measuring very low voltages reliably

[xaccto]

How do you measure low voltages like between 1mV and 20mV with precision ?

I mean sure a multimeter supposedly has that on its range, but can it be relied on?

and opamps have offset voltages in this region that would effect accuracy hugely, no?

Without having access to hugely expensive equipment, what would be a good approach ? a DIY method ?

Ultimately I want to build my own current shunt, so measuring resistance of possibly
0.001Ohm or less, and I expect I might be able to build a 10 or 20Amp current source.

 

[Mad I.D.]

Hi.
Yes, that is a real problem. Measuring very small DC (or low frequency) voltages is very difficult, as you said, because it requires a very precise OP-AMP.

The solution to your problem is to use an auto-zero or chopper op-amp. Google those terms. Auto-zero OP-AMP's can have a offset voltage as low as 1uV or less. For e.q. you can check this one : AD8551 (auto-zero)

 

[xaccto]

hmm, 1uV offset for AD8551, and not too expensive, ~$3USD
excellent, I will have to order some next time.
I found I have OP27, so 10uV offset, that may do for first experiments.

Actually I don't know how you calculate what the resulting error is, given the offset voltage,
is there an equation for this ?

One also needs precision resistors to set the gain, is 1% metal film enough though?
looks like you can buy more precise resistors......

 

[keith1200rs]

In general, for a DC measurement, the offset will be multiplied by the gain. You can buy resistors more accurate than 1% but how accurate do you need the result? Also, how accurate is your shunt resistor?

Keith

 

[Mad I.D.]

hmm, 1uV offset for AD8551, and not too expensive, ~$3USD
excellent, I will have to order some next time.
I found I have OP27, so 10uV offset, that may do for first experiments.

Actually I don't know how you calculate what the resulting error is, given the offset voltage,
is there an equation for this ?

One also needs precision resistors to set the gain, is 1% metal film enough though?
looks like you can buy more precise resistors......

You just multiply it with your gain. That's why offset voltage is a huge problem when measuring low DC. You need a high gain and you cannot block the DC with a cap since it's your voltage + offset.

For the shunt resistor you can easily use 0.1%.

 

[karesz]

Hi,
pls. check these apll. note too...
K.

 

[pauloynski]

Usually weighing scales need to measure very low voltages from a strain gauge (down to the nV region). So, I suggest your looking for special devices like the CS5522 from Cirrus Logic https://www.cirrus.com/en/pubs/proDatasheet/CS5521-22-23-24-28_F8.pdf or the LTC2440 https://www.linear.com/pc/productDetail.jsp?navId=H0,C1,C1155,C1001,C1152,P1771 Also take a look at a good op-amp at **broken link removed**
Regards

 

[karesz]

...Ultimately I want to build my own current shunt, so measuring resistance of possibly 0.001Ohm or less, and I expect I might be able to build a 10 or 20Amp current source.

Hali xaccto,
Seems me (I studied for my office past time some similar solutions...) that its a quasi standarad high current-shunts to design for i.e. 60 mV range...
OK, if its a DIY subject you can have other systems too, but a similar way as i.e. Hybrid car producer follows too,: will have more lighter to find the realisitions possibility, and components sortiments too for you...
K.