Isolation transformer with a high voltage winding... how to handle it?

[mlefe]

Hi guys, I'm working on a DIY isolation transformer to add to the bench (I got the transformer from a APC Line R 1200 voltage stabilizer ).

After a fair amount of analysis I was able to identify and separate all windings and choose the ones that I could use to have a primary and secondary.

The issue is that, from the way the transformer for the voltage stabilizer is built, if I put 110V on the primary, I can get 110V on the secondary just fine, but I also obtain in one set of windings more than 770V (7 times more than the voltage in the primary/secondary).

I think I can tape/insulate that pair of cables and leave them floating but I'm not really sure if that's safe (the whole thing will be in a box anyways) or if I have to take the "safest" path and completely dissassemble the transformer and remove the cables (and risk damaging the other windings and/or enamel of the core and then cope with the hum)

What do you think? What makes sense?

[retrolefty]

Leaving unused secondary windings disconnected is fine, just dress them carefully with proper voltage rated tape or sleeving.

[ArthurDent]

It sounds like what you're describing is a ferroresonant transformer  that should only be used if the proper capacitor is wired across the other secondary winding. These transformers when properly wired  generally produce a distorted sine wave output and a stronger external magnetic field than a conventional transformer although with the proper capacitor they will provide a stable voltage output.   

http://www.electroncoil.com/ferroresonant_transformers.php

[David Hess]

The APC Line-R 1200 uses a buck/boost autotransformer and not a ferroresonant transformer but I have no idea where a 700 volt AC winding fits with that.  There should be a bunch of low voltage windings or taps.

I suspect it has a dual winding 120/240 volt primary so the same transformer can be used at 120 and 240 volts.  The dual winding can be used as a 120 volt isolation transformer at the cost of increased copper losses so limit power to 600 watts continuous.  Isolation will not be as good as with an isolation transformer which has physically separate primary and secondary and there will be no electrostatic screen between the windings.

[ArthurDent]

The APC Line-R 1200 has a 3 position switch to set it for the line voltage in your area (110,120,127) but after you set it for your region, it is an automatic line regulator for 85-140 in a 120volt region. Here is a link to the manual that describes that.

http://www.apc.com/salestools/ASTE-6Z7V37/ASTE-6Z7V37_R1_EN.pdf

[thm_w]

It seems to just have a few buck/boost taps, based on the Vin vs Vout graph: http://stop-tarif.ru/shema-apc-line-r-1200.php

So is OP connecting 120V across the ~10V taps? That would explain where the 700V is coming from.
From what I see from example buck/boost transformer schematics there is only one high voltage winding and a number of low voltage windings.

I don't think this transformer will work for what OP wants to do.

[David Hess]

So is OP connecting 120V across the ~10V taps?

If that happened the transformer would saturate and draw massive current.

[ArthurDent]

It would appear from the link in post #5 that there is a 3 position switch positions to center the input voltage depending on your region. Then there is a sensing circuit that controls some relays that switch other taps so with an input range of about 170-260 (in this Russian model), the output will stay within 215-240. From the graph it appears that on the first step when the output voltage rises to 240 volts the relays switch taps and the output will drop back to 215 volts. If the input voltage continues to rise to 240 volts then the second relay will switch to drop the output back to 215 volts again. 

Here is a Google translation of some of the Russian manual. "I understand that this stabilizer honestly fulfills the claimed hackatery, now in the network in the stabilizer switched to a lower voltage and at the output of it in. Increase the switching speed of transistors SR. And the imported equipment is considered normal. The device is budgetary, the stabilizer is APC Line-R, everything is corny - autotransformer whose windings are switched by relay, relays are switched on with transistor keys, the keys are controlled by the compiler."

This does seem to agree with the David Hess theory about an autotransformer but if this is correct then there would be no 770 volt winding described in the first post. If it is a regular transformer then it is probably a single primary with taps marked for the 3 voltage ranges depending on region, then a low voltage center tapped tapped secondary that is relay switched in phase for the 2 boost steps and switched out of phase to lower the output. This would allow a smaller transformer to be used but there would be no isolation because this would be an autotransformer.

Where there are 2 equal voltage windings and the high voltage winding I would still wonder if it were a simpler ferroresonant transformer. The English manual I linked to states that the regulator should not be used with A.C. motors which wouldn't be a problem with a switched autotransformer but would be with a ferroresonant transformer because of the distorted waveform.

A photo of the actual transformer in question would help.

[mlefe]

I'm putting here two photos of the transformer, please consider that it's not how it was originally setup: I've separated some of the wires (I explain why down below).

The way I understood is like this: originally, the transformer has the primary neutral connected to the secondary neutral and then you have 3 pairs of windings that will generate 1/7 (+15V) of the main voltage (+110V).

I believe this exists because, depending on how you use your relays, you can use the first two pairs of windings (let's call them A and B) to increase the voltage by 1/7 or decrease it by 1/7 so you can regulate if there's too much voltage or too little coming in (thus stabilizing the secondary output).

The third winding (let's call it C and also with a 1/7 ratio) is used for the electronics part, so it gets rectified, filtered and used for the microprocessor, comparator, controlling the relays, etc.

