Well I decided to replace my 200ohm gain resistors on my f5 with 2x200, basically go back to stock.
I pull the boards, add another set of 200ohm resistors on top of the existing, put in new sil-pads, set the pots to 0 and fire it up. Seems good, 0V bias then one of the thermistors on a mosfet catches on fire. Burns up completely.
No resistors burned, Mosfets aren't grounded on the heatsink. What happened? Thermistor accidently touching something it shouldn't? Where to start?
I pull the boards, add another set of 200ohm resistors on top of the existing, put in new sil-pads, set the pots to 0 and fire it up. Seems good, 0V bias then one of the thermistors on a mosfet catches on fire. Burns up completely.
No resistors burned, Mosfets aren't grounded on the heatsink. What happened? Thermistor accidently touching something it shouldn't? Where to start?
check them all , replace what's needed, case solved
with so low number of parts , there is no use of playin' Poarot , if you already don't remember what was the cause
I know for F5s caught in smoke , when - due to overtemp (sissy htsnks) mosfet goes oped G-D , and that's causing adjacent NTC go puff
in your case ........ dunno
with so low number of parts , there is no use of playin' Poarot , if you already don't remember what was the cause
I know for F5s caught in smoke , when - due to overtemp (sissy htsnks) mosfet goes oped G-D , and that's causing adjacent NTC go puff
in your case ........ dunno
hopefully not turned in the wrong direction
personally I never tweak or experiment with amps that work fine
when people ask, did your try this or that cap, etc etc
I say, no why should I, it works fine as is
And build a lamp checker
Biasing up Class A with a "lamp checker" ? Are we checking for dim bulbs ?
Well, it was an amp that was up and running good. Guess I could have used a bulb tester.
I had 200ohm resistors on R5-R8 which I had read will increase gain by 6db. Since I am using a preamp now I decided I didn't need the extra gain so I added another set of 200ohm resistors in parrallel to the existing 200ohm to make 100ohm, the stock value.
Since I have to put in new mosfets most likely I'll just put in the proper 100 ohm resistors. Although I don't think that was the problem.
All I have on hand for mosfets is IRFs, I had fairchilds in the amp.
So, if a mosfet grounds with the heatsink is it likely the mosfet is fried? That's my best guess as to what happened.
I had 200ohm resistors on R5-R8 which I had read will increase gain by 6db. Since I am using a preamp now I decided I didn't need the extra gain so I added another set of 200ohm resistors in parrallel to the existing 200ohm to make 100ohm, the stock value.
Since I have to put in new mosfets most likely I'll just put in the proper 100 ohm resistors. Although I don't think that was the problem.
All I have on hand for mosfets is IRFs, I had fairchilds in the amp.
So, if a mosfet grounds with the heatsink is it likely the mosfet is fried? That's my best guess as to what happened.
Good info, thanks Zen.
Anything else I should consider? So, what exactly happens when the mosfet is not isolated? It's dumps V+ to ground? Is there potential for jfet/transistor damage when this happens as well? Nothing was hot after tis happened (except the thermistors) and all resistors look good.
Anything else I should consider? So, what exactly happens when the mosfet is not isolated? It's dumps V+ to ground? Is there potential for jfet/transistor damage when this happens as well? Nothing was hot after tis happened (except the thermistors) and all resistors look good.
You cannot bias up a ClassA amp while powered up via a bulb tester.
Even ClassAB amplifiers should not be biased up via a bulb tester.
Yes. Always use the bulb tester after EVERY modification, no matter how simple the mod appears.Not biasing. Smoke testing after a component change...
No............. class A/B will can be done. at low bias
The bulb increases the source impedance and the result is that the amplifier under test runs on a lower voltage than normal.
If the bias is turned from OFF to a little the voltage drops even more. This results in a false bias voltage.
If the bias is turned up to optimal ClassAB then the voltage will drop a lot and the amp could end up being biased when the rails are at 50% to 80% of "normal" supply voltage.
DO NOT try to bias any amplifier via the bulb tester.
Use the bulb tester to check for errors.
Well, the saga continues.
I replaced the gain resistors w/100ohm, installed new IRF mosfets and thermistors. Powered it up, no smoke, no fire.
But it won't bias up. P2 has no effect. DC offset is about -1V. Adjusting P1 brings the bias up w/little effect on offset.
I think the Jfets are also burned. That seem like a good diagnosis?
I guess I will order new ones.
So, I had Fairchild mosfets in the amp. I just replaced them w/IRFs. Since I have to order JFETs from spencer I could order some 2SK1530-Y/2sJ201-Y Mosfets too and put them in.
Any mods needed (resistors values, etc.) to install 2SK1530-Y/2sJ201-Y? I hear the toshibas are the prefered mosfet for the F5.
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.