I don’t want to mess this up for others; do I save my own version by simply renaming the project? Or is some other action required?
@namghiwook,
Looks like it was a dodgy connection on the MiniFit crimp connector. Do you have Molex compatible crimp pliers? Glad that the direct connection works. I agree with Vunce - a proper metal chassis for a Class A PSU would be more appropriate and fire-safe.
Looks like it was a dodgy connection on the MiniFit crimp connector. Do you have Molex compatible crimp pliers? Glad that the direct connection works. I agree with Vunce - a proper metal chassis for a Class A PSU would be more appropriate and fire-safe.
I use this plier. I don't know it is Molex compatible.
Actually many tube amps have chassis made of wood, so I didn't have any thought of safety of wood as material of amp chassis.
I wll try to change the 'wooden' chassis to a 'metal' chassis someday.
Actually many tube amps have chassis made of wood, so I didn't have any thought of safety of wood as material of amp chassis.
I wll try to change the 'wooden' chassis to a 'metal' chassis someday.
Your crimpers might be fine. They look like ones that do both wire and insulation at same time.
Look at your connections. Are they clean and tight?
I use AMP/Tyco/Delphi compatible crimpers:
https://a.co/d/e6wihV7
Look at your connections. Are they clean and tight?
I use AMP/Tyco/Delphi compatible crimpers:
https://a.co/d/e6wihV7
The 50VDC 1000uF Panasonic caps are on a long lead time at Mouser. Would the 35VDC ones be OK for my application; single rail, 24v in monoblocks so using 2 x single rail boards.
Last edited:
Yes that’s fine. It just needs to be rated less than DC volts after the ideal bridge rectifier.
Ok. Yes, the MiniFit Jr doesn’t like fat wires. Thinner pvc insulation might be better. In my photo that is 18ga silicone wire with fat insulation.
After using it for a year or two, the SLB powersupply in my USSA5 amp failed. The output of the negative rail died. The positive rail seemed ok, and the second SLB board in the same chassis worked fine. When I started looking for the issue, things went a bit strange... at some point the mains fuse was gone, so I replaced it. Immediately failed again. I disconnected the mains transformer from the SLB, and it was still blowing fuses. This means something is wrong with the mains transformer now (a Toroidy donut).
Questions:
Questions:
- I don't have another transformer at hand which I could use to further debug / test the SLB board. Can I test / debug the SLB board using a regulated DC bench opwersupply instead of the AC input from a transformer? Or will the active rectifier on the SLB board freak out with a DC input?
- Is is possible the SLB somehow "killed" the mains transformer? Any thoughts on how/why the mains transformer went kaputt?
Last edited:
@mbrennwa
Essentially, we need more info.
Take the Toroidy out of the enclosure and test it completely separately. Use a variac and slowly increase the voltage up to your 230V mains. Check what the voltages are on your secondaries. They should be what Toroidy has specified, maybe a little higher, since it is an unloaded measurement. Take pictures of your setup so we can see how you are implementing it. This should be a decent test to see if your Toroidy went kaput. Did you use a softstart with your USSA-5 build? If you did, which one? I hope your primary winding in your Toroidy didn’t short circuit to your core (especially if you didn’t use a softstart). That’s rare, except in builds where softstarts weren’t implemented in the first place. Everytime you turn on an amp (with a large toroid, i.e. 300VA or more), the surge currents are HUGE, and the enamel+insulation of the primary winding will wear after years of use if no softstart was used. Again, probably unlikely in your case. But caveat emptor.
Next, your SLB. You state that your negative voltage railing “died.” I don’t know what that really means, but I am assuming, you are getting zero volts. You need to take a good set of photos and show us if there any burned devices. In addition, did you use the Molex connector to connect to your cap mx pass transistor (the Toshiba transistors?)? Did the connector become loose or short circuit? Or did you directly solder the pass BJT transistor to the pcb (like I did in my NPXP build)? If the molex connector wires are loose, then you will see failure in this design. Specifically resistors R17 and R18 can burn up.
