A question about PWM IC's

[keeney123]

Re: A question about PWM IC's

Hi.
Excuse the delay.

I lifted a leg of the resistor and had a poke around measuring in both DC and AC for any ripple.

There's no working scope here so the testing was done with the DMM and revealed nothing on the DC although did appear to show some AC ripple when measuring across the filter caps and transformer primary.

It shouldn't be the caps as they are new replacements

If I'm honest, I'm not really sure what signs I should be looking for other than low voltages, but perhaps someone can enlighten me?

Did your replace the IC? I am going by your data sheets typical circuit. Is this what you have? The bridge rectifier is the 4 diodes group diamond shape. If the resistor leg is lifted to Vin and you measure the capacitor going to the bridge rectifier at capacitor pin1 you should have a DC voltage. If no DC voltage on the capacitor check pins 1 to pins3 of the bridge rectifier for you AC input Voltage. If you have the AC Voltage then check the diodes of the rectifier.

Again, I am going by the typical circuit in your data sheet. If you are measuring with a DVDM then on the AC scale you might pick up the ripple which should be a sawtooth wave form as AC Voltage. This sawtooth waveform rides on top of the DC which is normal without the IC connected. When the IC is connected an working you should have a pulsating DC square wave. Depending on the frequency and the meter you are using you may or may not be able to read it with that meter.

[keeney123]

Re: A question about PWM IC's

Here is the front end of your sample circuit. This is the first half cycle of what happens with the AC voltage. I showed the current flow of what happens in the diodes of the bridge rectifier. When the bottom half of the cycle of AC voltage happens, the top voltage goes negative with respect to ground. Then it will proceed through the other two diodes of the bridge rectifier. That then in one cycle. In the US this happens 60 times a second. In England I believe it is 50 times a sec. The diodes are just one-way switches. The line on the diode forms a letter K which stands of Cathode. A cathode needs a negative voltage to work. So that means the back end of the blackened arrow is a Anode. A anode will need a positive voltage to work. Each silicon diode drops 0.7 Volts DC in order to forward bias the diode so current will flow.

[Crystaleyes]

Re: A question about PWM IC's

Did your replace the IC? I am going by your data sheets typical circuit. Is this what you have? The bridge rectifier is the 4 diodes group diamond shape. If the resistor leg is lifted to Vin and you measure the capacitor going to the bridge rectifier at capacitor pin1 you should have a DC voltage. If no DC voltage on the capacitor check pins 1 to pins3 of the bridge rectifier for you AC input Voltage. If you have the AC Voltage then check the diodes of the rectifier.

The IC is due to arrive on Wednesday, and as for the circuit, it has a standard single piece, four pin bridge rectifier, which is testing fine with or without the IC connected, as are the two filter caps.
I also have one of those multi-component testers and the primary all seems fine (MOV, NTC, FET, caps, diodes, resistors etc.) except that the voltage across the transformer primary is just a constant DC. No frequency, and the DMM I have is good enough to detect that.

I'm eager to see how changing the IC will affect things.

[R_J]

Re: A question about PWM IC's

Is this the same power supply that had the blown source resistor? https://www.badcaps.net/forum/showthread.php?t=105220

If it is, then that was why the ic was damaged internally.

You won't measure any "ripple" on the main capacitors in the state is in now. The main filter caps will charge and without any load, they will not discharge, so you may want to discharge them with a large resistor. at least check the voltage across them before working on the circuit to make sure they are discharged.
Once the ic is installed, the main dc voltage will not change that much, likely just be slightly lower dc value.

[keeney123]

Re: A question about PWM IC's

The IC is due to arrive on Wednesday, and as for the circuit, it has a standard single piece, four pin bridge rectifier, which is testing fine with or without the IC connected, as are the two filter caps.
I also have one of those multi-component testers and the primary all seems fine (MOV, NTC, FET, caps, diodes, resistors etc.) except that the voltage across the transformer primary is just a constant DC. No frequency, and the DMM I have is good enough to detect that.

I'm eager to see how changing the IC will affect things.

That is what you are looking for a constant DC with "no" IC in place. It is so much easier with a Oscilloscope. So with the DC voltage on the transformer it tell you that the bridge circuit and everything that comes before is working. One can actually figure out what DC voltage will be on the Primary. One would take the peak to peak voltage and divide it in half then by 0.668. I am going on a 40 year old memory on the .0.668 I believe that to be very close. That would be called average power which is the DC equivalent of a sinusoidal waveform of 60 Hz better know as 60 cycles per second.

