1kW smps project (based on MicrosiM design)

CDJKuznets

New member
Silvio, wery big thanks for the help to me and time spent for me!

I don't know material of my transformer now. I think that N87.
Why you have on the picture a working frequency = 65kHz?
There can be 75?

Resistance of rds channel = 0.2 Om for w20nm50.

My stpr1520D diodes.
 

CDJKuznets

New member
My smps v2.1 by Ludo

Silvio, tell me that it isn't correct on my picture for calculation of the transformer?
 

Attachments

  • ИТ+-85V_10A_75kHz_E_55,28,21.jpg
    ИТ+-85V_10A_75kHz_E_55,28,21.jpg
    268.8 KB · Views: 94
  • ИТ+-85V_10A_75kHz_E_55,28,21_Correct.jpg
    ИТ+-85V_10A_75kHz_E_55,28,21_Correct.jpg
    262.5 KB · Views: 69
Last edited:

Silvio

Well-known member
Silvio, wery big thanks for the help to me and time spent for me!

I don't know material of my transformer now. I think that N87.
Why you have on the picture a working frequency = 65kHz?
There can be 75?

Resistance of rds channel = 0.2 Om for w20nm50.

My stpr1520D diodes.

Transformer Material PC40 or N87 are more or less the same.

Copper wire 0.5mm cannot be operated more than 68Khz due to skin depth. I chose frequency of 65Khz because of that.

I calculated the transformer for you for 1200 watts, because the switching transistors w20nm50 are only 20 amps. At this power you do not operate transistors with more than 12 amps 60% rated power.

The input capacitor need to be larger for more power. You will need min 1uf per watt across 320v with the capacitors you have you have a total of 1000uf across 320v (1000uf x 200v X 2 = 2000 in series with 1000uf x200v X 2 = 2000uf for a total of 1000uf X 400v.) For more power you need more capacitance in input.

regarding RDS ON this will make little difference in power output. The dead time must not be less than 1uS

My calculation for you for peak voltage because the amplifier will be working with that maximum voltage most of the time and smps is not regulated

The calculation in Excellent IT software is for output voltage at continuous full load and also with output inductor included. The output voltage will however decrease with full load. In an amplifier the load is not constant and will be between 30% and 60% duty cycle.

If you want more power from SMPS you have to put more capacitance on input and also change switching transistors with higher amperage like W45NM50FD .then you can get 2.5 or 3Kw, You will need to make more wire strands for more current in winding of transformer.

Regards Silvio
 
Last edited:

CDJKuznets

New member
Silvio, many thanks for the clear and developed answers for me.
As there will be free time, i will be continue work on this project.
Very big thanks!
 

P.S.RAJU

Member
can some one please tell me the winding detial of etd49 working at 70khz circuit is 1kv based on micro sim my voltage is -+60 volt dc i want the primary and secondary winding
turns thanks
 

Silvio

Well-known member
1) ETD 49 cannot go to 2Kw at 70Khz It will go a t 1.2 to 1.3 Kw at best
2) what is the material type of the transformer?
3) what topology are you using?
4) what is the mains AC input voltage?
5) What do you mean exactly by 2 step?
 
Last edited:

P.S.RAJU

Member
View attachment 6817
thanks for your reply silvio,
this is the circuit i have used ,1)the circuit is half bridge topology,2)the supply input volt is 230volt 50hz,3)transformer material is cf139cosmoferrites,
4)2step means i have made 2 different voltage ,-+60and,-+120volt dc.
5)i used igbt instead of mosfet to increase the watts. igbt no,30n60a4d
6)i want to get maximum power from etd 49,
7)so tell me what frequency should i keep to get 60volts and also the winding turns
8)the above image is my final design the circuit and design is perfect only the the transformer iam not able to do it properly

waiting for your reply
thanks p.s.raju
 

Silvio

Well-known member
QUOTE=P.S.RAJU
thanks for your reply silvio,
this is the circuit i have used ,1)the circuit is half bridge topology,2)the supply input volt is 230volt 50hz,3)transformer material is cf139cosmoferrites,
4)2step means i have made 2 different voltage ,-+60and,-+120volt dc.
5)i used igbt instead of mosfet to increase the watts. igbt no,30n60a4d
6)i want to get maximum power from etd 49,
7)so tell me what frequency should i keep to get 60volts and also the winding turns
8)the above image is my final design the circuit and design is perfect only the the transformer iam not able to do it properly

waiting for your reply
thanks p.s.raju[/QUOTE]

Hi Rayu I have worked out the windings on the data you gave me. I have posted a couple of pictures for you to see if I understand you correctly about the 2 step windings.
I have understood that the amplifier will be working most of the time on 60v and then when there is peak current draw on high volume setting a mosfet will switch the other half of the secondary to maintain the voltage. The first picture will show the winding configuration. The current in the second half of the winding will be somewhat lower due to the higher voltage. However I opted for a current draw of 8 amps for this winding and around 12 amps for the main 60v winding.

