EEVblog Electronics Community Forum
Electronics => Power/Renewable Energy/EV's => Topic started by: DELTA67 on April 17, 2024, 05:57:41 pm
-
Hi all,
I want to test the IR2104 MOSFET driver but without the mosfets. I built the circuit attached bellow.
The simulation in PROTEUS works great but the real circuit niet!!!
The frequency is about 31KHz.
I tried loading the HO and LO with 10k resistors with no success.
I've also tried with an IR2111 (which has a bit different pinout).
Please, what I'm missing?
Thanks
-
What I've done I simply connected VB to Vcc and VS to COM (GND).
The ATtiny85 code for 31KHz PWM @ 8MHz CPU clock:
//ATtiny85 PWM in CTC Mode
int main(){
DDRB =1<<PB0; // set PB0 as output to enable PWM generation
TCCR0A = 1 << COM0A1 | 1 << WGM01 | 1 << WGM00;
TCCR0B = 1 << CS00; //prescaler = 1
TCNT0=0;
OCR0A=150; //Generating PWM of 58.9% duty cycle from OC0A
while(1);
}
Dont forget to program the LOW fuse to 0xE2 (disable the divide by eight).
-
What made you think you can connect VB to VCC?
-
From the internal circuit I think we can do that, no?
Is there any other simple way to do the test?
-
It should work I guess but it's completely not clear what is what and at what voltage levels on scope picture.
-
What made you think you can connect VB to VCC?
why not? the drive for HO is the voltage between VB and VS
-
it's completely not clear what is what and at what voltage levels on scope picture.
The scope pic is from SIMULATION.
HO is the Yellow trace
LO the Blue one
both at 5V/div (Vcc is 11V).
HO is in phase with the In(put) signal as expected.
-
Connecting the chip like that is ok if you just want to see both gate drivers operating as low-side drivers. But that won't be capable of driving an actual half bridge, which I'm assuming is what you aim to do (otherwise there's be no point in selecting a part like this).
To properly test the high side driver, you need to connect actual transistors. To start, you should power the half bridge from a power supply with a low current limit (maybe 100mA), and no other circuits loading the bridge output, such that if you make a mistake you're much less likely to damage you components due to cross conduction.
-
If you want to test the high side driver in its floating supply configuration, you will need either differential oscilloscope probes, or use the math function on the scope.
-
I just want to test the chip is Ok or not, of course it's not the correct config to drive a high side MOSFET.
My problem it's not working where theoriticaly it does!!!
The circuit is mounted on a breadboard but I think it's not an issue because the working frequency is relatively low (31KHz).
Vcc is 11V but the IC works from 10V to 20V.
-
it's completely not clear what is what and at what voltage levels on scope picture.
The scope pic is from SIMULATION.
HO is the Yellow trace
LO the Blue one
both at 5V/div (Vcc is 11V).
HO is in phase with the In(put) signal as expected.
So you did not say at all how exactly is does not operate. Also on schematic it's not clear to what SD pin is connected.
-
I just want to test the chip is Ok or not, of course it's not the correct config to drive a high side MOSFET.
My problem it's not working where theoriticaly it does!!!
I had assumed you were unsuccessful in testing the full half bridge circuit, but you never clarified that this test setup also isn't working. What are you observing on the HO and LO pins?
The circuit is mounted on a breadboard but I think it's not an issue because the working frequency is relatively low (31KHz).
Vcc is 11V but the IC works from 10V to 20V.
11V should be enough to exceed the chip's UVLO threshold.
What logic levels are you applying to the IN and SD pins? Note these often don't have TTL compatible inputs, best to drive them with 5V logic. If you don't want to use the SD pin, tie it to VCC.
-
I've never said that I want to test a half bridge! I want to test the chip it self only.
as shown in the circuit joint in the 1st post, SD is tied to 5V and the INput is from an ATtiny85 uC.
What are you observing on the HO and LO pins?
No signal, just DC.
For IR2111 I've got:
LO 11V.
HO 2V
-
What are you observing on the HO and LO pins?
No signal, just DC.
For IR2111 I've got:
LO 11V.
HO 2V
I'm still confused. Your first post you attached a scope photo. I thought that photo was from your real scope? If so, your chip appears to be working perfectly.
-
I thought that photo was from your real scope?
Sorry if I've not been clear on that point.
The scope picture is from the simulation on PROTEUS.
If it was from the real device then there's no need to open this thread!!!
