Bridge rectifier overheating

Dodgydave

Joined Jun 22, 2012
11,285
Time to start measuring current and voltages,

1) whats the primary current with no output on the secondary connected.
2) whats the output voltage and current when the bridge rectifier and capacitors are connected.
3)same as 2) but with a led connected.
 

GopherT

Joined Nov 23, 2012
8,009
The diodes have about a 1 volt forward drop per leg. At 4 amps DC thats about 8 watts. You have it on thermally insulating cloth. It can run pretty hot, but it wants a heat sink. If you builld this into metal chassis, bolting it to that will be plenty.
You are using thick vinyl insulated wire as your secondary and it appears to be losely wrapped on the core. You should be using enabled magnet wire wrapped reasonably close to the core in a nice orderly fashion. The copper is also a heat sink to the core - and the rest of the magnet wire. The vinyl insulation is thermally insulating the core and, the loose wrapping means you have poor magnetic coupling and, therefore, low efficiency and heat. At least that is where I would look.
 

Thread Starter

Gabriell

Joined Jan 16, 2016
47
What happens if you have the secondary of the transformer disconnected from any load an apply power to the primary? Does the transformer still get hot?
So i took out my secondary coil and and the primary gets hot when i apply voltage.
I replaced my secondary with copper wire and applied voltage without load. But again it gets hot....
 

Thread Starter

Gabriell

Joined Jan 16, 2016
47
You are using thick vinyl insulated wire as your secondary and it appears to be losely wrapped on the core. You should be using enabled magnet wire wrapped reasonably close to the core in a nice orderly fashion. The copper is also a heat sink to the core - and the rest of the magnet wire. The vinyl insulation is thermally insulating the core and, the loose wrapping means you have poor magnetic coupling and, therefore, low efficiency and heat. At least that is where I would look.
image.jpeg
This is the secondary i used this time. It wrapped pretty tight. But still overheating....
 

GopherT

Joined Nov 23, 2012
8,009
@Gabriell

The each of the silicon-steel plates need to be electrically insulated. Once they all start conducting with each other, you will get Eddie currents and heating of the transformer's core (known as magnetic losses). Note, this only concerns the conduction between plates (they were insulated with an oxide coating or laquer at the factory). Scratches, rust or even just age can cause the conductivity. Eddie currents can heat the core even with no short from a winding to the core.
 

Thread Starter

Gabriell

Joined Jan 16, 2016
47
@Gabriell

The each of the silicon-steel plates need to be electrically insulated. Once they all start conducting with each other, you will get Eddie currents and heating of the transformer's core (known as magnetic losses). Note, this only concerns the conduction between plates (they were insulated with an oxide coating or laquer at the factory). Scratches, rust or even just age can cause the conductivity. Eddie currents can heat the core even with no short from a winding to the core.
The core it self does not heat up its the primary coil that gets hot and then trasferes the heat to the core. I was checking the core and it gets warmer only after my primary is hot.
 

GopherT

Joined Nov 23, 2012
8,009
The core it self does not heat up its the primary coil that gets hot and then trasferes the heat to the core. I was checking the core and it gets warmer only after my primary is hot.
So, the primary copper is heating before the plates near the primary? Then you must have a problem with the copper or it's insulation on the primary.

The Eddie currents are most intense on the plates close to the winding.
 

Thread Starter

Gabriell

Joined Jan 16, 2016
47
So, the primary copper is heating before the plates near the primary? Then you must have a problem with the copper or it's insulation on the primary.

The Eddie currents are most intense on the plates close to the winding.
Yes its definitely not the core. And the primary looks good. No cracks or scratches. I was very carful when i removed it from the core.
Does it matter where i put the primary? Top or bottom of the core and upside down or not?
 

GopherT

Joined Nov 23, 2012
8,009
Yes its definitely not the core. And the primary looks good. No cracks or scratches. I was very carful when i removed it from the core.
Does it matter where i put the primary? Top or bottom of the core and upside down or not?
Ok, now I see the vice. Tell us exactly how you deconstructed the old one and built the new one. Sketches would be helpful.

If you cut the steel plates, you have conductivity between plates.
Or,If you simply placed an iron core within the core of another transformer, and rely on induction to couple the magnetic flux in the primary to the core of the secondary - you likely have an issue.

Please explain your experiment. A video of you taking it apart would be highly informative.
 

Thread Starter

Gabriell

Joined Jan 16, 2016
47
Ok, now I see the vice. Tell us exactly how you deconstructed the old one and built the new one. Sketches would be helpful.

