Continue to Site

Welcome to EDAboard.com

Welcome to our site! EDAboard.com is an international Electronics Discussion Forum focused on EDA software, circuits, schematics, books, theory, papers, asic, pld, 8051, DSP, Network, RF, Analog Design, PCB, Service Manuals... and a whole lot more! To participate you need to register. Registration is free. Click here to register now.

Please Help me in smps fet blown up problem

Status
Not open for further replies.

Praful

Member level 3
Joined
Jul 3, 2005
Messages
54
Helped
0
Reputation
0
Reaction score
0
Trophy points
1,286
Activity points
1,665
smps ei 33 schematic

Hi, i made a smps with 3842 and 9n80 it works very well at no load or at very few load (its capacity 12v is 3amp)but as soon put an induction load its temp. rises and after some time i.e. within 5min. the fet blown up . What happend in circuit ..................
 

smps ei-33

is there some one who can told me where is the problem in my smps circuit
 

fet smps digram

Hello Praful,
This could be due to the control loop not being stable over the whole load range, the inductive load will probably also have an effect on this. This can sometimes be corrected by adjusting the feedback components slightly, usually one or more of the capacitances has to be increased a little. I cannot be specific as I know nothing about what topology your SMPS is or what primary control mode it is using. It is normal to design the feedback loop for some low intermediate load but there are exceptions to this. Design of the control/feedback loop is very important, this is often glossed over in manufacturers application notes because it is not always straightforward even though there is lots of info on magnetics design. Current control requires a simpler control loop than voltage control when applied to flyback converters runnig in discontinouous mode.

FET destruction can also occur if the core of the transformer saturates, this effect can occur some time after switch on with bridge converter topologies due to flux imbalance which builds up over time, this problem is fixed by incorporating a capacitor in series with the transformer primary to eliminate the direct dc path across the FET bridge, this only applies to bridge type forward converters, not to the flyback type.

Also is the heatsink big enough on the switching FET, you could try a FET with a lower Ron the 9n80 is about 0.9R this is quite low for an 800V FET but lower ones do exist? Your FET should not really be dissipating more than about 1W in your circuit so it should not really need much heatsinking but a small one should be incorporated to keep the FET junction temperature below 150 deg C.

Regards, I hope this is some help.
Bob.
 

where can i buy a ei 28 transformer

Can you attach a schematic? I would like to take a look at it.
The problem you are mentioning may be related to insufficient heatsinking, but there could be other reasons, too.
 

why mosfet blown open failures

dear vvv,
shortly I'll attach a schematic but thanks for ur help Please kindly help me to solve my problem.........
 

smps fet application

Hello,

your SMPS is just 36 Watts......to be honest , with these low power SMPS's you can get away with using little deliberate control loop compensation whatsoever......

just get a potential divider across the output and opto feed it back with TL431/4N35

(I assume this is an offline converter)


As long as you have set your current sense resistor correctly then you should have no problem.

Try different loads and see when it blows

Have you reduced all current loops?

Is you MOSfet switching gate drive signal nice and clean?.....or does it show noise turning the fet on and off at super high frequencies.

It is usual to use R's and C's on the TL431 error amp.....in say TYPE 1 or TYPE 2 configurations...eg single pole, single zero compensation.......but in actual fact...........if you just leave these components out altogether it usually works just fine.............if anything it just tends to work in burst mode when you leave out the compensation components, which is bad for EMC reasons but thats another story.

It can actually be quite difficult to design an SMPS that goes seriously unstable...........and remeber, if you set the current sense resistor right, there will be a maximum output power that it can give and no more....you can see this by removing feedback, putting it on max load...and seeing how high the voltage rises...assuming you significantly derated your output caps....(advise wearing eye protection when you do this)

Many feedback compensation books articles don't take into account the way the opto is set up in the feedback loop............

**broken link removed**

....the above article shows that proper isolated feedback compensation of an smps using TL431 and an opto is actually very involved indeed.

and even if your control knowledge is good...how do you know what the particular exact CTR of your opto-coupler is?

for a 36W converter, its just a case of going into the lab and using your general knowledge to swap R's and C's in and out and get the transient response satosfactory.
 

smps voltage too high

Dear DrBob13, VVV and other members,
This is my smps circuit diagram Now it works fine means it gives preffered voltage but it cant give desired current output it was only few miliAmps.
Please guide me to build a smps transformer with EI-33,EI-28,EE-28,EE-33 etc. ferrite cores.
Please give detailled data for 14v-5Amp. supply with wire guage and turns....................
 

testing smps opto potentiometer pin 3 4

I see that you didn't connect the output rectifier in flyback mode (provided the transformer wiring is drawn correct). This may be a sufficient reason for overloading the switch transistor. Furthermore a bad circuit layout of a basically correct designed SMPS may cause it's failure. We can't see this from your post.

I think, the manufacturer application notes have plenty of dimension suggestions, also for the transformer. You can also buy standard off-mains SMPS transformers from several manufacturers.

