Modifying transformer

lol, now I see why the welded it... anyone know?

If it's already close to saturation, removing primary turns could be dramatic. 30/24 = 1.25, and few transformers are designed 25% away from saturation.

Saturation isn't an absolute event, but losses climb radically as the core is driven harder. Smoking varnish is a reasonable expectation.

John

If I remember correctly it was only pulling 1A before without any secondary load. Now its about 2.5A and I've removed about 5 turns. That doesn't seem right but thats my readings. I think the extra current is because the core isn't welded so there is a great more deal of leakage. The voltage only went from 24.2 to 25.7VAC so I'd have to remove a lot more.

I'll remove some more turn to get me to about 27VAC and then weld it shut again and then try and run it through some stress tests. Unfortunately I can't easily add to the primary because of the huge wire it uses(I don't have any even close to it. I guess its about AWG 16-18 or something.

I suppose if I could add to the secondary that I wouldn't have any of these issues? Would it even reduce the saturation if I did? (I could always rewind the bobbins if I got the wire. I don't need the 15A the transformer is rated for but just about 8 or would be perfectly fine.

Thanks, Jon

It will be very nonlinear. I bet it's getting hotter, too.

I think the extra current is because the core

Probably not; it's likely just saturation.

Yup.

The primary determines saturation. If there's room for more secondary turns, maybe of finer wire, leave the original primary winding and work on the secondary.

John

It seems like a whole lot of bother to make a transformer do what it was not originally designed to do, and taking turns from the primary is the worst way possible. Once saturation occurs, the output voltage actually becomes somewhat regulated, so a large change in primary voltage produces much less change in the secondary. If you are not using a true-RMS meter, your readings will be false because of the increased distortion (clipping, or a flattened peak). Your current readings will also be erroneous because you will see even more distortion, but for current this will be sharp peaks corresponding to the flattened output voltage waveform.

#16 or #18 wire is not huge, but of course I am a high current guy and I'm used to #4-0 welding cable and 6" x 1/2" bus bar. If you are willing to go to the trouble of unwelding a transformer and pulling out primary turns, you should be able to find an old automotive alternator or generator with wire heavy enough to use for a proper secondary. A motor winding shop might give you some old wire, or maybe an entire motor, for the price of its recycle value.

Better yet, you should be able to find a more suitable transformer surplus or at a hamfest. The heavy items are often left behind. I've picked up (with some difficulty!) some big transformers and iron core chokes that were in the freebie piles and trash bins. Sometimes you have to pull them out of old tube equipment. If you can find a Variac or Powerstat, you basically have a toroidal core primary, and you can wind your own secondary for any voltage you want. I have a couple dozen of these with power rating of 500 VA to 2500 VA or more, and I've even repaired some that were burned, because all the windings are accessible.

Paul

Well, maybe a bother to you but was fun for me. Learned a little bit about the practical side of transformers.

Oh, the reason why I didn't want to mess with the secondary was that it is center tapped. I figured it would be much more complicated to mess with? I'd also have to worry about keep the taps even. (Although I suppose it wouldn't be hard to add turns but since I didn't have any wire handy I just decided to mess with the primary.

Well, I probably could have found something to use but wasn't interested in it so much. I kinda ruined the transformer as now its pulling 4A but I am geting my 30V on the secondary. Makes more noise than before but I didn't weld it all that well. Only took a min to cut it open and another min to weld it closed. The hardest part was getting the bobbin out.

Yeah, I'm going to get me a better quality transformer. I was kinda experimenting with this one. I'm sure I could always rewind it if I wanted. I assume that I could reduce the saturation point by using more windings on the primary?

The transformer died in a good cause though as atleast I have learned a great deal about them. Wasn't a good transformer to start with so no big loss.

Thanks, Jon

Oh, actually the secondary is about AWG 10 or 9(if there is a 9?). The primary is about 16. Not that it makes much of a difference.

Now I have 30VAC secondary and 4A primary. The transformer is pretty useless now unless I only plan on using it for a few mins or so(last time it took about an hour to heat up pretty good so now I would imagine that it would take a few mins)

yeah, after I welded it there wasn't any significant decrease in saturation... just noise.

Well, chances are this transformer sucked from the start? (from the other posts) It was already in saturation and I didn't want to mess with the secondary because it is CT'ed. I didn't want ot mess with adding wire to it because theres not much room left.