[keeney123]

Re: A question about PWM IC's

Is this the same power supply that had the blown source resistor? https://www.badcaps.net/forum/showthread.php?t=105220

If it is, then that was why the ic was damaged internally.

You won't measure any "ripple" on the main capacitors in the state is in now. The main filter caps will charge and without any load, they will not discharge, so you may want to discharge them with a large resistor. at least check the voltage across them before working on the circuit to make sure they are discharged.
Once the ic is installed, the main dc voltage will not change that much, likely just be slightly lower dc value.

I am thinking the path for the discharge and recharge are though the diode bridge as the AC voltage becomes less the cap voltage becomes less because the cap has access to ground through the bridge diodes. The cap will become a supply voltage when the bridge voltage becomes less on the down part of each half wave of the AC voltage. So it would be worth it for someone to set up a bridge rectifier an capacitor to see with a scope if there is a discharge. It has been a while since I did any of this work and would say not to take anyone's work for it but, to actually try it in a circuit.

One could first bypass the FET with a diode and put the transformer into play. The resistor on the bottom of the FET should protect the current draw. If this was OK then one could bias the FET on to see if that works.

[R_J]

Re: A question about PWM IC's

Yes your right the ripple will not appear without a load because the capacitor needs a path to discharge and recharge.
One could first bypass the FET with a diode and put the transformer into play. The resistor on the bottom of the FET should protect the current draw. If this was OK then one could bias the FET on to see if that works.

This would result in the resistor (0.12Ω) blowing open. The fet must be driven by high frequency drive signal, Bypassing the fet with a diode? this would cause excessive current through the primary and at a minimum blow the main fuse. You do not want the transformer in a dc circuit, it must be ac, and in this case it is designed for 20~60khz

In the OP other thread, the problem was the mosfet shorted causing the source resistor to go open, when this happens the sense input to the ic which is connected to the mosfet source via R16 is subject to the full dc voltage, 165~350vdc, the ic can never handle this.

[keeney123]

Re: A question about PWM IC's

This would result in the resistor (0.12Ω) blowing open. The fet must be driven by high frequency drive signal, Bypassing the fet with a diode? this would cause excessive current through the primary and at a minimum blow the main fuse. You do not want the transformer in a dc circuit, it must be ac, and in this case it is designed for 20~60khz

In the OP other thread, the problem was the mosfet shorted causing the source resistor to go open, when this happens the sense input to the ic which is connected to the mosfet source via R16 is subject to the full dc voltage, 165~350vdc, the ic can never handle this.

There is not AC going through the transformer but, a pulsating DC waveform with a varying duty cycle. This is what the bridge rectifier is putting out DC.
The cap with bridge rectifier should not be at a greater voltage than what the designer had intended whether it is DC or a pulsating DC square wave. So when voltage is applied to its max at either DC or pulsating DC one has the same current going through the primary. If the designer did not allow for a steady state to happen then I would say he under design the circuit. Remember I am going by the sample circuit as yet I have not seen anything else posted in this thread. Looking at the other thread you referred too I see no schematic in that thread either.
Also, understand that the DC voltage has a primary winding in it and a resistor under the FET. I do not know what the FET value is but, when it conducts it will have a very low resistance. Perhaps the bottom resistor is not correct and allowing to much current to flow. Perhaps there needed to be another resistor in the FET circuit to make a voltage divider for the sense. Or it may need a resistor above the FET to lower the current. I do not know as I do not have the valves and schematic to the real circuit.

[R_J]

Re: A question about PWM IC's

The cap with bridge rectifier should not be at a greater voltage that what the designer had intended whether it is DC or a pulsating DC square wave. So when voltage is applied to its max at either DC or pulsating DC one have the same current going through the primary. If the designer did not allow for a steady state to happen then I would he under design the circuit. Remember I am going by the sample circuit as yet I have not seen anything else posted in this thread.

You must have a smps around , try bypassing the fet with a diode like you suggest.
The difference between dc resistance and ac inductance is what you are overlooking. The DC resistance of the primary is likely about 2~10Ω, apply (using N.A. line input values) 165vdc, what would be the current?

[keeney123]

Re: A question about PWM IC's

You must have a smps around , try bypassing the fet with a diode like you suggest.
The difference between dc resistance and ac inductance is what you are overlooking. The DC resistance of the primary is likely about 2~10Ω, apply (using N.A. line input values) 165vdc, what would be the current?

