The last time I measured, J200 & K1530 have something like 5S at 1.6A bias. So we would just make 0.1R even without source resistor. If we wish to parallel devices to improve max power handling, then a 0.1R source resistor is probably close to minimum, and we start losing transconductance as the bias for each FET reduces.
http://www.diyaudio.com/forums/showthread.php?s=&postid=1507044&highlight=#post1507044
FQA19P20C has slightly higher transconductance at 7S under 1.6A, but there is no true complementary P-FETs to go with. The Toshiba's are really amazing in that respect.
http://www.diyaudio.com/forums/showthread.php?postid=1567443#post1567443
BJTs are of course totally different story.
Patrick
http://www.diyaudio.com/forums/showthread.php?s=&postid=1507044&highlight=#post1507044
FQA19P20C has slightly higher transconductance at 7S under 1.6A, but there is no true complementary P-FETs to go with. The Toshiba's are really amazing in that respect.
http://www.diyaudio.com/forums/showthread.php?postid=1567443#post1567443
BJTs are of course totally different story.
Patrick
EUVL said:
For example, here:
http://www.ant-audio.co.uk/Theory/N-channel D-MOSFET output stage with improved linearity.pdf
the circuit on Fig 6 has an output impedance of less than 0.15 Ohm at only 100mA idle. If it is biased at 200 mA the output impedance would be below 0.05 Ohm.
Alex
The hum probably comes from DC on the power line. This is common. Some power transformers are more sensitive to it than others, even from the same stock. AC coupling the power line with a large electrolytic cap is the best solution. It would be best to Google for a small company that provides such a device.
As far as power amplifier design is concerned, I think that high damping factor is overrated. I use a .4 ohm output Z to my 4 ohm WATTS and find nothing wrong with it. I also use a separate subwoofer with only .1 ohm or less output Z drive.
If I were to do this in the real world, I might find a used HCA-3500, run it at 1/2 the rated voltage and change the circuit to single sided folded cascode output. It would be the cheapest and best way to get low Z at reasonable idle current. However, I think that an idle current below 1A is a waste of time and effort, as you will never get the higher order harmonics out of the listening level sound, without at least that much current.
This circuit uses a complementary Darlington output using complementary power fets driving 8 pairs of complementary high speed output devices, per channel.
If I were to do this in the real world, I might find a used HCA-3500, run it at 1/2 the rated voltage and change the circuit to single sided folded cascode output. It would be the cheapest and best way to get low Z at reasonable idle current. However, I think that an idle current below 1A is a waste of time and effort, as you will never get the higher order harmonics out of the listening level sound, without at least that much current.
This circuit uses a complementary Darlington output using complementary power fets driving 8 pairs of complementary high speed output devices, per channel.
nope, a big, fat, juicy toroid.
as john suggested, a properly chosen series cap is a good solution especially if you can't get rid of the dc that is the source.
mlloyd1
as john suggested, a properly chosen series cap is a good solution especially if you can't get rid of the dc that is the source.
mlloyd1
hitsware said:
Do they have laminated xfmrs and mechanical hum ?
Have you tried it ? It sounds crude (and is) but
really does work in many instances (but not all)
- Status
- Not open for further replies.