An externally hosted image should be here but it was not working when we last tested it.
Please elaborate.
Hi Millwood,
CFPs ? Can you be more specific? Thank you!
Audiofanatic
P.S. I have the PCB's
millwood said:I haven't tried it. But CFPs on MOSFETs, especially the non-lateral types, are tough to get to work. they usually break into parasitic oscillation at the smallest disturbances.
Hi Millwood,
CFPs ? Can you be more specific? Thank you!
Audiofanatic
P.S. I have the PCB's
Complentary Feedback Pair=> a topology of 2 complementary transistors wired together a certain way.
See this thread were these things are discussed briefly:
http://www.diyaudio.com/forums/showthread.php?s=&threadid=25051
See this thread were these things are discussed briefly:
http://www.diyaudio.com/forums/showthread.php?s=&threadid=25051
Just a suspicion, but the oscillation that Millwood mentions may be the reason for the reason for the resistors between the emmitters of the BD139/40 and the drains of the FETs. In a CF-type BJT ouptut section, the emitters of the drivers and the collectors of the output devices are directly connected. Putting resistors in between may be intended to reduce the tendency to oscillate. Just a guess.
At the same time I notice that the gate stopper resistors are only 15-ohms. These also are intended to suppress oscilation, but that looks like an awfully low value. I'm more used to seeing 400-1k ohms used for these.
The placement of the rail fuses is a bit brave, at least from the perspective of one who has found more than one way to short an amp!
What is the origin of this schematic? Was/is it a commercial product?
At the same time I notice that the gate stopper resistors are only 15-ohms. These also are intended to suppress oscilation, but that looks like an awfully low value. I'm more used to seeing 400-1k ohms used for these.
The placement of the rail fuses is a bit brave, at least from the perspective of one who has found more than one way to short an amp!
What is the origin of this schematic? Was/is it a commercial product?
CFP!
Thanks JP!
Audiofanatic
jean-paul said:Complentary Feedback Pair=> a topology of 2 complementary transistors wired together a certain way.
See this thread were these things are discussed briefly:
http://www.diyaudio.com/forums/showthread.php?s=&threadid=25051
Thanks JP!
Audiofanatic
Yes,
it DOES work; I have 2 of them myself, and they work without any problems, no oscillation or whatsoever.
@rtirion: the "update" for the oscillation-problem was NOT for this amplifier, it was for the upgraded version of this amp with a pair of GT20D101/GT20D201 IGBT's.
Grtz, Joris
it DOES work; I have 2 of them myself, and they work without any problems, no oscillation or whatsoever.
@rtirion: the "update" for the oscillation-problem was NOT for this amplifier, it was for the upgraded version of this amp with a pair of GT20D101/GT20D201 IGBT's.
Grtz, Joris
Vigier said:Yes,
it DOES work; I have 2 of them myself, and they work without any problems, no oscillation or whatsoever.
@rtirion: the "update" for the oscillation-problem was NOT for this amplifier, it was for the upgraded version of this amp with a pair of GT20D101/GT20D201 IGBT's.
Grtz, Joris
any chanses you could tell me what the update was?
I cannot find it in elektuur at our library!
I build the IGBT-amp and tested it with a powersupply with current limit at 400mA and only + and - 20V to start with.
Guess what, the negative side of the powersupply alsways drops down to 4V and the amplifier pulls the lab-powersupply into currentlimit (both + and - side are giving 400mA to my amp).
I also measured oscillations at the output.
A fellow student build the same amp and faces the exact same problem.
Any help would be welcome... as the IGBT's cost me €27 each
I am also looking to put HEXFET's in the schepmatic...but which ones should I pick since my powersupply (normally for the IGBT-amp) is + and - 43V...probably too high for the original IRF540/IRF9540- design.
Hope anyone can help.
Bart
Not too high re voltage as long as it never goes over +-50 (though that would be the absolute maximum) but the 540/9540 will not be able to dissipate the heat.
Also, IRF540 and 9540 are NOT complementary (not even close). Unfortunately, in this family there is no good complementary pair for the current you need, though IRF9540 with IRF640 will give you about 2x less gain mismatch, but it will still be on the order of 15-17%.
My chice would be IRFP9140 and IRFP240. This is as close as you can get with IRF hexfet complementary pairs as long as <100V rail to rail voltage is enough. One pair will work fine up to 120+W assuming a good heatsink and proper mounting. The gain mismatch of these MOSFETs is less than 5% which is less than the variances between MOSFETs of the same kind, so with some selection you sould get almost ideal complements - for a fraction of the price of lateral FETs.
I am attaching a Gm (Id versus Gds) plot for all of the above MOSFETs so you can draw your own conclusions...
Also, IRF540 and 9540 are NOT complementary (not even close). Unfortunately, in this family there is no good complementary pair for the current you need, though IRF9540 with IRF640 will give you about 2x less gain mismatch, but it will still be on the order of 15-17%.
My chice would be IRFP9140 and IRFP240. This is as close as you can get with IRF hexfet complementary pairs as long as <100V rail to rail voltage is enough. One pair will work fine up to 120+W assuming a good heatsink and proper mounting. The gain mismatch of these MOSFETs is less than 5% which is less than the variances between MOSFETs of the same kind, so with some selection you sould get almost ideal complements - for a fraction of the price of lateral FETs.
I am attaching a Gm (Id versus Gds) plot for all of the above MOSFETs so you can draw your own conclusions...
Attachments
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.
- Home
- Amplifiers
- Solid State
- Did someone tryed this?