Ad620 INA for shunt measurement

Thread Starter

Jpnagaraj

Joined Mar 8, 2020
25
Hi i have using ad620 INA for shunt measurement. Actually the shunt is connected to 300v dc supply, im giving the differential shunt voltage to ad620, the issue is the output is jumping verry noisy i have sent my circuit please let me know any changes
 

Attachments

Reloadron

Joined Jan 15, 2015
6,704
Any reason you are doing high side current with a shunt rather than low side? As mentioned your AD620 has a common mode voltage which is not 300 volts.

Ron
 

Thread Starter

Jpnagaraj

Joined Mar 8, 2020
25
Any reason you are doing high side current with a shunt rather than low side? As mentioned your AD620 has a common mode voltage which is not 300 volts.

Ron
Thankyou Reloadron.... sorry i could not understand the the first part of the question...

As of the second one ad620 common mode voltage is not 300v, but that 300v ground is totally different from ad620 ground im just measuring the voltage difference between the shunt resistance so that i can get the current drawn in the 300 v line..for this the gain is not linear.
 

Reloadron

Joined Jan 15, 2015
6,704
Normally you would measure on the low side placing your shunt on the low side rather than have it at 300 V. I can appreciate your AD 620 has its own supply and does not share the same ground as the 300 volt line. However the AD 620 still has it's inputs floating at about 300 V above its own ground.

Ron
 

JohnInTX

Joined Jun 26, 2012
4,669
I have played with INA-170. That is modern day wonder. And solves the same problem.
Not this particular problem. The INA-170 is good for 60V common mode voltage, not 300V.

EDIT: Keep in mind that the amplifier measures the difference of the two input voltages with respect to the amplifier's ground. It does not measure the floating differential voltage as you would with a DVM. The maximum input voltage to ground is one of the things that determine the amplifier's common mode voltage spec.

If the 5V supply was isolated, you could reference its common to one side of the shunt but that would cause the amplifier section to ride at 300V - not a good thing. Ron’s advice to put the shunt in the low leg is what I would do, too.

http://www.ti.com/lit/ds/symlink/ina170.pdf
 
Last edited:

Thread Starter

Jpnagaraj

Joined Mar 8, 2020
25
Normally you would measure on the low side placing your shunt on the low side rather than have it at 300 V. I can appreciate your AD 620 has its own supply and does not share the same ground as the 300 volt line. However the AD 620 still has it's inputs floating at about 300 V above its own ground.

Ron
Yes but the shunt placement is not in my control the people who installed shunt have place it in the positive side and i have to just read the shunt value and give them the voltage across the shunt
 

Thread Starter

Jpnagaraj

Joined Mar 8, 2020
25
Not this particular problem. The INA-170 is good for 60V common mode voltage, not 300V.

EDIT: Keep in mind that the amplifier measures the difference of the two input voltages with respect to the amplifier's ground. It does not measure the floating differential voltage as you would with a DVM. The maximum input voltage to ground is one of the things that determine the amplifier's common mode voltage spec.

If the 5V supply was isolated, you could reference its common to one side of the shunt but that would cause the amplifier section to ride at 300V - not a good thing. Ron’s advice to put the shunt in the low leg is what I would do, too.

http://www.ti.com/lit/ds/symlink/ina170.pdf
Thankuou JohnInTX but the placement of shunt is not my choice it has been already installed i need to measure can i use INA170 with different ground from 300v shunt ground
 

JohnInTX

Joined Jun 26, 2012
4,669
Thankuou JohnInTX but the placement of shunt is not my choice it has been already installed i need to measure can i use INA170 with different ground from 300v shunt ground
Here's the deal: NO, you can't do that.
Conceivably you could power the amplifier from an ISOLATED supply referenced to the 300V supply BUT that means that the output of the amplifier will be referenced to 300V. On paper, that is not totally unmanagable but in reality, it means you have to pay strict attention to the ratings of the other devices in the system. Is this going to a uC / ADC with some readout and buttons to control it? Better make sure you have a 300V rating on the buttons, display window etc as they are part of the safety barrier between your 300V and the user. Consider what happens when you hook up an oscilloscope or PC/emulator to this thing.. it all will be referenced to 300V - don't touch anything... Even if you could find an amplifier with a 300V common mode rating, you still are connected to hazardous voltages and are presenting a serious hazard to yourself and your users. From a practical viewpoint for all of these reasons and more, I DO NOT RECOMMEND it. And the good news is you don't have to do it.

