My first build smps 1kw design by microsim..

vorezky04

New member

Attachments

  • asdfdf.jpg
    asdfdf.jpg
    20.2 KB · Views: 766
  • 26052012(002).jpg
    26052012(002).jpg
    204.3 KB · Views: 425
  • 26052012(001).jpg
    26052012(001).jpg
    223.6 KB · Views: 357

wally7856

New member
vorezky04, Give people a few days to read this to check for mistakes.

Guessing N87 material, there is a flux chart here. It is in a format that is hard to read but it looks to me like 1400 Gauss is best place to run this core.

http://www.epcos.com/web/generator/...F/PDF__N87,property=Data__en.pdf;/PDF_N87.pdf

170VDC nominal
70Khz
ETD49 Ae = 2.114 CM squared
1400 Gauss flux (.14T)

Primary turns = (Vpri x 100,000,000) / 4 x frequency x Gauss x Ae

Primary turns = (170VDC X 100,000,000) / (4 x 70,000 hz x 1400 Gauss x 2.11 Ae)

Pri turns = 20.55 turns, make 20 turns.

Secondary you want 2 x 70vdc output. 70vdc x .707 = 49.49vac + diode drop 2v = 51.49, about 52VAC.

Secondary turns = (20 Primary turns x 52VAC secondary) / 150VDC primary low = 6.93 turns. The 6.93 turns is the total turns in secondary, giving you about 10 volts per turn. You need an even number of turns for the center tap winding. So you can use 3 + 3, giving you about +-60VDC, or 4 + 4, giving you about +- 80VDC.

70Khz skin depth = 10.72mils x 2 = 21.445 mils, metric equivalent .5mm Diameter wire = 400CM

400CM / 350CM/A (hot design) = 1.14A per strand.

Want 10 amp output x 1.2 multiplier for diodes = 12 amps minimum.

I am uncertain if you have to add in more current for 70% efficency and power factor.

12amps / 1.14 = 10.52 strands of .5mm wire, make it an even 10.

P = I x E, 12 amps secondary x 52VAC secondary = 624W

624W / 170 primary volts = 3.67A, because of poor power factor and efficency, the primary will see about 30% more current. 3.67A x 1.3 = 4.78A primary current.

Using the .5mm wire at 1.14A per strand. 4.78A primary / 1.14A strand = 4.2 strands use 4 strands.

Primary 20 turns .5mm x 4 strands
Secondary 4 + 4 turns .5mm x 10 strands

Lets see how wire fits on the ETD-49 core.

Width of the winding area is 36.16mm, subtract 1.3mm per side for bobbin = 33.62mm, subtract 4mm per side for VDE voltage specification = 25.62mm for winding area.

Primary 20 turns .5mm x 4 strands = 40mm, to wide for single layer. Use 2 layers 10 turns each on inner and outer layers of transformer.

Secondary 4 + 4 turns .5mm x 10 strands = 40mm, to wide for single layer. Use 2 layers 4 turns each on inner layers of transformer.

We have 5.62mm left over on each layer we can use for winding’s. If we add more strands to fill it up you will have a cooler running transformer.

Primary 20 turns .5mm x 5 strands = 50mm, or 25mm per layer leaving only .62mm.

Secondary 4 + 4 turns .5mm x 12 strands = 48mm or 24mm per layer leaving only 1.62mm.

The new CM/A is.

Pri. 5 strands x 400CM = 2000CM / 4.78A Pri. = 418CM/A

Sec. 12 strands x 400CM = 4,800CM / 12A Sec = 400CM/A

This will be a very nice transformer. 500CM/A is considered a good design point for continuous use. And you are using this for music power so not as many CM (circular mils) are needed. I am very happy the way it turned out.

One last thing I did not have a bobbing to measure for the width i had to guess its dimensions. Check yourself and let us know how much room there is.
 
Last edited:

vorezky04

New member
Wow amazing your calculate..
May i want make voltage of primer 200 Vdc?
Why must be Vin minimum in primer?
 

wally7856

New member
“May i want make voltage of primer 200 Vdc? Why must be Vin minimum in primer?”

The Pri. Vin should be worst case high voltage. You have to decide what to design for. Some people may want to use international standards. I usually design for the voltages I have at my location 242VAC is the highest i have seen. Also the Pri. low is another choice of international standards or local power.
 

vorezky04

New member
I was print my PCB.
This is my PCB
 

Attachments

  • 461367_448146991864618_100000079620075_1714294_1822484066_o.jpg
    461367_448146991864618_100000079620075_1714294_1822484066_o.jpg
    249.2 KB · Views: 368

wally7856

New member
Good job on your PCB.

There may be a problem with your transformer. In the picture you show a copper foil strip around the transformer. It looks like you soldered the 2 ends together. This would cause a shorted turn on your transformer drawing infinite amps. The copper foil should stop just short of one full turn, or it should overlap with a layer of adhesive between the foil to insulate it and hold it on.