You are going to have to also look for some accidental ground connection in your USSA-5. I am curious to why your fuses are blowing when you have already disconnected the SLB. That doesn’t make sense unless there is a ground fault somewhere. Look for stray wires, etc…
You can use a DC bench supply to test the SLB, but you would apply the voltage right after the rectifiers, i.e. at the 1st capacitor, or alternatively, at the resistor (R3 and R4) in the CRC supply and insert the voltage there. Remember to use (+) volts for the (+) side of the supply and (-) volts for the (-) supply side. I would remove U1 and U2 to be on the safe side. But… I wouldn’t do any of this SLB ‘testing’ until you have a) tested the Toroidy, b) looked for accidental ground connections and c) checked to see that the Molex wiring and/or connections to the SLB pass transistor didn’t accidentally come loose. Look at the trials and tribulations of member namghiwook above. He had loose wires in the molex connector due to poor crimping (post 1538 and onwards).
It’s time to post detailed pics. Someone will spot something (usually Vunce or XRK971 😉 ).
Usually, when these things happen, you need to start looking backwards. Start at the end of the capacitance multiplier, then look at the CRC section and then look at the mains toroid. Somewhere, there will be an oops! I hope you get this rectified mbrennwa! Your contributions (OSMC and others) are most appreciated! 🙏
Best,
Anand.
Essentially, we need more info.
Take the Toroidy out of the enclosure and test it completely separately. Use a variac and slowly increase the voltage up to your 230V mains. Check what the voltages are on your secondaries. They should be what Toroidy has specified, maybe a little higher, since it is an unloaded measurement. Take pictures of your setup so we can see how you are implementing it. This should be a decent test to see if your Toroidy went kaput. Did you use a softstart with your USSA-5 build? If you did, which one? I hope your primary winding in your Toroidy didn’t short circuit to your core (especially if you didn’t use a softstart). That’s rare, except in builds where softstarts weren’t implemented in the first place. Everytime you turn on an amp (with a large toroid, i.e. 300VA or more), the surge currents are HUGE, and the enamel+insulation of the primary winding will wear after years of use if no softstart was used. Again, probably unlikely in your case. But caveat emptor.
Next, your SLB. You state that your negative voltage railing “died.” I don’t know what that really means, but I am assuming, you are getting zero volts. You need to take a good set of photos and show us if there any burned devices. In addition, did you use the Molex connector to connect to your cap mx pass transistor (the Toshiba transistors?)? Did the connector become loose or short circuit? Or did you directly solder the pass BJT transistor to the pcb (like I did in my NPXP build)? If the molex connector wires are loose, then you will see failure in this design. Specifically resistors R17 and R18 can burn up.
You are going to have to also look for some accidental ground connection in your USSA-5. I am curious to why your fuses are blowing when you have already disconnected the SLB. That doesn’t make sense unless there is a ground fault somewhere. Look for stray wires, etc…
You can use a DC bench supply to test the SLB, but you would apply the voltage right after the rectifiers, i.e. at the 1st capacitor, or alternatively, at the resistor (R3 and R4) in the CRC supply and insert the voltage there. Remember to use (+) volts for the (+) side of the supply and (-) volts for the (-) supply side. I would remove U1 and U2 to be on the safe side. But… I wouldn’t do any of this SLB ‘testing’ until you have a) tested the Toroidy, b) looked for accidental ground connections and c) checked to see that the Molex wiring and/or connections to the SLB pass transistor didn’t accidentally come loose. Look at the trials and tribulations of member namghiwook above. He had loose wires in the molex connector due to poor crimping (post 1538 and onwards).
It’s time to post detailed pics. Someone will spot something (usually Vunce or XRK971 😉 ).
Usually, when these things happen, you need to start looking backwards. Start at the end of the capacitance multiplier, then look at the CRC section and then look at the mains toroid. Somewhere, there will be an oops! I hope you get this rectified mbrennwa! Your contributions (OSMC and others) are most appreciated! 🙏
Best,
Anand.
