A question about PWM IC's

[keeney123]

Re: A question about PWM IC's

Thanks.

Classic quote from the first page seems to sum up my situation perfectly


“The way to succeed is to double your failure rate.”

- Thomas J. Watson

One of my teachers once quoted Thomas Edison on inventing something new. "It's 1% genius and 99% persistence."
Edison tried thousands of experiments to come up with the light bulb. He categorized everything he did. One of those things was the foundation for a vacuum tube.
My teacher was one of the people who made the first transistor computer that Westinghouse paid $2 million to develop.

[Crystaleyes]

Re: A question about PWM IC's

So here is an insight to wire wound resistors.

https://eepower.com/resistor-guide/r...und-resistor/#

Thanks.

Classic quote from the first page seems to sum up my situation perfectly


“The way to succeed is to double your failure rate.”

- Thomas J. Watson

[Crystaleyes]

Re: A question about PWM IC's

A quick update.

I replaced the Ic and used a 0.12 ohm ceramic resistor and switched it on.

No 24V DC output and no frequency on the gate pin of the IC.

Unfortunately I had been using hooks to measure the FET voltages and even though it was switched off, I hadn't waited for the filter caps to discharge before removing them which shorted it out, so until I get a replacement FET there is not a lot else to report.

All invaluable lessons, although I am now short of ideas on resolving this.

This PSU is not really so important, so I may just take the experience learnt and move on, if a replacement FET does nothing new.

[keeney123]

Re: A question about PWM IC's

The shops were all shut on saturday due to Carnival, but are you saying that this will cause the unit to fail?
Digikey is a long way from here and the 40 USD postage is too much, so it'll have to be something else, at least for now, until I get back to Europe.

The new chip is installed, and only the source resistor remains.

Can the PSU not be switched on for a short period of time, just to verify, or is it absolutely critical to use non-inductive resistors on the source of the switching Mosfets?

So here is an insight to wire wound resistors.

https://eepower.com/resistor-guide/r...und-resistor/#

[R_J]

Re: A question about PWM IC's

Any resistor should be ok for the source resistor, being such a low value I doubt there would much inductance in even a wire wound resistor. So you should be ok.
The part number I gave was from digikey KOA Speer electronics, but any supplier should be able to provide a suitable resistor

[Crystaleyes]

Re: A question about PWM IC's

Most non-inductive resistors have a black band as the last band

In this case I would just use a metal oxide type resistor, like a MOSX2CT52RR12J 0.12Ω/2watt from digikey

The shops were all shut on saturday due to Carnival, but are you saying that this will cause the unit to fail?
Digikey is a long way from here and the 40 USD postage is too much, so it'll have to be something else, at least for now, until I get back to Europe.

The new chip is installed, and only the source resistor remains.

Can the PSU not be switched on for a short period of time, just to verify, or is it absolutely critical to use non-inductive resistors on the source of the switching Mosfets?

[R_J]

Re: A question about PWM IC's

Nice one for the heads-up on those points.

I was unware on 'non-conductive' resistors, so that is for sure good to know.

There are three different ones here, two are wire-wound, one a fat, green 5W affair and the other looking like a wire-wound carbon.
The third is a square-edged, white ceramic axial, 2W

Looking into it, some non-inductive resistors have an extra yellow band, which none of these have. I'll see if I can't get back to the shops today.

Most non-inductive resistors have a black band as the last band

In this case I would just use a metal oxide type resistor, like a MOSX2CT52RR12J 0.12Ω/2watt from digikey

[keeney123]

Re: A question about PWM IC's

Thanks for posting this website link for this type of information
Very informative information

I once took an amplifier source from them for free. At the time they actually gave you a test and they graded it. So, your welcome.

[sam_sam_sam]

Re: A question about PWM IC's

As a last thing I can point you to in understanding this circuit and others is this site.

https://training.ti.com/pulse-width-...pwm-converters

I find the free ti training course offer a good fundamental understanding.