The reasons for circuit failure can be visisualized by usual engineers equipment, particularly an oscilloscope and the necessary accessories, e. g. a high voltage probe and, very important, a safety isolation transformer for the mains supply. If you have no access to these tools or don't know how to handle it, it's hard to make further suggestions.
 

monitor uc3842 smps schematics

hello , what value is the resistor 50E?.....if its too high your mosfet wont get on properly

and what value is that inductor on the output....i thought you should take feedback after the first cap...not after the inductor...which should nt be a big inductor

is it discontinuous or continuous mode?.....need to know for turns deciding.

if continuous....Vout = ND/(1-D)

If discontinuous..............make sure dt = L di/ Vsec can get discharged before switch turns on again.

are you sure you wound the turns right................you gotta take it by the corkscrew rule...........secondary puts field in same direction as when primary was on.......because its flyback.

wire gauge chart tell you how thick wire for 5 amps.....

number of turns depend on your frequency

too many primary turns mean primary current cannot get up high enough in your switch's on time..........too few turns mean you primary current go up too quick.....

if too quick (<120ns) then switch not get on before it should turn off again.
 

smps tl431 output voltage changes

Can anybody tell me what is the formula or equation for making of smps transformer...........because i took a transformer from another smps and attached to my given circuit of smps it works fine and gives preffered voltage and Amps but i cant able to open it because it was taken from my neighbours audio system.....so............please tell me about transformer design .............
 

blown fet

Here as per your schematics circuit works as forward convertor.Better change to flyback.

& take feedback before inductor
 
how to check mosfet in smps

what can i do to convert it to flyback but it works fine I just want Transformer data with EE33 or EI33.............Please..........

Added after 3 minutes:

Can anybody help me to get 50Hz frequency from SG3525.............Please reply with circuit diagram for 50 - 60Hz................
 

i can´t do smps work

Hi you can change the direction of secondry winding.
in simple -connect diodes to other end of secondry.
 

tl431 compensation

hello if you want transformer design, you have to give the frequency, max duty cycle,
Vin, Vout, also, the area of your EE33 core, its satuaration flux density, its winding window size, its AL value etc etc.

im not sure what you mean you say you want 50Hz from SG3525.

On your schematic , i dont seem to acces it now...but i saw an unusual secondary winding with a high voltage cap on it..this winding didnt seem to be doing anything and i wondered your explanation of it?
 

monitor problems smps with 3842

Hello Praful,
You could have a design error in your transformer but this is unlikely if it works in the way that you have said, I think that you have just connected the windings backwards, I agree with the other correspondents to this thread, to make your circuit work as a proper flyback topology you must connect the 14V secondary winding the other way round, this is what FvM and others have been trying to tell you. The other transformer design from your friends system that works will have been wound with the correct phasing, provided that you have designed yours correctly it will probably work as well if you simply swap the two connections of the output, the ones that are connected to the diodes and the secondary ground. This would explain most of the problems that you are experiencing.

For the benefit of the previous correspondent, the capacitor is providing a high frequency transient path between the primary and secondary grounds, this is often included with a resitor in series to modify the frequency response slightly.
Best regards,
Bob.
 

switch mode flyback waveforms fet fail

First of all, the output winding connections should be reversed. This could be just a schematic problem, since the P/S works with another transformer. Otherwise, this certainly could explain the blown MOSFET.

I will try to help you with the transformer design, but it's going to take a while, since I do not have that much time and such a transformer is not too easy to design. But it's good you gave me the core size.
How soon do you need it?
 

ee33,transformer,data

OK, here is my calculation. You should keep in mind that variations are possible, depending on what compromises a particular designer makes.
I made it discontinuous because the control loop seems to be for a discontinuous flyback. But a continuous one would be a better choice.
Anyway, if you have questions, ask me.

I would like to stress the importance of safety here. If you are going to build the transfoermer yourself, make sure you use the required creepage and insulation. Don't forget, your wire exists must be insulated with sleeving of at least 0.4mm thick.

From a circuit perspective, make sure you have the required airgap in the transformer, or it may not work as expected. The best way to adjust it is to actually measure the primary inductance.

Good luck!
VVV
 

reason for smps failure

Thanks VVV,
Now I,m try for Transformer and test the results................
 

hi pot smps

Dear friends,
now i found that the bise winding has 10 turnes for the supply of UC3842 pin no 7 but now a days the voltage increases upto 50 volts and the ic blown up Please somebody give me a solution on this problem.
 

smps topologies

So did you try to build the transformer as I described? Are you sure the winding starts are correct?

Which voltage goes up to 50V? The voltage at pin 7?
It sounds like the feedback loop is not working. Did you check the output voltage?
Speaking of the loop, the optocoupler is not pulled up by anything other than the IC's pin 1. Ad a 2.7kohm resistor from pin 1 to pin 8. Somwehow I missed that the first time I checked the schematic.
Also, remove that cap from pin 1 to GND, it slows down the loop too much. The loop should be made stable only by the components on the TL431.

Another thing: if you think the IC blows because the voltage at pin 7 gets too high, then temporarily disconnect the feedback winding and power the IC from a lab supply. Make sure you start at about 17-18V and then reduce it to 12V. With the IC powered this way it will not blow up, but the MOSFET still might, if the loop is open.

Until you get this P/S working, it's also a good idea to temporarily limit the duty-cycle. Change the timing components (see datasheet) such that the max duty-cycle is about 60%. That should better protect the MOSFET.

If you have a variac, then it's a good idea to vary the input voltage slowly while watching the waveform at the drain. If you can, please attach waveforms. I can then comment on those. It's very hard for me, not seeing the actual circuit or waveforms, to provide much advice.
 

Status
Not open for further replies.

Similar threads

Part and Inventory Search

Welcome to EDABoard.com

Sponsor

Back
Top