I can not exactly remember but i think that linear and similar hfe effects third harmonic and early voltage second, or was that the other way around ? Concerning hfe it is important that hfe IS LINEAR OVER CURRENT and THE SAME; BOTH NPN AND PNP and early voltage is HIGH for best speed.
You only need HIGH hfe when you want low base current ( low DC offset in the input ) and HIGH impedance in current sources and mirrors. Nevertheless the Paradise mirror is set up in such a way that hfe multiplies many times. The formula is aprox :
hfe x hfe x 2 x emitter resistor. So for hfe 300 we have : 300 x 300 x 2 x 100 =
18Mohm. That is already 3 orders of magnitude higher then the transimpedance RIAA.
Ricardo and MIIB know the exact formula but trust me : put that damn hfe challenged cripples in and be happy.
You only need HIGH hfe when you want low base current ( low DC offset in the input ) and HIGH impedance in current sources and mirrors. Nevertheless the Paradise mirror is set up in such a way that hfe multiplies many times. The formula is aprox :
hfe x hfe x 2 x emitter resistor. So for hfe 300 we have : 300 x 300 x 2 x 100 =
18Mohm. That is already 3 orders of magnitude higher then the transimpedance RIAA.
Ricardo and MIIB know the exact formula but trust me : put that damn hfe challenged cripples in and be happy.
Short update - all the boards are allocated right now, there were some latecomers that should be quite happy now, and some people increased, so thats all finished.
Just to remove a little ambiguity - I am no longer using the original list, that one was "stopped" when we closed the group buy, and I copied the content into another list.
That other list is NOT public, as it contains all the email and shipping addresses etc, a lot of personal data which I am not going to make public. But be assured that everybody who wanted boards and reacted to my PMs and mails is on the list alright, and all the latecomers have been notified as well (I am close to a couple thousand PMs and emails on this one...)
As for the transistor selection etc - we have a "stuffing guide" that is 95% ready, and it talks a lot about this and other topics. I will start a new thread for the assembly and post it there, so that this thread can be kept for the group buy only, ok?
Just to remove a little ambiguity - I am no longer using the original list, that one was "stopped" when we closed the group buy, and I copied the content into another list.
That other list is NOT public, as it contains all the email and shipping addresses etc, a lot of personal data which I am not going to make public. But be assured that everybody who wanted boards and reacted to my PMs and mails is on the list alright, and all the latecomers have been notified as well (I am close to a couple thousand PMs and emails on this one...)
As for the transistor selection etc - we have a "stuffing guide" that is 95% ready, and it talks a lot about this and other topics. I will start a new thread for the assembly and post it there, so that this thread can be kept for the group buy only, ok?
I’m sorry to inform you that I do not have a setup, I did buy the Lucky_Z curve tracer but I have only used it for matching power transistors that are being used in my power amplifier (that I am currently building). Work on all fronts is very slow here (one of the reasons I do not post that often at the moment) due to ‘great pressure’ generated at my work (we have a 5 year project in its final stages; it is going operational in December). With regard to the ‘Paradise phono’, I will not build this puppy until, probably, February or March next year. In the mean time I will make myself as available as possible for builder support (the PSU part), but from 6 to 27 October there will be little or no time (I’m sorry
to say).About matching the transistors, Joachim has actually posted several time about this. It is not critical to have exact matches; hfe matches within 5% will be fine. Using a simple matching setup using a multimeter is fine, using the ‘PEAK Atlas DCA55’ is highly recommended, it cheap and very usable (I think Joachim is using it).
See also: Transistor matching
And: PEAK Atlas DCA55 Semiconductor Analyser
&%^#$*(&*$)(*#)(*
Ricardo, you can parallel two 2SK170GR. No, more Idss is not a problem in certain limits.
What could happen is that the cascode Fets run out of steam. I would say up to 12mA it is fine. Even 15mA will work. The way it works is that the cascode Fets have higher Idss and higher Ugs then the main transistor ( 2SK170BL ). When the main transistor has too high Idss the regulation from the cascode does not work well any more.
What could happen is that the cascode Fets run out of steam. I would say up to 12mA it is fine. Even 15mA will work. The way it works is that the cascode Fets have higher Idss and higher Ugs then the main transistor ( 2SK170BL ). When the main transistor has too high Idss the regulation from the cascode does not work well any more.
many thanks, thats great! things move quicker around here than elsewhere - maybe the ATL guys are changing the laws of physics ? ;-)
While I listen to an IMHO fantastic recording of Beethoven's 9th Symphony (Karajan and the Berliner Philharmoniker) I do hope you are doing well.
Thanks for all the efforts!
Cheers,
Mathias
Last edited:
