Maker Pro
Maker Pro

Mosfet driver IR2104 High out not working

Status
Not open for further replies.

Sophie

Jul 17, 2013
11
Joined
Jul 17, 2013
Messages
11
Sir,
I am a final year M.Tech student, doing my project on grid connected inverter. My circuit is having 2 Power MOSFETs(IRF540) on same leg. For triggering, I am using Half bridge driver IR2104. There is no output signal in 'High side out' of the driver IC. But low side out is working properly. I checked everything, replaced driver IC, checked for many values of capacitors(0.01uf, 0.1uf, 1uf, 22uf, 47uf, 100uf, 2200uf etc.). But no use. I am attaching the datasheet of IR2104. Please help me in this issue by giving me technical advises.

http://www.irf.com/product-info/datasheets/data/ir2104.pdf

Expecting your intelligent response

Sophie
 

BobK

Jan 5, 2010
7,682
Joined
Jan 5, 2010
Messages
7,682
Have you grounded \SD? It will never go high unless IN is high and \SD is low.

Bob
 

Sophie

Jul 17, 2013
11
Joined
Jul 17, 2013
Messages
11
Thanks for reply. \SD is not grounded. IN is a pulse with amplitude 5V, time period 0.02sec and duty ratio 50%. \SD is high(5V). [as in timing diagram in the datasheet, for getting HO/LO, \SD should be high]

Sophie
 

Attachments

  • timing.jpg
    timing.jpg
    7.8 KB · Views: 1,445
Last edited:

BobK

Jan 5, 2010
7,682
Joined
Jan 5, 2010
Messages
7,682
Sorry, you are right, \SD should be high, not ground.

So you are not seing HI go high when \SD is high and IN is high?

Are you driving the IN with a square wave? This a bootstrap supply for the high side driver gate, and it will not operate unless it is changing constantly.

Bob
 

Sophie

Jul 17, 2013
11
Joined
Jul 17, 2013
Messages
11
It is not square wave sir. It is pulse(with high and zero state). I am giving this pulse to test the driver IC. But its not working. Only Low side out is getting. Sir you said that it won't work unless the signal is changing constantly, but I am giving pulse, it changes high- zero- high. What change should I make in the pulse (frequency?) or anything

Sophie
 

BobK

Jan 5, 2010
7,682
Joined
Jan 5, 2010
Messages
7,682
The cap should charge during a low cycle, then will supply the gate voltage for the high cycle. Try applying a series of pulses to see if it works any better. The cap should be something like 0.01 to 0.1 uF.

Bob
 

KrisBlueNZ

Sadly passed away in 2015
Nov 28, 2011
8,393
Joined
Nov 28, 2011
Messages
8,393
Hi Sophie and welcome to Electronics Point :)

As Bob says, the most likely reason for the top MOSFET not turning on is a problem with the bootstrap voltage on the VB pin (pin 8).

As long as you are driving the device with a pulse signal at about 1 kHz or higher, with a duty cycle between say 2% and 98%, the bootstrap circuit should work.

Check that your circuit matches the design in the data sheet. In particular, check that you have:
(a) a suitable diode between VCC (pin 1) and VB (pin 8);
(b) a capacitor between VB (pin 8) and VS (pin 6);
(c) a connection from VS (pin 6) and the MOSFETs.

Measure the DC voltage between VB and VS (pins 8 and 6). It should be slightly less than VCC.

If you haven't solved the problem, please tell us:

(1) What is the frequency of your driving signal?
(2) What are the minimum and maximum duty cycles of your driving signal?
(3) What is your VCC voltage?
(4) What is the part number of the diode from VCC to VB?
(5) What value capacitor are you using between VB and VS?
 

Sophie

Jul 17, 2013
11
Joined
Jul 17, 2013
Messages
11
How to charge the capacitor during the low cycle?
I tried with series of pulses. No high out
 

Sophie

Jul 17, 2013
11
Joined
Jul 17, 2013
Messages
11
Thank you KrisBlue.
Sir, now I gave the driving signal for testing purpose only. I already have
(a) a diode (1N4007) between VCC (pin 1) and VB (pin 8);
(b) a capacitor(checked with many values) between VB (pin 8) and VS (pin 6);
(c) a connection from VS (pin 6) and the MOSFETs.
I will check the DC voltage between VB and VS and reply.

(1) Now frequency of driving signal is 50Hz
(2) duty cycle 50%
(3) VCC is 12V dc
(4) diode 1N4007
(5) As I am not getting High out, I tried with many values of capacitor (0.01uf, 0.1uf, 1uf, 22uf, 47uf, 100uf, 2200uf)

Sophie
 

Sophie

Jul 17, 2013
11
Joined
Jul 17, 2013
Messages
11
And one more thing. I am operating 2 driver circuit which are connected to same dc source.
 

KrisBlueNZ

Sadly passed away in 2015
Nov 28, 2011
8,393
Joined
Nov 28, 2011
Messages
8,393
OK. VCC=12V and diode=1N4007 is fine.

