I have got a board to layout...hi luka and others,
iam about to complete the board.just transformer remaining.
here are the calculations for etd 49 50-0-50 5amp each.
vin low=254
vin high=382
highest average current=po/vin low
500/254=1.96amp
no of primary turns
np=vin min x10^8/4xfxbxac
=127x10^8/4x50000x2100x2.11
=14.33 turns or 15 turns
calculating skin depth
sd=6.62/sqrtF
=6.62/sqrt50
=0.0296cm
wire dia =2xsd=2x0.0296
=0.0592cm
bare wire area
3.14xwir dia/4
=3.14x(0.0592)^2/4
=0.00275 square cm
ill be using swg wires
so 23swg=0.610mm
bare wire area using 23swg=0.00292 square cm
primary bare wire area
awp=Iin pri x sqrt dmax/j
assuming 500cirmills/amp and duty cycle of 0.95
=1.96 x sqrt0.95/500
=0.00382cm2
no of strands required for primary
=awp/swg23=0.00382/0.00292
=1.30 or 2 wires of 23swg for primary
secondary turns
nsec=1.1(vout+vd) x npri/vin min x dmax
=1.1(50+1) x 15/127 x 0.97
=6.97 or 7 turns
no of strands for secondary=3.32 or 4 strands
are these calculations ok
can i use mur 160 instead of sb160.will post pics as soon as transformer gets completed.i will be using cosmo ferrites cf139 material.
http://www.cosmoferrites.com/material_info.html
many questions to come.
thanks in advance
ravs
be a vegeterian.......
iam about to complete the board.just transformer remaining.
here are the calculations for etd 49 50-0-50 5amp each.
vin low=254
vin high=382
highest average current=po/vin low
500/254=1.96amp
no of primary turns
np=vin min x10^8/4xfxbxac
=127x10^8/4x50000x2100x2.11
=14.33 turns or 15 turns
calculating skin depth
sd=6.62/sqrtF
=6.62/sqrt50
=0.0296cm
wire dia =2xsd=2x0.0296
=0.0592cm
bare wire area
3.14xwir dia/4
=3.14x(0.0592)^2/4
=0.00275 square cm
ill be using swg wires
so 23swg=0.610mm
bare wire area using 23swg=0.00292 square cm
primary bare wire area
awp=Iin pri x sqrt dmax/j
assuming 500cirmills/amp and duty cycle of 0.95
=1.96 x sqrt0.95/500
=0.00382cm2
no of strands required for primary
=awp/swg23=0.00382/0.00292
=1.30 or 2 wires of 23swg for primary
secondary turns
nsec=1.1(vout+vd) x npri/vin min x dmax
=1.1(50+1) x 15/127 x 0.97
=6.97 or 7 turns
no of strands for secondary=3.32 or 4 strands
are these calculations ok
can i use mur 160 instead of sb160.will post pics as soon as transformer gets completed.i will be using cosmo ferrites cf139 material.
http://www.cosmoferrites.com/material_info.html
many questions to come.
thanks in advance
ravs
be a vegeterian.......
Last edited:
Off Line Switcher
What Is Your Output Power Level??
Possibly a 2-Switch Forward Converter will work for you - it is a MUCH More Robust Topology with FAR fewer Complications and you only have 2 MOSFETS To Drive.
With a 2-Switch Forward your MOSFET Voltage Stress will be the same as the Full Bridge
Steve
What Is Your Output Power Level??
Possibly a 2-Switch Forward Converter will work for you - it is a MUCH More Robust Topology with FAR fewer Complications and you only have 2 MOSFETS To Drive.
With a 2-Switch Forward your MOSFET Voltage Stress will be the same as the Full Bridge
Steve
hi luka,
i used vin low 180 vac rectified 254vdc and half of it 127vdc for half bridge
which one you go with
thanks in advance
ravs
be a vegeterian....
Attachments
oh man, so sorry, I work crazy hours this days....
so, I get 15 turns for primary @ low input also, that is ok
with 50kHz you can still use 1mm wire, no problem, its what I got, only 1mm wire
One thing you need to understand here is this: you can use more then 4A for every 1mm2 of wire you have, since this peak is not constant, also you probably will never have that low input voltage... if you know you will, I would really think about PFC at this point
So by using 2 or even 3 primary wires, since you only have 15 turns... here I would go and use 16 or 18...
say you use 18, then I would use 8 for secondary.
You can use any fast diode, mur, sb,... that is for gate drive? even 1n4148 is ok there.
