I got a matched set of 24 FETs from a well known source and the boards from DIYAUDIO. All the parts are from Digikey and hificollective, all top parts. The amp is exactly as per schematic.
The gain board outputs 44v peak-to-peak which goes into the bias board then into the two output boards. The output on the output boards is also 44v p2p which is not much considering the rails are on 80v dc. I was expecting 60-65v p2p. I suspect something is wrong.
The gain board outputs 44v peak-to-peak which goes into the bias board then into the two output boards. The output on the output boards is also 44v p2p which is not much considering the rails are on 80v dc. I was expecting 60-65v p2p. I suspect something is wrong.
well , invest some efforts here
post some pics , same as schematic what you exactly did , especially concerning values of rail voltages
bloody FE isn't intended for 40V rails , as is
post some pics , same as schematic what you exactly did , especially concerning values of rail voltages
bloody FE isn't intended for 40V rails , as is
Also, isn't the output stage a buffer (and hence no voltage gain from
what comes out of the FE)?
what comes out of the FE)?
Yes you must be right. The output stage is just an input buffer with a current gain, not voltage so the only way to get more out of this is to increase the output of the gain stage. Increasing the value of R13 does increase the sensitivity but the gain stage still clips at 44v. I will attempt with the Bride of Zen pre-amp instead unless anyone has other ideas?
rails voltage , input stage current , output stage current
all that speaking of FE only
besides that few pics
just for kicks , double value of output resistor value and recheck max output Upp
all that speaking of FE only
besides that few pics
just for kicks , double value of output resistor value and recheck max output Upp
You mentioned you're building the single-ended version and that the amp
is clipping. Is it the output stage that is clipping? Do you have enough
bias there to support the higher voltage swing you want?
is clipping. Is it the output stage that is clipping? Do you have enough
bias there to support the higher voltage swing you want?
Hi Dennis.
The gain board is clipping at 44v p2p and this is reflected by the output boards too. The output stages are all followers, so they only swing what the front end does, and I have +/-24V supplies on the front end, so that sets the limit.
I think if the gain board was giving a stronger signal to the output boards, the output boards would probably cope.
There is no simple solution, you need total redesign of the FE or insert an additional stage with 2x (or more) voltage gain between FE and OS.
well, after two days of me begging for more details , you're saying crucial thing
Don't worry, I will get the measurements tonight 🙂
no need
what is important is that you at last said that FE is using +/-24V rails .......... and it's completely logical that you can't have swing higher than rails
so , 44-46Vpp is normal
you can have more only if you cascode input stage and use higher rails
what is important is that you at last said that FE is using +/-24V rails .......... and it's completely logical that you can't have swing higher than rails
so , 44-46Vpp is normal
you can have more only if you cascode input stage and use higher rails
Not so fast. 10Vpp or less is more likely since 28V is at the gate of cascode transistor and 20-30V drop across drain load resistor from 30ma bias.