Electronics > Projects, Designs, and Technical Stuff
Transformer power supply stuff
(1/1)
sahko123:
I'm making a linear power supply for a project. I want to filter the mains in case the project is used in a place with shitty mains (like my home lab). Would the secondary voltage by a good representation of how the waveform would look at the primary side? I don't have a proper differential probe and cant really buy one to own yet so i want to somehow probe the mains without probing the mains.
Ian.M:
Maybe!
It depends on what's wrong with the incoming mains waveform. Iron core transformers have limited bandwidth, so fast spikes, powerline communications databursts, much higher harmonics etc. will be severely attenuated on the secondary side, Also they will start to saturate if overvoltaged so the output may not be proportional to the input.
OTOH if you don't have access to a CAT II or better isolated differential probe, its going to give you at least a rough idea if there is excessive waveform distortion without a significant risk of electrocution or blowing up your scope. Also, from the project's point of view, it doesn't really matter what the mains supply actually looks like as long as the secondary side waveform is reasonably clean near the peaks of the waveform, as that's the only part of the waveform that actually contributes to the ripple on unregulated DC rail.
David Hess:
For practical purposes, what you see on the secondary is all you have to deal with and a line frequency transformer filters its input pretty well. Line input filters are normally added to linear power supplies to prevent interference going in the other direction.
Doctorandus_P:
First I assume you have a scope and that it's grounded.
You can make a 1:10 probe by putting 9 1M resistors in series, and putting a small capacitor over each.
Make sure voltage ratings for both capacitors and resistors are adequate.
Put some heat shrink tube over everything and put it in a plastic tube from for example a pen, and solder a needle to the hot end of the probe.
Frequency compensation as normal for any 10:1 probe.
You can quite safely poke around with this tip as long as the voltage is <1kV.
Then do not use any additional GND connection of your scope, because your scope itself is (should be) grounded already.
For the transformer option:
Toroidal transformers tend to have a much higher bandwidth then E/I transformers.
All mains rated transformers are usually close to saturation, but if you put the primaries of 2 (identical) transformers in series then each only sees half the mains voltage, and measurements will be more lineair.
Navigation
[0] Message Index
Go to full version