Hello all,
http://sound.westhost.com/project89.htm >> the schematic of the SMPS in question.
i have used snubber in it(68r/2w+100nf).
without these the output voltage is much more than what the turns ratio permits,using these the output voltage is perfect but the resistance heats up and ultimately burns out in seconds.i then used 5W res also but to no avail.
Sagar
http://sound.westhost.com/project89.htm >> the schematic of the SMPS in question.
i have used snubber in it(68r/2w+100nf).
without these the output voltage is much more than what the turns ratio permits,using these the output voltage is perfect but the resistance heats up and ultimately burns out in seconds.i then used 5W res also but to no avail.
Sagar
your edit was seconds late.
The cap was bead capacitor and rated at ??volts.
i replaced it with polyester(philips) at 105volts.
But in either case the res burns out.
Primary voltage is given in the schematic of the SMPS.(12v square wave)
Sagar
The cap was bead capacitor and rated at ??volts.
i replaced it with polyester(philips) at 105volts.
But in either case the res burns out.
Primary voltage is given in the schematic of the SMPS.(12v square wave)
Sagar
sagarverma said:
if you don't know the ringing frequency and the peak to peak charge held by the capacitor there isn't much point to the discussion -- you can't just pull capacitor and resistor values out of a hat (which many are wont to do) --
the snubber resistor must be able to dissipate P >= C f V^2
usually a value in the low nF's is specified, you are using a value 100x greater so the dissipation factor is 100x greater.
knowing the inductance of the trafo primary and secondaries, the junction capacitance of the switching devices will give you a first approximation of the snubber values --
Dear Jackinjj
i havent taken arbitary res value.
these are mentioned in the circuit.i have made excatly that circuit.Even the switcher res and caps are the same.
regarding 68R instead of 56R,i tried 56R also but to no avail.
Sagar
i havent taken arbitary res value.
these are mentioned in the circuit.i have made excatly that circuit.Even the switcher res and caps are the same.
regarding 68R instead of 56R,i tried 56R also but to no avail.
Sagar
can you measure the ringing frequency and the transformer's primary leakage inductance? measuring the ringing frequency isn't trivial, it is going to be 10x or more times the switching frequency.
cut the capacitor value down to 22nF and your snubber won't blow out.
without a value for the leakage inductance this is just a guess, here are some values from Power 4-5-6
Leakage L Snubber Cap Snubber Res Power
10uH 2.53n 62.8R 0.32W
1uH 25,3n 6.3R 3.2W
100nH 253n 0.63R 32W
10nH 2.5u 0.06R 323W
cut the capacitor value down to 22nF and your snubber won't blow out.
without a value for the leakage inductance this is just a guess, here are some values from Power 4-5-6
Leakage L Snubber Cap Snubber Res Power
10uH 2.53n 62.8R 0.32W
1uH 25,3n 6.3R 3.2W
100nH 253n 0.63R 32W
10nH 2.5u 0.06R 323W
i have made the circuit as it is...whatever the freq comes out from that combination of Rt and Ct, i dont know.
i will change the cap and post the result
Sagar
i will change the cap and post the result
Sagar
i replaced the caps with 22nf.although the res didnt burn out but became very very hot.
replaced the caps with 33nf polyester caps and the result was good,but the res became very hot in this case too(68R/2W)
so what should be the ideal values to be used here?
should i increase the wattage ratings of the res(although 68R/10W would be difficult to find and will be costly too)
Sagar
replaced the caps with 33nf polyester caps and the result was good,but the res became very hot in this case too(68R/2W)
so what should be the ideal values to be used here?
should i increase the wattage ratings of the res(although 68R/10W would be difficult to find and will be costly too)
Sagar
Quote from the original article:
<quote>
That it does work if built as described is certain, that you will be able to achieve the same results is not. If you do not have (or at least have access to) an oscilloscope - don't even think about trying to make the supply, as it will not be possible to ensure that the duty cycle of the controller is exactly 50%, or that there is no severe overshoot or ringing at the output.
</quote>
Have you looked at waveforms?
Cheers
Rob
<quote>
That it does work if built as described is certain, that you will be able to achieve the same results is not. If you do not have (or at least have access to) an oscilloscope - don't even think about trying to make the supply, as it will not be possible to ensure that the duty cycle of the controller is exactly 50%, or that there is no severe overshoot or ringing at the output.
</quote>
Have you looked at waveforms?
Cheers
Rob
thanks for the info..but i suppose u would like to see this>>http://www.diyaudio.com/forums/showthread.php?postid=756648#post756648
i have built two before.this is the one thats creating probs and i dont feel it cant be rectified.
i havent seen the waveform....but jus like before,i took a DMM and set that to AC,saw the voltage o/p as 10v for applied 12 v
Sagar
i have built two before.this is the one thats creating probs and i dont feel it cant be rectified.
i havent seen the waveform....but jus like before,i took a DMM and set that to AC,saw the voltage o/p as 10v for applied 12 v
Sagar
So, if it worked two out of three times, you only have to figure out what you did differently this time. If the original design worked you wont fix this one by changing component values.
You have to locate the fault and rectify it (no pun intended).
Realistically though, you were lucky the first two times but you are flying blind without a 'scope to see what's really going on.
Cheers
Rob
You have to locate the fault and rectify it (no pun intended).
Realistically though, you were lucky the first two times but you are flying blind without a 'scope to see what's really going on.
Cheers
Rob
Hi Rob
how would u explain the reduction in the heat dissipiation on changing the cap value from 22nf to 33nf?
if u think there is really a prob in the pwm then i will use my pc as osciloscope to see the waveforms but thats a very boring process🙄
Sagar
how would u explain the reduction in the heat dissipiation on changing the cap value from 22nf to 33nf?
if u think there is really a prob in the pwm then i will use my pc as osciloscope to see the waveforms but thats a very boring process🙄
Sagar
I am not trying to be argumentative as I'm no expert on switching power supply design. There are several others more than amply qualified in that area.
I'm only trying to apply simple logic to your situation.
Looking at waveforms boring? That's how we learn.
Anyway your PC will be useless for looking at waveforms unless your supply is running at less than 2 KHz. Also if you don't have a proper front end interface for it, you stand a good chance of damaging it.
I'm only trying to apply simple logic to your situation.
Looking at waveforms boring? That's how we learn.
Anyway your PC will be useless for looking at waveforms unless your supply is running at less than 2 KHz. Also if you don't have a proper front end interface for it, you stand a good chance of damaging it.
- Status
- Not open for further replies.