So about a wire...if you like it, like me, test it, use only 1 primary wire, and see what you get, then use 2... you will then see how smps will behave...if you have scope to check... but in any case you MUST use copper, and not leave trafo empty... did do that myself a lot of times and none of them were as good and use bit more wire
so, I get 15 turns for primary @ low input also, that is ok
with 50kHz you can still use 1mm wire, no problem, its what I got, only 1mm wire
One thing you need to understand here is this: you can use more then 4A for every 1mm2 of wire you have, since this peak is not constant, also you probably will never have that low input voltage... if you know you will, I would really think about PFC at this point
So by using 2 or even 3 primary wires, since you only have 15 turns... here I would go and use 16 or 18...
say you use 18, then I would use 8 for secondary.
You can use any fast diode, mur, sb,... that is for gate drive? even 1n4148 is ok there.
So about a wire...if you like it, like me, test it, use only 1 primary wire, and see what you get, then use 2... you will then see how smps will behave...if you have scope to check... but in any case you MUST use copper, and not leave trafo empty... did do that myself a lot of times and none of them were as good and use bit more wire
hi luka,
belated happy new year to you and all.
yesterday i started powering the control section of smps.without any powe device and ac.
on giving 12 volt to sg3525 and ir2110 i get the following voltages since i have no scope.iam using 47u instead or 33uf for start up cap.
on pin 14 , 11 of sg3525 iam getting 5.39 and 5.40 volts and when i insert ir2110 i get
5.40 and 10 volts on pin no 7 and 1 or ir2110 is this normal.sg3525 voltages remain the same.
thanks in advance
ravs
be a vegeterian.....
.
belated happy new year to you and all.
yesterday i started powering the control section of smps.without any powe device and ac.
on giving 12 volt to sg3525 and ir2110 i get the following voltages since i have no scope.iam using 47u instead or 33uf for start up cap.
on pin 14 , 11 of sg3525 iam getting 5.39 and 5.40 volts and when i insert ir2110 i get
5.40 and 10 volts on pin no 7 and 1 or ir2110 is this normal.sg3525 voltages remain the same.
thanks in advance
ravs
be a vegeterian.....
.
Last edited:
I can't find this part... my guess is, that no, you can not change it, since that is dedicated ballast driver if it is
Happy new year too you to!
Now,
you can't meter pins 14 and 11 with normal meter, you could measure freq., if meter has that option, same goes for pins 1 and 7 of IR... you can't do nothing without scope or scope meter... go and find some used, old one, even if it is <<<20MHz
My guess would be that it works ok, but no way to tell if you can't see waveforms
hi luka,
i blown my 2110 and irfs.with bulb in series.everything seems to be ok.bulb blinks for a
less than a second.first i soldered the transformer and mosfets plugged through the bulb.
again bulb lights for a second.but i couldnt measure the ac voltages on the transformer
output.now i soldered the diodes and filters .the readig goes out of 600 volts dc range.
of meter . then i connected a load of 200 watt bulb on output.reading comes to be 70 volts.both bulbs were lighting continuosly.i thought because it is loaded thats why the
is lighting continuosly.i measured voltages on pin 1 and 7 f ir2110 in 20 volt dc range they came to be 5.40 and other goes out of range .i removed the series bulb and plugged directly the fuse blown it was only 1 ampere i replaced it with 3 amp now nothing happens for a second then with a little hiss new year ffireworks starts .
i have seen in many dc-ac invertors both side mosfet get the same level drive voltage.
luka could you please measure these voltages for me in your smps.since i have no cro.
ill try again.
thanks in advance
ravs
be a vegeterian....
i blown my 2110 and irfs.with bulb in series.everything seems to be ok.bulb blinks for a
less than a second.first i soldered the transformer and mosfets plugged through the bulb.
again bulb lights for a second.but i couldnt measure the ac voltages on the transformer
output.now i soldered the diodes and filters .the readig goes out of 600 volts dc range.
of meter . then i connected a load of 200 watt bulb on output.reading comes to be 70 volts.both bulbs were lighting continuosly.i thought because it is loaded thats why the
is lighting continuosly.i measured voltages on pin 1 and 7 f ir2110 in 20 volt dc range they came to be 5.40 and other goes out of range .i removed the series bulb and plugged directly the fuse blown it was only 1 ampere i replaced it with 3 amp now nothing happens for a second then with a little hiss new year ffireworks starts .
i have seen in many dc-ac invertors both side mosfet get the same level drive voltage.
luka could you please measure these voltages for me in your smps.since i have no cro.
ill try again.
thanks in advance
ravs
be a vegeterian....