Skewed sine wave in PSUD?

[whoaru99]

Playing around with PS for tube amp and when running one of the iterations I see this slight skewing of the wave when I zoom in to look at the ripple. This would be the node for the plate supply.

The ripple level itself I believe will be fine for my use, but any cause for concern with that slight skewing?

 

[Alan0354]

Some context would be helpful. I have no idea what is this.

 

[BinaryMike]

I'm pretty sure you're seeing a filtered sawtooth waveform here, which would be entirely normal at the first and possibly second filter stages downstream of the rectifier.

 

[megasat16]

Ripple voltages is not sine wave. It is more like a sinusoidal (sawtooth) shape wave. Yours looks pretty low and pretty similar to sin. It is very low level and seems conduction angle is not that bad. Is that LC filter or CR filter or a combination of both?

Mike beats me to it!

 

[whoaru99]

Some context would be helpful. I have no idea what is this.

This power supply circuit, at the I1 load. It's a way zoomed in view of the curve out in the flat section, maybe around 17-second area.

 

[Alan0354]

I think you have a very quiet supply, if I read the first diagram correct, those are ripple that is so low.

 

[kward]

This power supply circuit, at the I1 load.

Check the ring at your output transformer tap point. Usually you will exacerbate the ring situation when you make both caps in the CLC filter the same size. Maybe make the first cap 100 uF instead. Or add a little series resistance after the rectification stage but before the first cap,

You can check ring by making the current sink step to a slightly higher value after say five seconds of simulation time. This would more or less simulate an LF transient pulse, like say a bass drum beat.

 

[whoaru99]

There is a little bit of wiggle there when the load is stepped from 165 to 200mA with 550uF in both places. And, when unloaded between the same values it wiggles the opposite way.

 

[Alan0354]

Of cause, you have an unregulated power supply, it will droop when you increase current drawn. Just be thankful you got rid of all the ripple!!!!

 

[whoaru99]

FWIW this is tentative for my Kegger KT88 UL PSE amp.

I can spec the caps to be 550uF at the first node and 1200uF at the 2nd node for not much more money, thereby dropping the ripple down to ~1mV with no apparent ring on stepped load.

Is there a point of capacitor absurdity?

 

[whoaru99]

Well, ... Did it start with a 3uF > choke > 50uF? How much cap does it take to replace the choke? Does one built according to the original plan not perform well or is it just this sim you have going throwing doubt into the design? Do diodes create a higher need for big caps? Where did I put my Magic 8 Ball?

I have not built the amps yet, and I've changed my direction on the PS. Thanks for your input.

 

[kward]

That wiggle doesn't look bad at all. I assume the series resistance of the caps and power transformer are modeled accurately.

 

[whoaru99]

That wiggle doesn't look bad at all. I assume the series resistance of the caps and power transformer are modeled accurately.

The PT specs are what PSUD comes up with when I put in the loaded/unloaded voltages from AnTek datasheet. The series resistance of the caps is double the cap's datasheet value as I'm going to use two in series (to get plenty of voltage tolerance even when the PS is unloaded/no tubes installed).