From symmetrical to N-channel only

Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.
bigpanda said:
Here is the modified circuit. I am making a wild guess so laugh if you want to, but just tell me if I am going to kill my amp (or even myself) by doing this. Thanks for watching this comedy.

In short: as darwn, it will definitely kill the fuse(s). If you short the fuses, YES you are likely going to make a lot of smoke - and kill the speaker attached to the amp for sure.
My point is that you cannot go modifying circuits without minimum basic knowledge of the operation of their parts. For instance: the original output stage is a follower, so gain is 1, i.e. at rst, you have 0V input and 0V output. To turn on a N-ch MOSFET to bias current levels, you need about +3V on it's gate compared to the source. This is done by the Vbe multiplier in the original schematic adding about 3V to the input for the top N-ch MOSFETs, and subtracting about 3V for the bottom, P-ch MOSFETS - for them everything is the same except with oposite polarity.
In your modification, for 0V input, the sources of your bottom MOSFETs are at -V rail (say -40V) and the gates are at 0V less 3V, since that has not changed from the PN version. This makes the gate at +37V WRT source. Disregarding for a moment the fact that maximum gate-source voltage for these parts is 20V and that with 37V it is only a matter of a short time till the gate is destroyed and the whole transistor turned into a crater, if 3V turns the MOSFET on to quiescent current levels, and about 6V turns it on all the way, making it essentially a short, what do you think 37V will do? Answer: for 0V in, your amplifier will produce a short to the -Vcc rail. I'm sure that was not so difficult to figure out? I mean, you would not go figuring out how a guillotine works by sticking your head in it, would you? Then why figure out how a N-ch MOPSFET works by putting it in place of a P-ch with D and S reversed?
 
bigpanda said:
Here is the modified circuit. I am making a wild guess so laugh if you want to, but just tell me if I am going to kill my amp (or even myself) by doing this. Thanks for watching this comedy.

Naw, you'll just fry the MOSFETs. You can't use the same drive circuitry with all NMOS outputs.

You want a quasi-complementary design. There is a long running thread on this very idea, because the designer had a bunch of N channel MOSFETs lying around. Go here:

http://www.diyaudio.com/forums/showthread.php?s=&threadid=43331&perpage=40&pagenumber=1

Here is the "final schematic" from that one.
http://www.diyaudio.com/forums/attachment.php?s=&postid=502698&stamp=1099193939

He even published a board for the design:
http://www.diyaudio.com/forums/attachment.php?s=&postid=505269&stamp=1099541934

But it might be cheaper just to buy some complementary devices, because most quasi amps I've seen either use boot strapping or auxiliary voltages to get rail to rail outputs. Good luck.
 
Tks guys,

Like I said : I have a bunch of the 240 at hand. Can they be used in Quasi's circuit. I attempt to change the original one to a N channel cause I was trying to make use of the old PCB board so that I don't have to lay the board all over again. Tks again.
 
Then you might want to have a look at passdiy.com, where you can find the MF12, Nelson's modified HK Citation 12 power amp. It is a quasicomplementary desigh using NPN BJTs originally, and Nelson modifies it to use two N-ch MOSFETs. Have a look at the output stage, it may prove interesting to you.
 
Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.