Author Topic: How to measure a lower current with a shunt from 0A to 1A  (Read 2924 times)

0 Members and 1 Guest are viewing this topic.

Offline ats3788Topic starter

  • Contributor
  • Posts: 26
How to measure a lower current with a shunt from 0A to 1A
« on: February 20, 2018, 03:31:49 pm »
Hello
I try for a while to find a solution to measure current with a shunt I tried the INA138 but this approach was not really precise.
The MAX472 start working with 3Volt. Davids µCurrent is to expensive to measure  a positive Voltage.
Does someone know an approach what is simple a precise. 
 

Offline Gibson486

  • Frequent Contributor
  • **
  • Posts: 324
  • Country: us
Re: How to measure a lower current with a shunt from 0A to 1A
« Reply #1 on: February 21, 2018, 02:06:00 am »
Are you saying you want the measurement to be more accurate in the lower range of your 0 to 1A measurement? Or are you saying that you can get 0 to 1A, but the resolution in that whole range is not enough?
 

Offline Bruce Abbott

  • Frequent Contributor
  • **
  • Posts: 627
  • Country: nz
    • Bruce Abbott's R/C Models and Electronics
Re: How to measure a lower current with a shunt from 0A to 1A
« Reply #2 on: February 21, 2018, 05:35:10 am »
Davids µCurrent is to expensive to measure  a positive Voltage.
What is your budget, and what currents do you want to measure?

Quote
Does someone know an approach what is simple a precise.
An ordinary digital multimeter is plenty precise enough for all my needs.
 

Offline danadak

  • Super Contributor
  • ***
  • Posts: 1875
  • Country: us
  • Reactor Operator SSN-583, Retired EE
Re: How to measure a lower current with a shunt from 0A to 1A
« Reply #3 on: February 21, 2018, 12:28:02 pm »
Your measurement low side or high side, I assume its high side ?

What is being used for A/D to do the measurement ?

Is this a one off design or a product design ? In either case range of temperature,
resolution you want in measurement.


Regards, Dana.
« Last Edit: February 21, 2018, 12:30:37 pm by danadak »
Love Cypress PSOC, ATTiny, Bit Slice, OpAmps, Oscilloscopes, and Analog Gurus like Pease, Miller, Widlar, Dobkin, obsessed with being an engineer
 

Offline ats3788Topic starter

  • Contributor
  • Posts: 26
Re: How to measure a lower current with a shunt from 0A to 1A
« Reply #4 on: February 21, 2018, 12:39:56 pm »
Thx for your answers
I just need a schematic where I can measure the Voltage from 0..20Volts, what is easy and the current with a Shunt Resistor from 1mA..1A Range.
I struggle with that for the last two years. I don't find a proper           
Solution. The cost is not important, but it would be great to find a Solution for a couple of bucks, I already ask this question in a German Forum but nobody could give me a Hint
Davis approach sounds great, but is there a less precise OP Amp then the MAX4239 what is doing the Job.
 

Offline Siwastaja

  • Super Contributor
  • ***
  • Posts: 8110
  • Country: fi
Re: How to measure a lower current with a shunt from 0A to 1A
« Reply #5 on: February 21, 2018, 02:17:50 pm »
Market is full of current measurement amplifiers. Your need of unidirectional current only makes the selection even wider. Look at the parametric search of any parts supplier.

For good accuracy, exact layout from the shunt resistor to the amplifier is important.
 

Offline Gibson486

  • Frequent Contributor
  • **
  • Posts: 324
  • Country: us
Re: How to measure a lower current with a shunt from 0A to 1A
« Reply #6 on: February 21, 2018, 02:20:56 pm »
The precision of the op amp is not that important for your application. Your op amp you selected is great if you wanted to take a transducer and ensure you have limited voltage output variation from chip to chip (ie, 300 counts will usually only be 300 counts +/- your error). That being said, there are also ways you can deal with those variations as well. What you should be concerned about is the common mode voltage between the inputs. The easiest and quickest way to complete your design is to get an actual differential amplifier or even further, try a current sense amplifier.  I recommend Linear Tech. Yeah, they are pricey when compared to the competition, but their datasheets are well thought out.

.
 

Offline carljrb

  • Contributor
  • Posts: 24
  • Country: ca
Re: How to measure a lower current with a shunt from 0A to 1A
« Reply #7 on: February 22, 2018, 05:50:53 am »
INA series current sense amplifiers like you're already using are actually great. I'm using the INA226 in several designs. In one of them, using a 3mOhm shunt, I can measure up to a bit above 27A, in sub-1mA increments. With basic calibration, its readings are at least as good as the ones from a B&K Precision 2831E while also being able to read a higher current range. It also measures bus voltage in 1.25mV increments which is also quite precise for its range (0-36V). The voltage readings are spot-on too.

