Original question:
Hi Darryl,
Current can flow both ways, you may be violating the absolute max rating, please check the following table and ensure you aren't exceeding the specs.
Best,
Mallika Senthil
Hi Darryl,
Firstly, the INA219 isn't suitable for a PWM operation, I'd recommend the INA240 or INA241 instead. Secondly, if you'd like to continue using the INA219, connect it in high-side or low-side instead of in series, it may work then.
If this doesn't resolve your question, please respond with schematics, scope shots or a block diagram of your setup. You can add messages by clicking on the "insert" button.
Best,
Mallika Senthil
Hi Mallika To an electronics dummy , you would have no idea that the INA219 could NOT be used with PWM... Yet it does work... as you can see 
in the set up the INA219 is set in the positive leg to the coil on the load side of the PWM driver thru to the proportional coil and back to the PWM driver... without doing anything to the wiring, the three position switch on the right hand side is switched over which would drive the current backwards thru the INA219 and immediately smokes the chip!
I have looked at the INA240 and yes it makes a special point of it being for PWM and I have ordered a few to try...
I have verified the current readings from the INA219 with my Fluke DMM and the readings are compatible so I think it's fare to say that the application of the INA219 in this case is okay as long as you don't reverse the current despite the data sheet saying that it can!
When I get the INA240's I'll let you know how they go
Regards
Darryl
Hi Darryl,
Thank you for the video and image of your setup.
From your setup and description, you appear to be connecting the INA219 in line with the PWM signal. As mentioned above, it won't work for PWM because when you try to reverse the current, you will violate the common mode voltage ( -0.3V) and this leads to the device smoking up.
The device works when PWM is forward biased (current flowing as shown in red), but when you switch of the PWM signal, the residual current will attempt to flow in the blue path. In this case Vcm will be that developed across the bottom left diode = -0.7V. This is below the -0.3 abs max value, hence the device breaks.
As a solution, you can move the INA device to either of the two locations shown in yellow, making it high side or low side. Now, the Vcm will either be referenced to supply or ground, preventing violation of Vcm specs.
If you'd prefer to keep the INA in line with the motor, then the INA240 is your solution.
For more information, please refer to the following app note, specifically section 7:
https://www.ti.com/lit/an/sboa536/sboa536.pdf
Best,
Mallika Senthil
Hi Darryl,
I'm happy to answer your questions. The "semi-circles" with diodes are mosfets with body diodes. They represent your PWM switch.
Best,
Mallika Senthil
Hi Mallika
Thanks for the prompt response, that makes a bit more sense...
This morning I have been investigating the functionality of the PWM driver I am using and made some interesting discoveries to find it is basically as below
In between the mosfet and the motor we have a DPDT selector switch with the OFF position and the voltage table below of the motor when FWD or REV was selected was initially VCC on both legs at zero speed THEN, as you wind the potentiometer up to increase the speed the voltage falls away to zero whilst the input voltage holds at VCC which is exactly as you would see according to above schematic.
I did not fit the INA219 for this test because I did not want to cook another one but the question back to you if I may please, does this verify the function of the INA219 and the fry ups I have experienced? From what I see on my DMM and that the INA219 is bi-directional I cannot see what PWM has got to do with the failures...
A close up photo of the PWM driver you will see the 555 clock, LR7843 mosfet the MBR1045CS back to back diode
Trust this further information helps
Regards
Darryl
Hi Darryl,
Thank you for sending the oscilloscope videos, they make the debug so much easier. When you reverse the switch, you can see the SDA line is dropping to a very negative value, violating abs max of (GND - 0.3) V. You're not smoking the INA219 at 5V, but if you keep operating the device in this manner, it will be damaged eventually. The device smokes at 24V because SDA is probably dropping to a even more negative voltage, that's causing an immediate failure.
Now in case of the INA240, it is built to handle these negative spikes so you shouldn't see the same issue.
Best,
Mallika Senthil
Sorry I didn't login first...
Verified this is the way the chip is connected...
Now I'll run the 24V PWM reversible driver thru it and record some oscilloscope results for your information.
Regards
Darryl
Hi Darryl,
Sorry to hear you're still facing the issue on the INA240, could you please check how large the negative spike is and whether it is within -6V (the absolute max spec).
Best,
Mallika Senthil
