Three phase transformer winding question / General Science and Electronics / Forums

Forums

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

« Previous topic | Next topic »

Three phase transformer winding question

1 2 next

Move Thread

LAN_403

Dr. H.

Wed Apr 22 2015, 09:32AM

Registered Member #931 Joined: Mon Jul 30 2007, 05:25PM
Location: Bulgaria
Posts: 486

Hi guys,

for the mill I am making, the spindle motor I found is a 24V at 150+Amps. So naturally the problem is providing that power. I've found a pre-wound 5kVa 3 phase transformer - but for around 100V output voltage. The secondary will be rewound for the needed voltage/amperage.

The primary is connected in star configuration, the center will be grounded. The question is how to wind the secondary - in delta or in star. I will be using a 3 phase rectifier after that and around 60mF capacitance.

If it is configured in Delta - the current in each of the secondary will decrease (but more windings will be needed). Also the 3rd harmonic is suppressed better. So I think it is the better choice.

Am I missing something fundamental here ? 3 phase is mister for me. Is the star connection better in any way for this application ?

Cheers

Dr. Slack

Wed Apr 22 2015, 11:28AM

Registered Member #72 Joined: Thu Feb 09 2006, 08:29AM
Location: UK St. Albans
Posts: 1659

Star will allow you fewer turns on the secondary for the same output voltage after the rectifier, so may be easier to wind. However you will be using fat wire so this advantage may be moot. The cente-tap used in the star config allows you an easier analysis of how your bridge recitifier is working, so that's the 'obvious' one to go for. It also allows you to use the phases unbalanced for other things, like powering some electronics, again the simpler choice to make. If you connect three star windings wrongly, the load doesn't see the full voltage. If you connect three delta windings wrongly, the transformer smokes. I think star is winning 3-0.

You can be clever in several ways. a) Configure your 3 windings in delta OR star switchable to give you root(3) difference in output voltage. b) Wind a delta set AND a star set, and use a 3 phase bridge rectifier on both. With the right ratio of windings, you have a '12 pulse' recitifier which gives you very little ripple, rather than the '6 pulse' rectifier using one set alone (ordinary single phase gives you a 2 pulse recitifier for comparison). Lower ripple means less heating in your motor, slightly, the difference may not be worth having. See wikipedia rectifiers.

Don't worry too much about harmonic distortion, unless you are selling equipment to a spec. Both delta, and star with no connection to the star point, refuse to allow 3rd and 3N harmonic currents to flow. It's only an issue when a supplier puts a flimsy neutral connection in, as it will carry 'little current' when the balanced load currents subtract in the neutral, and then sees it overheat when all the loads draw a large 3rd harmonic current waveform which adds in the neutral conductor.

Dr. H.

Wed Apr 22 2015, 12:04PM

Registered Member #931 Joined: Mon Jul 30 2007, 05:25PM
Location: Bulgaria
Posts: 486

Hi Dr. Slack,

Thank you for your helpful reply. I liked the point with "no smoke"

Making multiple sets of winding is a bit too much, but I definitely liked this one:

Dr. Slack wrote ...

b) Wind a delta set AND a star set, and use a 3 phase bridge rectifier on both. With the right ratio of windings, you have a '12 pulse' recitifier which gives you very little ripple, rather than the '6 pulse' rectifier using one set alone.

I guess this is due to the 30 degree phase shift when you have a star-delta configuration ?

Dr. Slack wrote ...

Both delta, and star with no connection to the star point, refuse to allow 3rd and 3N harmonic currents to flow. It's only an issue when a supplier puts a flimsy neutral connection in, as it will carry 'little current' when the balanced load currents subtract in the neutral, and then sees it overheat when all the loads draw a large 3rd harmonic current waveform which adds in the neutral conductor.

I couldn't understand the above - so not to have a 3rd harmonic you need to use a thick neutral wire or leave it floating ?!

I was planning to connect the center of the primary star to neutral, and was wondering what to do with the secondary star center ?

Cheers

Dr. Slack

Wed Apr 22 2015, 03:05PM

Registered Member #72 Joined: Thu Feb 09 2006, 08:29AM
Location: UK St. Albans
Posts: 1659

Dr. H. wrote ...