If you cut the steel plates, you have conductivity between plates.
Or,If you simply placed an iron core within the core of another transformer, and rely on induction to couple the magnetic flux in the primary to the core of the secondary - you likely have an issue.

Please explain your experiment. A video of you taking it apart would be highly informative.
Sorry but i cant upload a video. I do have some pictures hope it helps.

So first i cut the weld on each side of the core. Then i use a piece of wood and a hammer to get out the coils. I try to avoid hitting the primary and i only hit the secondary to push everything out from the core. I dont force it very hard. Then i clean it prepare the new secondary and put everything back. Some of the plates of the top of the core ( I peace ) is separated but usually when i clamp it together its tight enough so there is no space.
image.jpeg
This the core and from left to right on the top it the primary and 2 secondary. I tried both but still overheats.
image.jpeg
This is the clamp i use to hold it together. Please note that i have tried without the clamp as well and heat is still the issue.
image.jpeg
This is the open core

image.jpeg
Core with copper secondary.

If you need any more info please let me know i will do my best!!
 

Attachments

Thread Starter

Gabriell

Joined Jan 16, 2016
47
So i just salvaged 2 more transformers. Both really good condition. I have not touched the coils nor the core. No scratches or cracks. I connected it to the power and again it gets hot. I left both on for about 5 min. The core and the primary coil is hot but i can touch it. But just barely.
Are they supposed to run hot? Can i keep using it?
I want to make power supply out of them and they should stay on for about 10-12 hours per day. Will that damage the transformer?
 

shortbus

Joined Sep 30, 2009
10,045
@Gabriell

The each of the silicon-steel plates need to be electrically insulated. Once they all start conducting with each other, you will get Eddie currents and heating of the transformer's core (known as magnetic losses). Note, this only concerns the conduction between plates (they were insulated with an oxide coating or laquer at the factory). Scratches, rust or even just age can cause the conductivity. Eddie currents can heat the core even with no short from a winding to the core.
I know this is what people say. But in reality it doesn't make sense to me. Any of the MOT's I've seen have the I and E lamination's welded together. And most induction motors do the same to the stator lamination's, a line of weld down one side. Wouldn't that short one all lamination's together?
 

shortbus

Joined Sep 30, 2009
10,045
@Gabriell, while I don't have any proof, stop and think how a microwave oven is used. It's only on for a few minutes most of the time, and then it has a fan cooling it. Maybe this is why it is heating up when trying to do what you want. This is the part I cant prove, the core may be limited in volume of core lamination's. If your trying to get more VA from a transformer than it is designed for, heat would rise proportionally.
 

GopherT

Joined Nov 23, 2012
8,009
I know this is what people say. But in reality it doesn't make sense to me. Any of the MOT's I've seen have the I and E lamination's welded together. And most induction motors do the same to the stator lamination's, a line of weld down one side. Wouldn't that short one all lamination's together?
Yes, I guess it would. I hope the manufacturing guys are talking to the design team. It may be that the magnetic flux closest to the core is key? Doesn't make sense to me why they would bother insulating the plates. They spend a good deal of time and money on the oxide costings. 30 to 60 cycles of (magnesium alkoxide/alcohol dip, dry, steam, dry, repeat). I guess the lack of conductivity closest to the core is key? No idea why they would weld it other than, as you said, losses are ok since the usage cycle is low and nobody wants the transformer hum noise of the unwelded plates? Who knows.
 

shortbus

Joined Sep 30, 2009
10,045
In my limited/hobby experience MOT transformers are the only ones that are welded. A standard transformer is usually inter leaved with the I's and E's reversed in each layer.
 
So i just salvaged 2 more transformers. Both really good condition. I have not touched the coils nor the core. No scratches or cracks. I connected it to the power and again it gets hot. I left both on for about 5 min. The core and the primary coil is hot but i can touch it. But just barely.
Are they supposed to run hot? Can i keep using it?
I want to make power supply out of them and they should stay on for about 10-12 hours per day. Will that damage the transformer?
MOTs are designed for intermittent duty. Low cost design for that purpose results in a transformer that is running its core well into saturation. This means that the core will heat up a lot even when the transformer is not supplying a secondary load.

It looks to me like you have extra room on the secondary for more copper. The primary is wound with well defined layers; count the turns in one layer and multiply by the number of layers to get the total primary turns. If you add about 10% more turns to the primary the maximum flux density will decrease and the core and primary won't get so hot. Wind that many turns in the secondary space (in a single layer if possible), with magnet wire (the kind you're using in the latest secondary) of the same diameter (or nearly so) as the existing primary. Wire those extra turns in series with the existing primary with the proper phase. The heating of the core and primary winding should decrease substantially.

