Home
JAQForum Ver 20.06
Log In or Join  
Active Topics
Local Time 06:26 20 Sep 2021 Privacy Policy
Jump to

Notice. New forum software under development. It's going to miss a few functions and look a bit ugly for a while, but I'm working on it full time now as the old forum was too unstable. Couple days, all good. If you notice any issues, please contact me.

Forum Index : Electronics : Bruce Inverter build

     Page 3 of 3    
Author Message
Revlac

Guru

Joined: 31/12/2016
Location: Australia
Posts: 526
Posted: 10:33am 05 Sep 2021
Copy link to clipboard 
Print this post

Now with 12 turns, I can see how it adds up, quite dramatic.
The little LC meter, Calibrated it each time as usual, tested with 12 turns and returned a reading of 74uH or something, so re-calibrated and tested 2 more times and got the desired result, apparently a wire contacting a finger will stuff up the readings, so best to check if not sure.
The air gaps are the same, I think the edge of the spacers are a little frayed.
94uH @ 391568Hz Close enough to the 97uH  would be some error margin with these meters.


Cheers Aaron
Off The Grid
 
Warpspeed
Guru

Joined: 09/08/2007
Location: Australia
Posts: 4200
Posted: 09:12pm 05 Sep 2021
Copy link to clipboard 
Print this post

There can be problems measuring any inductor, as it always has to be done with ac at some particular measurement frequency.

Now any inductor will also have inherent stray self capacitance, and hence a self resonant frequency. Things can go really screwy around the self resonance point, or even close to it. Errors can be huge and the results meaningless.  So don't be too surprised if three different types of measurement instruments produce three different inductance values.

Try to fit on as many turns as you can possibly squeeze in there, using the thinnest wire you dare use at full maximum current. That will give you the greatest combined inductance and saturation figures.  Then tweak the air gaps trading off inductance against final saturation.
Cheers,  Tony.
 
renewableMark

Guru

Joined: 09/12/2017
Location: Australia
Posts: 1663
Posted: 10:24pm 05 Sep 2021
Copy link to clipboard 
Print this post

That's the same meter I have, they measure a odd way which gives a consistent, but much lower reading.
From memory you can't change the frequency.
Have a look here https://www.thebackshed.com/forum/ViewTopic.php?TID=12027&P=33 Bottom of the page.
Cheers Caveman Mark
Off grid eastern Melb
 
Revlac

Guru

Joined: 31/12/2016
Location: Australia
Posts: 526
Posted: 11:39am 07 Sep 2021
Copy link to clipboard 
Print this post

  Quote  Try to fit on as many turns as you can possibly squeeze in there

Some time back I tried a large choke from an old battery charger, heavy wire, large air gap, lots of turns etc, checked it with this meter and it was 685uH and produced the nastiest waveform ever seen and high idle current, under light load it got much worse, just my findings on a previous build, results for others may differ.

Think I will stick with the sweet spot with this one, and wind it up with a jig on the lathe, did one by hand......fun it was not.

  Quote  That's the same meter I have, they measure a odd way which gives a consistent, but much lower reading.
From memory you can't change the frequency.

They are helpful for this even if not perfect, have tested a few capacitors as well, been very handy for the price .
You have reminded me of another project that needs to be finished, the little Charge controller, most of it is done and will post one day.
Edited 2021-09-07 21:41 by Revlac
Cheers Aaron
Off The Grid
 
Warpspeed
Guru

Joined: 09/08/2007
Location: Australia
Posts: 4200
Posted: 05:01pm 07 Sep 2021
Copy link to clipboard 
Print this post

  Revlac said  
Some time back I tried a large choke from an old battery charger, heavy wire, large air gap, lots of turns etc, checked it with this meter and it was 685uH and produced the nastiest waveform ever seen and high idle current, under light load it got much worse, just my findings on a previous build, results for others may differ.

I would expect to see a horrible wobbly distorted waveform under no load, exactly as you describe.

The reason being the high series impedance of the choke decouples the transformer from the PWM waveform, which is what the choke is supposed to do. The greater the inductance the more free the transformer is to "do its own thing".

The transformer will have a natural resonance of typically a few hundred Hz, and that can be excited into ringing from a harmonic of 50Hz. These wobbles will be stationary because they are a direct harmonic multiple of 50Hz.

The resonant wobble below looks to be at about the tenth harmonic, probably 500Hz.
This is bad enough, but it can be a lot worse than this if its a lower harmonic.