*You* don't have a problem with 3rd harmonic, the supply company does. If you, and the rest of the district, throw too much back at them, then their neutrals could overheat, if the neutrals are sized assuming a reasonable degree of phase balance. So you can pretty much forget about 3rd harmonic as a domestic user. Because it causes the supply company problems, there is often a limit on how much 3rd harmonic commerically sold equipment is allowed to fling back. That's why power factor correction electronics is becoming more prevalent, it's mandatory on loads above, I think, 300W, in many jurisctictions. A capacitor input rectifier, because it only conducts at the peak of the waveform, is good at causing bad 3rd harmonic.

If you don't sell stuff, then who's going to know? Your few amps of 3rd is not going to embarass any supply transformers. You might, out of the goodness of your heart, put a bit of inductance in the phases to improve the power factor by increasing the conduction angle. However, that improves high harmonics first, you need a lot of inductance to make any improvement in 3rd, so it's not really worth it.

You connect the secondary star to (a) the approrpiate point on your output caps. If, as is typical, you generate +/- supplies and have 6 caps, 3+ve, 3-ve, then the neutral star connects to the centre tap of these, the 0v connection . Then, (b) you need to decide what secondary earthing practice you use, where you connect your output 0v to. The transformer provides isolation. You could leave 0v floating, or you could connect it to a ground spike, or to mains earth. All three have their merits, it depends on the particular circumstances which is best to use.

Dr. H.

Thu Apr 23 2015, 07:53AM

Registered Member #931 Joined: Mon Jul 30 2007, 05:25PM
Location: Bulgaria
Posts: 486

Thank you for your nice feedback.

I guess I will go with the star configuration on the secondary.

Best Regards

Dr. Slack

Thu Apr 23 2015, 10:58AM

Registered Member #72 Joined: Thu Feb 09 2006, 08:29AM
Location: UK St. Albans
Posts: 1659

It occurs to me that perhaps I have over-sold the star to 0v configuration. There is quite a lot of ripple on each of the +ve and -ve, each is fed by a 3-pulse rectifier. However, these pulses are anti-phased, and the differential output from +ve to -ve has the much lower 6 pulse ripple.

So the avilability of the centre tap is more of a convenience for other circuitry.

Dr. H.

Thu Apr 23 2015, 01:03PM

Registered Member #931 Joined: Mon Jul 30 2007, 05:25PM
Location: Bulgaria
Posts: 486

The biggest advantage of the Delta configuration is that the phase current decreases (Iph = I(line)/sqrt(3) .... so this is quite a lot for this application

Dr. Slack

Thu Apr 23 2015, 04:04PM

Registered Member #72 Joined: Thu Feb 09 2006, 08:29AM
Location: UK St. Albans
Posts: 1659

Dr. H. wrote ...

The biggest advantage of the Delta configuration is that the phase current decreases (Iph = I(line)/sqrt(3) .... so this is quite a lot for this application

I'm not sure I understand what you're asserting here? You would not be saying that like for like, for the same line input voltage, and rectifier output power, that the delta configuration achieves a factor of 1.732 lower current consumption. That would mean such a collossal difference in efficiency that star would never be used. So it must be something else.

hen918

Thu Apr 23 2015, 04:53PM

Registered Member #11591 Joined: Wed Mar 20 2013, 08:20PM
Location: UK
Posts: 556

Dr. Slack wrote ...

Dr. H. wrote ...

The biggest advantage of the Delta configuration is that the phase current decreases (Iph = I(line)/sqrt(3) .... so this is quite a lot for this application

In star, the windings are put two in series phase to phase with 0V in the middle. In delta, only one winding goes between the phases, and so has a greater volts per turn and greater voltage. If the voltage increases, the current decreases, to the extent that Dr. H described.

Dr. H.

Fri Apr 24 2015, 05:36AM

Registered Member #931 Joined: Mon Jul 30 2007, 05:25PM
Location: Bulgaria
Posts: 486

Hi guys

here is a good explanation of what I mean about the decreased phase current -

I am still looking for opinions what I should use

Best Regards

1 2 next

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