The IRFP 250 is basically two IRFP240s in one package. Substituting it for the IRFP240 would be the same as using two IRFP240s for the output with poorer thermal capability.
Basically you can try any vertical fet with a Vgs in the neighborhood of 3-4V. You may need to adjust the differential pair current and/or its drain resistor to get the output stage to bias properly.
Rather than looking for higher current capability, (you won't be pushing that limit at all) look for devices with low Ciss - input capacitance. Irfp140 or IRFP044 might be more interesting
Basically you can try any vertical fet with a Vgs in the neighborhood of 3-4V. You may need to adjust the differential pair current and/or its drain resistor to get the output stage to bias properly.
Rather than looking for higher current capability, (you won't be pushing that limit at all) look for devices with low Ciss - input capacitance. Irfp140 or IRFP044 might be more interesting
point is that mosfets with lower voltage capability are usually with lower amount of parasitic capacitances
What happens to the signal when taking MosFETs with higher Ciss, let´s say about 3000pF? And where do i change the drain resistors reffering to gray´s schematic? R14 and R16 in the same amount?
simply saying - more capacitance will swallow highs ..... 
first - tell why you need to change drain resistors ?
for detailed explanation of every usual Aleph part - study this file :
first - tell why you need to change drain resistors ?
for detailed explanation of every usual Aleph part - study this file :
Think i´ll go with IRFP150 in the output-section and Elna Silmic 2 for the caps. Erm... Just two questions: Don´t you need DC-protection caps at the input? The MPSA18 / ZTX450 are for the short-circuit-protection? Would it be possible to use BC550C instead? Thanks-a-lot 😉
Only one of the MPSA18/ZTX450 is for short circuit protection. The other controls the active current source. I don't see a reason why you couldn't use any general purpose NPN transistor in those spots. For the current source, low noise would be important.
The bias set by the active current source would vary depending on the device characteristics, but once set should remain stable. Similarly, the protection current limit will be different.
I don't know of anyone reporting testing the sonic character of various devices in the active current source. If Papa changed to Zetex, one would think there is a reason other than having them on hand, though.
The bias set by the active current source would vary depending on the device characteristics, but once set should remain stable. Similarly, the protection current limit will be different.
I don't know of anyone reporting testing the sonic character of various devices in the active current source. If Papa changed to Zetex, one would think there is a reason other than having them on hand, though.
Last edited:
Hi Bob,
I've used all three and couldn't detect a difference. The Vbe is a little different though. And watch the pinouts.
I think the ZTXs are more robust, and NP probably bought thousands and uses them for everything.
I've used all three and couldn't detect a difference. The Vbe is a little different though. And watch the pinouts.
I think the ZTXs are more robust, and NP probably bought thousands and uses them for everything.
I realize through superdumbosmosis - that Papa have few drekloads of ZTX parts , same as 9V1 zeners
One question about the supply:
Would a CRC with 15.000µF - 5x 0R33 - 15.000µF be enough for a Mini-Aleph on ~16V Rails?
Would a CRC with 15.000µF - 5x 0R33 - 15.000µF be enough for a Mini-Aleph on ~16V Rails?
I would think 5 X .33ohms is too low a resistance. The capacitance would be enough depending on the bias current.
Why don't you download the Duncan Amps Power Supply Designer software from their website. It's free.
Best, Bill
Why don't you download the Duncan Amps Power Supply Designer software from their website. It's free.
Best, Bill