No actually I have none of that. I do not like working on these things. Instead I enjoy doing art. AC inductance you will have your apparent power, real power. Real power is resistive power. Apparent is power is figured by Xc. Xl and resistance. It is figured in a sinusoidal wave frequency. Real power is actually what is going through the circuit. However, we have pulsating DC. So when the circuit gets turned on it is full on not a gradual sign wave this will affect the circuit differently. The moment the voltage is applied the transformer it will see all voltage. The current will lag behind by 90 degrees in a sinusoidal waveform. With pulsating DC with a varying duty cycle I would not begin to try and figure that one. I am sure there are equations for this an long explanation. I do not wish to go into that as I have no desire to be an electrical/electronic engineer. Still the current will see the resistance of the coil and the resistance of the FET and the resistance of the bottom resistor. The real current will be resistance and this will give the real power. Phase shifting the Voltage as oppose to Current effects only the apparent power. The apparent power will always be greater than the real power.

[keeney123]

Re: A question about PWM IC's

You must have a smps around , try bypassing the fet with a diode like you suggest.
The difference between dc resistance and ac inductance is what you are overlooking. The DC resistance of the primary is likely about 2~10Ω, apply (using N.A. line input values) 165vdc, what would be the current?

So your telling me that the peak to peak voltage on the AC line is 486 Volts? We do not know what the resistance of the primary is. We do not know what the value of the FET is. We do not know what the value of the bottom resistor is. If you give me these valves I can figure out the current in the circuit at 165 Volt DC I can tell you whether the FET can handle that current. If it can not handle that current the circuit needs a FET capable of doing it. If none go up that far then the circuit needs to be re-design to accommodate a lower current. In this way one does not blow up the circuit when something goes bad.

[Crystaleyes]

Re: A question about PWM IC's

Hi fellas.

That was an informative few posts. Thanks.

As for the values, the original source resistor is unknown however looking at almost identical boards with the exact same output values, then I believe it to be 0.12Ω. The transformer has two primaries which both measure super low 0.1 ish and the original fet was a 12 60 of some sort (don't remember right now).

Given the question around the source resistor, would it not be worth first switching it on with the gate disconnected, in order to check that the chip output is fluctuating correctly?

[keeney123]

Re: A question about PWM IC's

Hi fellas.

That was an informative few posts. Thanks.

As for the values, the original source resistor is unknown however looking at almost identical boards with the exact same output values, then I believe it to be 0.12Ω. The transformer has two primaries which both measure super low 0.1 ish and the original fet was a 12 60 of some sort (don't remember right now).

Given the question around the source resistor, would it not be worth first switching it on with the gate disconnected, in order to check that the chip output is fluctuating correctly?

I would go by what R_J is telling you about the FET and DC voltage. I read through the chips information and they do all sorts of things to protect from over current and overvoltage. R_J is an active technician doing repairs and should know how these things work. My intention is not to take that away from him. My intention is to provide you with what basic theory of electronics is. With that I still believe that the mosfet in the primary winding circuit needs to handle the full DC current running through it. This is a condition that the designers are trying to avoid with all the over voltage and current protection built with-in the chip. However, we know that in your case this did not work. When Engineers I have worked with have come across such a problem they go back to the circuit and find the cause and then make the circuit more robust so the same condition is not done again.

[R_J]

Re: A question about PWM IC's

So your telling me that the peak to peak voltage on the AC line is 486 Volts? We do not know what the resistance of the primary is. We do not know what the value of the FET is. We do not know what the value of the bottom resistor is. If you give me these valves I can figure out the current in the circuit at 165 Volt DC I can tell you whether the FET can handle that current. If it can not handle that current the circuit needs a FET capable of doing it. If none go up that far then the circuit needs to be re-design to accommodate a lower current. In this way one does not blow up the circuit when something goes bad.

How do you come up with this? The line voltage in N.A. is 120vac, with a full wave bridge and a filter capacitor the DC voltage across the filter capacitor is approximately 165VDC. The primary DC resistance was just a guess, I went and checked one I have and it measured about 1Ω.

[keeney123]

Re: A question about PWM IC's

Hi fellas.

That was an informative few posts. Thanks.

