4-Coil Transformer? Possible? Dumb idea? Any help appreciated

Thread Starter

52brandon

Joined Sep 9, 2018
15
Brand new here so I apologize if this is the wrong place or if this is just a plain stupid question. I'm building an isolation transformer out of a couple transformers that I got from old stereo receivers (these things are amazing for build components. Pots, switches, jacks, MOSFETs, relatively large transformers - these 2 also happened to not alternate E and I sheets, so I can easily break em apart, large heatsinks, large capacitors, and more). They're nearly identical (same manufacturer, same dimensions, nearly the same color wires even) so I decided they'd make good candidates for an isolation transformer. Unlike MOTs, these have the secondaries wound around the outside of the primaries rather than being stacked. So my first thought was to just tear off the secondaries and I sections of the cores and then attaching both E sections together with the mains primary coils making it an EE core transformer with identical primary and secondary coils of mains voltage. Which to my understanding is exactly what I want for an isolation transformer. While planning it out, I was considering how I could add in cooling fans and thought that would be simple if I could keep the secondaries, meaning it would be a 4 coil transformer. Now, I've heard of these existing, so I know the concept isn't impossible. But is it wise? I've not been dabbling in this field for very long (the "field" I'm referring to is "building custom electrical devices", which is probably evidenced by the fact that I don't have an isolation transformer already). I've been really getting into it for just shy of a year now and collected a good bit of components from devices people were throwing out/giving away/or I found super cheap. This isolation transformer build is using all parts that I've accumulated except for the enclosure that I am likely going to print on my 3d printer.

Any and all guidance would be greatly appreciated. And just let me know if you need any other details/pictures, etc. Thank you ahead of time.
 

LesJones

Joined Jan 8, 2017
2,638
I would suggest just using the transformers back to back. Find the secondary winding that looks like it has the highest power rating. This will be the one that powered the amplifier section. It will probably be a centre tapped winding to supply a positive and negative rail. Connect the ends of this winding to the endings of the same winding on the other transformer. Insulate the centre tap of this winding and the ends of all the other windings. You now have an isolation transformer. I you are not sure that the transformers are IDENTICAL then confirm that the voltage from the chosen windings is the same when the primaries are powered from the mains. The output voltage of an isolation transformer made from two transformers connected this way will be slightly less than the input voltage.

Les.
 

Thread Starter

52brandon

Joined Sep 9, 2018
15
Thank you for your input. I've seen a handful of people put together isolation transformers that way and considered it but decided to go the way I did because I can't imagine how it would be more efficient to go through the 120v primary, a thinner @60v secondary, a wire connection over to a completely different transformer, a second @60v coil, a second 120v coil, and finally out to the device as opposed to in the 120v primary, over to an identical 120v secondary and then out to the devices. I just can't imagine any way that the second option wouldn't be a more efficient and reliable circuit. Or am I missing something? I have very little experience, so I've had to go based on theories I put together based solely on information. But I thought it was a pretty safe bet to assume removing the hundreds of feet of wire it would use going through both secondaries and the jumping from one core to the other would net a better circuit. I have a couple other questions about that approach. I guess first and foremost, the transformers are NOT identical. They are NEARLY identical. By that, I mean they're made by the same manufacturer, the cores are the same size (down to even the same number of iron sheets), they had identical spec capacitors (a pair of 71v 10000uF) on the output (these capacitors were actually the reason I cannibalized these receivers to begin with), the tape and bobbins are identical... eh, I stopped my reply here and edited up some pictures that show em next to each other and then detailed pics of each wire coming off

Both.png 1 of 2.png 2 of 2.png
 

Thread Starter

52brandon

Joined Sep 9, 2018
15
What do the labels on them say?
I included pictures that have the labels in my last post. Are they not legible for you? I'm sorry. I'm not at home at the moment so I can't see em right now. But when I get back I'll try to get a better picture, as I don't know what significance the positioning of the info on the label has, so I think it's better to just include a picture so you can see everything that I can
 

shortbus

Joined Sep 30, 2009
7,670
I just can't imagine any way that the second option wouldn't be a more efficient and reliable circuit. Or am I missing something?
The something I can see is, can you live with the noise that may come from the lamination's after you put them back together? one Of the last steps in making a transformer is to 'vacuum impregnate' them with varnish. Taking the lamination's apart will and rebuilding will remove the varnish . The varnishes biggest use in a transformer is to keep it quite, to cut down on the vibration as the polarity changes, since they are in an AC circuit.
 

ebp

Joined Feb 8, 2018
2,332
It looks to me like it would be impossible to fit two windings on the EE core assembly, even if the bobbin flanges were carefully cut way, since wires exit on both ends of the windings - but it's hard to tell for sure from the photos.

Any imperfection in the mating of the ends of the Es will have much greater effect on primary to secondary coupling in the proposed configuration than would similar imperfection between the original E and I stacks of the original.
 

