latfets

What people are working on at the moment
User avatar
Nick
Site Admin
Posts: 15706
Joined: Sun May 06, 2007 10:20 am
Location: West Yorkshire

#16 Re: latfets

Post by Nick »

jack wrote: Tue Jan 23, 2018 9:35 am Moving to dual rail gets rid of the biasing issue for the opamp but then you'll need a DC offset for the FET!
Yes, thats why I was thinking of keeping the 431 but now using it to set the DC conditions for the fet via the op amp.
Whenever an honest man discovers that he's mistaken, he will either cease to be mistaken or he will cease to be honest.
User avatar
jack
Thermionic Monk Status
Posts: 5493
Joined: Wed Dec 29, 2010 8:58 pm
Location: ɐılɐɹʇsnɐ oʇ ƃuıʌoɯ ƃuıɹǝpısuoɔ
Contact:

#17 Re: latfets

Post by jack »

Nick wrote: Tue Jan 23, 2018 9:50 am
jack wrote: Tue Jan 23, 2018 9:35 am Moving to dual rail gets rid of the biasing issue for the opamp but then you'll need a DC offset for the FET!
Yes, thats why I was thinking of keeping the 431 but now using it to set the DC conditions for the fet via the op amp.
Yep. Much cleaner approach - allows DC coupling and does away with X1 & R5 as well as positive feedback via Vcc.
Vivitur ingenio, caetera mortis erunt
User avatar
ed
retired
Posts: 5384
Joined: Thu Jun 21, 2007 4:01 pm
Location: yorkshire
Contact:

#18 Re: latfets

Post by ed »

thanks chaps....I've been all over this today..and have come away mighty confused now......

I might be misunderstanding your references to DC coupling( was thinking about DC coupling in valve terms) but..

did you mean bias the opamp so that the output sits on a dc value to bias the latfet and do away with the voltage divider bias on the latfet?????

if so I don't see how I'm going to do that with what I've got....

I want 8 and a bit volts peak(17-18v pp) out of the opamp(i.e. gain of 3) in order to satisfy the fet when its choke loaded. To bias the fet at around 1.6 amps I need about 4 and a bit volts on the gate....sooo with the opamp having 4 volts superimposed on the input as soon as it tries to swing the output it bottoms out....i.e cant swing 18vpp with a bias at 4v....This all makes sense in my head but I'm guessing I've missed something fundamental.

I am open to ridicule at this point.
There's nowhere you can be that isn't where you're meant to be
User avatar
Nick
Site Admin
Posts: 15706
Joined: Sun May 06, 2007 10:20 am
Location: West Yorkshire

#19 Re: latfets

Post by Nick »

Not thought it through in general, but if you have (say) a 9v supply, I was suggesting using a 9v -> 15v + 15v DC to DC converter, that then gives you enough to drive the op amp with a split rail supply, and it will simplify a lot by adding the complexity of the converter.

With what you have you are going to almost have the same problem as it can't drive the fet below 0v, so at most the output can swing no further down than 4v (if thats the op amp bias)
Whenever an honest man discovers that he's mistaken, he will either cease to be mistaken or he will cease to be honest.
User avatar
ed
retired
Posts: 5384
Joined: Thu Jun 21, 2007 4:01 pm
Location: yorkshire
Contact:

#20 Re: latfets

Post by ed »

I think I'm going to go with this to begin with. I nearly forgot what I was trying to do. I've got the tl431 in which is about as far as I can go with decoupling the bias. If there turns out to be feedback then I'll consider the extra £30 for a pair of dc converters(to keep the spirit ofdual mono).

I've gone back to the 2sk1058 as theres slightly more headroom....ecx has about 3.5v bias for 1.6 amps and the 2sk1058 has about 4.2v...all according to the models I'm using obviously.....the proof of eating will be in the pudding or somesuch. I will put the diode back in when I'm measuring the real thing to get a feel for the difference....the devil is in the detail and for this spice wont dig him out...
2sk1058 se.jpg
and the output, which I make to be about 3.33 watts.

2sk1058 op.jpg
diet: theres a link missing between stages...it's just for testing the fet bias..I forgot to put it back. between c4 and r7
more diet: It has been suggested that I'd get a bit more oomph from BUZ900s but they are T03 and wont easily go on the sinks I have....there's more mileage in this for the future if it works
There's nowhere you can be that isn't where you're meant to be
Post Reply