Differences in bias current per output device
I've just started testing my assembled Peter Daniel PCB's for the first time. I am using matched 240's and 9240's from tech-diy, as well as the input devices.
I did the tests recommended earlier without the Mosfets in circuit and everything tested as expected. I installed the output devices, then started bias at minimum without problems and increased it as it warmed up. When I switched my meter to check the other 5 output devices, I was surprised to discover there was significant variation across the 0.47R's. Q3 was 250mv, while a couple were almost 100mv less (150-160mv). This channel provides good signal input to output without any obvious distortion and I played music through them no problems. A few millivolts of DC offset, which was easily adjusted out once warmed up.
What is a reasonable amount of bias current variation between the Mosfets?
Is this a matching issue, bad Mosfet or is something else going on?
Tom
I've just started testing my assembled Peter Daniel PCB's for the first time. I am using matched 240's and 9240's from tech-diy, as well as the input devices.
I did the tests recommended earlier without the Mosfets in circuit and everything tested as expected. I installed the output devices, then started bias at minimum without problems and increased it as it warmed up. When I switched my meter to check the other 5 output devices, I was surprised to discover there was significant variation across the 0.47R's. Q3 was 250mv, while a couple were almost 100mv less (150-160mv). This channel provides good signal input to output without any obvious distortion and I played music through them no problems. A few millivolts of DC offset, which was easily adjusted out once warmed up.
What is a reasonable amount of bias current variation between the Mosfets?
Is this a matching issue, bad Mosfet or is something else going on?
Tom
There is definitely something wrong. Check the resistor values first. If one of the resistors is way off it could cause a problem. But the bias across each resistor should be very close. All of mine are within a few millivolts.
I would start with the resistors then, yes you may be looking at a bad mosfet.
I would start with the resistors then, yes you may be looking at a bad mosfet.
Andrzej Sochon said:
Cool it, sorry, I am not very familiar with contemporary audiophile parts and trends, so I am just asking to be sure what I found on net this is it what Mr. Pass suggests to play with...
Best regards,
Andrzej
Just joking!!
...Found the Banzai place very easy to deal with and very reasonable ,all things considered. In the states we often find it difficult to obtain the more exotic parts .
Handmade, http://www.hndme.com/ also has Elna caps, though not quite as wide a selection of the "silks" as Banzai. . .
Steve Z
Steve Z
tms0425 said:Thanks. I was afraid of that. I re-checked all of the gate (150R PRP's) and source (0.47R Mills) resistors and they look correct. I guess it's time to start swapping mosfets. Unfortunately I don't really have any matched extras
I wouldn't get excited about it. Do make sure that the device
with the higher bias figure is properly tightened down, and then
enjoy it.
Thanks! I did exactly that and the one that was a bit out of sync came back in range. I am using those Berquist K10 pads initially to test with, so I think I better just get some mica and goop and be done with it. The mosfets seem fine. I don't even have 40c on the sinks so far at 200mv. I still have one about 20mv higher, but others are within a few now, so I'll recheck that one.Nelson Pass said:
I'm pretty excited as I haven't built anything this ambitious since the original A40 with the big Signal trafo in Audio Amateur many moons ago. This has been BIG fun and the best part is I already have four F4 channels playing in the "lab" room (laying across the bench). It's gone together fairly easily, mostly because of the elegant simplicity of this design, which I love. The music room is next on my Emerald Physics CS2 OB's (100db efficient HF compression drivers mounted to waveguides).
As always, thanks again for your contributions and counsel
Tom
My first F4 is alive!
My first of two F4's is finally up and running in a "mule" sort of case. I'm currently running it stereo on the HF of my biamped OB Emerald Physics CS2's with great results so far (100db efficient compression driver in a waveguide). Gain is provided from a Behringer DCX with Jan Didden output board and a Burson Buffer (fixed to +12db). Not quite enough to swing the full voltage but seems plenty for my HF drivers at the levels I need.
I used two Plitron 300VA dual 18vac trafo's in it with two Peter Daniels PS boards, 2 bridges/4 caps each per channel. I originally planned one of these transformers and one PS board for each stereo amp, but it was irresistable to go ahead and set it up this way. The rails are at about 24.5vdc (hopefully a little bit high is ok?). With 200ma across the source resistors, it's only running 41C on the Conrad MF35-151.5 heat sinks after a couple hours or so. Once warmed up, it settled right in and really sounds terrific. Can't wait to do some further tweaking with various parts to see how far you can really take it.
Next up is the second monoblock and an Aikido gain stage to provide the full swing if I need it.
Again, as usual, thanks Nelson for sharing this!
Tom
My first of two F4's is finally up and running in a "mule" sort of case. I'm currently running it stereo on the HF of my biamped OB Emerald Physics CS2's with great results so far (100db efficient compression driver in a waveguide). Gain is provided from a Behringer DCX with Jan Didden output board and a Burson Buffer (fixed to +12db). Not quite enough to swing the full voltage but seems plenty for my HF drivers at the levels I need.
I used two Plitron 300VA dual 18vac trafo's in it with two Peter Daniels PS boards, 2 bridges/4 caps each per channel. I originally planned one of these transformers and one PS board for each stereo amp, but it was irresistable to go ahead and set it up this way. The rails are at about 24.5vdc (hopefully a little bit high is ok?). With 200ma across the source resistors, it's only running 41C on the Conrad MF35-151.5 heat sinks after a couple hours or so. Once warmed up, it settled right in and really sounds terrific. Can't wait to do some further tweaking with various parts to see how far you can really take it.
Next up is the second monoblock and an Aikido gain stage to provide the full swing if I need it.
Again, as usual, thanks Nelson for sharing this!
Tom
