Continue to Site

Welcome to EDAboard.com

Welcome to our site! EDAboard.com is an international Electronics Discussion Forum focused on EDA software, circuits, schematics, books, theory, papers, asic, pld, 8051, DSP, Network, RF, Analog Design, PCB, Service Manuals... and a whole lot more! To participate you need to register. Registration is free. Click here to register now.

[SOLVED] IR2110S Driver failed when suddenly increasing duty cycle to 70 percent

Status
Not open for further replies.

BoopathiS

Member level 2
Joined
May 18, 2018
Messages
44
Helped
1
Reputation
2
Reaction score
0
Trophy points
6
Activity points
484
Hi.. I'm facing one issue with IR2110S driver while driving 3 winding BLDC motor(24V,8A rating with Hall sensor feedback) using ifr3710 mosfet(Half bridge configuration).

Issue :
When varying duty cycle from 0 to 100 present with 5 percent interval, motor running fine. but when varying duty cycle(PWM) from 0 to 70 percent suddenly, IR2110S driver IC getting failed (with more than 60 percent duty cycle interval observing this effect). there is no IC heat. MOSFET not damaged.

Operating parameter :

Switching frequency is 50kHZ.
bootstrap capacitor is 47uF electrolytic with 0.1uF ceramic in parallel, Used 1K resistor between between gate and source. 47uF and 0.1uF capacitor connected to H bridge 24VDC input.

Is this due to transient voltage from G and S of MOSFET affecting driver ?

If yes, Can i use 15V TVS diode between Gate and source of each mosfet to suppress transient voltage from G and S of MOSFET to driver? Now only 1K resistor connected across G and S.

Please provide me suggestion to solve this issue...
 

Hi,

Problems with IR2110 are frequently discussed here.
Read through other IR2110 threads.

In most cases it´s a problem of bad PCB layout, wrong schematic, or wrong drive signals.
Thus: Read the datasheet. Follow the application recommendations. Read through application notes.

If this doesn´t help..
* Show your schematic (not the datasheet one).
* Show your PCB layout or a photo of your circuit, where we can see signal paths and GND paths.

Klaus
 

Please check the schematic.

driver issue.png
 

The schematic doesn't show a connection between DC- and IR2015 ground. How does it look on PCB?
 

usually too much wiring inductance around the half bridge leg can cause the mid point of the leg to go below gnd and kill the driver - try putting a bjt emitter follower pair in between to buffer the gate drive - this usually fixes it - remember to put some anti parallel schottkies across the CE of the xtors, 40V 1A.

Unless you layout is tight with minimised current loops you will get all sorts of problems...
 

Hi,

usually too much wiring inductance around the half bridge leg can cause the mid point of the leg to go below gnd and kill the driver
true.

try putting a bjt emitter follower pair in between to buffer the gate drive - this usually fixes it - remember to put some anti parallel schottkies across the CE of the xtors, 40V 1A.
For me this means: I know what causes the problem, but instead of removing the problem I add additional circuitry to remove just the symptom.

Thus my recommendation is: Read datasheet and application notes and follow the recommendations to remove the problem.

Klaus
 

When you say 70%, do you mean 70% of the half cycle that the fets can be on for?
 

The schematic doesn't show a connection between DC- and IR2015 ground. How does it look on PCB?

Hi thanks for all your reply,

Sorry i missed to mention current sensing block schematic, please find the attachement. From lower Mosfset COM pin(Source pin) connected to ground through 0.05ohm resistor(which is connected in another small board using some wire has 10cm length, due to shortage of board space). Is this return path will affect driver performance ?

Its a six layer board so top layer has three high side mosfet and Bottom layer has 3 lower side mosfet. As per application note high side mosfet and low side mosfet connected within few mm(<1.6mm).

Duty cycle 70 percent means as per hall sensor feedback, high side MOSFET will ON 70 percent of 1/50k sec duration. when increasing suddenly to higher duty cycle, driver getting damaged. observing on low side mosfet drain and source terminal some pulses. RED LED gets turn ON and stops the PWM to all half bridge low side mosfet.

