12v 250w Car SMPS based off SG3525

A

AMSA

New member
I wally, thanks a lot.

You are right, taking the average that Ae result gives the same as in the datasheet.

About the Ae, you are right. I should have painted the center leg for sure. However, nothing stops me in using that right leg, for example, right? Another question regarding the core. I can use the U type core, right? I know that this type of core is more suitable for flyback, but nothing stops me as well to use it in forward converter right? This kind of core has larger value of B when compared to the E type, right? There is any difference in using one or the other? I have that type of core too. I got it from a TV High-voltage transformer.

About the topology, maybe you are right. Using the configuration of two transistor forward may be more wise.

I read that the reset winding may be the same as the primary winding (wire size, number of turns, etc), and for best results that should be wrapped together with the primary winding. Is that correct?

It is difficult to do the reset winding?

Regards!
 
W

wally7856

New member
“About the Ae, you are right. I should have painted the center leg for sure. However, nothing stops me in using that right leg, for example, right?”

Well using the end leg for the example will mislead people.

“Another question regarding the core. I can use the U type core, right? I know that this type of core is more suitable for flyback, but nothing stops me as well to use it in forward converter right? This kind of core has larger value of B when compared to the E type, right? There is any difference in using one or the other? I have that type of core too. I got it from a TV High-voltage transformer.”

You can use any core type, if the winding’s fit. B is flux and a property of the ferrite material.

“About the topology, maybe you are right. Using the configuration of two transistor forward may be more wise.”

There are advantages and disadvantages to all topologies. The two transistor forward is another one that almost no one uses, so it will also be hard to find example schematics. Also it needs a high side driver.

The push pull is very common and you will find many examples. It does not need a high side driver but it needs 2 primary winding’s.

The half bridge is very common and you will find many examples. It need’s a high side driver but only uses 1 primary winding.

“I read that the reset winding may be the same as the primary winding (wire size, number of turns, etc), and for best results that should be wrapped together with the primary winding. Is that correct?”

Yes i think you are correct about winding the 2 winding’s together, but the reset winding can be about 1/2 the diameter of the primary winding.

“It is difficult to do the reset winding?”

It sounds like we have it figured out so it would not be to hard. The last thing is to make sure of is that you get the polarity correct for the reset and primary winding as marked in the schematics of the transformer.
 
A

AMSA

New member
Hmm, I see. In fact there are no many examples on forward converter, even with 1 or 2 transistors. Maybe it would be better to do a push-pull configuration?

Wally, I saw somewhere, in a book or in a website, the primary calculations for the forward converter involving the duty-cycle of the converter. Why that is done? For example, with your method, we don't use that and even then the transformer will do for sure the job.

Do you understand what I mean?
 
W

wally7856

New member
AMSA

If you want to make a simple circuit this looks like a good one to me. Look at page 10, fig 20.

http://www.onsemi.com/pub_link/Collateral/TL494-D.PDF

It is uses a very common TL494 control chip and a low voltage push pull circuit. Because of the low voltage you would have a low amount of turns and not have to worry about VDE and insulation. The TL494 is available almost anyware in the world.

Many people on this forum are experts with this control chip and topography so it would be easy to get help.

One thing i noticed about that circuit is it has a bad part number for the Ferroxcube transformer. So ignore the part number and the information for it.

Also because of the few turns and low voltage and current this would be very easy to do with a toroid, if you can find one. But not a toroid made for an inductor, it has to be power ferrite.

Last thing, if you keep the frequency down below 20khz you could even build this with cheap low frequency 1N4000 series diodes.
 
W

wally7856

New member
“Wally, I saw somewhere, in a book or in a website, the primary calculations for the forward converter involving the duty-cycle of the converter. Why that is done? For example, with your method, we don't use that and even then the transformer will do for sure the job.”

I have seen that also. When an example is given i compare their answer to my answer. I do not get the exact same number of turns but i get close enough.
 
A

AMSA

New member
Okay wally.

Well, I'm gonna study now the converter, and your design procedure in order to make the transformer!
 
A

AMSA

New member
Hi there.

Wally, I tried to dismount a transformer that I had here, using boiling water. After 15min / 20 min I tried to separate both parts with 2 pliers and the core has broken. I'm gonna try again with another transformer that I have here but I'll let that be in the water more time. However, I think that this other transformer that I'm gonna try to separate is for flyback application.

Is there any problem in put some glue and get together those parts? The core will lose his properties?

BTW, what is the difference between ferrite core and Iron powder core?

Regards!
 
Last edited:
A

AMSA

New member
Hi again.

