Power supply simulator. / General Science and Electronics / Forums

Forums

4hv.org :: Forums :: General Science and Electronics

« Previous topic | Next topic »

Power supply simulator.

first previous 3 4 5 6 next last

Move Thread

LAN_403

2Spoons

Tue May 08 2018, 10:20PM

Registered Member #2939 Joined: Fri Jun 25 2010, 04:25AM
Location:
Posts: 615

Ash Small wrote ...

EDIT:- Unfortunately I still can't measure the DC resistance of choke, as the only meter I have available at present won't measure less than 0.5 Ohm.

put an amp through it and measure the voltage - surely your meter can measure millivolts?
Its the only realistic way to measure small resistances, otherwise you have to start compensating for lead and contact resistances.

Ash Small

Wed May 09 2018, 03:28PM

Registered Member #3414 Joined: Sun Nov 14 2010, 05:05PM
Location: UK
Posts: 4245

Well, I've rigged up a test circuit, transformer, rectifier, 4700uF capacitor, 2.3mH choke and another 4700uF capacitor.

All this into a 5.2 Ohm load.

Voltage ripple on the first capacitor is 2.5V, on the second it's less than 1V.

Seems like the choke is working.

It's been running for nearly an hour and a half, and is getting warm to the touch, but not very warm. PVC insulation is good for 105C, so everything seems ok.

I do have a 250 gramme spool of 14SWG here, so might try winding one with that and see how far I get.

The image below shows both traces, 'scope is set to 0.5V per division.

DerAlbi

Wed May 09 2018, 05:23PM

Registered Member #2906 Joined: Sun Jun 06 2010, 02:20AM
Location: Dresden, Germany
Posts: 727

If thats such a nice sine, why not short circuit it with a 50/100 Hz notch filter

(L-C series connection from VCC to GND with 50/100Hz self resonance, 50H or 100Hz depending on your rectification)

Edit: i just simulated..
470uH (low L-value to minimize series resistance) + 5400uF (parallel some for low ESR) provides 20dB attenuation @ 100Hz given 50mOhm overall ESR of C+L.
Compared to no notch filter.

Sulaiman

Wed May 09 2018, 06:19PM

Registered Member #162 Joined: Mon Feb 13 2006, 10:25AM
Location: United Kingdom
Posts: 3141

try simulating with the new 470 uH at the rectifier output followed by the 5400 uF capacitance to 0v,
then the existing filter, then the load .....

Ash Small

Thu May 10 2018, 03:11AM

Registered Member #3414 Joined: Sun Nov 14 2010, 05:05PM
Location: UK
Posts: 4245

Sulaiman wrote ...

try simulating with the new 470 uH at the rectifier output followed by the 5400 uF capacitance to 0v,
then the existing filter, then the load .....

I tried simulating that last week, the voltage on the rectifier spikes up to 220V.

I'm considering a notch filter, but it needs very precise values. Electrolytics have very wide tolerances, so it basically comes down to trial and error. I calculated that for a 1.5mH choke I'd need 6750uF.

DerAlbi

Thu May 10 2018, 05:14AM

Registered Member #2906 Joined: Sun Jun 06 2010, 02:20AM
Location: Dresden, Germany
Posts: 727

It depends on what kind of performance you want to achieve. The higher the ESR of the notch, the worse its quality factor the wider its notch the more tolerant to derivations from optimal values it is. ESR is a very important part of a filter, you cant neglect parasitics.
In Simulation, Notch + C-L-C-PiFilter is worse than C-L-C-PiFilter + Notch in transient and AC-simulation because the notch does not work well without a source impedance. The Pi-Filter gives the notch some source-impedance so it actually can short circuit something without being driven by force...

Ash Small

Thu May 10 2018, 06:56PM

Registered Member #3414 Joined: Sun Nov 14 2010, 05:05PM
Location: UK
Posts: 4245

DerAlbi wrote ...

It depends on what kind of performance you want to achieve. The higher the ESR of the notch, the worse its quality factor the wider its notch the more tolerant to derivations from optimal values it is. ESR is a very important part of a filter, you cant neglect parasitics.
In Simulation, Notch + C-L-C-PiFilter is worse than C-L-C-PiFilter + Notch in transient and AC-simulation because the notch does not work well without a source impedance. The Pi-Filter gives the notch some source-impedance so it actually can short circuit something without being driven by force...

I figured the bast place to put it was in parallel with the second cap in a Pi filter

Ash Small

Sat Jun 30 2018, 08:29PM

Registered Member #3414 Joined: Sun Nov 14 2010, 05:05PM
Location: UK
Posts: 4245

Well, I've wound a couple of chokes, both using the 3" type 26 D coils (D denotes double width), both with 14 SWG magnet wire, one with 250 grammes, the other 500 grammes.

Winding the 500 gramme one was a nightmare, but I finished last night. First layer was very slow going, lots of birds nests to untangle. It became easier during the second layer, but the wire started to work harden as I started the third layer, and became worse as I progressed.

It came to exactly four layers.

I've just tested it and am getting some very surprising results.

The first one, with 250 grammes on wire is around 1.75mH, with twice the wire on the second one (less than twice the number of turns, due to increasing circumferance, I'd expect maybe 3mH (1.75^2), however, the results Im getting are indicating 9mH (certainly well over 8,5mH)

While some may be due to variations between cores, I think the main factor here is vastly reduced stray inductance, due to more layers and tighter winding.

I knew this was a black art (I've been studying it for two decades) but this has taught me loads about stray inductance (assuming that's the factor involved here).

I've just wound a choke that's 10mH +-10-15%, I'm amazed

EDIT: the SWG wire is apparently rated for 9.6 Amps (~10 Amps), I only plan on putting half that through them, certainly initially.

Next job is to test it under load.

Ash Small

Thu Jul 12 2018, 07:46PM

Registered Member #3414 Joined: Sun Nov 14 2010, 05:05PM
Location: UK
Posts: 4245

I came across a Tektronix test capacitor while sorting a box of stuff I'd been given, accurate to +1/4%.

Thought I'd run a test with it, using the 1uF section, and it pretty much agreed with my other tests, so I'm presuming it is good (my decent DMM still hasn't turned up since moving house, so I've no way of measuring exact capacitance).

Assuming it is good, I re-tested the latest choke, and get a reading of 9.3mH, that's 10mH +/-10%. I'm more than happy with that.

Sulaiman

Thu Jul 12 2018, 10:19PM

Registered Member #162 Joined: Mon Feb 13 2006, 10:25AM
Location: United Kingdom
Posts: 3141

All toroid cores have a specified Al, inductance factor,
inductance will then be Al x (number of turns squared)
this applies from one to many many turns.

Also any core has a maximum energy storage limit, 0.5 x L x I^2
which is related to the amp.turns = magnetic saturation limitation.

In other words, I believe that the variable results that you got are almost entirely determined by the cores and number of turns.
=======================================
Abo ve a few MHz cores tend to be of low relative permeability so do 'leak' magnetic flux,
self-capacitance due to multi-layer windings or litz wire is usually avoided.

first previous 3 4 5 6 next last

Moderator(s): Chris Russell, Noelle, Alex, Tesladownunder, Dave Marshall, Dave Billington, Bjørn, Steve Conner, Wolfram, Kizmo, Mads Barnkob