Electronics > Projects, Designs, and Technical Stuff

Noob plays with transformers

<< < (4/8) > >>

ArthurDent:
Years ago, when I was a kid, old radios that were thrown out were tube type with power transformers and there was a rating tag on the back so you knew how many watts the transformer was good for. Almost all these radios had tubes with 6.3 volt filaments except for the rectifiers which were generally 5.0 volt filaments. (The really old TVs had much larger transformers as did some industrial test equipment.) If I took all the ‘E’s and ‘I’s out of the transformers then unwound the low voltage secondary filament windings and counted the turns I could calculate the turns per volt needed when I rewound my windings on the existing primary.  The high voltage windings could be unwound without counting because I already had the turns per volts from the low voltage windings I removed so I didn’t need to count a few hundred more turns of wire.

I would leave the insulating stiff paper on top of the primary winding and also save any other insulating paper to reuse. I would anchor the start of my new winding with sturdy fabric tape and evenly wind the required number of turns for my new secondary, generally putting insulating paper between layers, just in case. The wire size I needed could be calculated from knowing how much total current the tube filaments used and that could be found from looking up all the tubes used in a tube manual. Using a wire gauge I could measure the wire size and look up the circular mils for that wire size in a wire table then calculate the number of circular mils per amp. With this information I could use the wire table to calculate the wire size I would need for my winding. Knowing the wattage of the transformer from the rating tag on the radio would tell me if the transformer was large enough for my needs. For instance, if the tag said 50 watts and I wound my secondary carefully I should be able to wind a 25 volt 2 amp winding on the existing primary. 

There was a 16 volume set of ‘Popular Mechanics Do-It-Yourself Encyclopedia’ from the 1960s or 1980s I had that went into detail on winding transformers that I used back then. There is one reference I found on line that probably goes into more detail than you need at:
 
https://www.electronicdesign.com/power/build-your-own-transformer

but is pretty good. As to the rust on the ‘E’s and ‘I’s you had, I would have just brushed the loose flakes off because as others have mentioned, polishing them could increase eddy currents and heating. Generally the better transformers alternate the ‘E’s and ‘I’s rather than put them in clumps. Some transformers do put all the ‘E’s in one side then lay the ‘I’s across the top and weld where they join but I believe that makes for a higher watts-loss design even though it is cheaper to assemble that way.   

rebelrider.mike:
Wow, lots of responses, thanks for the interest folks!

So getting the core sheets off of the spool. The Is popped out pretty easy. I did manage to tear up a couple of the Es. I've hammered them back into shape as best I could, but I don't know if they'll be reusable. As for cleaning, I've been using a scotch bright wheel on a grinder. Takes the rust and old insulation right off without damaging the iron underneath.

As for trying to guess the number of primary turns without math. Once the basic math is learned, it's super easy to run the calculation and get very close to the actual number of turns I'll need. In my humble opinion, way faster than trial and error. Using a light bulb in series is a very good idea though. Keeps a dead short from happening if there are not enough turns on the primary, or if there is a short in the winding. Without some kind of short, the light bulb will act as an ordinary resistor, and not much will happen. If a short occurs, rather than melting something, tripping the breaker, or some other annoyance, the light bulb will simply light up as a short would provide sufficient current to light it up. It's also a convenient signal that something's wrong.

Isolating the secondary from the primary winding is the difference between a "regular" transformer and an autotransformer. The windings share the core, but they are not physically connected to each other. I have no idea what creepage distance is.

As far as re-insulating the plates, I'm not sure what to use yet. I was thinking either some kind of urethane, or maybe trying to make a black oxide layer. I will be alternating the Is and Es as I put it back together. I only have 40 plates left to do out of 194. So I'm going to have to figure this out soon! :)

And thanks everyone for the links. I've read them all. In fact, the "Build Your Own Transformer" article from ElectronicDesign.com is one I'm already familiar with.