The PDF file will show all the details of the windings including number of turns, number of wires 0.5mm for different current rating (9 wires of 0.5mm for main 60v and 6 wires for 120v. You need 12 wires (18 amps) for the primary

Please note that the workout voltage on the software is written down at 50v but this is the loaded voltage with an inductor in the output included, however the peak voltage will be around 61v at the rated input voltage. You can see the real unloaded voltage peaks in the small box in the lower middle of the software. The amplifier will be working around the peak voltage most of time.

I have also included +-15v for you and also output for fan motor

If you have difficulty in understanding please ask and I will help you.

Rayu.JPG
 

Attachments

  • rayu 2.pdf
    1.4 MB · Views: 217
Last edited:

Silvio

Well-known member
thanks silvio i will try this calculation
p.s.raju

If you are using any auxiliary windings use also 0.5mm wire as in the data I forgot to change the wire diameter for these windings. However a single strand of 0.5mm is enough.
 

P.S.RAJU

Member
hi silvio thanks for your kind help i tried your winding data the voltage is coming perfectly with out load , but when i connect a load of amp of 1000watt the power supply
igbt30n60a4d blows off and the supply becomes dead.
 

P.S.RAJU

Member
i also got the smps transformer winding tool (excellent(7100) from the forum ,and some one helped to how to use it ,
in my calculation also the igbt blows off
View attachment SMPS_Transformer_Winding.pdf
the above pdf is a file found from the fourm this is etd49 winding detial out voltage is 50+-
can you tell the frequency of this detial
the above pdf data of transformer works perfectly with load also
thanks
 
Last edited:

Silvio

Well-known member
i also got the smps transformer winding tool (excellent(7100) from the forum ,and some one helped to how to use it ,
in my calculation also the igbt blows off
View attachment 6820
the above pdf is a file found from the fourm this is etd49 winding detial out voltage is 50+-
can you tell the frequency of this detial
the above pdf data of transformer works perfectly with load also
thanks

1) The switching frequency is 65Khz
2) send some scope shots of the wave forms at the gates of the igbt. (measure one gate at a time and isolate the scope)
3) load the smps gradually and use a lamp in series until you get things stable.
4) post also schematic of the circuit you are using.
5) Probably the IGBTs are blowing due to incorrect gate drive or because of noise entering the gates at high load.
6) Post some pics of your pcb from top and bottom.

Regards Silvio
 
Last edited:

welwel

New member
Protection Board Schematics and PCB
Thx to you and sir microsim for this smps design, i want to ask you sir or sir microsim
If i only build this smps based on the parts refer to your layout on first uploaded finished file, can it work sir?
cause im not good reading sch.
 

CDJKuznets

New member
Silvio, hallo. Can you help me with the checkups?
Time appeared, I continue my project smps xp 5000 (post 1321).
I have a ferrite core torroid:
PC-40 63x38x25, 4 cores.
Wire 0.1x400 and wire 0.1x600 litzendrat
Conversion frequency 120-130 kHz,
220v input AC
Output:
+/- 95v 30A (2 cores of 15A left, and 2 cores of 15A right)
+/- 24v 2A
+/- 12v 1A
Calculation in the program:

9 turn primary (9x591)
7+7 turn secondary (7x363)
2+2 turn secondary
1+1 turn secondary
It is correct?

Post of 1323. Capacitances and transistors are taken into account

Very big thanks you.
 

Attachments

  • IMAG1871.jpg
    IMAG1871.jpg
    1.4 MB · Views: 21

Silvio

Well-known member
Hello again, I made some calculations and found out that one core can give the power needed. I would rather go for full bridge at this power. I cannot read Russian and do not know exactly the parameters. I cross checked with my Excellent IT (english) and found these results

Switching frequency 120Khz
output voltage un-regulated +/-95v
output current 30A
RDS on 0.125 ohms
voltage diode drop 2.5v

According to the software it is saying that the 1 core is capable of delivering 10KW???? (rated transformer power)

However your total output power is 5KW.
I am not sure what is your smps intended for but take a look at my calculation at full bridge. The primary current is half and I guess you will be better off using 1 transformer and avoiding complications



https://imgur.com/xKIWXgM


R core calc.jpg
 
Last edited:
Top