has 12inch 8 inch , 1 inch long fin for 62w. For 1Kw dissipation, I would need 3.3m long heat sink!
Since many main stgream high end speakers's power requirement comes from not the overall low efficiency, but very high current requirement in low frequencis due to dips in impedance and phase rotation, one should make careful consideration of their speaker's requirements.
Of course if what you want is to drive 80db efficiency speakers to 120db, that's different stroy, but that's really not my thing!!
Last edited:
I mean that this is the Papa forum.
For sub-optimal Class AB you are in the dead wrong place, and me no explain why so, therefore the answer is Burp !
Guys,
Back in the loop. Had no internet access yesterday when I was away.
Anyhow, here is the single Mosfet output board.
Andy
Back in the loop. Had no internet access yesterday when I was away.
Anyhow, here is the single Mosfet output board.
An externally hosted image should be here but it was not working when we last tested it.
Andy
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.