Thanks for posting this website link for this type of information
Very informative information

[Crystaleyes]

Re: A question about PWM IC's

As a last thing I can point you to in understanding this circuit and others is this site.

https://training.ti.com/pulse-width-...pwm-converters

I find the free ti training course offer a good fundamental understanding.

Cool.

Thank you for your patience.

[Crystaleyes]

Re: A question about PWM IC's

Just a note on the source resistor, it should be a non-inductive type once you do the final repair. Also without a load on the power supply, the circuit may start but be in burst mode until a load is placed on the secondary. This is a complex, green mode ic, and the circuit operates at high frequency. The whole circuit is fairly simple as the ic does a lot of the work, I suspect and hope the ic should fix the smps.

Just a note on the source resistor, it should be a non-inductive type once you do the final repair. Also without a load on the power supply, the circuit may start but be in burst mode until a load is placed on the secondary. This is a complex, green mode ic, and the circuit operates at high frequency

Nice one for the heads-up on those points.

I was unware on 'non-conductive' resistors, so that is for sure good to know.

There are three different ones here, two are wire-wound, one a fat, green 5W affair and the other looking like a wire-wound carbon.
The third is a square-edged, white ceramic axial, 2W

Looking into it, some non-inductive resistors have an extra yellow band, which none of these have. I'll see if I can't get back to the shops today.

[keeney123]

Re: A question about PWM IC's

Interesting indeed.

Your input is appreciated

I now have the IC only these last few days haven't had the spare time to install it. Should be over the weekend

As a last thing I can point you to in understanding this circuit and others is this site.

https://training.ti.com/pulse-width-...pwm-converters

I find the free ti training course offer a good fundamental understanding.

[Crystaleyes]

Re: A question about PWM IC's

Interesting indeed.

Your input is appreciated

I now have the IC only these last few days haven't had the spare time to install it. Should be over the weekend

[R_J]

Re: A question about PWM IC's

If I'm honest, I don't really follow you when you say things like, "check out the winding of the circuit by placing a short across the FET", so I'm just gonna put in a 0.12Ω 5W resistor and see if replacing the IC brings it back to life.

From what I have found elsewhere, the original Source resistor would have been 2W.

Just a note on the source resistor, it should be a non-inductive type once you do the final repair. Also without a load on the power supply, the circuit may start but be in burst mode until a load is placed on the secondary. This is a complex, green mode ic, and the circuit operates at high frequency. The whole circuit is fairly simple as the ic does a lot of the work, I suspect and hope the ic should fix the smps.

[keeney123]

Re: A question about PWM IC's

If I'm honest, I don't really follow you when you say things like, "check out the winding of the circuit by placing a short across the FET", so I'm just gonna put in a 0.12Ω 5W resistor and see if replacing the IC brings it back to life.

From what I have found elsewhere, the original Source resistor would have been 2W.

I would not lower the source resistor from what it is intended for. That resistor was calculated along with the other components to handle so much current. If you allow more current, then the design allows then most likely you will have fireworks and at the least stress the components for an early grave.

[keeney123]

Re: A question about PWM IC's

If I'm honest, I don't really follow you when you say things like, "check out the winding of the circuit by placing a short across the FET", so I'm just gonna put in a 0.12Ω 5W resistor and see if replacing the IC brings it back to life.

From what I have found elsewhere, the original Source resistor would have been 2W.

So the designer gives you a 2Watt resistor. With a 40 percent safety factor. You take the 2W X 0.60 = 1.2W This is maximum allowable power that the designer intended for that resistor.

Using Ohms Law I sq R = P , transposing this law to Isq = P/R , Then finally the sq root of I=I

Isq = 1.2W/0.2 Ohms, Isq = 6, sq root of 6 = 2.45 , I therefore equals 2.45 amps

2.45 Amps is the allowable continuous current to flow in that circuit.

The FET should be rated for this current.