I suppose, if I'm really eager to mess with it, is I can just rewind the whole thing?

If I use smaller secondary and primary windings I can probably keep it out of saturation(it was about 100 to 110VAC before it saturated at 120VAC)? Since I don't need the large secondary current I can use smaller wire on both... and since it won't saturate as much that means I can even decrease the size farther?

I'm not sure if its the laminates that are the main issue or the turns? I understand that there are better laminates but can this be offset by using larger turns or is there some upper bound that is reached?

The way I see it is that I would need to rewind the primary using a smaller gauge and if I keep it out of saturation since its about AWG16 I might be able to go with AWG 18 or even 20? Of course I'd then have to worry about the secondary and its number of turns. If its about a 6:1 ratio then that means I can get about 6 amps out on the secondary for 1amp on the primary.

The problems I forsee is that if I use to small of a gaugue to reduce saturation and have a large number of turns then I also have to have a large number of turns but larger gauge on the secondary.... Its possible that I might end up without the room on the bobbin for the secondary. Although I guess I could always remove primary turns as long as it doesn't cause saturation.

I guess I need to figure out what size wire I need on the secondary a given wire size on the primary at a certain max current? Once I know this I can then estimate if the secondary will fit on the bobbin given the turns on the primary?

Surely theres some software for this type of stuff?

Thanks, Jon

For a given core, it's primary turns that cause saturation; fewer turns makes higher flux density and heads for saturation. It sounds like this one was borderline to start with.

Rewinding sounds like a huge hassle. There are lots of surplus transformers around.

John

good! it's not.

yup.

it might be better to re-wind the secondary with more turns (maybe use thinner wire)

if the weld is thin (not right across the core) it doesn't form a shorted turn for eddy currents.

--

Bye.
   Jasen

yeah, but what I'm saying is that if I would have added turns to the primary then I could have reduce the saturation? Of course I would have decreased the secondary voltage too but...

Well, I will try to look into that and see. In any case its a good learning experience and things always make more sense when I get my hands dirty.

Thanks, Jon

If the core is indeed saturating, you'll only make it worse by removing turns. It's anyone's guess how it will turn out - mine would be badly, but if it is only to be used occasionally for short periods it may work. The transformer is rated at 15 amps at 24 out raising to

30 will lower the output rating as well. 100 watts of waste is a fair amount of heat and cost if it is to run for protracted periods.

Putting a small 4 amp 6 volt transformer phase wired to boost the secondary is safer and probably easier.

Transformer manufacturers do take some short cuts and welds are one of them - you see it a lot in el-cheapo 12V battery charger transformers. microwave ovens, and other bulk produced consumer stuff. Not as bad as it may seem at first because the weld usually has very shallow penetration and is on a part of the core where the field is weaker - but it will waste some power and I wouldn't try to re weld it. It is also hard to do unless you have a TIG setup and are very good and fast.

You can add turns to the secondary if there's room.

--

----== Posted via Newsfeeds.Com - Unlimited-Unrestricted-Secure Usenet News==----
http://www.newsfeeds.com The #1 Newsgroup Service in the World! 120,000+
Newsgroups
----= East and West-Coast Server Farms - Total Privacy via Encryption =----

As a learning experience what you are doing is fine, and maybe for a "rough and dirty" power supply it will be OK. I looked at your previous post in Mid-March (although the nym was BobJones, so maybe not you), the transformer was a 12 VCT at 15 A, and another was 25 VCT at 2 A. The purpose there was for a variable supply 5 V to 12 V.

Anyway, much depends on what you intend to use the transformer for. A variable DC supply will be more efficient if you run the primary off of a Variac or Powerstat. At low outputs you will be below saturation and it will be very happy. You can also boost the 120 input to 140 and push the output at the expense of saturation.

You could also use a phase modulated triac controller designed for inductive loads.

If you just want to get higher DC voltage out, you can connect the output in various ways. A full wave bridge with 12 VCT will give you about 15-18 VDC, while a full wave CT will give you half the voltage and twice the current. You can make a doubler circuit using two capacitors and two diodes to get 30-35 VDC. The regulation will be poor but such a circuit is also somewhat current-limiting, which may be good. You will need about 2000 uF per ampere.

Have fun with your experiments.

Paul