None of the pictures show where the ground thermistor was originally connected - if it was to the chassis, then there must have been voltage on the other pole, which is GND, which is hardly possible. Could it be that the chassis-side of the thermistor was accidentally connected to one of the rails instead? There are only red and black cables, so it's easy to mix them up. But maybe I'm wrong, of course.
Last edited:
After all the capacitors have been removed checked we can use an ohmmeter to check the board for any defects.
Dan
Dan
If it's a faulty cap (which I've come to agree that it might be), it should probably be one of the caps in front of the thermistors. In other words, one of the caps out of the two 4-capacitor-banks right after the rectifiers. Otherwise one of the thermistors in the DC supply would've also seen some hefty current.
So either CT to Rect+ on the positive or CT to Rect- should measure as a short. That would narrow the search down to four caps.
You should measure if there's any residual charge in the PSU caps (simply measure voltage over cap pins) and bleed it off before desoldering, if you measure any.
So either CT to Rect+ on the positive or CT to Rect- should measure as a short. That would narrow the search down to four caps.
You should measure if there's any residual charge in the PSU caps (simply measure voltage over cap pins) and bleed it off before desoldering, if you measure any.
The check should include checking for DC bias on the mains that you connect your power supply to.
See https://sound-au.com/articles/xfmr-dc.htm .
Looking at that again lead me down a rabbit hole: in https://sound-au.com/xfmr2.htm#s122, the same author discusses asymmetric inrush current, and in https://sound-au.com/project39.htm talks about inrush current limiting, and why a thermistor alone might not work (for long).
So the power supply could be doing this (creating offsets and destroying thermistors) all by itself.
The shorted capacitor is rendering the PSU inoperative, so remaining capacitance is inconsequential.
If the bad capacitor can be determined, it’s easy enough to replace.
If the bad capacitor can be determined, it’s easy enough to replace.
The shorted cap will short out the rest of the ones in parallel with it.
But if it is shorted, without removing it from the board, you will know which group it is in.
Sometimes the structure of the snap in leads can cause a short to a trace underneath,
on the top of the board, if there is one (which there shouldn't be, to avoid a short).
But if it is shorted, without removing it from the board, you will know which group it is in.
Sometimes the structure of the snap in leads can cause a short to a trace underneath,
on the top of the board, if there is one (which there shouldn't be, to avoid a short).
Routinely a leaking failed cap will heat up if the circuit operates for a little while.
And the mains at the outlet are ok? No AC offset between ground and neutral? No noticeable DC offset?
I just reread the note for the F5m amp, and a remark that I assume applies to other amps as well seemed at odds with your symptoms: The notes say the primary fuse should be dimensioned such that it blows before the thermistors overheat: "The AC line fuse would be chosen at a rating lower than the current rating of the Thermistors." (from https://cdn.shopify.com/s/files/1/1006/5046/files/F5M_DIY_AMPLIFIER_R1.pdf?v=1709576510) .
You are observing the opposite, right? First the secondary side ground connect thermistor overheated, and after you disconnected that, the thermistor on the primary side turned red hot and cracked, and only then (after several seconds, I assume) did the primary fuse blow. What are the specs of the thermistors and the fuse you are using?
Also, how big is the transformer (I mean in VA or kVA)? And it measures high DC resistance (MOhm) between the primary and secondary side?
I just reread the note for the F5m amp, and a remark that I assume applies to other amps as well seemed at odds with your symptoms: The notes say the primary fuse should be dimensioned such that it blows before the thermistors overheat: "The AC line fuse would be chosen at a rating lower than the current rating of the Thermistors." (from https://cdn.shopify.com/s/files/1/1006/5046/files/F5M_DIY_AMPLIFIER_R1.pdf?v=1709576510) .
You are observing the opposite, right? First the secondary side ground connect thermistor overheated, and after you disconnected that, the thermistor on the primary side turned red hot and cracked, and only then (after several seconds, I assume) did the primary fuse blow. What are the specs of the thermistors and the fuse you are using?
Also, how big is the transformer (I mean in VA or kVA)? And it measures high DC resistance (MOhm) between the primary and secondary side?
Hi Guys,
I am going to desolder my caps and check them in the capacitor mode.
My multimeter doesn't have that functiom, so I have to go out and get one.
I'll get back to you when I get a result.
Thanks again,
Dave
I am going to desolder my caps and check them in the capacitor mode.
My multimeter doesn't have that functiom, so I have to go out and get one.
I'll get back to you when I get a result.
Thanks again,
Dave
You may be able to check the capacitors without removing them.
Try it first without removing them. See this:
https://www.electricaltechnology.org/2013/06/how-to-check-capacitor-with-digital.html
The paralleled capacitors at the DC input are paralleled. If one of the paralleled capacitors is shorted, that will affect the measurement. Measure the V+ capacitors and compare them to the V- capacitors.
Try it first without removing them. See this:
https://www.electricaltechnology.org/2013/06/how-to-check-capacitor-with-digital.html
The paralleled capacitors at the DC input are paralleled. If one of the paralleled capacitors is shorted, that will affect the measurement. Measure the V+ capacitors and compare them to the V- capacitors.
If you have or have access to a thermal camera, they're pretty amazing pieces of gear for troubleshooting. Small variations in temp show up like a neon sign
I'll let others chime in here if they so choose. You should be able to tell if a capacitor is defective with a normal ohmmeter. No need for a fancy one. Since it appears that 10-15 amps is flowing through the one thermistor there are a limited number of things that can cause this. The community at large (and I agree) seems to think it is a shorted capacitor. IMHO the next logical step is to remove all capacitors and check them one by one. Ben Mah offered and excellent link on this.
Keep going. A solution is near.
Regards,
Dan