H bridge Problem

[keets]

Please read +24v instead of -12v.
I am measuring 8.16v across the two leads of the motor.
I am getting 8.30v at out1 & 0.15v at out2.The upper mosfet getting very hot while lower remains at more or less at room temp. There is drop of 10v across the drain/source of upper mosfet.Is it that upper mosfet is not getting on properly.
what would be the ideal gate-source voltage for upper/lower mosfets to get on?also ideal voltage between vb-vs?

 

[ronsimpson]

Please read +24v instead of -12v.

There are two spots "+12". Are they now +24V or are they +12V?

 

[keets]

yes they are 12vonly H bridge supply is 24v

 

[ronsimpson]

If I had a oscilloscope I would look to see if the supply voltage on the top FET driver is collapsing. The voltage on pins 5 to 6 must not discharge when the top FET is on. It is not easy to measure this voltage.

Again: I would place a 10uF cap across the two little 0.01uF caps.
Your switching frequency is 1/10 to 1/50 of what I use. This means the FETs must be on 10 to 50 times longer. The two caps should be 10 to 50 times larger than normal.

Do you have a oscilloscope? The top FETS should pull to 24 volts. I think they start out pulling up hard then drop back to about 12 volts before the cycle is over.

 

[keets]

Yes I have. out1 shows continuous 12v dc while out2 shows sawtooth for on time of the magnitude of 5V w.r.t bridge gnd.
connected 10mf cap. between vb-vs. 39e gate resistance

 

[ronsimpson]

Yes I have. out1 shows continuous 12v dc

Not good.

while out2 shows sawtooth for on time of the magnitude of 5V

You need a bigger cap!

w.r.t bridge gnd.

What?

connected 10mf cap. between vb-vs. 39e gate resistance

You need a bigger cap! Please add more cap! See picture. Much more cap!

 

[johansen]

10Khz is plenty fast enough for a 10uF cap.

there must be something else wrong.

for example you should not see a sawtooth waveform. you should see the mosfet turn on for some time and if there is something discharging the bootstrap capacitor then you will see the fet turn off after some time.

I have been able to use a 100uF cap on an IRS2453 all the way down to ~10-15Hz before running into problems with the bootstrap cap discharging.
even if you have two orders of magnitude more leakage current you should have no problems at 1Khz, muchless 10.

 

[ronsimpson]

Johansen,
I don't think he is using a 10uF cap. I think he is using little ceramic cap. 0.01uf.
If the cap is too small the top FETs will run hot and you could easily get a sawtooth ramp waveform as the top driver stops working.

 

[ronsimpson]

The data sheet on page 10 shows the "top FET driver" at room temperature draws 125uA typical, (225uA worst case).
Assuming the MOSFETs are not damaged. Assuming the ir2110 is not damaged. Assuming the ir2110 is not hot.........
If the ir2110 was not driving a MOSFET, the voltage on the top driver will decay at a rate of 225uA/0.01uF and for a time of 100uS. That said; I think the little cap should work. Form experience if the top FET is not on well you will have a sawtooth waveform and hot top FETs.

The original drawing had errors, Some of the MOSFETs were backwards! This might kill the gates of the MOSFETs.

 

[ronsimpson]

What does PWMforward and PWMreverse look like?
>I think one is 75% square wave and the the other is 0.
>Then for reverse it is 0 and 75% square wave.
>Are you driving both at 25% and 75%?
>Is one not driven, 0 or 1 all the time?

Maybe it time to see if the IR2110 is OK.
Remove the MOSFETs.
Look at the bottom gate drive. good waveform?
Short VS to ground.
Look at the top gate drive. good signal?

Next add back the MOSFETs but with no motor.
The two outputs= ?? 0 volts and 24 volts?

 

[keets]

I removed the MOSFETSs and saw the waveforms as you said.The bottom gate signal is clean square wave of amplitude of 12V as per the duty cycle.
Also the top waveform is also a square wave amplitude of 11V of the programmed duty cycle with the upper part of the waveform dropping slightly by 1V as it progressed.
Then I connected the MOSFETs again without motor. Output1 was at 0 V but the output2 originally at 12V slowly discharged to 10v and remained steady.
Then I shorted the upper MOSFET'S D-S & connected resistive load res of 56 ohm across the motor terminals. out1 show 24v while out2 waveform was switching between 0 v & 2v max i.e squarewave of 2v amplitude having the desired PWM frequency.Then I tried with fresh MOSFET same result was found.
Ishorted lower MOSFET's D-S . Both out1 remained at 0v. & out2 at 3.48v

 

[ronsimpson]

I can not see that you are doing.

Please start over.
Please send your schematic with the errors corrected.
Send wave form on the two inputs.
----____-----______------
------____-------_____----
What MOSFETs are you using?
Send picture.

 

[keets]

View attachment 65545View attachment 65543Sir here are few photosView attachment 65533View attachment 65533View attachment 65534View attachment 65535View attachment 65536View attachment 65537View attachment 65538
i AM USING IRF540N

 

[ronsimpson]

It is hard to trouble shoot when the schematic is wrong.
The schematic is wrong.
The PCB is not the same as the schematic.
The PCB: how does the Source of U18 and U19 get to ground? PCB question. R30??

On the oscilloscope traces. Please show both inputs at the same time!

What type of resistor is R30? Why R30?

 

[ronv]

Ir2110

Maybe this will help you guys get on the same page with the pictures.

 

[ronsimpson]

Thanks RonV.
I hope that helps him.
I can't get from him if he is turning on the top and bottom FETs at the same time. I do understand 75/25. I can not get the phase between the two inputs. If the two inputs are in phase he will not get a output. (much like what we see)
It appears the top FETs do not turn on. I can not see of the little diodes are in backwards.
The schematic is not what he built! There is missing information.

 

[ronv]

I bet your right. It may just be no activity on one side while the other is PWming.

 

[ronsimpson]

How the micro drives the two inputs is very important. I can not get a clear picture of that.

It is possible to turn on the top and bottom MOSFET at the same time. This shorts out the power supply. PWM both inputs.

I think one input should be low. This should turn off (top left and bottom right). PWM the other input (PWM top right and bottom left).
To reverse the motor reverse the two inputs.
------------------------------------------------
Also there are hardware issues. There are parts on the PCB, not in the schematic, that will effect how the bridge works.

 

[keets]

There maybe some drawing mistakes as I am very poor in it that you must have noticed. I don't have dual trace oscilloscope with me so I can't show you both waveforms simultaneously.But what I am not able to understand is that when I connect only one MOSFET in the circuit i.e lower one U19. & shorting drain and source of upper MOSFET U17., instead of getting same PWM waveform with increased magnitude I am getting the waveform as in the picture 8 from left. That shows that the MOSFET is not turning off completely
View attachment 65558picture showing PWM waveform from PIC to Lin of IR2110.
View attachment 65559 Picture shows low output at pin 1 of IR2110
View attachment 65560 Picture shows waveform at drain of U19 i.e out1 out2 shows 24v.
Please give your remarks over it.

 

[johansen]

you will have to make a new circuit board, or cut a few traces per half bridge and put some jumpers in.

the first thing i noticed is the boot strap capacitors are connected wrong. the negative side of the capacitor has to be connected to the source of the high side mosfet and the appropriate pins of the chip, you have the capacitor connected to ground.