What I did was identify the neutral wires and separate them (so I can have isolation) and then used the two first pairs (A and B) as my isolation transformer's primary and secondary --> I've put 6VAC on one of the pairs, resistors on the other pair (to limit the current and ensure the voltage is not ghost) and got around 5.7VAC in both pairs of windings (isolated). The third winding also get's 110V but I'm simply ignoring it.

The problem is that the original winding (that in the original configuration was the primary at 110V) has 7 times more voltage because simply of the number of turns --> I get around 40V on it when I put 6V on one of the other pairs.

If I put the whole thing on 110V, this will generate 770V in that winding.

As you can see from the picture, the wires are beefy and most of them are doubled, so I don't see any issues to handle big currents, but perhaps I'm missing something: what do you think?

[mlefe]

I believe it's not a ferroresonant (I've uploaded pictures, perhaps you can identify it properly) --> my guess is what thm_w and David mentioned (I've tapped the lower voltage's windings)

[mlefe]

Can you help me understand why it saturates it? Can you point me at where I can learn a bit more about this effect? Putting limiting resistors make sense or the whole thing won't work no matter what I do? (I'm sure I haven't seen the effect because I've used a very low 6VAC to test the thing and never connected it to mains)

[mlefe]

It seems to just have a few buck/boost taps, based on the Vin vs Vout graph: http://stop-tarif.ru/shema-apc-line-r-1200.php

So is OP connecting 120V across the ~10V taps? That would explain where the 700V is coming from.
From what I see from example buck/boost transformer schematics there is only one high voltage winding and a number of low voltage windings.

I don't think this transformer will work for what OP wants to do.

I believe this version works in a similar way, but has some differences in the components (I see some things populated in mine that I'm not seeing in the pictures mentioned in post)

I'm uploading a picture of my version, in particular, there's a big white cap where in the link shows a smaller yellow one.

[tkamiya]

I'm sure you will be encasing the whole setup in a suitable insulated case.  If that is the case, what you propose is safe.  Just make sure you use materials that can withstand the voltage to isolate that high voltage wires.

Having said that, I'm not sure if it's safe for YOU.  US being tinklers, I would imagine you'd be opening that box from time to time.  Possibly in a hurry and mind elsewhere.  You'd be now faced with that high voltage.  Perhaps you've forgotten it's there by this happens?  By looks of it, the transformer does not appear to be that large.  At least in US, isolation transformers are cheap and plentiful.  (smaller ones are....)

I had potentially fatal high voltage encounters before.  I never thought I would but I did. 

I think, if I were in your position, I'd just get a proper transformer, a larger one, and construct the isolation box.  Then save that 700v transformer when high voltage is necessary.

[ArthurDent]

Here's my theory #2. The 770 volts you assumed never made much sense no matter what the transformer type was. You were testing the transformer properly to get ratios but you weren't sure which winding was which. What I 'think' now is explained in the simple schematic below. What you believed was a 770 volt winding is actually the 110 (or 120) volt winding. This makes the 2 identical windings about 16 volts each and these were used to power the circuitry and relays. The heavy leads were low voltage secondary windings that would either subtract or add to the line voltage wired in series through the relays, depending on the phasing of the windings. thm_w was close in their explanation but you were only doing low voltage ratios not putting line voltage on a low voltage winding.

Check this out and see if it makes sense and see if you can power what you thought was the 770 volt winding as the primary through a 40 watt bulb for protection and if it doesn't light up, this is probably correct. If it is correct, this cannot be used as an isolation transformer but could be used like a buck/boost autotransformer.

[Jwillis]

Are you sure that isn't an auto transformer.Since the secondary shares the neutral of the primary and can be a single winding with taps. that means that all wires are the same gauge . If you run your voltage into secondary the primary will become very high voltage and other secondaries can be the same as the input voltage.If it is an auto transformer it won't be very good for isolation.
Check continuity between taps .If you get continuity between suspected secondaries and suspected primaries then you probably have an auto transformer.

[mlefe]

I'm pretty sure that it was wired as an autotransformer, but I've separated the neutrals... so I guess it became isolated?
I tend to think that it's like Arthur is saying (the 770V is the original 110V) although I still haven't put together the test to see if the winding for the new primary/secondary will work or simply doesn't have enough turns to avoid saturation (I've been reading a bit on this)

What I can say is that I can measure the resistance and inductance because, believe it or not, I've purchased a functional HP 4261A for only 17USD!!

I must learn how to use it, of course, but I believe that knowing the inductance should help to see if it saturates with 110V? (perhaps I can put a rough model in LTSpice?)

I still didn't create the test with the light bulb because I wanted to create a box for the variac I've also purchased and use that to slowly ramp-up the voltage while measuring the current... also I don't have a lightbulb (can you imagine that? I have a 4261A and don't have knowledge nor light bulbs... I know, life is very unfair sometimes!)

[mlefe]

Hi guys, as you have suspected, this was a transformer with windings that were only able to provide less voltage than what the primary has (+-10%) and thus, once I tried to energize those windings with my variac, quickly saturated the core and the current went to the roof. I've put some different resistances in series but in the end, I've realized that they will have to be huge and dissipate lots of power to have a usable 5A transformer.
In any case, despite the failure I've learned a lot and that's what this is all about for me 

Thank you for your comments!