A real solution is to use an Analog Isolation Amplifier with appropriate ratings. These have a high voltage amplifier separated by some intrinsically safe barrier to a low voltage side and are commonly used for just this kind of thing. Here is some info to get you started.

https://www.analog.com/en/products/analog-functions/isolation-amplifiers.html

It hooks up like this from pp4 of MT-071 (below) using an AD-210 isolation amplifier and AD620 instrumentation amp for the shunt:
1584041374781.jpeg
Note that the shunt amplifier AD620 is powered by the isolated supply provided by the AD210 Isolation Amplifier. All of that is referred to the high side i.e. hot and dangerous but is completely isolated from the output. The whole thing is powered by a more sane +15V that is referenced to system ground which also provides save, isolated power to the shunt amp.

If you can't relocate the shunt, that's really the only way to go.

Good luck!
 

Attachments

Last edited:

JohnInTX

Joined Jun 26, 2012
4,669

cmartinez

Joined Jan 17, 2007
7,746
Assuming you are sensing current, how about dispensing with the shunt altogether and using a hall effect current sensor? No electrical contact at all with the sensed circuit, just pass the wire through. It's probably cheaper than an isolation amplifier.

View attachment 201329
Example:
https://www.magnelab.com/products/dc-hall-effect-current-sensor-ht500m-series/

Info:
https://www.nktechnologies.com/engineering-resources/current-sensing-theory/

I've used these too.
Which also begs the question, how much current do you want to sense?
 

TeeKay6

Joined Apr 20, 2019
572
Thankuou JohnInTX but the placement of shunt is not my choice it has been already installed i need to measure can i use INA170 with different ground from 300v shunt ground
It seems to me that this is a classic case for a "high side current monitor" (do web search for) IC. Just as an example (I am NOT making a recommendation of a specific device), take a look at the Microchip Technology (Supertex) HV7800 series devices. Very low cost. Easy to use.
Link to datasheet: Datasheet for HV7800
 

Thread Starter

Jpnagaraj

Joined Mar 8, 2020
25
Here's the deal: NO, you can't do that.
Conceivably you could power the amplifier from an ISOLATED supply referenced to the 300V supply BUT that means that the output of the amplifier will be referenced to 300V. On paper, that is not totally unmanagable but in reality, it means you have to pay strict attention to the ratings of the other devices in the system. Is this going to a uC / ADC with some readout and buttons to control it? Better make sure you have a 300V rating on the buttons, display window etc as they are part of the safety barrier between your 300V and the user. Consider what happens when you hook up an oscilloscope or PC/emulator to this thing.. it all will be referenced to 300V - don't touch anything... Even if you could find an amplifier with a 300V common mode rating, you still are connected to hazardous voltages and are presenting a serious hazard to yourself and your users. From a practical viewpoint for all of these reasons and more, I DO NOT RECOMMEND it. And the good news is you don't have to do it.

A real solution is to use an Analog Isolation Amplifier with appropriate ratings. These have a high voltage amplifier separated by some intrinsically safe barrier to a low voltage side and are commonly used for just this kind of thing. Here is some info to get you started.

https://www.analog.com/en/products/analog-functions/isolation-amplifiers.html

It hooks up like this from pp4 of MT-071 (below) using an AD-210 isolation amplifier and AD620 instrumentation amp for the shunt:
View attachment 201323
Note that the shunt amplifier AD620 is powered by the isolated supply provided by the AD210 Isolation Amplifier. All of that is referred to the high side i.e. hot and dangerous but is completely isolated from the output. The whole thing is powered by a more sane +15V that is referenced to system ground which also provides save, isolated power to the shunt amp.

If you can't relocate the shunt, that's really the only way to go.

Good luck!
Thankyou it was great help to me i think the analog isolation is a must and i will work on it
 
Top