Autotransformer question

Ignoramus28517 fired this volley in news:jK6dnX51o462LrXRnZ2dnUVZ snipped-for-privacy@giganews.com:

Iggy, providing a non-isolated supply to the drives might or might not be harmful depending upon their local grounding. The schematics will tell that.

However, I think you might not know that a transformer must receive a certain minimum excitation voltage to efficiently operate. At roughly

1/4 the input at which it's designed to operate, there's a good chance that core losses are going to go sky-high, resulting in overheating (probably even without any load).

Because switching power supplies have become so inexpensive relative to the cost of new iron, you'd be better off either finding a switcher rated at the output you need, or thrash around on the surplus market for a while to locate a transformer. Running a 460V winding on 125V most likely won't work satisfactorily for you.

LLoyd

That's huge. Almost exactly what I use in my Hardinge CHNC. Don't need isolation, just separate circuit and keep wiring as far as possilbe from rest of control.

here's a tutorial on DC supplies:

Karl

You mean the transformer is huge?

And you think that it will work for the intended application?

I will read that right now.

Thanks

i

Industrial controls are a little different from standard electronics practice, they have added problems with long noisy wire runs, ground loops and some idiot spearing the control box with a forklift, then stepping off into a puddle of coolant.

jsw

Well, at home I may have an idiot, but no forklift.

i

I just buy 2:1 transformers and step 110 down by 1/2 for the A8 series of AMC drives, then a bridge recticifier and an elctrolytic capacitor. My mill uses a 1KVA, that's 1000 watts. I used an old 5 KVA on the hardinge. Divide by voltage to get amps and allow for some loss. I'm sure a EE type will chime in and give the exact formuals.

These things are real simple to build. Even I can do it.

Karl

I meant that as an example of the different rationales for the design rules.

BTW I got to keep the damaged Variacs from a similar forklift incident, and shuffled their parts into a few good ones which I built into power supplies. The cores roll off too fast with frequency to make subwoofer crossovers.

jsw

I think you have this backwards. At 1/4 the input voltage the core losses will be much less than at full voltage.

=20 Dan

" snipped-for-privacy@krl.org" fired this volley in news:ad72df97- snipped-for-privacy@u7g2000yqm.googlegroups.com:

Not below the minimum required excitation, Dan.

LLoyd

I have decided that I have enough problems already and do not want to create any more. I will buy a toroidal 1000 VA 50v transformer at Digikey. It converts 115 VAC into 50 volts, or 123 VAC (my typical voltage) to 57 VAC, which is basically perfect for me. It has isolation and I can wire it to share the same ground as signals, etc. The cost is $168 and I am sure that it will be money well spent.

It produces 20 amps at 50 VAC.

i

Ignoramus28517 fired this volley in news:hKmdnQ4FuLJ5T7XRnZ2dnUVZ snipped-for-privacy@giganews.com:

Indeed. Good on ya'.

LLoyd

You've already bagged the idea, but to answer your question:

The kVA rating of a piece of AC gear is the maximum design RMS voltage (i.e. 220V) times the maximum design RMS current (i.e. 14A), when it's hooked up normally.

The reason that it's a "kVA" rating and not "watts" is because when you are driving reactive loads (i.e. inductive or capacitive) you can have lots of current without any real power being consumed -- yet in an electrical machine, much of the loss (and hence heat build up) is caused not by the power going through it, but by the applied voltage causing core losses, and the current through the coils causing I^2*R losses.

Had you hooked it up backwards as you had planned, you would have had to assume that the wire up to the 380V terminal was lighter than the wire up to the 220V terminal, and derated that 3kVA rating. The safe rating would have been whatever the current would have been out of that 380V terminal -- a PE with experience with that sort of hacking might be able to figure out a higher current rating than that, but would have to take care that the transformer wouldn't experience spot over heating.

To make 20-80VDC for your drive? That should be good. What current do the drives pull? You need to size your caps so that the instantaneous voltage to the drive never falls below 20V -- I calculate from the peak voltage to 20V in 1/120th of a second, assuming maximum line sag (usually 100V if it's a 120V line). You really have more than 120th of a second, but that's an easy number to remember and it's conservative.

A surplus place may have something for less money; since it's a one-off you won't care if it's not a production part. I like Surplus Sales of Nebraska

and Herbach & Rademan I've never had a problem ordering from them. I took a quick look expecting to find joy and they may not have exactly what you want -- but it's not a bad idea to look.

Yes, 20A at 80 VDC. Drives are limited to 20-80 VDC. (they refuse to work if voltage is outside this spec)

15A per drive continuous, 30A per drive "peak".

I will have 20,000 uF.

It does not look so.

A lot of used toroidals (and transformers in general) are sold with bundles of wire stocking out, with no clue as to which are primary, secondary etc.

I will just get one at DigiKey and will get on with my life. Hopefully by the end of this week, I will have a working power supply and will not need to worry about it ever again.

i

Thats pretty much what an OmniTurn CNC lathe uses.

Check on the Left side of the control

Gunner

One could not be a successful Leftwinger without realizing that, in contrast to the popular conception supported by newspapers and mothers of Leftwingers, a goodly number of Leftwingers are not only narrow-minded and dull, but also just stupid. Gunner Asch

If you full-wave rectify and filter it with just a capacitor, it will give you more like 12A at 60-70VDC without overheating.

You could also consider something like this (probably overkill)

Looks good.

i

Why would it not give me 20A?

i