If you take the total DC voltage availably 108DC and you take the maximum current allowed a safe resistor can be calculated. 108/2.45=44.08 ohms

The problem then becomes trying to work with a lower wattage resistor than the 44.08 ohms allows.

One would have to use a 3000 ohm resistor rated at 5 watts source resistor so cost of the test resistor would not be to much.

We can now measure with a meter the voltage drop across each component and the voltage drop across the supply voltage. This then will tell the resistance of each component including the supply voltage.

[keeney123]

Re: A question about PWM IC's

If I'm honest, I don't really follow you when you say things like, "check out the winding of the circuit by placing a short across the FET", so I'm just gonna put in a 0.12Ω 5W resistor and see if replacing the IC brings it back to life.

From what I have found elsewhere, the original Source resistor would have been 2W.

I am not telling what to do on this repair. Do as you have planned.

I am trying to explain how the circuit is working and how to figure out maximum current that could flow.

I am explaining this in the sample circuit in your data sheet with your 0.2 Ohm resistor in place of the sample source resistor.

I am explaining how to calculate how much of maximum continuous current the designer has intended to flow through the circuit.

Plus, his safety value that is calculated into his design.

I am telling you how to figure out the value of a test resistor so the winding that everything from the winding on backwards can safely be tested.

Only after all the above can be understood should you test the circuit by replacing the 0.2 ohm resistor with the calculated test resistor. Not in your present circuit but in a test circuit for understanding.

Because I do not know what the FET is I cannot try to calculate how this would be turned on.

The test resistor needs to be in place to lower the current in the circuit at the maximum allowable continuous current the designer had intended for continual use.

This current will have a 40% safety factor figured by the designer.

Then putting a short across the FET will be safe to do so the current will flow allowing the circuit to be tested out.

If you do not understand what I am explaining, then do not worry about it. Just continue on with what you do understand.

[Crystaleyes]

Re: A question about PWM IC's

If I'm honest, I don't really follow you when you say things like, "check out the winding of the circuit by placing a short across the FET", so I'm just gonna put in a 0.12Ω 5W resistor and see if replacing the IC brings it back to life.

From what I have found elsewhere, the original Source resistor would have been 2W.

[keeney123]

Re: A question about PWM IC's

Nice little video. Clarifies nicely how inductors help to smooth out a pulsing circuit.

Thank you very much

Understand that this circuit you are working on is a current feed circuit. The DC voltage is constant. When the FET conducts which closes the circuit the current starts to overcome the counter Electro Motive Force of the primary winding. Once it completely overcomes the counter EMF then it is Maximum current. I believe this circuit has a greater than 50% duty cycle which makes it more difficult to control. If they were using photo optic transistor, they would have better control of the circuit. However, they claim that the optic transistor fails more often than the Mosfet. From what I have read usually the power mosfet is driven by another mosfet with-in the IC. I also read that typically the sense input has a 100K ohm resistor inside the IC. Which is going to reduce the current going in. I believe the current divider would be the 0.2 ohm as oppose to the 100K ohm internal resistor. So at MAX DC volts. A inverse ratio is set up for current. The mosfet will drop some voltage but, if we considered it as a short and 108 Vdc was at the source resistor you would have 540 ampere through the resistor and 1.08 milli ampere through the sense line. This is why they can only let the full current for a moment. If you know what wattage the source resistor is one can calculate out how much continuous current can go through the circuit. Once that current is figured then one multiplies that by 0.6 to get the actual current. Engineers always calculate a 40% factor in for protection. From the calculated 60% of maximum continuous current one can then calculate a test resistor to put in place of the source resistor and check out the winding of the circuit by placing a short across the FET.

[Crystaleyes]

Re: A question about PWM IC's

How long is a piece of string? When you say it is reading "much" higher, can you give a value?
..

Measuring with various testers it sits between 0.3 and 0.4. The only tester with .xx digits reads 0.38Ω, and that seemed too high to me

I have various value resistors here so can mix and match if necessary.

I'm starting to see the importance of the value of this resistor now.