Now I've changed the power Vcc to 14V, the chip get's very hot.
Can the uC drive directly the chip or I must use a buffer?
are these chips(IR2104/2111) cloned?
-
I thought that photo was from your real scope?
Sorry if I've not been clear on that point.
The scope picture is from the simulation on PROTEUS.
If it was from the real device then there's no need to open this thread!!!
Now I've changed the power Vcc to 14V, the chip get's very hot.
Can the uC drive directly the chip or I must use a buffer?
are these chips(IR2104/2111) cloned?
If it's from China, there is a high chance it's fake and even could be relabeled different unrelated IC.
-
If it's from China, there is a high chance it's fake and even could be relabeled different unrelated IC.
Yes, I think so!
Forget it, I'll skip this project for the moment until finding a genuine chips.
Thanks for all.
-
Can the uC drive directly the chip or I must use a buffer?
Actually now that I think about it, this could be your problem.
I'm pretty sure the input PWM must swing from 0V to 11V or whatever your IR2111 is connected to.
I assume your ATtiny is only connected to 3.3V? No wonder the IR2111 output wasn't switching!
In other words, you'll need to voltage translate the ATtiny to whatever the IR2111 is using. ATtiny driving an open collector with pull-up, should do the trick.
-
Can the uC drive directly the chip or I must use a buffer?
Actually now that I think about it, this could be your problem.
I'm pretty sure the input PWM must swing from 0V to 11V or whatever your IR2111 is connected to.
I assume your ATtiny is only connected to 3.3V? No wonder the IR2111 output wasn't switching!
In other words, you'll need to voltage translate the ATtiny to whatever the IR2111 is using. ATtiny driving an open collector with pull-up, should do the trick.
It basically needs to be Vcc voltage for IR2111 but IR2104 is completely fine with 3.3V or 5V logic out. Actually there is no MCU or other circuit creating switching signal required at all, as bootstrap circuit is unused you can just apply low/high level DC to its input and see if output changes.
-
I'm pretty sure the input PWM must swing from 0V to 11V or whatever your IR2111 is connected to.
Thanks for this detail !!
Yes, in fact the uC is 5V powered so it's output swings from 0V to 5V.
But the IR2104 is 3.3V compatible, I don't know why it's not working.
-
Is there a Vcc bypass capacitor near the IC? Some driver ICs can self-oscillate and damage themselves.
-
Is there a Vcc bypass capacitor near the IC? Some driver ICs can self-oscillate and damage themselves.
Yes
-
Now I've changed the power Vcc to 14V, the chip get's very hot.
If a chip like this is noticeably warm in a test like this then it's been damaged, replace it and try again.
Can the uC drive directly the chip or I must use a buffer?
The datasheet says it's 3.3V compatible, but also gives a max VIH of 3V. IMO that makes 3.3V too marginal, 5V should be plenty though. The drive current strength of ordinary MCUs should be plenty though.
are these chips(IR2104/2111) cloned?
It's possible, but I doubt that alone explains what you're seeing. Most cloned/counterfeit chips actually do function similarly to the genuine article. At the very least they should not exhibit clear signs of damage (like self-heating in your case).
-
you'll need to voltage translate the ATtiny to whatever the IR2111 is using. ATtiny driving an open collector with pull-up, should do the trick.
YES, it works with the IR2111 @ 11V and no warming, Thanks!
I've read the datasheet many times but I've never paid attention to the IN(put) level details. I thougt it's 3.3V compatible like the IR2104.
For the IR2104 I get only the LO output working with or without the level shifter.
No signal on the HO output.
-
you can just apply low/high level DC to its input and see if output changes.
A very good idea!
Here's the results @ 11V:
1- For the IR2111:
IN = H (11V) ===> HO = 11V, LO = 0V
IN = L ===> HO = 0V, LO = 11V
2- For the IR2104 (SD=5V):
IN = H (5V) ===> HO = 11V, LO = 0V
IN = L ===> HO = 2.9V, LO = 11V
-
Looks working fine.
-
Looks working fine.
Yes, the only downside in the DC test is that the HO output of the IR2104 don't goes low down to 0V. the low level of this output is about 3V.
in the AC test there's no signal at this output as I said before (it stays high or low, I don't remember the measurement I've done) .
-
Something is still wrong with the IR2104 then. Its outputs should pull strongly all the way down to ~0V.
If you tie SD to GND, does that cause both outputs to pull low?