So I completely unwound 12awg secondary (160 turns which has linear dimension of abt 40 meters) from 1:1 toroidal line isolation transformer as first step in process of rewinding secondary as 80 turns of two paralleled wires for step-down configuration (120V, 12A Pri / 60V, 24A Sec). (Now @Hypatia's Protege and everybody else thinking how stupid I was plz read disclaimer at end of post!)

So I know enamel insulation is vry good quality cuz it didn't crack or flake or degrade at all when I unwound secondary unto 1 meter _shuttle stick_ but I still need 2b vry careful not to fatigue Cu wire!

So my plan is to fold wire in half, permanently connect loose ends and wind it back on transformer (not really as _bifilar_ winding just two electrically and physically paralleled wires for 2X current capacity).

So needless to say winding 80 turn secondary from 20 meter long pair of 12 gauge wires is going to be a PITA no matter what but it can be a lot easier if I _lightly_ twist pair together first.

So here's question: Since twisting wires together forms helix (so axis of wire will be at slight angle to axis of core) I'm worried abt coupling problems cuz just thinking on it, coupling decreases from max (when wire is parallel with axis of core) to theoretically uncoupled at angle of π/2. So calculating effect on coupling is just simple trig based on angular relationship of core and wire axes which I totally get. But I just don't have intuitive sense and so no practical grasp of how significant the effect of a slight curl (by which I mean just enough to make pair manageable as single wire) will be to coupling?

Also I know higher freq xfmrs are sometimes wound with Litz wire (which is also formed of twisted insulated strands) but at high freqs decreased coupling is prolly way more than compensated by reduction of skin and proximity effect losses. So there's no net benefit for me at 60Hz, just loss!

So thanks in advance for help (which includes explaining errors in my reasoning if I'm missing something)!

Here's disclaimer (sry if it's too clunky):
Just so HP and other ppl don't think I'm as dumb as unwinding secondary sounds I need to explain that I TOTALLY KNOW usually much easier way of what I'm doing is avoiding un/rewinding by just finding electrical center of secondary by exciting primary (with low voltage for safety) then with o'scope set to display sum of two vertical channel inputs connected to opposite ends of secondary, just probe for center (using like _half-interval search_ pattern) with lead from scope common (so o/c electrical center is point for 0V flat-trace display cuz of π phase relationship between halves of secondary). So anyhow once center is found, winding can be cut to two secondaries which can be electrically paralleled to halve secondary turns and resistance. (So o/c u could also just add center tap but I don't like taped FW rectification for vry good reasons!)

Anyhow easy way wasn't option cuz transformer was like _scramble_ wound (so turns had vry poorly defined _sort order_ so searching would mean taking way too many samples damaging insulation) also cuz of multiple layers, center was prolly buried in winding anyhow!

 

Wouldn't it be best if you were to replace the #12 ga wire with, say, a #10 ga instead? It'd probably be easier to wind up, and less cumbersome.

 

Reduction in coupling will be pretty much non existent. Almost all coupling in a toroid happens inside the hole. You can take a single conductor, put it through the hole, largely without any regard for its position therein, and couple quite effectively (consider clamp on current meters with large openings in the clamp - the clamp is the core of a topological toroid and the secondary winding is all bunched up at one place).

You will find getting a nice uniform twist of twenty metres of wire to be a challenge, especially if it isn't very nicely straight to begin with.

That is a lot of turns and it will be a bit tedious. I have hand wound powdered iron and MPP toroids with an inner diameter of about an inch with 11 AWG magnet wire. It isn't fun, but it is doable.

 

One help is to wind the wire around a length of ruler ot plastic pipe or..
Then you can thread it through the toroid like a needle, and take a turn or two off the needle as you go. That helps not having to pull loads of wire through the core each time. Make sure you don't kink the wire around the needle.
There is no need to twist the wires together, in fact I would say it is better not to. Then you lesson the risk of wires rubbing through the insulation as they cross.

 

if you were to replace the #12 ga wire with, say, a #10 ga instead?