People see this and realise its being caused by the choke, which it is, and think the choke is crap. That is entirely the wrong conclusion.
A good high inductance choke offers many benefits, but also creates an interesting new problem.

If you add capacitance across the transformer secondary, this obviously lowers the transformer resonant frequency. All that does is make the wobbles larger, as it now resonates at a lower harmonic where there is even greater energy to cause the wobbles.

But there is a trick !!

If you move the resonant frequency down to EXACTLY 1.5 times the inverter frequency (75Hz for 50Hz inverter) any slight resonant buildup of energy in one cycle ends up being out of phase and cancelled by the following cycle.
Its a kind of automatic self damping effect, and its very effective in reducing wobbles to the point where they are no longer visible.

This tuning process needs to be pretty precise to be effective, maybe 75Hz +/-1Hz works very well.  Just experimenting with a whole bunch of random capacitors will only make it worse.
The tuning process must be very deliberate and precise set to an accurate frequency.

The wobbles are a resonance, which is worst at zero load, or light load.
Under heavy inverter load, the wobbles will be somewhat damped out by the load.

Very severe wobbles will definitely increase idling current, because the resonant buildup of energy in the transformer is fighting the inverter. Once the transformer is properly tuned, idle power will drop, you will have a clean sine wave, and all will be well.
Edited 2021-09-08 03:08 by Warpspeed
Cheers,  Tony.
 
brucedownunder2
Guru

Joined: 14/09/2005
Location: Australia
Posts: 1512
Posted: 10:46pm 09 Sep 2021
Copy link to clipboard 
Print this post

I have fitted the shelf on brackets, plenty of room for choke and other parts . the coil goes bottom left ,choke on top .
Had a bit of grief posting the pic ,must be something astray with my I-pad or me ,we'll blame me !.  I'm a bit slow at working on this inverter, many other pressing things to fix around the house. Frig carked it, water seeped through to my solar racks,, washing machine decided to rattle and Hum !, hot water system wanted a new heater installed, and ,of course, I still have my concrete water tank to tape-up the cracks, before the rain comes.

So, you can imagine ,with all these hobbies AND me going to hospital visit every week on and off ,not much time left. But, I'm feeling fine this last year , so the dreaded "C" is taking a rest...

Bruce
 

Edited 2021-09-14 11:41 by brucedownunder2
Bushboy
 
Revlac

Guru

Joined: 31/12/2016
Location: Australia
Posts: 526
Posted: 08:59am 15 Sep 2021
Copy link to clipboard 
Print this post

That looks better Bruce, I hadn't thought about using angle there, as I didn't know if it would be in the way of the wiring, but it look like the thick perspex made a good easy mounting point.

The choke should fit in there ok, and looks like enough room for one of these other chokes if need be.
Cheers Aaron
Off The Grid
 
Revlac

Guru

Joined: 31/12/2016
Location: Australia
Posts: 526
Posted: 09:05am 15 Sep 2021
Copy link to clipboard 
Print this post

  Quote  This tuning process needs to be pretty precise to be effective, maybe 75Hz +/-1Hz works very well.  Just experimenting with a whole bunch of random capacitors will only make it worse.
The tuning process must be very deliberate and precise set to an accurate frequency.

The wobbles are a resonance, which is worst at zero load, or light load.
Under heavy inverter load, the wobbles will be somewhat damped out by the load.

Found a photo of that choke when tested, can't find the notes I had written, but the photo was taken back about 2018 sometime, so there is a good chance that I never did check the resonance on that particular build at the time.
It was a lot worse before increasing the gap, before this photo was taken.

Each time the air gap was increased the sine wave improved, that gave me the idea that the uH was too high, there may well have been more going on than first thought at that time.....(still learning).
I plan on revisiting that one later.

Unwound the 2 smaller chokes the other day, the wire is still good, will weld up the end brackets to make a clamp, then get some scrap wire and run a few test's and see how that turns out, they might be better suited to this single transformer.
Cheers Aaron
Off The Grid
 
Warpspeed
Guru

Joined: 09/08/2007
Location: Australia
Posts: 4200
Posted: 09:37am 15 Sep 2021
Copy link to clipboard 
Print this post

That waveform looks like magnetic saturation, not sure what is saturating though.

The only way to know for sure what is going on is to build a version of the choke saturation testing circuit.
Cheers,  Tony.
 
     Page 3 of 3    
Print this page


To reply to this topic, you need to log in.

© JAQ Software 2021