Guru
Joined: 13/10/2014
Location: AustraliaPosts: 2039
| Posted: 09:50am 16 Jan 2023 |
Looking at noise on the Inverter AC out.
I monitor the AC out via what appears to be a small current transformer? pulled from the old SMA inverters I scrapped, 600V to 7.5V 66mA VN30.15/00858 25mm square potted block, quite weighty for its size. Anyway, I use this to give me ground isolation between the CRO probe ground clip and other test gear, or when looking at more then one input on the CRO, and to help with earth loops around the inverter.
I left the gold foil shield on the AeroSharp test toriod, I wanted to see if it made any change to the common kinks in the wave form? No it didn't. Was it useful in stopping noise coupled between the primary and secondary? Yes it certainly was.
However a simple 240V filter on the secondary side of the transformer does exactly the same thing as the transformer shield. This could change at high power?
The small amount of noise reduced quite a lot, and there was no longer this weird noise induction appearing on the AC out when I touched the choke lead. I thought that was being induced into the CRO leads, but since the shield stopped that too, it's likely more just radiated noise getting into the primary windings and my test layout.
I have found a few interesting results in this current test setup:
1: Crossover point glitches get worse as the resonating CAP value get larger.
2: Placing a CAP at the output side of the primary Choke does remove some noise, but it does the same thing as increasing the secondary CAP value, it causes the crossover glitch AND it also increased idle current, so there may be some resonance interaction at play?
3: A lower value of resonating CAP did not increase idle current, but that may be because the lower value of CAP [1.2uF] was needed to resonate the transformer at 76 Hz.
4: The Choke from blue powdered iron mix cores at only 40uH, let a tiny amount of 24 kHz switching noise through to the secondary. However:
A: It does NOT increase idle current [11 watts], therefore it is not causing eddy current loss in the toriod, well at least not at low power loads and idle.
B: It removes most of the remaining glitches on the sine wave.
C: The sine wave no longer change shape under light loads.
D: The transformer, choke and load are now silent at idle to 300 watts (may test limit) and zero startup noise.
E: Using a choke with "more" inductance causes a slight startup growl and the Toriod starts to get a low frequency growl even at idle.
This is with the transformer shield, 40uf choke and 1.2uF CAP.
I've just modded the EG002 board and I will get it connected to the batteries tomorrow.
Edited 2023-01-16 20:20 by KeepIS