But thicker dia!
14g is rated at 24a for single conductor, 12g 34a for single.
https://www.engineeringtoolbox.com/wire-gauges-d_419.html
Max.

 

Kind friends:

@Aleph(0) wished me to reply to responses to this thread in her absence (exactly how one might be involuntarily 'absent' from the 'Web' in this day and age is a question best left to her)

But to continue -- Inasmuch as @Aleph(0) is well aware of my stance upon vandalism ahem... removal of the entire secondary --- (IMNSHO cutting the conductor and un-winding less than ten turns near center, scoping out the 'mid point' followed by trivial rewinding, etc... should have 'done the trick' at considerably less effort!)---

Taken with the fact that your responses would appear to fully and competently address her concerns--- I leave the replies to her (please expect her 'return' Ca. Mon 16 Jul)

Best regards and many thanks
HP

 

Big time thanks for replies! Sry for absence and delayed response Also just so u know I read responses everyday! It's that I can't logon from some locations! (Which is NOT an AAC problem at all! Just my security paranoia about some ISPs)

Wouldn't it be best if you were to replace the #12 ga wire with, say, a #10 ga instead? It'd probably be easier to wind up, and less cumbersome.

Cmartinez ur totally correct that it would be way easier to just CUT old secondary off core and rewind it with heaver gauge wire! But our surveys keep telling us most of our readers are total skinflints with hobbies and independent study and like that So anyhow we try to help them _make do_ with materials on hand

So anyhow Here's my progress so far:
◊ Unwound old secondary with 1 meter lath shuttle.
◊ Drew wire out full length (40m+).
◊ Groomed (by which I mean _de-kinked_ and straightened) it with DIY PVC _die_
◊Folded it in half and drew pair taught and taped it to itself at 15cm intervals.

So end result is cable made from two totally parallel wires half length of original secondary

Reduction in coupling will be pretty much non existent. Almost all coupling in a toroid happens inside the hole. You can take a single conductor, put it through the hole, largely without any regard for its position therein, and couple quite effectively (consider clamp on current meters with large openings in the clamp - the clamp is the core of a topological toroid and the secondary winding is all bunched up at one place).

ebp That's vry interesting and totally correct cuz I tested idea with just fluke 337 on 800A bus rod and like you say there's no real difference in reading from axial to like 70° (which is just physical limit of clamp to bus rotation). So I totally admit I don't get it? So anyhow I know I have a lot to learn abt magnetics!

You will find getting a nice uniform twist of twenty metres of wire to be a challenge

There is no need to twist the wires together, in fact I would say it is better not to. Then you lesson the risk of wires rubbing through the insulation as they cross.

So ideas from those two quotes taken together are good enough reason NOT to use twist at all! Cuz paralleled wires in electrical contact (cuz of insulation damage) at same point won't cause inefficiency (cuz phase and voltage would be equal) but twisted pair aren't guaranteed 2b totally electrically parallel and track exact same distance from their ends! So HUGE tnx for heads up on that!

One help is to wind the wire around a length of ruler ot plastic pipe or..
Then you can thread it through the toroid like a needle, and take a turn or two off the needle as you go. That helps not having to pull loads of wire through the core each time.

Dendad tnx! So that's exactly how I unwound it (_needle_ is what I meant by _shuttle_ in OP) but problem I'm anticipating is that keeping fully wound shuttle girth small enough to fit though core aperture (which is initially abt 7cm) means minimum shuttle length has 2b abt 1 meter which is awkward cuz each turn length is only abt 26cm So hopefully I'm just missing something that makes long shuttle not a major problem?

But thicker dia!
14g is rated at 24a for single conductor, 12g 34a for single.
https://www.engineeringtoolbox.com/wire-gauges-d_419.html
Max.

