I bet by now we have scared off the OP.. To him I say modify the pre pre as DF96 suggested and you should be good to go. I'd probably ignore the balance of the commentary. Lots of opinions here and little direct experience. (Yes I acknowledge the obvious shortcomings, but for some strange reason the thing actually works just fine.)
Ummm... This statement only shows:
- The very poor quality of the human spectrum analyzer. You are actually not listening to the objective performance, but to the name of Hiraga. Should that design being authored by Hashimoto or Joe Sixpack, you would certainly be more suspicious.
- The stories about the cartridge current are nothing but another fetish promoted by the audiofool purists. I'm sure there are right now some 5,000 quid supa dupa cartridges connected to this design, when it's probably the worst way known to mankind to cancel the cartridge current (including the temperature and bias dependence). If Hashimoto or Joe Sixpack would promote this schematic, nobody would touch it with a six feet pole... But Hiraga...
BTW, it takes 4 more transistors (unfortunately need to also be low noise) to implement a input current cancellation schematic for the differential mode. Not worth the complexity, the AC performance will still be crap.
Didn't say it was something I would choose for myself as I prefer either the multi-tan head amp or SUTs I mentioned. Signal levels are so low that linearity seemed perfectly ok in the short audition I heard - I did not know it was the Hiraga design until afterwards. It did not stand out one way or the other. I heard better and worse that day.
Yes I agree the quality of the human spectrum analyzer, that is what we have instrumentation for.
Yes I agree the quality of the human spectrum analyzer, that is what we have instrumentation for.
Going back to what DF96 said, input impedance is determined by transconductance, which is a product of Rb and hFE. So input impedance will be possibly wildly different at each input. Clearly up there with the best!
It is possible to design valve circuits that meet the requirements of HiFi, but when we look at the cost to performance ratio compared to well designed solid state amplifiers, it becomes apparent that the extra cost is for novelty purposes.
Clearly up there with the best!It is possible to design valve circuits that meet the requirements of HiFi, but when we look at the cost to performance ratio compared to well designed solid state amplifiers, it becomes apparent that the extra cost is for novelty purposes.
Did you not read my post 12, or not understand it? Input impedance would be almost identical at each input, because current is almost identical at each input. BJT transconductance is set by current. Similarly, voltage gain will be almost identical for each channel.monty78pig said:
It may be a poor circuit, but let us not criticise it for things it gets right!
Only when the current is high or the base resistance is unusually high. As I said, to a first approximation transconductance is Ic/Vt.
If Ic is high (which it isn't in this case) then you have a second-order correction for base resistance too. I would guess that transistors with lowish base resistance have been chosen, as that reduces noise. There is also a possible correction for Early effect, but that should be small in this circuit as the voltage gain is fairly low.
If Ic is high (which it isn't in this case) then you have a second-order correction for base resistance too. I would guess that transistors with lowish base resistance have been chosen, as that reduces noise. There is also a possible correction for Early effect, but that should be small in this circuit as the voltage gain is fairly low.
The input impedance is;
Re||(Beta/gm)||(1/gm)=Re||[Beta/(Beta+1)/gm]~Re||gm~gm
where Re is the emitter resistor. If you draw the equivalent small signal circuit, this will be immediately obvious; Beta/gm is indeed in parallel with the input, but so is the VCVS, which lowers the input impedance to almost exactly 1/gm and makes it pretty much independent on the transistor type and parameters.
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.