Switching Supply (sorta) Problems

[codex653]

k so i just finished building i guess what could be called a single switch forward converter. Pretty simple stuff, the only problem is that the voltage stresses on the transistor are pretty high. Now, i've posted the schematic i'm using at the bottom.
the PWM is coming from this circuit here:

**broken link removed**

i've used the circuit before on a motor and it seems to work pretty good.
using some simulations (since i don't have a scope), the pulse duration (on time + off time) is 20.15µS...i don't know what frequency that is...but the cap i use is pretty close to the 800pF i used in my simulations. i think its like 832pF or something on the timer?

now here's where the problem comes in..i was testing out the switching supply on the transformer (which i wound myself from a core taken from a 400watt PSU), and at 12vdc, it was drawing 3.14A! not good if i'm gonna eventually want to use this with mains voltage! but i believe all that happened after the mosfet broke...cause before the mosfet broke, i had 6vdc (pulsed with the 555) on the primary and i was getting around 0.5vac on the secondary (yah i know i still have to tweak the transformer turns ratios)....i shorted the secondary and was able to get around 600mA, so atleast i could draw current ...then i started turning up the voltage and that's when the problems started..I heard something go "click" and then the mosfet started to heat up rapidly and the transformer core started vibrating with a 60hz hum....i then stopped everything, disconected the mosfet and checked for any shorts....I found that the Gate-Drain resistance was at 12 ohms. That's not supposed to happen! So...does anybody have some ideas as to what went wrong??

 

[crutschow]

What you are building appears to be a flyback converter.

It's no surprise the transistor has high voltage stresses. The problem is the inductor L. It has no useful purpose in a flyback circuit, and perversely will generate a large inductive spike across the transistor each time the transistor turns off, which is likely what zapped it. Such an inductor can smooth the output of a rectified DC supply but here you want minimum inductance between the filter caps and the transformer.

 

[codex653]

ahhh really? ok! i knew that it would generate a voltage spike, but i had thought that the Diode/resistor across the primary would help suppress that ....but yah i had put the inductor in there in order to smooth the dc supply voltage

 

[ericgibbs]

ahhh really? ok! i knew that it would generate a voltage spike, but i had thought that the Diode/resistor across the primary would help suppress that ....but yah i had put the inductor in there in order to smooth the dc supply voltage

hi,
I would also suggest that you connect a 1meg resistor across each of those 200V caps.
This should help equalise the voltage across them

 

[codex653]

heh, i'm gonna have to go do some digging through the parts bin then lol....i technically could use 100k resistors as long as i don't exceed the power dissipation correct? the only downside would be that the 100k's would waste more energy than the 1meg

 

[ericgibbs]

heh, i'm gonna have to go do some digging through the parts bin then lol....i technically could use 100k resistors as long as i don't exceed the power dissipation correct? the only downside would be that the 100k's would waste more energy than the 1meg

Use Ohms law say 300V / 200K = 1.5mA ! === to 0.45Watt total, so use 0.5W or 1W resistors.

 

[codex653]

haha well for me, i'm in the USA so my rectified mains is around 160v so a little over a half watt i guess

 

[crutschow]

ahhh really? ok! i knew that it would generate a voltage spike, but i had thought that the Diode/resistor across the primary would help suppress that

If you look at the circuit you will see that the current through the inductor is from left to right. When the transistor turns off, the inductor will try to keep the current flowing. This will generate a large positive voltage on the right side of the inductor which will then appear across the transistor to ground. The diode/resistor suppresses the transformer ringing but has no effect on that inductor spike since it is reversed biased to that voltage and also has no path to ground.

 

[codex653]

If you look at the circuit you will see that the current through the inductor is from left to right. When the transistor turns off, the inductor will try to keep the current flowing. This will generate a large positive voltage on the right side of the inductor which will then appear across the transistor to ground. The diode/resistor suppresses the transformer ringing but has no effect on that inductor spike since it is reversed biased to that voltage and also has no path to ground.

mhm! and that's why i am about to go and take the inductor out right now i'll give an update later letting you all know how it went