Maxheadroom Tnx so cuz of exponential relation of AWG sizes, 2 parallel 12awg conductors are equivalent to single 9 AWG (which I know is non-standard size) which is perfect margin for abuse in hands of newbies and just students in general

Inasmuch as @Aleph(0) is well aware of my stance upon vandalism ahem... removal of the entire secondary --- (IMNSHO cutting the conductor and un-winding less than ten turns near center, scoping out the 'mid point' followed by trivial rewinding, etc... should have 'done the trick' at considerably less effort!)---

HP I say you need to relax! Being totally serious if I had it to do over again I'd SO follow your advice! But since scramble-wound windings are a total mess I'd prolly be right where I am anyhow and it's for sure a lot of ppl following projects will have same problem! So I say knowing technique for totally rewinding secondary is vry good idea!

So anyhow summarizing, here's new question:

Now that I have a straightened and taped pair (which can be physically and electrically thought on as single wire), how best to wind it on core? Cuz it just seems like I'm going to have problems maintaining tension with 1m shuttle and nasty fact that core can't be clamped in vise (cuz of damage to primary)?

So huge tnx for above replies and also major advance tnx for any further help!

 

HP I say you need to relax! Being totally serious if I had it to do over again I'd SO follow your advice! But since scramble-wound windings are a total mess I'd prolly be right where I am anyhow and it's for sure a lot of ppl following projects will have same problem! So I say knowing technique for totally rewinding secondary is vry good idea!

Fair enough!

Now that I have a straightened and taped pair (which can be physically and electrically thought on as single wire), how best to wind it on core? Cuz it just seems like I'm going to have problems maintaining tension with 1m shuttle and nasty fact that core can't be clamped in vise (cuz of damage to primary)?

FWIW The most satisfactory approach I've yet encountered is the 'gated hoop' and 'slider' method -- Having locked the hoop about the core, the former is wound with the requisite lengith of conductor then 'unwrapped' to the core (à la Swant, Vergara, Waterfield et al...)

As an aid to comprehension - presented below is an industrial 'Swant-esk' winder in action...

Of course low-voulme (e.g. avocational) 'production' may adaquately be acomplished via a manually minuipulated (do pardon the redundancy) gated hoop, and, I hasten to add, no small amount of patience!

Very Best regards and please enjoy the remainder of your 'vacation'! -- Heaven knows you've earned it!
HP

 

In another post: https://forum.allaboutcircuits.com/...-and-construction.113504/page-69#post-1248540

HP described the capability of the low voltage power supply: "...the circuit under discussion being, as it is, after the fashion of a 'battery eliminator' continuously adjustable from 0 Vdc-50 Vdc @30A)."

The transformer she was proposing to use had a (unmodified) rating of 16 amps, whereas the transformer you're discussing here is only 12 amps. Has the current requirement diminished, or are you just practicing rewinding with what you had on hand?

I think there is a problem looming. I haven't seen any discussion of the need to derate a transformer used in rectifier service. I discuss the details of the need for derating here: https://forum.allaboutcircuits.com/threads/derating-transformers-for-rectifier-use.68839/

The transformer example in that thread was a small transformer typical of one a hobbyist might use, about 35 VA rated. In that case the derating factor was about 1.6 which is a rule of thumb value often seen in hobbyist power supply projects.

Unfortunately, for larger transformers, and especially toroidal transformers, the derating factor is larger. For a 1.92 kVA transformer like HP was using, or 1.44 kVA as in this thread, the derating factor can easily exceed 2.

This means that the transformer in this thread, with its secondary rewound to a 24 amp rating, would only be able to supply 12 amps or less DC at the filtered rectifier output when the secondary current was 24 amps RMS. A (rewound) secondary current of 24 amps RMS would heat the transformer to its maximum allowable temperature rise above ambient; to supply 24 amps of filtered DC would seriously overheat the transformer.

Similarly, using HP's transformer one could only get a DC output of 16 amps or less.

 

@The Electrician

Well hey! -- Many thanks for your kind input and observations! Please know such are very much appreciated!

The transformer she was proposing to use had a (unmodified) rating of 16 amps, whereas the transformer you're discussing here is only 12 amps. Has the current requirement diminished, or are you just practicing rewinding with what you had on hand?

