uCurrent Gold problem

[ANTALIFE]

Hello

I recently got the uCurrent gold in hopes of using it to characterise some solar cells; I know that you can use a parameter analyser to do the same, but I just wanted to see if I could make an analyser board myself.
Anyway, the circuit schematic of the uCurrent gold (

) shows that the -ve pin of the current input (J2) is connected to the -ve pin of the voltage output (J4) via the virtual ground (VGND). This is a problem for me, as with my setup (

) I am connecting the output of the uCurrent gold to an Arduino board via analog pin 1 (A1), and am measuring the voltage across a rheostat using the same board on analog pin 0 (A0). This setup forms a short effectively lowering the developed voltage across the rheostat to 0V; meaning that when ever J4 is connected to the ground of arduino, I can't measure the developed voltage across the rheostat.
Now, I'm thinking that it's possible to solve this problem by using a buffer amplifier on the J4 output (or maybe the ground of the rheostat); or even buffer both J3 & J4.
But my sugestion on the future versions of the uCurrent is that perhaps its a good idea to isolate J2 from J4; or what do you fellas think, is there another way to solve this issue?

Cheers, Anton

[mij59]

Use a inverting opamp, referenced to J2.

Amplification of the opamp depends on the voltage of the solar cell.
The input of the inverting opamp is negative pole of the solar cell.
The output of the inverting opamp is connected to the arduino A1 pin.
The opamp must be  rail to rail, and capable to operate on a supply voltage of 3V

 

[ANTALIFE]

Sorry am a bit confused, is this what you are suggesting?

[mij59]

Sorry am a bit confused, is this what you are suggesting?

Almost, put a resistor between the inverting input of the opamp and the negative pole of the solar cell.

[ANTALIFE]

Yup forgot that, but am still confused, what is the output of the inverting amplifier going to be relative to? Could you please draw a circuit diagram of what you mean?

[mij59]

Yup forgot that, but am still confused, what is the output of the inverting amplifier going to be relative to? Could you please draw a circuit diagram of what you mean?

The output of the opamp is referenced to J2 and J4 witch is GND on Arduino.
 
The schematic is correct, just remove the connection from J2 to A1 on arduino.
The supply voltage on the opamp should read +1.5V and -1.5V if you want to use the lithium battery.

[ANTALIFE]

Ok, cheers for that, so like this:

Other question, how would I go about making the inverting amplifier single sided/supplied?

[ANTALIFE]

Ah no worries mij59, have just used a two cell 7.4V LiPo battery to make the -ve rail.
The inverting amplifier works like a charm Though there are a few what look to be noise related issues that I have to figure out.
Also would be interesting to see Dave's opinion on this.
Anywho, thanks for the awesome help 

[mij59]

Ok, cheers for that, so like this:

Other question, how would I go about making the inverting amplifier single sided/supplied?

Even for a rail to rail opamp the output voltage can't be exactly zero.

Looks great, only 100 Ohm is a bit  low, would go for 1k Ohm.

BTW, no inverting amp is needed if the load is in the positive lead of the solar cell.

[ANTALIFE]

Ah righto, why can't the output be zero? I though with a gain of -1 the output is equal to the -ve of the input voltage; so shouldn't a 0V input give a 0V output?
Yuppo, though the same about the resistances, changed to 100k.
And roger that, took me a while to figure out why you recommended using an inverting amplifier lol

[mij59]

Ah righto, why can't the output be zero? I though with a gain of -1 the output is equal to the -ve of the input voltage; so shouldn't a 0V input give a 0V output?

The actual properties of the opamp important, a opamp does't comply to a simplified mathematical model.
Take a look at the data sheet of e.g. the MAX4239, things like Voltage Output Swing etc.

[ANTALIFE]

Ah yup, is that where offset voltage comes in?

[rs20]

As an aside, this isn't a design flaw because:
a) You're using this part in a way that wasn't intended (intention = battery powered).
b) To make it work the way you were hoping would involve considerably more cost/space/power consumption.

To me, it seems like a better solution to all this is to use a isolated DC-DC to just power the uCurrent. Then all your problems go away, you can interface your current output and measure the voltage (via a simple voltage divider) to your Arduino with no need for op-amps at all. Referenced to VGND, of course (which is connected to both negative terminals of the uCurrent).

[ANTALIFE]

As an aside, this isn't a design flaw because:
a) You're using this part in a way that wasn't intended (intention = battery powered).
b) To make it work the way you were hoping would involve considerably more cost/space/power consumption.

To me, it seems like a better solution to all this is to use a isolated DC-DC to just power the uCurrent. Then all your problems go away, you can interface your current output and measure the voltage (via a simple voltage divider) to your Arduino with no need for op-amps at all. Referenced to VGND, of course (which is connected to both negative terminals of the uCurrent).

What do you mean by isolated DC-DC?

[mij59]

Ah yup, is that where offset voltage comes in?

Also things like:

Offset voltage
Offset voltage drift
Bias current
Common mode voltage
Max. supply voltage
Max. output current
Max. die temperature
Bandwidth
Slew rate

[rs20]

What do you mean by isolated DC-DC?

Like this: http://www.analog.com/media/en/technical-documentation/data-sheets/ADuM6000.PDF
Or this: http://www.ebay.co.uk/itm/1W-UNREG-DIP-3-3V-IN-3-3V-OUT-DC-to-DC-Converters-Isolated-Board-Mount-/281610496943?pt=LH_DefaultDomain_3&hash=item41914b2baf

You put 3 or 5V in one end of it, and you get 3.3V (or whatever) at the output, but completely galvanically isolated from the input. So you can pretend that the output is like a 3.3V battery; completely isolated.