Last edited:
Excellent troubleshooting advise Anand👍🏻
Mbrennwa,
Also check the USSA5 boards for working properly with your DC bench lab supply.
Mbrennwa,
Also check the USSA5 boards for working properly with your DC bench lab supply.
Thank you @poseidonsvoice for the advice. I'll look into this slowly, step by step. In the meantime, to make a good plan for my attack:
Basically, the question I am trying to answer is: was the transformer winding on the "dead" negative rail already kaputt before I started messing with the transformer? If a broken AC winding was the reason for the "dead" negative rail, the SLB might still be fine. My idea was to verify/check this with a DC bench PSU, and I was hoping I could just feed the DC into the AC input terminals of the SLB board.
What happens if I apply the DC to the the terminals where the transformer windings normally go? I am asking because my build is a bit like a very tight 3D puzzle. Here's a photo from when I built the thing. I'd like to test things as much as I can without taking it apart if I don't have to.
Basically, the question I am trying to answer is: was the transformer winding on the "dead" negative rail already kaputt before I started messing with the transformer? If a broken AC winding was the reason for the "dead" negative rail, the SLB might still be fine. My idea was to verify/check this with a DC bench PSU, and I was hoping I could just feed the DC into the AC input terminals of the SLB board.
Do you have another transformer you can use to power up the SLB board in question?
Disconnect the USSA5 amp board and power the SLB from another trafo and check output for even +/- voltages
The test trafo could be any size or voltage (below rated capacitors) for testing purposes.
Disconnect the USSA5 amp board and power the SLB from another trafo and check output for even +/- voltages
The test trafo could be any size or voltage (below rated capacitors) for testing purposes.
Sure, I would use another transformer to test the SLB if I had one. But I don't... that's why I am thinking to use DC input from a bench PSU. Can I do that, or will I screw up the active rectifier on the SLB?
Ahh, darn.
I’m not sure about feeding DC through the active rectifier, don’t have an answer for that, sorry.
I’m not sure about feeding DC through the active rectifier, don’t have an answer for that, sorry.
Hi Mbrennwa,
Sorry to hear about your troubles. You mentioned the second SLB in the same chassis works fine.
You can swap SLBs and see if it still works. Sounds like you have a Toroidy with 4 secondaries for dual mono from one trafo?
A blow the fuse type situation usually is a dead short or secondary phase connections reversed. A dead short is easy to check with an ohm meter on your windings.
Did you happen to remove any trafo connections and reinstall reverse phase (dots) incorrect?
I have never had a trafo short on me that was properly installed.
I would check ohm meter on windings to see if internally shorted.
Then swap working SLB with the one in the failed circuit.
Be ready to flip power off if you hear a huge hum sound, smell smoke or fire and stand back with safety glasses.
A variac is a good idea. I don’t think the active bridge works with DC input.
Good luck!
Sorry to hear about your troubles. You mentioned the second SLB in the same chassis works fine.
You can swap SLBs and see if it still works. Sounds like you have a Toroidy with 4 secondaries for dual mono from one trafo?
A blow the fuse type situation usually is a dead short or secondary phase connections reversed. A dead short is easy to check with an ohm meter on your windings.
Did you happen to remove any trafo connections and reinstall reverse phase (dots) incorrect?
I have never had a trafo short on me that was properly installed.
I would check ohm meter on windings to see if internally shorted.
Then swap working SLB with the one in the failed circuit.
Be ready to flip power off if you hear a huge hum sound, smell smoke or fire and stand back with safety glasses.
A variac is a good idea. I don’t think the active bridge works with DC input.
Good luck!
Yes, four separate secondaries on one transformer.
The chassis of this amp has been closed for more than a year, but the issue only started last week. It would take a very special type of gremlin to reverse the connections in there... ;-) Anyway, I don't think the relative polarity of the secondaries matters much. Looking at the SLB schematic, I see the AC inputs to the positive and negative rails are separated from each other by the rectifier circuits. The AC windings are therefore not connected to each other.
Ok, I should stop procrastinating and get down to work...