That'd be down to me (my bad) - Sincere apologies for the ambiguity -- FWIW, had I my 'druthers', I'd strongly advise connection of two 16A PowerVar (120V isolation) transformers sans modification for 0-120V @ 32A continuous output (pre-rectification) -- While 'series' connection of the primaries (for 240V input) is, of course, optimal - 'parallel' primary connection (for 120V input) may be acceptable where a dedicated 20A 'branch circuit' is available (or, alternatively, via observation of appropriate current limitations).

While such an arrangement is with several advantages (i.e. greater output EMF range, compatibility/ready adaptability with/to most residential electrical service 'parameters' (i.e. 120V and 240V - note also that said transformers are specified at full continuous output over the common line frequency range [i.e. 50 to 60 Hertz]) --- to say nothing of obviation of the unutterably irksome task of modification) --- Alas! It seems a mere $1,200 ($600 per PowerVar unit) is more than the over-taut 'purse strings' of many experimenters/hobbyists will bear! -- Hence our attempt to 'meet them half way' via modification of a single 12A PowerVar transformer (transformer cost ≈ $200) -- That said, I do not advise 'R&Ring' the entire secondary! --- Methinks Aleph's penchant for occasional bouts of impatience-spurred-impulsiveness (can't imagine how she came by that!?) - has come home to roost, as it were

I think there is a problem looming. I haven't seen any discussion of the need to derate a transformer used in rectifier service. I discuss the details of the need for derating here: https://forum.allaboutcircuits.com/threads/derating-transformers-for-rectifier-use.68839/

The transformer example in that thread was a small transformer typical of one a hobbyist might use, about 35 VA rated. In that case the derating factor was about 1.6 which is a rule of thumb value often seen in hobbyist power supply projects.

Unfortunately, for larger transformers, and especially toroidal transformers, the derating factor is larger. For a 1.92 kVA transformer like HP was using, or 1.44 kVA as in this thread, the derating factor can easily exceed 2.

This means that the transformer in this thread, with its secondary rewound to a 24 amp rating, would only be able to supply 12 amps or less DC at the filtered rectifier output when the secondary current was 24 amps RMS. A (rewound) secondary current of 24 amps RMS would heat the transformer to its maximum allowable temperature rise above ambient; to supply 24 amps of filtered DC would seriously overheat the transformer.

Similarly, using HP's transformer one could only get a DC output of 16 amps or less.

Thanks for that -- Indeed we are aware of said considerations (though [Re: de-rating] I had thought perhaps 30% -- but fully 50%! Ay yai yai!) - That said, the (relatively) low 'current-demand duty cycle' nature of our intended applications (loads) may represent a bit of a 'saving grace' on that 'score' -- Even so - your point is well taken! -- I too am dubious of the safe continuous output current restrictions inherent to the modified 12A transformer -- seems keeping costs 'hobbyist friendly' may require oil immersion cooling, etc... Many thanks for the heads up!

FWIW Please be advised that @Aleph(0) anticipates return to 'on-line activity' Ca. mid-July -- wire-blistered fingers notwithstanding

Very best regards and many sincere thanks!
HP

 

Can you give a link to descriptions and prices of the PowerVar transformers; a vendor perhaps?

 

I use Antek, he manuf. and sells Toroidal Txr's, also on Ebay under Antec-inc.
http://www.antekinc.com/.
Max.

 

I use Antek, he manuf. and sells Toroidal Txr's, also on Ebay under Antec-inc.
http://www.antekinc.com/.
Max.

You have solved HP's transformer problem!

 

I use Antek, he manuf. and sells Toroidal Txr's, also on Ebay under Antec-inc.
http://www.antekinc.com/.
Max.

I also have bought from him over the years. Can't get a better transformer for the price. Not sure but think he will wind what is needed and the price drops with the number ordered. The set up is what costs, like in any mechanical/machine operation.

 

Thank you for your continued interest!

Can you give a link to descriptions and prices of the PowerVar transformers; a vendor perhaps?

FWIW I've always purchased the units as a whole... Come to that I'm uncertain the transformers are available for individual sale?...

Here's a link to the product line...


Below are some (potentially) helpful images of the devices in question:

Post continued after images...