To determine the proper phasing (there are two different ways you can connect the additional winding in series with the existing primary) connect an incandescent light bulb in series with the grid voltage before connecting to this new arrangement. The light bulb will light up brighter with the incorrect connection of the additional primary turns. Leave it connected such that the light bulb is least bright. Then remove the light bulb from the circuit and run the unloaded transformer for a while to see if it heats up substantially less.

It would be good to measure the unloaded primary current before and after you add extra turns in series with the primary. If you do this, report to the forum the before and after values.

Before you add a secondary to this new arrangement, wrap some tape around the extra primary turns to insulate them from the secondary. This will work out best if you wind those extra primary turns carefully in a nice single layer.
 

Thread Starter

Gabriell

Joined Jan 16, 2016
47
MOTs are designed for intermittent duty. Low cost design for that purpose results in a transformer that is running its core well into saturation. This means that the core will heat up a lot even when the transformer is not supplying a secondary load.

It looks to me like you have extra room on the secondary for more copper. The primary is wound with well defined layers; count the turns in one layer and multiply by the number of layers to get the total primary turns. If you add about 10% more turns to the primary the maximum flux density will decrease and the core and primary won't get so hot. Wind that many turns in the secondary space (in a single layer if possible), with magnet wire (the kind you're using in the latest secondary) of the same diameter (or nearly so) as the existing primary. Wire those extra turns in series with the existing primary with the proper phase. The heating of the core and primary winding should decrease substantially.

To determine the proper phasing (there are two different ways you can connect the additional winding in series with the existing primary) connect an incandescent light bulb in series with the grid voltage before connecting to this new arrangement. The light bulb will light up brighter with the incorrect connection of the additional primary turns. Leave it connected such that the light bulb is least bright. Then remove the light bulb from the circuit and run the unloaded transformer for a while to see if it heats up substantially less.

It would be good to measure the unloaded primary current before and after you add extra turns in series with the primary. If you do this, report to the forum the before and after values.

Before you add a secondary to this new arrangement, wrap some tape around the extra primary turns to insulate them from the secondary. This will work out best if you wind those extra primary turns carefully in a nice single layer.

So i added about 15 turns to primary from a spare coil i had ( both the same size ) but the heating issues remains. I didnt put in the secondary this time.
One thing i dont understand!
I was trying to measure the current and it keeps going up and down between -0.06amps and +0.06amps
How is that possible?
image.jpeg
image.jpeg
image.jpeg
 
So i added about 15 turns to primary from a spare coil i had ( both the same size ) but the heating issues remains. I didnt put in the secondary this time.
One thing i dont understand!
I was trying to measure the current and it keeps going up and down between -0.06amps and +0.06amps
How is that possible?
You're measuring AC current on a DC range. I can't see the connetions to the meter very well. Back off with the camera so I can see the whole meter and where the leads are connected.

Did you do the trick with the light bulb? Can you measure the current (and it must be the AC current) with both possible ways to connect the extra turns in series with the primary?

I'm returning to the big city, so I'll be on the road for a few hours. I'll check in later.
 

Thread Starter

Gabriell

Joined Jan 16, 2016
47
You're measuring AC current on a DC range. I can't see the connetions to the meter very well. Back off with the camera so I can see the whole meter and where the leads are connected.

Did you do the trick with the light bulb? Can you measure the current (and it must be the AC current) with both possible ways to connect the extra turns in series with the primary?

I'm returning to the big city, so I'll be on the road for a few hours. I'll check in later.
I didnt do the light bulb trick as i dont have one around me but i will hint one down.
My multimeter cant measure ac current ( i think ).
I have. To go to work now but i will post some pictrures when i get home!
Thanks for your help so far!!!
 

Thread Starter

Gabriell

Joined Jan 16, 2016
47
You're measuring AC current on a DC range. I can't see the connetions to the meter very well. Back off with the camera so I can see the whole meter and where the leads are connected.

Did you do the trick with the light bulb? Can you measure the current (and it must be the AC current) with both possible ways to connect the extra turns in series with the primary?

I'm returning to the big city, so I'll be on the road for a few hours. I'll check in later.

I connected a light bulb to the circuit. I tried to wire it in any possible way but the light has the same brightness.
I still cant measure ac current ( i have to buy a good enough meter ).
There is one thing i noticed.
My primary coil is at the bottom and the extension is sitting on the top.
Connecting L and N either pushes the coil up or pulls it down depending on how i connect them together. That would be the magnetic flux i assume. Is there a right and wrong way of connecting it? Does the flux has to flow in one direction or it doesn't matter at all?!
image.jpeg

image.jpeg

image.jpeg
 
Top