Hello!
Is it possible to make a little bipolar 35v smps capable of delivering about 2.5A, that runs on 12vdc with a UC3844, four paralelled 2SC2625 transistors and some other components required around the transformer ?
The transformer i'll wind myself, is the winding ratio the same for a flyback type smps as for a pushpull one ? How big does the core need to be to make a 100-150w smps ?
I have this idea to use it with the 1982 elektor amp.
Is it possible to make a little bipolar 35v smps capable of delivering about 2.5A, that runs on 12vdc with a UC3844, four paralelled 2SC2625 transistors and some other components required around the transformer ?
The transformer i'll wind myself, is the winding ratio the same for a flyback type smps as for a pushpull one ? How big does the core need to be to make a 100-150w smps ?
I have this idea to use it with the 1982 elektor amp.
TL494
EDIT: UC3844 needed 16v to start so lets go with TL494 instead.
Is this possible ?
How thick wire do i need to use to get 35v 2.5A out ?
EDIT: UC3844 needed 16v to start so lets go with TL494 instead.
Is this possible ?
How thick wire do i need to use to get 35v 2.5A out ?
Nice page 🙂 But it only calculates one output winding, i need two as a amp needs bipolar power supply.
You can take into account the total output power and efficiency.
For example if you want +35V/2.5A and -35V/2.5A, then your output power is Pout = 175W.
In order to use the program, you can set, for example, the output voltage equal to 35V and for output current use
Iout = Pout/(Vout*efficiency)
With this settings, program will return a value for the primary current which is close to reality. The value for primary wire diameter could be used.
For secondary windings, you have to set Iout = 2.5A and the program will do the rest. The specified value for secondary wire diameter is a good choice.
Have fun !
For example if you want +35V/2.5A and -35V/2.5A, then your output power is Pout = 175W.
In order to use the program, you can set, for example, the output voltage equal to 35V and for output current use
Iout = Pout/(Vout*efficiency)
With this settings, program will return a value for the primary current which is close to reality. The value for primary wire diameter could be used.
For secondary windings, you have to set Iout = 2.5A and the program will do the rest. The specified value for secondary wire diameter is a good choice.
Have fun !
Something like this ?
An externally hosted image should be here but it was not working when we last tested it.
An externally hosted image should be here but it was not working when we last tested it.
At the sort of output power you're asking for, you'll probably need to run your flyback supply in continuous-conduction mode, so make sure your feedback stabilisation is happy with operation in both discontinuous and continuous mode. (The app notes for a TL3843, UC3843 etc will explain this).
Don't use a TL494, horrible old things! TL3843s cost pennies.
Don't use a TL494, horrible old things! TL3843s cost pennies.
First prototype running
Now i have built a first prototype around a UC3844, two IRFI640 and a transformer made with a emi filter toroid.
The thing gives out around 20.2 volts and can actually drive the elektor amp, tho the voltage is alittle too low but i have tried to turn the bass up with no prob, i took it quite loud and voltage only drivved to 19v, pretty impressive for a first prototype built around a emi filter toroid.
Heres some pics:
A little video:
smps and amp running
Now i have built a first prototype around a UC3844, two IRFI640 and a transformer made with a emi filter toroid.
The thing gives out around 20.2 volts and can actually drive the elektor amp, tho the voltage is alittle too low but i have tried to turn the bass up with no prob, i took it quite loud and voltage only drivved to 19v, pretty impressive for a first prototype built around a emi filter toroid.
Heres some pics:
An externally hosted image should be here but it was not working when we last tested it.
An externally hosted image should be here but it was not working when we last tested it.
An externally hosted image should be here but it was not working when we last tested it.
An externally hosted image should be here but it was not working when we last tested it.
A little video:
smps and amp running
It will be nice if you post converter's key data.
As far as I understood, you need -35V / 0 / +35V at 2.5A, so your output power is around 175W. For this power level and selected input voltage range (10Vmin - 15Vmax), the primary current has a pick value of 82A at 10V input and assuming 70% efficiency. I'm affraid that flyback is not suitable for this application. I suggest a push-pull topology.
