martin manning wrote: ↑Sat Dec 25, 2021 2:22 pm
If two different values are used then they will want to share the voltage inversely with their capacitances, and the voltage ratings and balancing resistors would be chosen accordingly.
Again, I see no reason for this. I think that the voltage ratings and resistors should be chosen such that the WVDC for each capacitor is not exceeded under the worst case scenarios.
The voltage drops across series capacitors are not necessarily inversely proportional to the capacitances of the series capacitors. In fact, it would be wildly coincidental if the voltage split according to the capacitances alone. The leakage current of a capacitor is given by this equation: (x * C * V) + y where x is a multiplication factor and y is a fixed amount of microamps. If C were the same for both capacitors, then it would be the WVDC for each capacitor that determines the voltage split.
The leakage current, and the resulting equivalent resistance, of the two 50uF capacitors strapped together is not the same as a single 100uF capacitor. Therefore, the voltage drop across the strapped 50uF capacitors is not the same as the voltage drop when a single 100uF capacitor is used.
martin manning wrote: ↑Sat Dec 25, 2021 2:22 pm
Ultimately they will settle somewhere in between, which means there will be some transient behavior related to the difference in time constants introduced.
Minor differences in tau brought on by minor differences in balancing resistors are going to be swamped once the power tube plates are connected.
I picked 500V as an arbitrary example, with the assumption that in the worst case the voltage rating of the capacitors will not be exceeded. The key is that the balance resistors are sized to pass several times the expected leakage current of the caps so they will dominate the paralleled resistances and therefore the voltage split. You can read all about how to select the values in app notes published by capacitor manufacturers. Here we are talking about a stack with dissimilar capacitances, which will want unequal resistances.
Here is an example (Fender "'65 Deluxe Reverb"):
You do not have the required permissions to view the files attached to this post.
Looks good! Do you have a little silicone under the new axial caps to secure them? What technique did you use to remove the excess solder around the old filter cans?
martin manning wrote: ↑Tue Dec 28, 2021 12:47 pm
Looks good! Do you have a little silicone under the new axial caps to secure them? What technique did you use to remove the excess solder around the old filter cans?
No I don't have anything holding them in. Truth is, I was having a hard time figuring out how to mount them AND make it look neat and tidy too, with the way they mounted the old dynamite sticks in there I didn't want to just replicate that, but try to figure out a cleaner layout. To be honest I'm not quite happy with it and I think I'm gonna try to clean it up. I had nothing to melt the solder pools, my dad had a 200 watt solder gun that still wouldn't touch it, so I read and hammer and chisel works, and it did. He did offer me a big soldering iron that you use a torch to heat up, and looked like it is used to brand cattle, I just kind of laughed and said no thanks lol.
I think what you have done looks great. It would be a good idea to squeeze a bead of silicone under each cap to keep them from vibrating, which could make noise and/or cause the leads to fatigue and break. Be aware that the metal cases of the axial caps in the stacked filters may be sitting at half of B+.
