- Ask a related questionWhat is a related question?A related question is a question created from another question. When the related question is created, it will be automatically linked to the original question.
Well Marks, if you drill down here and click on the circled line it will open the zip files including the .lib version - where it works or not in LTSpice is the next question,
And here is that file if you can import directly from here,
Thanks Michael,
I will try the .lib file to include it at LT Spice.
Meanwhile - I want to limit the sense voltage on pin3 and pin4 cause the external circuit could allow a higher current dropping a high voltage on Rs (in my case 10...22R).
Any problems at the INA if I use R1=R2=200...1kR with a diode in forward direction to limit the voltage to 0.7V?
Thanks & BR,
Markus
Hi guys,
thanks for the advice. Have integrated the lib file within LT spice. Thanks for the advice with your tools. Had never a look on it. But normally we use a lot the LTspice here or Pspice what I dont have.
1.) That one seems to work in principle. When V1 is positive I have no issues. An if R1 and R2 are low enough:
2.) It doesnt work when R1, R2 are getting higher. It seems there is a huge current flow into the INA?! What is the input impedance here at that sense pins? Thought pretty high resistive. I have a scenario where I have a voltage >=100V + together with the protection diode I need to limit the diode current. That is lets say in the kR-Range but that makes it impossible to sense properly cause of the voltage drop on R1,R2...
3.) It doesnt work when I use a negative voltage vor V1. I changed the sense pins that the +Pin is more positive then the -Pin for sensing. Shouldnt it work also?
To summerize:
* is the current consumption into the sense pins really existing?
* how can I protect the sense pins from a higher voltage than 500mV, 2V if not with a diode + R1,R2?
* why is it not working when I my supply is a negative voltage?
BR,
Markus
Marks,
Have a look at this forum post. This might be the current spike you are seeing?
Regardless, the resistance on each of the lines is only 5kΩ's, as shown in the datasheet (these are technically transimpedance amps, so the input impedance is small and the output impedance is large):
The addition of resistors R1 and R2 are going to significantly change the input structure here, and are not recommended with this family of amplifiers (we recommend output filtration, since the output impedance is large here). Do you require a current out response here?
ok...for the negative common mode voltage issue I`m targeting more the LMP8601.
For the positive high voltage application I`m still interested in the INA138.
Where I have now answeres to is why is the Vout higher than the Vs=5V? That shouldnt be the case right?
Markus,
I'm not quite sure what is going on in your environment, but I suspect something did not import correctly in the model. Specifically, I think the saturation point is not present. I ran your test setup as it is in TINA-TI, and came out with a different result, showing the device in saturation at roughly 1V below V+, which corresponds to the datasheet worst case condition. In your setup, I have a measured 808.06mV at the sense pins, which is indicating that the diode has turned on and is clamping the pins to that voltage:
Analyzing this number, the 808mV at the sense pins would take the place of IsRs in the output equation (what's measured at the shunt on the other side of the 100 ohm resistors is irrelevant, its what's at the pins that matters), so the expected output at the Vout pin would be .808*50k/5k = 8.08V, which is roughly what you are seeing in your simulation. That indicates to me that your model is simply outputting the actual calculated output voltage rather than showing the device entering output saturation of ~4V. That shouldn't happen, and it would seem that something in the model most likely didn't import correctly, as I do not see this behavior in the TINA model.
Markus,
Regardless of the state of V+, if the INA138 were to see a common mode of -15V, there's a chance you could damage the device, as per the abs max ratings table:
Even if the input pins were to survive the event, in either case, as the device is not rated for this type of common mode voltage, the output would not be linear, and would not be valid information to be used in the system. I would expect the device to potentially saturate against the bottom rail in response to the sudden change in the common mode, but I cannot say with certainty. The expected behavior here is not something we typically characterize, due to the fact that this is outside of the absolute maximum ratings.
Markus,
Glad to hear it. While I was digging further into this this morning, I also came across this in section 8.1.1 of the datasheet, so be aware:
So it looks like if the diode is allowed to forward bias (as well as a small amount of headroom prior to turn-on, which looks to be around .648V > .5V), the information from the output of the device will no longer be guaranteed to be accurate.