Can you post a schematic?
As far as I understood, you need -35V / 0 / +35V at 2.5A, so your output power is around 175W. For this power level and selected input voltage range (10Vmin - 15Vmax), the primary current has a pick value of 82A at 10V input and assuming 70% efficiency. I'm affraid that flyback is not suitable for this application. I suggest a push-pull topology.
Can you post a schematic?
I hate measured amps my shorting the outputs with the amp meter and on the 2A setting it went to the bottom = more than 2A, it also gives 20.3V unloaded, if i turn the bass and volume up on the amp it only drops to 19v. This may sound impossible but its true!!
Btw according to the smps calculation site it will be 25A input with a flyback config for 35v 2.5A out.
Heres the schematic of my first ever prototype:
Btw according to the smps calculation site it will be 25A input with a flyback config for 35v 2.5A out.
Heres the schematic of my first ever prototype:
An externally hosted image should be here but it was not working when we last tested it.
Now i have made a new transformer from the core off a tv flyback transformer and it works mutch better. i accedently hit 60v per rail because i forgot to turn the vfb pot down to minimum.
This took out the diodes i used and destroyed a cap, and it may have hurt a component on the amp because everything over 20v makes it sound like a mad cat and oscillating @ 1hz. Now the thing has some power and capability of destroying components if you dont pay attention😱
This took out the diodes i used and destroyed a cap, and it may have hurt a component on the amp because everything over 20v makes it sound like a mad cat and oscillating @ 1hz. Now the thing has some power and capability of destroying components if you dont pay attention😱
There was nothing wrong with the amp, it was the emi filters that were made for mutch lower switch freqs.
Yes it is true that for 35V/2.5A at the output, the primary peak current is around 28 for unity efficiency.
Now i have built a third prototype that works very well, only issue is that when i start to reach 35v and above the amp starts sounding like *FFFFFFFFF* Why i dont know. Also the amp has the issue taking so mutch amperage in the start that i have to turn the smps vfb pot down otherwise my 12v 8A ps gets overloaded so the smps controller looses power and shuts down.
With the smps the amp also has alittle crisp treble, on some freqs the treble sounds alittle like a broken tweeter, but good bass.
Some pics:
I have changed the board alittle since i had done an error there after i took the pic:
A transformer from a 200W pc ps rewound:
And finally the schematic:
With the smps the amp also has alittle crisp treble, on some freqs the treble sounds alittle like a broken tweeter, but good bass.
Some pics:
An externally hosted image should be here but it was not working when we last tested it.
I have changed the board alittle since i had done an error there after i took the pic:
An externally hosted image should be here but it was not working when we last tested it.
A transformer from a 200W pc ps rewound:
An externally hosted image should be here but it was not working when we last tested it.
And finally the schematic:
An externally hosted image should be here but it was not working when we last tested it.
Low input voltage means MOSFET with BVDSS>3VCC available, class E resonant topology may be used -- extreme efficiency & clean waveform.
That sounds interesting, what is it ?
Now im making a new transformer that should work better than the one im using now that actually is from a pushpull topology ps, the new one is from a 200w flyback atx ps.
Now im making a new transformer that should work better than the one im using now that actually is from a pushpull topology ps, the new one is from a 200w flyback atx ps.
Now i have settled for the board from the second prototype and have descided to finalize it together with the bobbinless core from the flyback trasf as a finished product smps.
Gonna do some final tests and adjustments and then box it and call it finished.
Gonna do some final tests and adjustments and then box it and call it finished.
I have came up with a problem that i dunno what do do.
If i don't turn the vfb pot all way to zero the smps just shorts out the 12v source i use. What is causing this and what options do i have to solve it ?
Anyone having any ideas ?
This problem is preventing me from putting the smps in a box.
If i don't turn the vfb pot all way to zero the smps just shorts out the 12v source i use. What is causing this and what options do i have to solve it ?
Anyone having any ideas ?
This problem is preventing me from putting the smps in a box.
- Status
- Not open for further replies.