Hi,
Attached is a simple current source for measuring DC resistance of small inductors (0-100mH). The constant current is 100mA and therefore max voltage is 100mV since I will be measuring up to 1Ω.

Using a 10-bit ADC with a Vref of 100mV would give me at least a resolution 1milliΩ. Having very little components on my PCB and waiting for any transients in the circuit to settle down, would it make this method suitable and immune to noise? If not what is the best alternative of amplifying. I will not increase the current as I plan to measure very small inductors.

The presented schematic actually worked when measuring with a DMM giving a resolution of 0.1mV.

 

The method looks fine.

I would add a diode across the inductor (anode to ground) to absorb any inductive transient when you turn off the power to the inductor.

 

It looks like it should work. If not, I've used these little guys before with a fair amount of success. The 1083 lets you adjust the gain as well, so if you want to use internal voltage references on your PIC (let's say Vdd, 3.3V), you can set the gain resistors to give you an output voltage swing of 0-3.3V instead of 0-100 mV. I've done it before and it works out pretty well.

 

Can you re-link your 'three little guys' pls, because only IC current monitor opened.

thanks again

 

Are you sure your ADC has 10 bit accuracy when the LSB is 100uV?
Jwilk13 wrote "these little guys", not "three little guys".

 

Ohh you're right. thanks for that. In fact today I tried it on my breadboard and the lowest I could get the ADC value was 20-30. With just reading for every sample without averaging the values were drifting from 20-30 (2-3mV).

I thought that it is because of the stray capacitance of the breadboard and poor connections and maybe some groundloops.

But now I started to think that I would be better off with an amplifying the signal or using the current monitor suggested by Jwilk13.

What would be the best configuration for my application. For my Vref I can use a maximum of +3.3V if i am not mistaked. As I can set +Vref = Avdd.

thanks

 

I received an email by subscription with your message;
"To resolve 100uV, you need an amplifier with input offset <<100uV.
Can you use a surface mount amplifier, of do you need a through-hole device?"
some how it did not appear on this thread.

If it is possible, I need a through hole unless it is feasible to use an adapter.

 

What is a common name for these devices so I can read about them an maybe find some through hole types.

thanks

 

I received an email by subscription with your message;
"To resolve 100uV, you need an amplifier with input offset <<100uV.
Can you use a surface mount amplifier, of do you need a through-hole device?"
some how it did not appear on this thread.

If it is possible, I need a through hole unless it is feasible to use an adapter.

I deleted it because the first sentence was probably unreasonably restrictive.
Here is an amplifier that should work for you. The gain adjustment takes the place of the current adjustment pot.

 

thanks a lot Ron_H. Today I will build it and let you know how it performed .

You know what will happen If i used the current monitor? I think there will be significant voltage drops between the tiny crocodile clips and the inductor terminal which will effect the accuracy of my reading unless the inductor is soldered. But soldering the inductor would not be the best idea on a measuring instrument.

 

thanks a lot Ron_H. Today I will build it and let you know how it performed .

You know what will happen If i used the current monitor? I think there will be significant voltage drops between the tiny crocodile clips and the inductor terminal which will effect the accuracy of my reading unless the inductor is soldered. But soldering the inductor would not be the best idea on a measuring instrument.

The current monitors could probably work. You would need to configure your current source as a sink, and place it where the load is shown in the datasheet. If given a choice, I prefer low side measurement, which is what the circuit does that I posted.
The Zetex current monitors are not available in through hole packages. The MCP6021 is.
Regarding croc clips, if you are worried about the resistance of measurement leads, you should use the Kelvin 4 wire technique.

 

Thanks for the interesting source. In fact that technique was what I had in mind - without knowing the exact term. That is why I though that with the current monitor would not be possible.

Today I bought the OP77 with an ultra low offset voltage of 10-25μV. What do you think? since the microchip op amps are hard to find here. I didn't have time to build it as I had trouble getting my 16x2 LCD running but now it's oki. I will build the circuit tomorrow.

Can you explain briefly what exactly would the low pass filter be filtering?
I guess johnson and white noise would be useless as they are normally distributed. If I calculated properly the filter has around 4Hz -3bB. Also the instrument would be battery operated.

thanks again

 

Thanks for the interesting source. In fact that technique was what I had in mind - without knowing the exact term. That is why I though that with the current monitor would not be possible.

Today I bought the OP77 with an ultra low offset voltage of 10-25μV. What do you think? since the microchip op amps are hard to find here. I didn't have time to build it as I had trouble getting my 16x2 LCD running but now it's oki. I will build the circuit tomorrow.

Can you explain briefly what exactly would the low pass filter be filtering?
I guess johnson and white noise would be useless as they are normally distributed. If I calculated properly the filter has around 4Hz -3bB. Also the instrument would be battery operated.

thanks again

That's a great op amp, but it is only spec'ed at ±15V. You could probably use it at ±5V. It doesn't have the input or output range to operate from a single supply. Do you have ±5V available?
The beauty of the MCP6021 is it has rail-to-rail input and output voltage range, and will operate from 3.3V, so it protects your ADC from overvoltage automatically. You will have to add protection if you use the OP77.
There is probably another op amp that you could get which is better suited for this application. What mfrs/vendors do you have access to in Malta?

 

You are right. I should have studied the datasheet better before buying it.
Below is a page with a list of amplification semiconductors where I shop.

If you can suggest a suitable single supply op amp it would be of great help.

http://www.fabian.com.mt/semiconductors/ic_list.html

For now can I experiment with the OP77 if create a dual supply with 2 zeners. In the datasheet it ranges from +-3v to +- 22V. I have 12v and 9V battery that I can use.

 

You are right. I should have studied the datasheet better before buying it.
Below is a page with a list of amplification semiconductors where I shop.

If you can suggest a suitable single supply op amp it would be of great help.

http://www.fabian.com.mt/semiconductors/ic_list.html

For now can I experiment with the OP77 if create a dual supply with 2 zeners. In the datasheet it ranges from +-3v to +- 22V. I have 12v and 9V battery that I can use.

That is a huge list! I might be able to find a suitable part if you post a sub-list of just the op amps.

 

What do you think about the LT1077 by Linear technology. I can order it via RS.

 

What do you think about the LT1077 by Linear technology. I can order it via RS.

That looks pretty good. You would have to run it off +5V and protect the ADC input with a series resistor and Schottky diode to +3.3V, unless your ADC input is 5V tolerant.

You might also want to consider LT1366 and LT1490, which have rail-to-rail outputs and will work on 3.3V.

 

It seems like I will have to stick to the LT1077 because the 3.3V types with input offset less than 100uV are hard to find. Am I right. How is the configuration you mentioned for ADC protection. Can you refer me to a schematic as an example pls?

 

It seems like I will have to stick to the LT1077 because the 3.3V types with input offset less than 100uV are hard to find. Am I right. How is the configuration you mentioned for ADC protection. Can you refer me to a schematic as an example pls?

Here is the protection scheme. It allows you to remove the inductor without blowing your PIC - unless the PIC input is 5V tolerant, in which case you don't need the resistor and the Schottky diode.

 

Hi,
I built the circuit with the op-77 for now with 2x 9V batteries until the LT1077 arrives. The amplification with adjustable gain worked quite well.

My problem is that when testing with 2 inductors of 100μH's I am noticing oscillations varying from 55kHz to 100kHz and with some 1.8Vrms with a DMM. How can I go about to eliminates these oscillations? I did not feed these to the op77. I only fed voltages from low resistances to be safe.

If I add some components to increase the current slowly instead of switching directly the 100mA would it fix the problem or do I have to do some magic with my PIC?