Current sensing circuit problem

[originator]

I went over thiis plan in detail with an Analog Device engineer, and it appears I have the exact same schematic, but somehow this isn't working:

link 1 is my schematic, link 2 is the AD drawing.

http://melrosemagazineonline.com/ad628schem.jpg

http://melrosemagazineonline.com/ad628.gif

This concept was ok when I was using a BB2604 (audio op amp) in the same manner. I could get that older version to output a volt or so which was proprotional to the voltage drop across the shunt.

Now with the AD device, there is still a volt dropping across the shunt, but no change on the output of this new "fancy" IC I decided to use instead. The output is sitting a 7.84 volts , and drops maybe .02 when I strain the motor. It should be going up not down.

I tried a diffrent IC but no change.


Any suggestions on what went wrong?

Thanks

Todd Chapman

 

[itsmoked]

Your "output" being specifically pin 5?

You have checked your wiring right? (Where I botch it up).
Weird scuzmatic on the motor.. I don't understand it. Does it go to ground somewhere.

You get what when you short R4?

Short R1 see what you get.

Keith Cress
Flamin Systems, Inc.-

http://www.flaminsystems.com

 

[originator]

Yes pin 5 is the out that goes to the comparator. I have went over it a hunder times. I floated the REF and it went to a fixed 24volt output.

Shorting the 1.2 ohm shunt resitor gives the same output!

Shorting R1 gives 23 volts. Pin 1 is the 24v supply. Pin 8 is off the motor side of the shunt, and is dropping from 23.85 down to 22.4 when a heavy load is on the motor.

The output pin five with the current gain set is putting out 7.88, and DROPS .02 to 7.86 on a heavy load. This is not right as it should go up, which would make it seem the inputs are reversed rights?

I pulled out the fileter cap just to check but ne change. I changed R1 to 15k, it just dropped the output to 1.7 volts, with still only the -.01 change on a load.


Makes no sense. I pulled the comparator out too just to make sure it wasn't screwy.


I'll call them tomorrow and see whats up. I let this one slip putting an unfamiliar ic on these new boards without breadboarding first...

 

[originator]

I did just notice in a lot of application notes that they are showing -Vs where I am only using a single rail 24v and Gnd. That could be an issue I suppose.

 

[itsmoked]

Cept their battery checking example seems to be that design.

Keith Cress
Flamin Systems, Inc.-

http://www.flaminsystems.com

 

[sreid]

Originator,

I think that you have guessed the problem in that the common mode input voltages do not incluce voltages near the negative rail.

 

[originator]

http://www.analog.com/Analog_Root/s...eTools/ad628/ad628.html?diffamp=AD628 +/- 15V

Worked it out with AD that Vref has to be a posotive voltage. I had it at ground for both the pin 3 and off the Rext2 location as well. Nice working chip, programmable gain plus a filter.

Thanks for the tips.