Right that! But 2 outputs per side and maybe going balanced. Still a lot more to read 🙂Like this?
EUVL,
I did not mean in any other way. For my limited knowledge of electronics, I cannot make informed changes to the circuit. I can only do or faithfully follow what others such as you have done. Some of these variations including the balanced design of yours require in depth understanding of the circuit to be able to implement well.
Therefore i wanted to restrict myself to the stock design and implement whether it is F5 or Aleph J, that way there is lot more help and experiences available from forum members.
Cheers.
What I was trying to say is that even the Toshiba's is 3x the price of the IRFP's, they amount in the end to say 20 USD extra or so, which is nothing compared to the cost of the entire amplifier including PSU, case, .....
And if you can hear an improved difference straight away, I would immediately spend the 20USD, for this is the cheapest improvement you can make. Any further improvements (like exotic passive components, etc.) will always end up costing you more.
Regards,
Patrick
And if you can hear an improved difference straight away, I would immediately spend the 20USD, for this is the cheapest improvement you can make. Any further improvements (like exotic passive components, etc.) will always end up costing you more.
Regards,
Patrick
now looking at TO-220, IRF520, IRL530, IRF530, IRF630, IRF9520, IRF9530, IRFZ14, IRFZ24, IRFZ34, IRF9Z24, IRF9Z34, IRL3215, IRL6215, etc
well, maybe we could say that Toshibas are like 300B tubes, and these smaller fellows like EL84
well, maybe we could say that Toshibas are like 300B tubes, and these smaller fellows like EL84
Patrick,
I agree. No regrets about spending the additional amount on the Toshibas. They certainly look much better built and rugged and also sound good. I just used your suggestion on MOSFET degeneration technique on 5ohms in the source path of 2sj74bl. Don't know what to tweak further.
Cheers.
Nothing wring with TO220.
But they do have a much smaller thermal contact area to the heatsink (via the insulator), compared to TO247, and especially to the 2SK1530/2SJ201.
Patrick
But they do have a much smaller thermal contact area to the heatsink (via the insulator), compared to TO247, and especially to the 2SK1530/2SJ201.
Patrick
I would say that it is a safe bet that the Toshiba transistors will sound different. Better is a personal preference, so that is a firm maybe.
Nelson has also treid all 3 types (Tos, IRF, FQA) and posted his opinions here.
Why not search for his comments.
Patrick
Why not search for his comments.
Patrick
Thinking about PSU size for dual mono F5, what can and cant you do.
I was thinking going for less total capacity but instead focusing on ripple current and low ESR.
Wouldnt a lower total capacity than what pass uses in his PSU, but high ripple current and low ESR be better than a huge total capacity but less ripple current and higher ESR?
Can get some very high ripple 2200 uF 50V 105 C caps, around 2.5-3.5A pr cap. Ten of these pr rail would give you 22000 uF pr rail but 25-35A total ripple current and very low ESR.
Am I completely on the wrong track here or do I have a point??
I was thinking going for less total capacity but instead focusing on ripple current and low ESR.
Wouldnt a lower total capacity than what pass uses in his PSU, but high ripple current and low ESR be better than a huge total capacity but less ripple current and higher ESR?
Can get some very high ripple 2200 uF 50V 105 C caps, around 2.5-3.5A pr cap. Ten of these pr rail would give you 22000 uF pr rail but 25-35A total ripple current and very low ESR.
Am I completely on the wrong track here or do I have a point??
You are absolutely on the right (but very expensive) track :
http://www.diyaudio.com/forums/pass-labs/172770-balanced-f5-question-8.html#post2349852
The PCBs that connects all these caps together are 110x80mm, double sided 210um (6 oz. I thnk) copper. Copper costs money.
Patrick
http://www.diyaudio.com/forums/pass-labs/172770-balanced-f5-question-8.html#post2349852
The PCBs that connects all these caps together are 110x80mm, double sided 210um (6 oz. I thnk) copper. Copper costs money.
Patrick
.... and now more than ever! Look how this (precious) metal increase its value in 2010. 😡
Ah, thats just, insane. 😱
But think I'll stay at 2-3 oz. Should be sufficient.
Most likely making a CRC supply, with 5 x 2200 uF caps - resistor - 5 x 2200 uF caps pr rail, as a minimum, might go nuts and add even more caps, but lets see later on. Caps can be really expensive, even small ones, if you want high ripple current. 😱
Hi,
what's wrong with assembling using copper wire, solid core or stranded?
Too much inductance?
What about rCRCRCC?
r=secondary resistance
C=4*2200uF
R=1r0//1r0//1r0//1r0
Buy 100off 2200uF and negotiate a big discount.
what's wrong with assembling using copper wire, solid core or stranded?
Too much inductance?
What about rCRCRCC?
r=secondary resistance
C=4*2200uF
R=1r0//1r0//1r0//1r0
Buy 100off 2200uF and negotiate a big discount.
Last edited:
Digi-Key - P10335-ND (Manufacturer - EEU-FC1H222)
At 1000 pcs, they cost less than 1 USD per piuece.
Patrick
At 1000 pcs, they cost less than 1 USD per piuece.
Patrick