Ok, it's not the transformer.
I cut one leg of the rectifier heater windings - nothing.
Next I cut ONE leg of the 660 AC (T1-4)... powered it up and the fuse that I bought from the hardware store never went, but R5 did !! Nicely, sort of like a sparkler... a few days too late though.
R5? That's interesting. B- is drawing too much current. Why would it go with only one phase attached to it... Could be that the fuse I used was extra slo-blow.
I'd suspect the mosfet, but it didn't measure shorted and there is no B+ anyway.
I need ideas at this point...
I cut one leg of the rectifier heater windings - nothing.
Next I cut ONE leg of the 660 AC (T1-4)... powered it up and the fuse that I bought from the hardware store never went, but R5 did !! Nicely, sort of like a sparkler... a few days too late though.
R5? That's interesting. B- is drawing too much current. Why would it go with only one phase attached to it... Could be that the fuse I used was extra slo-blow.
I'd suspect the mosfet, but it didn't measure shorted and there is no B+ anyway.
I need ideas at this point...
OK, so with R5 "gone" you can hook the transformer all back up to the PCB and it doesn't blow fuses? That's a good sign. While R5 is only a 2W part, it's only 270 ohms while the 5W part is 10k. So for a given current through the two, Ohm's law says R5 can handle about 4x the current that R6 can before reaching its dissipation limit. That was my line of thinking anyway. Can you check the resistance of R6 to see if it is still in spec? If it somehow shorted internally (pretty rare, but who knows), maybe it is passing much more current than it should.
If the power supply side checks-out, then yeah...the MOSFETs would be the next thing to look at. Maybe try desoldering one leg of R14 and R25 and then replace R5. If you can power up like that (no tubes, of course), then the MOSFETs are likely the culprit.
EDIT: the above might be such a good idea...maybe George can chime-in here but lifting those resistors might cause C6 and C7 to see too much voltage due to the drop in current draw on the B- rail. In this circuit, I measure about 200k between the source and the drain (pin 3 and 2). Put positive lead from yoru meter on the source.
If the power supply side checks-out, then yeah...the MOSFETs would be the next thing to look at. Maybe try desoldering one leg of R14 and R25 and then replace R5. If you can power up like that (no tubes, of course), then the MOSFETs are likely the culprit.
EDIT: the above might be such a good idea...maybe George can chime-in here but lifting those resistors might cause C6 and C7 to see too much voltage due to the drop in current draw on the B- rail. In this circuit, I measure about 200k between the source and the drain (pin 3 and 2). Put positive lead from yoru meter on the source.
rknize said:
I haven't hooked the transformer back up yet...
I would agree with you that it could be the mosfets, but since there is no B+ I just don't see a current path there...
My meter says that they are both still diodes. Although one is dropping more than the other (.8v vrs .6v). The caps appear to be still caps, since there are no shorts.
I have seen these diodes measure good with an ohm meter but still break down when high voltage is applied. With R5 removed and the 5AR4 removed but the transformer reconnected, a blown fuse would indicate bad diodes.
Shorted mosfets will generally make R6 hotter than normal but not bother R5. It will put positive bias on the output tubes causing them to get upset and possibly blow fuses.
Test as above (no tubes installed). No R5. Blow fuse, replace diodes. No blown fuse, Replace R5 and try again. Blown fuse or smoking R5, replace caps. No smoke, verify negative voltage present. Voltage OK, install 5AR4 (no other tubes). Measure voltage at output tube grids. It should be adjustable via the bias pots. If one channel does not adjust and is at a positive voltage, replace the mosfet for that channel and check the grid stopper resistor.
Shorted mosfets will generally make R6 hotter than normal but not bother R5. It will put positive bias on the output tubes causing them to get upset and possibly blow fuses.
Test as above (no tubes installed). No R5. Blow fuse, replace diodes. No blown fuse, Replace R5 and try again. Blown fuse or smoking R5, replace caps. No smoke, verify negative voltage present. Voltage OK, install 5AR4 (no other tubes). Measure voltage at output tube grids. It should be adjustable via the bias pots. If one channel does not adjust and is at a positive voltage, replace the mosfet for that channel and check the grid stopper resistor.
tubelab.com said:
Thanks George - sounds like a plan. I'll get to some of this today, but replacing parts - yeah, I didn't order extras.
One thing - I was using a 5U4-GB/A old used tube for the rectifier. Any chance this had something to do with smoking the diodes/caps etc (assuming that is what the problem is)
PaulyT said:
Err, yeah, No diode. I remember seeing it but I already bought my parts and I don't have an equivalent one lying around...
You know - I do have an inrush current limiter in circuit... this takes a few seconds to warm up which delays B+ even longer - perhaps that slightly negative voltage is more damaging that it would be normally...
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.