It's variable at this point. The absolute limit is about +/-20V.
@5mA/jfet we have 100W/ch push-pull Class A? No heatsink required
It would dissipate 100 watts. Output power would
probably be about 20 watts.
still cool...
20W is voltage limited at 8 ohm.
50W peak is 25W RMS. That would be something like 3.5A, 14V for 4 ohm. I max will be 12mA per JFET, which means V grade or some forward bias.
50W RMS at 4R requires 17mA per FET. You need to use V grade AND forward bias. And dissipation at the required 20V rail will be a bit high, so that the case temperature will be a touch higher than the reported 45°C.
30W rms at 6 R is 3A 18V. 180mW dissipation per FET.
Patrick
50W peak is 25W RMS. That would be something like 3.5A, 14V for 4 ohm. I max will be 12mA per JFET, which means V grade or some forward bias.
50W RMS at 4R requires 17mA per FET. You need to use V grade AND forward bias. And dissipation at the required 20V rail will be a bit high, so that the case temperature will be a touch higher than the reported 45°C.
30W rms at 6 R is 3A 18V. 180mW dissipation per FET.
Patrick
Patrick, shouldn't that be: 50W peak is 25W average ?
My professors taught me that you get RMS value by dividing peak value with square root of 2 i.e 1.41. So 50W peak should be 35.46W RMS
Sorry Patrick, I don't know why I asked at all - absolutely unimportant because this calculation assumes pure resistive load and sinusoidal signal - pretty much useless in real life.
Peak voltage/current value might be informative - it determines the clipping limit, the rest is of no use, unless we want to go into really complex modeling...
This is correct. The square of the square root of 2 is equal to 2...
Prms=(Vrms)^2/R
since Vrms=Vpk/sqrt 2 then:
Prms=(Vpk/sqrt 2)^2/R; this then equates to:
Prms=(Vpk^2/2)/R)=Vpk^2/(2*R) or the customary =(Vpk^2/R)/2
As an example with the known values Vpk=20V and R=8
Prms=(20^2/8)/2=25 watts
At this time I have 5 pc boards completely stuffed, ready
for action. Four of them perform as previously described,
and one of them has a little extra distortion.
I began searching for the bad device by measuring the THD
at the Source pin of each of the 588 devices on the board,
and I got half way through before I had to go work on other
things for a while.
In actual testing the best performance looks to be with
supplies in the +/-15V area - battery operation is very
feasible. To get the power up I can run two pc boards
in balanced mode, and this looks like what I may do,
creating one stereo pair at about 40 watts and a single
board suitable for two channels at 10 watts or less.
I may have it a BAF, then again I might have something
else.
for action. Four of them perform as previously described,
and one of them has a little extra distortion.
I began searching for the bad device by measuring the THD
at the Source pin of each of the 588 devices on the board,
and I got half way through before I had to go work on other
things for a while.
In actual testing the best performance looks to be with
supplies in the +/-15V area - battery operation is very
feasible. To get the power up I can run two pc boards
in balanced mode, and this looks like what I may do,
creating one stereo pair at about 40 watts and a single
board suitable for two channels at 10 watts or less.
I may have it a BAF, then again I might have something
else.
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.