Last thing, how did you put all those pictures together in one big picture.
 

vorezky04

New member
it's not completed.
It just foto.
My trafo don't i winding.
May be wait a moment about 3 days.
I put all picture together with my software was instaled in my PC..
 

norazmi

diysmps Senior Member
why not you wind yourself vorez? We will try to help you until u get the smps work. Nice build and nice pcb :D
 

vorezky04

New member
hehehe, lol..
Okey Norazmi.
Thank you about your partisipation.
So what the firs i do to winding this ETD49 if i want make +/-70 Vdc with curent 10A?
Help me full
 

MyFirstSMPS

New member
So what the firs i do to winding this ETD49 if i want make +/-70 Vdc with curent 10A?
Help me full


WTF?:D +-70V/10A it is a certainly not " smps 1kw".
Why is this transformer in pictures, such that it seems to be old winding just cut off?
In this case, the wire must be somewhere between carried out and this is :x::x: ... the wrong approach, a real pornography
 

norazmi

diysmps Senior Member
thats ok, hehehehe. Assume that u want take out full power from your trafo, ETD49. But i think that trafo u take out from TV right? So take a look at it may have gap at the center leg. Because many of TV trafo using flyback mode smps. I assume also this is the first time u wanna make hi power smps too, so to be simply ur not knowing much about the calculation for the winding, working freq and more. So with this situation u cannot guaranteed u can get your target power output from this project, at least if u can get more than 600w is good :D . Because this smps is not regulated so this is good for you to test output with only 1 round of coil with simple winding. Simple winding u can make is wind all primary first. Then continue with secondary.

As u c an see at page 1, Wally give sample of calculation for primary and secondary winding. But u have little problem there, u didnt know what material your core is, but its ok still can give it a try with simple winding. one more thing before we go too far, did you know how to wind trafo? For example primary have 2 section clockwise and anticlockwise :D .

regards.
 

gloriel

New member
is it posible to wind secondary together? for example,
first wire a ------------------ b
second c-------------------d

wind as bifilar then solder b to c as centertap, a as + d as -

pls advice, i am building this IR2153 smps, from norazmi schematic
 

norazmi

diysmps Senior Member
trafo winding

Ok look, here i put some picture how to wind trafo with simple explanation.

as u can see i`ve put 2 cable, which called A and B. if you look at the picture there is 2 point of arrow and 1 point at center trafo which is your wind direction. At start point will be cable A and B put together at trafo and wind together at same way like arrow show at center trafo and will be ending at A1 and B1. So A1 and B1 is the ending of winding from cable A and B which mean cable A is A1, and its label with A1 because its end of wind cable A, maybe before winding u need to mark with something because wrong connection will blow ur power mosfet. so same as cable B. So to make center tap connect A1 with B, or B1 with A, you can choose. If we make A1 and B is center tap(tight together), so A will be first connection, A1+B = center tap which 0 volt or we can call it GND and B1 is second connection.

A = first connection
A1+B = center = G
B1 = second connection

here to explain for primary winding, with primary winding offline smps. The center tap is exist but not connect to any point, it will be hide and cover with tape, the only connection exist is A and B1, if you refer to IR2153 smps, "A" will goes to "m" and "B1" will goes to "o". So the turn is depends on calculation. Maybe u get 6 + 6 for primary and if u wind together 2 cable at once u need to wind 6 turn, which provide 6 + 6, A=6 turn and B1=6.

For secondary winding, we will make same example and refer to the above explanation ok. So we rename is C and D, C1 and D1 which end of wind each cable. Mark it with tape or something before winding. Then start wind both cable together until finish, let say we want to make D and C1 as center tap, tight together cable D and C1 and become Gnd(0V) , and C will be first connection (first output), D1 will be second connection (second output) and refer to the ir2153 smps ,
"C" goes to "q" , "D1" goes to "s" and center tap D + C1 tight together goes to "r".

I hope this explain better, and please understand very well, one mistake of winding trafo may blow ur power mosfet and other component, ur working with high voltage u may wear some safety equipment or glass and for first testing(power up) smps please use light bulb rate 60w or 70w it will protect ur smps from blow or burn.

regards
Azmi
 

Attachments

  • trafo.jpg
    trafo.jpg
    31.3 KB · Views: 255

norazmi

diysmps Senior Member
Vorez, in halfbridge we need to have trafo without airgap, so u need to cut with circular saw or something left and right leg and should same with center leg. I already test before trafo with airgaps will provide buzz sound or something and not efficient when driving with halfbridge.
 

vorezky04

New member
Because wait a moment come diode MUR1520 i was buy a multimeter with current 20A..
 

Attachments

  • 04062012.jpg
    04062012.jpg
    313.6 KB · Views: 96
Top