Reading application note, i got return path should be less. driver COM connected to GND plane, this is same like what driver manufacturer recommended and low side MOSFET source pin connected through ground through sensing resistor which is placed in external board. this connection am having doubt. can you tell me this will cause any problem on driver ?
current sensing.png
 

the ckt as shown with the shunt R (not shown) could go into a linear state with the 2N222 not driven fully on, and thus not triggering pin 2, thus instead of pin 3 going high, it remains low, keeping the mosfet on until it goes west. The whole concept described in the ckt is not really robust enough for motor drive ...
 

Hi,

You don't show PCB layout - maybe there is a reason for it?
low side MOSFET source pin connected through ground through sensing resistor which is placed in external board
In mye eyes a "no-no".
This causes way too much inductance to GND. The voltage drop influences the gate drive signals.

The current measurement method is not suitable. There are some issues.
* wiring to external shunt may not give reliable operation
* I assume there is a lot of power dissipation
* pot adjustment range is not suitable
* pot adjustment affects RC time constant
* is it fast and stable enough?

Klaus
 

Hi,
I took some time for testing above suggested points.

I'm thinking MOSFET driver failure problem is due to current sensing circuit. i have removed current sensing circuit and extenal resistor board and connected lower side MOSFET source pin directly to ground . Now i can switch from 0 to 90 percent in 1 sec period not instantaneously.

if am switching directly 0 to 90 percent, During MOSFET switching on time observed input voltage 24V dipping by 5V. but not causing physical damage. so i provided ramp up time 1 sec (time to reach final duty cycle(90%) from Intial) and then verified functionality, its working fine.

Like same i have tested with current sensing circuit, its not working and driver getting damaged frequently.

Now with 1 sec ramp up time, i didn't observed any failure. I had more doubts on current sensing circuit and external resistor earlier, thanks for Easy peasy, KlausT to confirm it.
Current sensing circuit not giving stable operation. so am Learning now to redesign current sensing circuit which is suitable to my application and trying to avoid voltage dip during mosfet turn on for higher load current.

Voltage Dip observed on high side mosfet drain terminal during switch on to higher load(0 to 90% duty cycle) :
Voltage dip at HS mosfet drain.jpg

If you have any idea suggest me.

Once again thanks all.
 

Hi,

... Maybe the function of your current control cir uit causes the problem.

But I think it's more likely that the increased series inductance in the Mosfet Source path causes the problem.

*****
I wonder why you can' t switch immediately to 90% duty cycle.
I see no reason for this behaviour.

Klaus
 

Hi,

Yes, its due to external resistor board connection also which is causing high loop inductance. I have removed both External resistor board connection and current sensing circuit at same time and observed good improvement. I will check with only current sensing circuit and identify the root cause.
 

You finally revealed the hidden current sense circuit, but the exact overall ground and power supply wiring is still unclear. It must be taken into account if you want to understand the driver failure cause.

You reported that the driver has been damaged but MOSFETs stayed healthy. If this is true, you must have managed somehow to feed the voltage transients generated by the source resistor wiring to the driver IC, but the published schematics don't indicate how. One possible explanation is that the actual wiring is a bit different. Undershoot of the VS against COM terminal is a "popular" way to kill a bootstrap driver, review IRF AN97-3.
 

You finally revealed the hidden current sense circuit, but the exact overall ground and power supply wiring is still unclear. It must be taken into account if you want to understand the driver failure cause.

You reported that the driver has been damaged but MOSFETs stayed healthy. If this is true, you must have managed somehow to feed the voltage transients generated by the source resistor wiring to the driver IC, but the published schematics don't indicate how. One possible explanation is that the actual wiring is a bit different. Undershoot of the VS against COM terminal is a "popular" way to kill a bootstrap driver, review IRF AN97-3.

Changes done :
1) Source resistor connection to external board is removed. MOSFET Source is directly connected to ground now. Before MOSFET source pin connected to external board having two 0.1ohm series resistor.
2) removed 10K potentiometer in current sensing circuit diagram to isolate the current sensing circuit from driver.

As per application note study only, i got doubt on External source resistor circuit connection. After above discussions, i have removed the source resistor and directly connected source to ground.
 

Status
Not open for further replies.

Part and Inventory Search

Welcome to EDABoard.com

Sponsor

Back
Top