As for the values, the original source resistor is unknown however looking at almost identical boards with the exact same output values, then I believe it to be 0.12Ω. The transformer has two primaries which both measure super low 0.1 ish and the original fet was a 12 60 of some sort (don't remember right now).

Given the question around the source resistor, would it not be worth first switching it on with the gate disconnected, in order to check that the chip output is fluctuating correctly?

I you lift up one leg of the resistor connected to the FET which I believe you said was the source leg and measure it then you will know it's values. The reason for the two primaries is explained in your data sheet. When working on a circuit it is important to know exactly what you are working on. To know the schematic and the components and how they interact with one another. This is a key to successfully repairing electronic so they do not come in your door more than once. I trust R_J's experience and I believe he will finish up with you in working on this circuit. If you have greater desire of the actual theory behind circuit I would suggest that you lean toward becoming an engineer. In this case it would be best to learn this in a school.

[R_J]

Re: A question about PWM IC's

Hi fellas.

That was an informative few posts. Thanks.

As for the values, the original source resistor is unknown however looking at almost identical boards with the exact same output values, then I believe it to be 0.12Ω. The transformer has two primaries which both measure super low 0.1 ish and the original fet was a 12 60 of some sort (don't remember right now).

Given the question around the source resistor, would it not be worth first switching it on with the gate disconnected, in order to check that the chip output is fluctuating correctly?

There is only one primary winding, that is the one connected between the main DC voltage and the mosfet drain. The other winding on the primary side is the feedback winding, it will supply the ic' vcc voltage (via a diode and cap) to keep the ic working after startup.
I would not disconnect the mosfet gate to check the ic output. The ic needs to have the feedback vcc voltage as well as the source voltage to operate otherwise it will go into protection.
If you are going to check the ic output you need an ISOLATED scope, so the scope must be battery operated, If the scope is line powered you need an isolation transformer to connect the power supply to. If you don't do this, as soon as you connect the scope ground you will blow the power supply main fuse and usually the bridge rectifier as well.

[keeney123]

Re: A question about PWM IC's

How do you come up with this? The line voltage in N.A. is 120vac, with a full wave bridge and a filter capacitor the DC voltage across the filter capacitor is approximately 165VDC. The primary DC resistance was just a guess, I went and checked one I have and it measured about 1Ω.

You simply can not get more power out of something that you put into it.125VAC bridge rectifier will never put out 165VDC. I do not see a voltage multiplier on the input. We have to ask is the 120VAC a RMS voltage or a Peak to Peak voltage. RMS voltage is figured from Peak to Peak voltage by divided the Peak to Peak voltage in half and then multiplying by 0.707. Average voltage which is the DC equivalent can be figured by dividing the Peak to Peak voltage in half and then multiplying by 0.678. And through algebra one can find the correlation between the two.

[keeney123]

Re: A question about PWM IC's

There is only one primary winding, that is the one connected between the main DC voltage and the mosfet drain. The other winding on the primary side is the feedback winding, it will supply the ic' vcc voltage (via a diode and cap) to keep the ic working after startup.
I would not disconnect the mosfet gate to check the ic output. The ic needs to have the feedback vcc voltage as well as the source voltage to operate otherwise it will go into protection.
If you are going to check the ic output you need an ISOLATED scope, so the scope must be battery operated, If the scope is line powered you need an isolation transformer to connect the power supply to. If you don't do this, as soon as you connect the scope ground you will blow the power supply main fuse and usually the bridge rectifier as well.

How is the ground of the primary winding circuit being risen above earth ground? Earth ground is usually connected to the neutral line at the fuse box and it is the ground of a plug.

[R_J]

Re: A question about PWM IC's

You simply can not get more power out of something that you put into it.125VAC bridge rectifier will never put out 165VDC. I do not see a voltage multiplier on the input. We have to ask is the 120VAC a RMS voltage or a Peak to Peak voltage. RMS voltage is figured from Peak to Peak voltage by divided the Peak to Peak voltage in half and then multiplying by 0.707. Average voltage which is the DC equivalent can be figured by dividing the Peak to Peak voltage in half and then multiplying by 0.678. And through algebra one can find the correlation between the two.

Do you have a bridge rectifier and a electrolytic cap of around 10~100µf with a voltage rating of 200vdc. connect them up and check the voltage. without the filter cap the raw dc output will be around 108vdc

[R_J]

Re: A question about PWM IC's

How is the ground of the primary winding circuit being risen above earth ground?

I give up!