I'm pretty convinced now that the power rating of a transformer is only limited by the maximum current that can be run through the wires. And that is more a limitation of the insulation getting too hot than anything else. So yesterday, I did a little experiment. I took a meter of the aluminum winding wire and ran a current through it with a CV CC power supply. One of those really low cost ones that can be found on Amazon: https://smile.amazon.com/KNACRO-Converter-Regulator-Constant-Voltmeter/dp/B01N7CG71M?keywords=CV+CC+power+supply&qid=1539923210&sr=8-38&ref=sr_1_38

Anyway, I slowly turned up the Amps to see when the wire would get hot. I got all the way up to 4.5A before giving up. My little power supply was getting very hot (it has a limit of 5A, and I didn't want to push it) and the wire was just barely maybe getting warmish. I think it may get a little warmer all wound up tight in a transformer, but I'm pretty sure I can get 540W out of it at 120V and be ok. The transformer was originally rated at 1000VA with the same wire. Also, I don't plan to use anywhere near that much power.

Another thing to keep in mind is that the transformer was originally wound for both 110V and 220V, and I'm only winding it for 120V. So I should have plenty of room for larger gauge wire if I decide I need more Amps in the future.

I'm curious about the different winding methods people use. Most DIY transformers I've seen are the same as what I'm doing. Wind the primary closest to the core and wind the secondary on top of the primary. But I've also seen other people wind the primary on the bottom half of the spool, and the secondary on the top half. I've also seen people wind half the primary, then wind the secondary, and then wind the other half of the primary. Like a sandwich. I wonder what the pros and cons are to doing it each way?

bsfeechannel:

--- Quote from: rebelrider.mike on October 19, 2018, 04:56:09 am ---I'm curious about the different winding methods people use. Most DIY transformers I've seen are the same as what I'm doing. Wind the primary closest to the core and wind the secondary on top of the primary. But I've also seen other people wind the primary on the bottom half of the spool, and the secondary on the top half. I've also seen people wind half the primary, then wind the secondary, and then wind the other half of the primary. Like a sandwich. I wonder what the pros and cons are to doing it each way?

--- End quote ---

Normally you start with the winding that has the thinnest wires. In a step-down xformer the primary wire is the thinnest. A split bobbin provides better insulation between primary and secondary and lower capacitive coupling. The "sandwich" winding (technically called interleaving) is normally used in output transformers for tube amps. That way you reduce the leakage inductance, which is responsible for limiting the high frequency response of the transformer.

The Electrician:
You probably shouldn't have polished the laminations.  Transformer steel is usually coated with a very thin layer of magnesium silicate, called Carlite.
See: http://nikomag-europe.com/index.php/markets/grain-oriented-electrical-steel

Here's another good read: https://en.wikipedia.org/wiki/Electrical_steel

The C5 coating is Carlite.  The minimum coating (C0) is also called steam oxide; they run the steel through a furnace with exposure to steam.

Also see: https://www.quora.com/How-does-CRGO-steel-reduce-core-loss-in-transformers

You might be better off if you took all the laminations outside, sprayed them with water, and left them for a day or so until they get a little rusty.  :)

Varnishing them will give you a coating that will be so thick that you will not have a good stacking factor--you probably won't be able to get all the laminations back in.

BradC:

--- Quote from: xavier60 on October 17, 2018, 05:59:52 am ---Cleaning the core laminations will increase eddy current losses, by how much, I don't know.

--- End quote ---

As clean as those are, quite a *lot*. As indicated by The Electrician in the post above, the insulating layer on a core is extremely thin. It's also extremely important. Transformers do not like shorted turns.

I'd wager you'll never get an insulating coating thin enough to get them all back in and not have a shorted turn with any form of applied synthetic coating. If you manage to get a decent oxide coating on there that'll do the job, however I will be seriously impressed. So please don't take this as a "toss it and start over". Just keep us posted on what you are trying and how it's working out. It's all good learning.

Navigation

[0] Message Index

[#] Next page

[*] Previous page

There was an error while thanking
Thanking...
Go to full version
Powered by SMFPacks Advanced Attachments Uploader Mod