Well, success. I manage to separate the core with the boiling water method. I left more time the transformer in the boiling water. Them I took out from the pot and with a towel I made a little bit of pressure and they were easily separated. However, I think that this core will not be useful to my forward converter since the center leg has a gap, right? Those kind of cores are only for fly-back application right?
 
W

wally7856

New member
“Wally, I tried to dismount a transformer that I had here, using boiling water. After 15min / 20 min I tried to separate both parts with 2 pliers and the core has broken. I'm gonna try again with another transformer that I have here but I'll let that be in the water more time. However, I think that this other transformer that I'm gonna try to separate is for flyback application.”

You can still disassemble the transformer and see how it was constructed with turns and insulation.

“Is there any problem in put some glue and get together those parts? The core will lose his properties?”

I have read story’s that if you do a good job gluing with a close fit the transformer will still work.

“BTW, what is the difference between ferrite core and Iron powder core?”

Ferrite is for power transformers (high permeability) and iron powder for inductors(low permeability).
 
W

wally7856

New member
Well, success. I manage to separate the core with the boiling water method. I left more time the transformer in the boiling water. Them I took out from the pot and with a towel I made a little bit of pressure and they were easily separated. However, I think that this core will not be useful to my forward converter since the center leg has a gap, right? Those kind of cores are only for fly-back application right?”

Usually for flyback’s but one and two transistor forward converters can use them also. What topology was used when you traced out the circuit.
 
A

AMSA

New member
Well, about where the transformer was, I can't really remember. It's been a while since I have got dismounted that transformer.

What is needed to make the fly-back transformer, that I have dismounted recently, work with the one transistor forward converter?
 
W

wally7856

New member
“What is needed to make the fly-back transformer, that I have dismounted recently, work with the one transistor forward converter?”

I never tried to design a fly-back circuit. But i think you would have to try to figure out what transformer you have by calculating the WaAe and measuring the gap and try to figure out what you have by looking at the fly back design equations. Maybe try to measure the inductance value also.
 
A

AMSA

New member
Hello there wally and others.

Wally, I'd like to ask you, what kind of core material and type (size, for example a E55, E42, ETD49, ETD39 is the best for general purpose application? I have looked in to the farnell site and I have seen EPCOS and FERROXCUBE 3C90, N27, N87.

Regards!
 
W

wally7856

New member
“Wally, I'd like to ask you, what kind of core material and type (size, for example a E55, E42, ETD49, ETD39 is the best for general purpose application? I have looked in to the farnell site and I have seen EPCOS and FERROXCUBE 3C90, N27, N87.”

There is no wright answer. Some ferrite materials are high flux, some are lower flux, some are more temperature sensitive than others etc. You have to look in the data curves to see what properties each one has. Some will be better or worse in each area.

Core size is a problem of getting a reasonable amount of strands to fit. The wire fit on a transformer depends on the number of turns and current you need. Each time you design a smps it has different current needs and therfore needs a different amount of wire. Again what i find the easiest to do is make a list of all of the transformer cores i can buy and then try to fit the wire i need on each core. Each core type has different advantages. EE cores can have wider winding areas. ETD cores usually have higher winding height.

Do not buy a core until you have made certain your winding’s will fit.
 
A

AMSA

New member
I see. I will look at ebay.co.uk which core are available and then I will give a look at pressmans book in order to compare the cores with those tables that he has.

Regards.
 
A

AMSA

New member
Wally, there is any difference between using a E I core and a EE type? The procedure is the same? I have got I core E I from a computer power supply, but I don't know if that method to design the transformer is suitable for this type of core.
 
W

wally7856

New member
“Looking into this datasheet of core material N87 from EPCOS, can you tell what is the Bmax? I can't figure it out.”

For what frequency?

“Wally, there is any difference between using a E I core and a EE type? The procedure is the same? I have got I core E I from a computer power supply, but I don't know if that method to design the transformer is suitable for this type of core.”

The type of core does not matter.
 
A

AMSA

New member
Okay. Frequency, 100kHz.

Regarding the EI type: In the datasheet of suppliers, for example, ferroxcube, the specification in those datasheet are only for one E half, right?

Now, the use of your method wally, can be applied to the I type? That part that goes on top of the E core? Where the I part enters in the calculations?
 
A

AMSA

New member
Wally, the EI cre type that I have with me has the following measures: (accordingly to the pictures that you have posted)

A: 33,07mm;
B: 23,75mm;
C: Can't take because there is only one halve;
D: 12,66mm;
E: 9,60mm,
G: Can't, for the same reason;

The I type part has:

Length: 33,1mm;
Thickness: 5,04mm;
Width: 12,69mm;

What is the power that he can handle? I have tried the pressman book but I didn't managed to find.

Regards.
 
You must log in or register to reply here.
Top