Couldn’t be more pleased with how this turned out. As I have mentioned, I have worked on reducing the 60/120 cycle hum in my Biacrown Hiwatt 50. In an earlier post, I discussed the printed circuit board I designed. I assembled a new one today and placed 100uf 400 volt caps in instead of the 33 I used before. This leads to an effective 50uf since they are in series. The scope tells the tale, and a photo shows how hansom it looks.
The input and output of the cap multiplier. We have the usual sawtooth at 4.04 volts in and a smooth output with the ripple reduced to 240 millivolts. Its sitting on about 525 volts and there is about 3 volts across the multiplier. That is a 16.8 to one reduction or 24.5 decibels! It does not show in this image, but the input voltage wanders up and down at something like one cycle per second–presumably due to variations in the 120 volts input.
I figured out how to use the math functions in my scope, so here is the voltage across the device. As would be expected, its voltage mirrors the input signal as its resistance changes to smooth out the input voltage. If the RMS is showing what I think it is, there is only one volt across the device for the purpose of determining how much power it is consuming and how much heat it is generating.
Here we have the multiplier sitting in the Hiwatt chassis which is gracious enough to leave plenty of space for adding things. It’s the purple printed circuit board in the upper left-hand corner. in the upper middle we have the scope probes I used to get the photo above. On the left of the top circuit board, you see the avalanche diodes I used to replace the original diodes. On the lower left of the chassis, you can barely make out Harry Joyce’s signature as the one who wired the amp, with the letters ‘joy’ barely visible upside down. In person its clear to see.
This is my first effort at building one. Proof of concept–it worked.
This is the one I had in the Hiwatt first. Those wire-to-board screw down connectors did not tighten down properly and I decided that inside the amp I wanted the multiplier to be soldered in, which is what I did in my pcb design. I also decided that this sort of breadboard was fussy to work with.