Thread Starter

52brandon

Joined Sep 9, 2018
15
Well I've already cut the I portions off and cut the bobbin down on one of them and it looks to me like it should fit perfectly fine. I've almost got the bobbin cut down on the second one, it's just taking a while because I'm using my little dremel instead of angle grinder because it's a much more precise job
 

Thread Starter

52brandon

Joined Sep 9, 2018
15
20180910_212649.jpg
The one on the left is the one that's already completely flat on the open side. Once my dremel finishes charging I'll finish the one on the right too. As convenient as it is for a lot of things, buying a Dremel Micro as my only dremel was a poor decision. That thing is not meant to be a primary rotary tool. But I should be done with the 2nd transformer within the hour.

As for them making noise. I'm not terribly worried about that. I'm also making an arc welder right now and have gotten good results out of the worm screw hose clamps with corners bent into them, while using steel epoxy to hold the cores together. And after that, I put MANY coats of coil dope on it that I made from styrofoam and xylene. I can't imagine there's much room in there that it can vibrate. But should I be wrong when I go to test them, I can always fill em completely with epoxy
 

Thread Starter

52brandon

Joined Sep 9, 2018
15
20180910_235730.jpg
Updated pic. I think it fits pretty well. Now I'm going to make the band clamps and then test it out after I strap it together
 

Thread Starter

52brandon

Joined Sep 9, 2018
15
test fit med.png
Here it is with the steel band clamps holding it together. I can't see any gap between the cores but it's not sealed together yet so I don't expect the best results yet. Hopefully it's enough to see that it will work though
 

Thread Starter

52brandon

Joined Sep 9, 2018
15
Oh yeah. I really should have clarified this earlier rather than solely focusing on the transformer. In terms of powering the isolation transformer, should I use a 2-prong or 3-prong cord? My basic understanding is that I don't want the output to be grounded. But do I want the input to be grounded?
 

shortbus

Joined Sep 30, 2009
7,670
The most important point in your core is the center leg of it. Any gap there and you have made a "gaped inductor". I say the center leg because that is where most of the magnetism is in a transformer. Are you going to weld the outer core legs like in a MOT?
 

Thread Starter

52brandon

Joined Sep 9, 2018
15
The most important point in your core is the center leg of it. Any gap there and you have made a "gaped inductor". I say the center leg because that is where most of the magnetism is in a transformer. Are you going to weld the outer core legs like in a MOT?
I'm leaning towards putting a couple tacks on the corners. Otherwise, I will be using metal epoxy on all 3 locations where the cores attach to one another. And I'm actually about to check the center of the core to make sure they do make good contact now. The wires were ziptied up and obstructed my view. But I just cut the zipties so I can hopefully get a good look. Though it may be necessary to sand down a little bit more of the bobbins. I guess we'll see

ETA: Yup. The center of the cores do make solid contact. After moving the wires out of the way, shining a flashlight through the back, I could see gaps where the bobbins sit, but solid black at the core
 

shortbus

Joined Sep 30, 2009
7,670
I will be using metal epoxy on all 3 locations where the cores attach to one another.
Won't that, the epoxy, be considered a gap? Not real sure what you mean by metal epoxy, but if you mean one that has metal in it, I don't think that makes it conductive. I would make some straps an use bolts to hold the two cores together. Even though it doesn't seem like it, there is a lot of vibration in the core of a transformer, more than I would trust epoxy to.
 

Thread Starter

52brandon

Joined Sep 9, 2018
15
Won't that, the epoxy, be considered a gap? Not real sure what you mean by metal epoxy, but if you mean one that has metal in it, I don't think that makes it conductive. I would make some straps an use bolts to hold the two cores together. Even though it doesn't seem like it, there is a lot of vibration in the core of a transformer, more than I would trust epoxy to.
Well I planned on clamping it together after I apply the metal epoxy so that the epoxy only fills the gaps between the iron sheets and around them while the iron sheets themselves are still butted up against one another. It's really only being used as a strong adhesive and filler to reduce vibration as much as possible. If you look at the last picture I posted, I've already made steel band clamps specifically for this transformer. And they're on there VERY tight. After I epoxy the cores together, I will be adding liberal (understatement. It'll damn near be considered potted) amounts of coil/corona dope to fill in any other gaps that may still exist. And should this somehow not work, I still have the I sections and can just put em back the way they were. Granted, the bobbins are now mangled. But that won't affect operation. Though I REALLY wanna make this work. Not only because I'm lazy and loathe putting in extra work just to get back to where I was before, but also because I haven't seen an isolation transformer project like this out there before, and it would be nice to add one more option for people in the future undertaking an isolation transformer build
 

LesJones

Joined Jan 8, 2017
2,638
If the metal epoxy is conductive you will create a shorted turn which may get very hot and /or cause the primary current to be higher than it should be. It will depend on how conductive the epoxy is.

Les.
 

shortbus

Joined Sep 30, 2009
7,670
If the metal epoxy is conductive you will create a shorted turn which may get very hot and /or cause the primary current to be higher than it should be. It will depend on how conductive the epoxy is.

Les.
Hi Les, if that is the case, how do they get away with putting a bead of weld on microwave transformers and induction motor stators? I've got a couple of each that have a weld bead connecting all of the laminations that way. I too thought it would act like a shorted turn but doesn't seem to.
 
Top