help to build a transformer to car amp.

[Dj BASS AMP]

Hi All...

I Want to build a power supply to Car amp. i Need the supply in the end 40V +/- of 15A.

now i build all the circuit but i the transformer i want to build to.
how can i do that.

*that direction of winding is matter??
*what the size of the wire in need to winding in the transformer
i need to put in the input and that the output i need to put to get a 15A of 40V + -. thanks !

the circuit is build from 12Xirfz44 (6 for side). it is good?

thanks.
Tamir

[Workhorse]

Hey DJ I only know how to fabricate low frequency 20-400Hz E-I Iron Core Transformers not high frequency ferrites.
If interested mail me!
A+m+p+m+a+n =?????

[ChocoHolic]

Hi DJ!
I think without schematic of your SMPS it will be hard to give
suitful hints.
My proposal:
Put the schematic and hope that EVA is in the right mood!

Cheers
Markus

[Dj BASS AMP]

something like that :

[ChocoHolic]

As far as I can see you must partially take care about the winding polarity of the windings.
The polarity of both primaries must match together
and
the ploarity of both secondaries must match together.
My proposal. Wind the transformer always with the same
direction of rotation. The points at the winding in the schematic
will mark the starting of the windings.
Polarity of primary to secondary should not be important in this design.
Transformer design does not require an airgap, from my understanding.

The size of the core depends very much on the used frequency.
Sorry I did not find time to check all details of the schematic
and find out the frequency....

The turns ratio is identical to the voltage ratio.
The absolute number of required turns can be determined by the voltage time product and the core geometry.
U x ton = N x deltaB x A

As an example:
operating frequency: 30kHz , ==> T=33us, ton = 16.7us
material: ferrite N67, suitable for +/- 250mT in this application and frequency (I think so, but NOT rechecked with data book...)
cross section area of the ferrite: i.e. 100mm^2
In this push pull circuit you will always switch 12V to the primaries
and the flux may run between +/-250mT ==> deltaB=500mT
N = (U x ton) / (deltaB x A)
Please take care for the units:
1mm^2 = 1E-6 m^2
1mT = 0.001T etc....
For the primaries this means
N = at least 4 turns in this example.

The wires should be suitable multistrand HF-litz.
0.1mm strands are popular and should give acceptable results
up to 100kHz in an ungapped design.

.....Eva? Where are you?
Please jump in and help, if you see mistakes or feel
that more details are required by DJ.

Bye
Markus

[destroyer X]

Just to put my fingerprint on this, i have seen 1 milimeter diameter wires on that kind of transformer. Also i now this special wire you told chocoholic, the one that is strong and is mixed with other non conductive fibers...maybe silk fibers or cotton fibers.
I also like when Eva appear. She is clear, direct and have deep knowledge... and wonderfull nails too!.
My wife is from Spain.... people there has a very strong personality.
You are talking in ferrite... unfortunattely whe do not use them in audio transformers (do not know why), also they not use E shape plates, is another assembling that is more adequated to audio.
When Israel ligths goes down, lamps became yellow, may be Mr. Dj Bass testing something!
regards,

Carlos....HELLO EVA, HELP US!

[Eva]

This shcematic is intended to oscillate at 150Hz [300Hz osc] to be used with a standard or custom designed 50Hz transformer. For high frequency SMPS I think there is a lot of info in the other SMPS thread

I see no point in mantaining two threads about the same topic at the same time

[ChocoHolic]

Uhps 150Hz....
Yes Eva is right,.. Ct is large (100nF) and Rt high (56kOhms).
This determines the low oscillator frequency.
Then forget all my HF-proposals like ferrite and HF-litz.
A 50Hz iron core will do it. Massive wire is also fine for our
low power. (At 50Hz, splitting the strands or "Roebeling" is only required at very high power , above 10kW)
But the core size will be big!!
If you want +/-40V and 15A (1.2kW) at 150Hz, then a core size
of a normal 750W/50Hz transformer should be fine, as given
in your schematic.
The points about winding direction and # of turns are
also valid at low frequencies....

My feeling about a SMPS for car Audio at 150Hz?
Well, big & heavy, not state of the art..
But less difficult than a high frequency design.

If you dare to do a HF design, you should really have a look to the thread which Eva mentioned.
toroid for car amp smps

Another link which might help you
http://sound.westhost.com/project89.htm

Cheers
Markus

P.S.
Hi Eva,
...charming comment...
But, well I must admit...
I was lazy and your expertise seemed to be the perfect solution...

[ChocoHolic]

Another hint:
If you use a 50Hz iron core you can go for higher flux densities.
But I would prefer to be careful.
I would expect that the circuit will fail if the core runs into saturation!
Saturation will happen above approx. 1T for normal iron cores.
But you should not try to use that full swing of +/- 1T in this design, because of two reasons:
1. The core losses are fine at 50Hz, but at 150 Hz they will be much higher (overproportional)
2. During start up the core will start from 0T and then the
core can easily run into saturation.

My proposal would be to run the core at +/-400mT nominal.
==> deltaB = 800mT= 0.8T.
This should be save for start up and also for a non stable battery
voltage.....

Bye
Markus

[ChocoHolic]

Wow!
The SG3525 features a slow ramp up of the MosFet signals.
So you will probably have no issues with saturation during
turn on....