0
\$\begingroup\$

I have got a design for milliohm measurement circuit that I have found on YouTube.

Also avaliable in pdf format.

This design uses the INA106U diff amplifiers and I want to replace it with something else less expensive.

I have found a wide range of diff amps. and also a wide range of instrumentation amps, I want to know which is better for this specific application and why?

Also I want to know what is the most important parameter\s that I should look for when choosing the amplifier, like for example should the Input bias current be as low as possible or should it be the offset voltage or maybe the input/output noise I really not sure about those params.

Edit
I want to be able to measure down to 1 milliohm with 0.5 milliohm max. error margin. Also, ambient temperature will be fixed and around room temperature(25C).

For the measurement range from 0 ohm up to 10 ohm, For the resistors that I will be using it will be 0.1% and around 25 ppm, For the stability I will zero the instrument every time I use it so I suppose this a short term stability. And lastly, for the power supply, I will go for whatever the components in my hands will require!

\$\endgroup\$
12
  • 1
    \$\begingroup\$ Appreciate the editing, but why did you remove "thanks"!! \$\endgroup\$ – Muhammad Nour Nov 30 '16 at 11:30
  • 1
    \$\begingroup\$ Because that's just time-wasting fluff. Read the rules. \$\endgroup\$ – Olin Lathrop Nov 30 '16 at 11:45
  • \$\begingroup\$ Time-wasting Fluff!!!, oh it's Olin speaking, not the first time, I have reviewed the rules and couldn't find anything saying not to write thanks at the end of your question, and also found something saying "BE NICE" electronics.stackexchange.com/help/be-nice \$\endgroup\$ – Muhammad Nour Nov 30 '16 at 11:52
  • 2
    \$\begingroup\$ OK, you see it like that, it is your point of view but not everyone think about it like this, only one word of appreciation to the VOLUNTEER which will give his time for free would be the least to say to him, I may have more than a 100 words in my question, adding another word for appreciation wouldn't waste anyone time!!!, it is only one word!! And I didn't see any part in the rules that prohibit this very specific action, please mention the reference to this portion! \$\endgroup\$ – Muhammad Nour Nov 30 '16 at 12:11
  • 1
    \$\begingroup\$ If you insist: electronics.stackexchange.com/help/someone-answers. I did not forbid you anything I just told you why I have removed the "thanks" and what should you do instead of it. Also "!!!!!!!!!" seems very emotional to me. \$\endgroup\$ – Bence Kaulics Nov 30 '16 at 19:30
1
\$\begingroup\$

The qualities of the amplifier that you look for are governed by what you expect of the product it's used in, and how you are going to use it.

It is a milli-ohmeter, so the source impedance is very low. You could have uA of bias current for all the difference that would make to your accuracy.

Offset voltage, initial. If you are going to zero it, this doesn't matter. If you want to build and forget, then it must be low enough to meet your accuracy expectations.

Offset voltage change with temperature. If you are going to zero it from time to time, this doesn't matter. If you want to zero it once, then use it summer or winter, then it must be low enough to meet your accuracy and environmental expectations.

Offset voltage drift with time. If you are going to zero it ... (I hope you're getting the general idea by now).

Noise. This is unlikely to be a dominant source of error, and you can always wait a little longer for the reading if it is. Thermal emfs at the contacts are likely to be a bigger problem, as they're DC so can't be averaged out like noise.

Don't forget to use those 4 leads in a 4-terminal configuration. If you connect the current source directly to the voltage sense and measure the resistor as a 2 terminal device, that will be your biggest source of error.

\$\endgroup\$
3
  • \$\begingroup\$ Thanks, Can I ask you what do you mean by source impedance is very low? \$\endgroup\$ – Muhammad Nour Nov 30 '16 at 15:09
  • \$\begingroup\$ The amplifer bias current has to flow from whatever you are measuring. If (say) 100nA flows through a (say) 1Meg resistor, that will drop 100mV and trash your accuracy. You are building a milli-ohmeter. That means the resistor that your amplifier bias current is flowing through is presumably in the milli-ohms, or at most ohms. That is the source impedance, a very low resistor. So low, you do not need to worry about any practical amplifier bias current. \$\endgroup\$ – Neil_UK Nov 30 '16 at 15:14
  • \$\begingroup\$ Yes, it is now clear, I was expecting this actually, thanks \$\endgroup\$ – Muhammad Nour Nov 30 '16 at 15:34
1
\$\begingroup\$

I want to know what is the most important parameter\s that I should look for when choosing the amplifier

As with any design, your specification for what it does and what performance it is meant to achieve is fundamental to deciding what the best device to use is. I'm not about to guess what you feel is important (or critical) in the design because that design you linked may not entirely suit your needs and you may not realize that but without a definition of what performance YOU expect, there is no specific choice to make and no specific guidance other than.... develop a list of requirements or a spec or a list of performance criteria that has to be met.

I have found a wide range of diff amps. and also wide range of instrumentation amps, I want to know which is better in this specific application and why

This is impossible to answer without you stating what performance you have in mind.

\$\endgroup\$
11
  • \$\begingroup\$ I want to measure down to 1 milliohm with error margin -+ 0.5 milliohm MAX \$\endgroup\$ – Muhammad Nour Nov 30 '16 at 13:20
  • \$\begingroup\$ Then that needs to be stated in your main question along with expected ambient temperature range, the precision of resistors you are wanting to use, the full resistance range of what you want to measure, the power supply regime you prefer, long term drift expectations, any reference voltage used and its stability. Measurement designs require all the I's dotting and the T's crossing. \$\endgroup\$ – Andy aka Nov 30 '16 at 13:39
  • \$\begingroup\$ I thought when I mentioned the link to the Youtube video and the pdf most of those parameters is already clear, but anyway, I will do it. \$\endgroup\$ – Muhammad Nour Nov 30 '16 at 14:42
  • \$\begingroup\$ You must understand that most engineers won't watch youtube videos provided as links on EE. Also the PDF doesn't indicate resistor precision in terms of ppm/degC drift, ambient temperature range, full range of input, your long term drift expectations and the reference voltage stability. In fact taking a look now and I would be very concerned about the stability of the two potentiometers. \$\endgroup\$ – Andy aka Nov 30 '16 at 14:49
  • \$\begingroup\$ Yes you are right, my wrong, I am adding them by now! \$\endgroup\$ – Muhammad Nour Nov 30 '16 at 14:51

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.