NO .... I would not recommend increasing the number of ouput devices beyond that which Nelson has used in the commercial versions.
Increasing the number of devices simply serves to share the power load and heat dissipation, hence prolong the life of the amp. NP was charging respectable amounts of money for these, so they are designed "bullet proof" as is!
If you read more widely on the Pass site you will find that THD decreases as the bias current through the device increases, somewhere (in the ZEN I think) he shows measurements of a single device at various current levels to prove this point. Also, as you parallel devices you effectively increase the gate capacitance of the "device" and enhance high-end roll-off.
So, the number of devices chosen is a balance of conflicting desires .....
cheers, mark
Increasing the number of devices simply serves to share the power load and heat dissipation, hence prolong the life of the amp. NP was charging respectable amounts of money for these, so they are designed "bullet proof" as is!
If you read more widely on the Pass site you will find that THD decreases as the bias current through the device increases, somewhere (in the ZEN I think) he shows measurements of a single device at various current levels to prove this point. Also, as you parallel devices you effectively increase the gate capacitance of the "device" and enhance high-end roll-off.
So, the number of devices chosen is a balance of conflicting desires .....
cheers, mark
So that would mean that the smallest Aleph should sound the best and the unit with the most output devices sounds the worst since all of them have the same internal topology except for bias point and voltages. However in Pass literature in 1996 I've read: "Aleph 1.2 does have more authority then Aleph 2 in the low frequencies and appears to be even smoother in the mid and high frequencies". Aleph 1.2 has twice the amount of output devices than Aleph 2. Any comments on that?
More output devices allows for more bias while maintaining the same die temperature.... more bias improves the sound it doesnt worsen it .... if you have more devices you generally need to increase the drive current from the front end to overcome the higher capacitance.... more output devices also means a better damping factor.... check out the Volksamp 30 .... it's an Aleph 3 with more of an input stage and has more output devices than the original ... if you compare this schematic with that of the Aleph 3 and work from the output back toward the input it'll give you some the changes they have made. You dont have to use the Volksamp input stage ... thats up 2 you.
Sorry, I didn't make myself clear. I agree, more bias will improve things - toa point. However, I assumed "Harry" intended keeping the same overall bias current, but increasing device number, thereby reducing the bias current per device ....... hence my comments against this.
Sort of what I did with my Aleph4 "in reverse", where I inadvertently made the amp with 3-FET sets by mistake, then reset the overall bias to the same by altering Rs, effectively doubling the bias current per FET.
BTW, the comparative curves are in the "ZEN Revisited" article, where NP also looks at the effect of different devices and parallel amps.
cheers, mark
Sort of what I did with my Aleph4 "in reverse", where I inadvertently made the amp with 3-FET sets by mistake, then reset the overall bias to the same by altering Rs, effectively doubling the bias current per FET.
BTW, the comparative curves are in the "ZEN Revisited" article, where NP also looks at the effect of different devices and parallel amps.
cheers, mark
- Status
- Not open for further replies.