External, frontal view 12 and 16 Ampere conditioners (external appearance is identical save for the fact that the 16A unit is equipped with a 20A power-cord plug)

Internal view of the 16A device (notably larger and heaver transformer by comparison with the 12A unit)

Detail of the 16A conditioner's transformer...

Spec tag on the 16A units -- Oh my! Seems there's no mention of 50Hz operation after all!? --- Jeeeze! By the time we've done 'de-rating' I have to wonder if we'll be up to 'pen-light' service

FWIW -- IMO very, very little (save for comparison to the 16A transformers) pictured below is @Aleph(0) 's 'faux pas' she's so kindly placed in my hands for a 'solution' while she larks about the West Country...
Front to back:
1) Segment of the (shuttled) 'prepared' secondary conductor.
2) US $20 banknote (perspective).
3) The 12A transformer core (inner winding intact).
4) Outer winding's bunting...

Note that the outer winding was employed as the primary (and vice versa) in the original application...

I use Antek, he manuf. and sells Toroidal Txr's, also on Ebay under Antec-inc.

I also have bought from him over the years. Can't get a better transformer for the price. Not sure but think he will wind what is needed and the price drops with the number ordered. The set up is what costs, like in any mechanical/machine operation.

You have solved HP's transformer problem

Thanks! -- Will check it out

@The Electrician
@MaxHeadRoom
@shortbus

Am I to understand that --in your considered opinions-- you feel Antek is in PowerVar's 'league' as regards the 'balance' of quality, cost and capacity? -- Please be advised this is an absolutely sincere question sans prejudice or presumption of any sort! - I have no experience with Antec whatever -- the 'take away' from my experience with PowerVar is merely that they significantly and reliably outperform their ratings (i.e. tolerate significant protracted overload conditions sans overheating, etc...) in their intended application (line isolation transformers)...

Many thanks and very best regards
HP

 

@The Electrician
@MaxHeadRoom
@shortbus

Am I to understand that --in your considered opinions-- you feel Antek is in PowerVar's 'league' as regards the 'balance' of quality, cost and capacity? -- Please be advised this is an absolutely sincere question sans prejudice or presumption of any sort! - I have no experience with Antec whatever -- the 'take away' from my experience with PowerVar is merely that they significantly and reliably outperform their ratings (i.e. tolerate significant protracted overload conditions sans overheating, etc...) in their intended application (line isolation transformers)...

Many thanks and very best regards
HP

I can't speak from actual experience with these transformers, but they look like they are built well with standard methods.

The price is unbeatable and you can get the turns ratio you need.

The manufacturer says in the description: "These transformers have heavier gauge wires then the normal requirement to avoid the copper lost during the full power output. The dielectric test is more than 3500V in between primary and secondary coils. Please see the test data for short circuit and open circuit. In most of the cases, this transformer can be output 20% more power from its rating at 60Hz power source without any problem."

MaxHeadRoom and shortbus apparently have used the actual units, so let's see what they say.

 

Never had any problems, even stuck a small overwind on for a low voltage low current supply in a pinch.
Max.

 

I just used them as power for stepper motors. So can't really comment past that. But will say they look as good or better than the ones HP showed in those units. They are a very professional company.

 

Greetz!
HP wants everyone to know she's ecstatic about (Yeah.. Well.. Very pleased with) Antek (as are the entire development team including Julie and I)! @HP will be on this weekend to tell you all about it!

But will say they look as good or better than the ones HP showed in those units.

Sacrilege! Sorry, @HP! But for all their "MIL-SPEC", "Medical-Grade" "wonderfulness", I'm afraid Powervar transformers' haphazard winding patterns look downright sloppy! Yeah, well.. I dig it! It doesn't really matter! Try not to let my iconoclasm bother you! That beauty should be more than cabinet and functionality deep is just my "non-sloanie-ism" talking!

Thanks!

 

That beauty should be more that cabinet and functionality deep is just my "non-sloanie-ism" talking!

you're not alone in that respect ... personally, I find beauty a prime mover of human nature ... cheers!