For operation at 50 Hz, the bootstrap capacitor should be about 1 uF. I recommend a 50V X7R ceramic capacitor like http://www.digikey.com/product-detail/en/C3216X7R1H105K160AB/445-1423-1-ND/569089 (surface-mount) or http://www.digikey.com/product-detail/en/FK16X7R1H105K/445-8356-ND/2815286 (through-hole). That will keep the top MOSFET gate drive voltage above 10V at all times.

The capacitor charges through the diode when the bottom MOSFET is conducting, and discharges into the VB pin when the top MOSFET is conducting. You don't need to charge it manually yourself. This happens automatically while the driver is switching.

The whole circuit needs to be fully connected in order for the top MOSFET to switch. If you have disconnected part of the circuit for testing, this may prevent the driver IC from generating the gate drive voltage for the top MOSFET.

What is the supply voltage on the drain of the top MOSFET? Is that fed from your main 12V supply as well?

Please install a 1 uF component for the bootstrap capacitor, and measure the voltage across it, with the drive signal present and the MOSFETs fully connected.

Also if you have an oscilloscope, look at the drive signals on the LO and HO outputs (pins 5 and 7). The LO signal (pin 5) should be a 50 Hz, 50% duty cycle square wave that swings from 0V to 10~12V. The HO signal (pin 7) should be a 50 Hz, 50% duty cycle square wave that swings from 0V to some higher voltage, about 10~12V higher than the voltage on the drain of the top MOSFET.

If those signals are not as I described, please describe the waveforms and the minimum and maximum voltages on those two pins.
 

Sophie

Jul 17, 2013
11
Joined
Jul 17, 2013
Messages
11
Ok sir I understood that. I am checking every output terminal with Digital signal oscilloscope. I will check with 1uf capacitor and reply soon. It is a H bridge inverter circuit. Supply voltage on the drain of the top MOSFET(IRF 540) is 25V dc. Is there any problem in that?
I am attaching the datasheet of IRF540.

http://www.irf.com/product-info/datasheets/data/irf540n.pdf

And now I am using a rheostat(1k) as load of the inverter output
As it is not working, the current in the DC source increases to high value(3-4A) and voltage drops. So I used to shut down the entire circuit suddenly.
Now I think my driver IC is damaged.
How to protect the driver IC from being burned?
what is the value of the resistors at the output of HO & LO
suggest me sir

Sophie
 

KrisBlueNZ

Sadly passed away in 2015
Nov 28, 2011
8,393
Joined
Nov 28, 2011
Messages
8,393
Ok sir I understood that. I am checking every output terminal with Digital signal oscilloscope. I will check with 1uf capacitor and reply soon. It is a H bridge inverter circuit. Supply voltage on the drain of the top MOSFET(IRF 540) is 25V dc. Is there any problem in that?
No, that's fine.
And now I am using a rheostat(1k) as load of the inverter output
You don't need a load on the MOSFETs. Just make sure the rest of the circuit is fully connected up.
As it is not working, the current in the DC source increases to high value(3-4A) and voltage drops. So I used to shut down the entire circuit suddenly. Now I think my driver IC is damaged.
If the 12V supply is going into current limit, the driver IC may be damaged. If it's in an IC socket, remove it and see whether the overload disappears. If the overload disappears, the IC is probably damaged. If the overload remains, then there is a fault somewhere else in the circuit.
If the 25V supply is going into current limit, you may have a faulty or damaged MOSFET, or a short somewhere else in your circuit.

Check your circuit very carefully. Double-check the pinouts of the MOSFETs and the pin numbering of the IC.

How to protect the driver IC from being burned?
It shouldn't need any protection, as long as it's connected correctly. You can add a zener diode from VCC to GND - for example, a 15V, 5W zener diode, with its cathode to VCC and anode to GND, to protect the driver IC against high VCC voltage.

what is the value of the resistors at the output of HO & LO
Around 10 ohms should be fine.

Can you post a photograph of your circuit? That would be helpful.
 

Sophie

Jul 17, 2013
11
Joined
Jul 17, 2013
Messages
11
Thank you all. My driver circuit is working perfectly now.
special thanks to KrishBlue sir.
Problem was with the capacitor used. When I replaced 2200uf with 0.1uf, it worked.

Sophie
 

BobK

Jan 5, 2010
7,682
Joined
Jan 5, 2010
Messages
7,682
I hate to be a grump, and mean to take nothing away from Kris, who is excellent. But gave you that answer in post #6 before Kris ever entered the thread.

+ you claimed in your original post that you had already tried that value, were you lying?

Edited to add: Oh, and any 1'st semester student could have told you why 2200uF would not work.

Bob
 
Last edited:

muhammad anees

Feb 7, 2016
5
Joined
Feb 7, 2016
Messages
5
sir in that case we are needed to generate LIN signal as well using microcontroller
i want to use IR2104 but prblm occurs
 

Theevan

Mar 3, 2016
1
Joined
Mar 3, 2016
Messages
1
(1) What is the frequency of your driving signal?
1KHz
(2) What are the minimum and maximum duty cycles of your driving signal?
50%
(3) What is your VCC voltage?
200V
(4) What is the part number of the diode from VCC to VB?
ZHCS750TA
(5) What value capacitor are you using between VB and VS?
94uF

Our problem is we are not getting 200v+12v(input to driver) at gate of High side MOSFET...
Please help me to sort it out
 
Status
Not open for further replies.
Top