If your readings aren't good, then the problem is most likely your design (and possibly fabrication too). There's many ways to go wrong: bad shunt choice, poor footprint (not kelvin sense), ground offset between the INA and the ADC reference, noise (many possible sources here), poor tolerances, poor placement and layout, classic ADC issues... You name it. Expensive parts or switching vendors won't fix any of this.

If you know what you're doing (reading the datasheet thoroughly helps a lot), that you pick the right shunt, have the right layout and connections, optimal ADC settings for the application (conversion time, averaging, etc), decent software (things like filtering and over-sampling could further help) and basic calibration then the results should be pretty fantastic.
« Last Edit: February 22, 2018, 05:56:19 am by carljrb »
 

Offline Gibson486

  • Frequent Contributor
  • **
  • Posts: 324
  • Country: us
Re: How to measure a lower current with a shunt from 0A to 1A
« Reply #8 on: February 22, 2018, 02:30:40 pm »
INA series current sense amplifiers like you're already using are actually great. I'm using the INA226 in several designs. In one of them, using a 3mOhm shunt, I can measure up to a bit above 27A, in sub-1mA increments. With basic calibration, its readings are at least as good as the ones from a B&K Precision 2831E while also being able to read a higher current range. It also measures bus voltage in 1.25mV increments which is also quite precise for its range (0-36V). The voltage readings are spot-on too.

If your readings aren't good, then the problem is most likely your design (and possibly fabrication too). There's many ways to go wrong: bad shunt choice, poor footprint (not kelvin sense), ground offset between the INA and the ADC reference, noise (many possible sources here), poor tolerances, poor placement and layout, classic ADC issues... You name it. Expensive parts or switching vendors won't fix any of this.

If you know what you're doing (reading the datasheet thoroughly helps a lot), that you pick the right shunt, have the right layout and connections, optimal ADC settings for the application (conversion time, averaging, etc), decent software (things like filtering and over-sampling could further help) and basic calibration then the results should be pretty fantastic.




I agree. Looking closer at his specs, he needs 1mA to 1A. With a 10 bit ADC (ie, an arduino), he will just barely meet spec, so at the very least, he needs something higher than a 10 bit ADC.  Also, this would be a very good design to look over the concept of ground current. It is crucial in these circuits.
 

Offline Bruce Abbott

  • Frequent Contributor
  • **
  • Posts: 627
  • Country: nz
    • Bruce Abbott's R/C Models and Electronics
Re: How to measure a lower current with a shunt from 0A to 1A
« Reply #9 on: February 22, 2018, 02:57:57 pm »
Looking closer at his specs, he needs 1mA to 1A. With a 10 bit ADC (ie, an arduino), he will just barely meet spec,
If the spec is met then where's the problem?

Quote
so at the very least, he needs something higher than a 10 bit ADC.
Or he could oversample. Simply accumulate 64x 10 bit samples in a 16 bit unsigned int, then divide down to the resolution you need. With this technique I was able to get stable readings with 13 bit resolution.   
 

Offline ats3788Topic starter

  • Contributor
  • Posts: 26
Re: How to measure a lower current with a shunt from 0A to 1A
« Reply #10 on: February 22, 2018, 03:00:02 pm »
Thank for you answers
Next Issue from the Dyode Magazin will have a story about Mulimeter maybe the mention this Topic. I will check out the INO226.
 

Offline Gibson486

  • Frequent Contributor
  • **
  • Posts: 324
  • Country: us
Re: How to measure a lower current with a shunt from 0A to 1A
« Reply #11 on: February 22, 2018, 04:37:15 pm »
Quote
Quote from: Gibson486 on Today at 01:30:40 AM
Looking closer at his specs, he needs 1mA to 1A. With a 10 bit ADC (ie, an arduino), he will just barely meet spec,
If the spec is met then where's the problem?

Quote
so at the very least, he needs something higher than a 10 bit ADC.
Or he could oversample. Simply accumulate 64x 10 bit samples in a 16 bit unsigned int, then divide down to the resolution you need. With this technique I was able to get stable readings with 13 bit resolution.

He needs to go from 1mA to 1A. On a 10 bit ADC, that is approx 1mA resolution. The lowest readings are essentially in the noise when you take into account ADC error, especially if the design does not take into account ground current. Super sampling will not fix that. If he is OK with not having the lower end be accurate, then it is a flaw he can deal with.
« Last Edit: February 22, 2018, 04:38:46 pm by Gibson486 »
 

Offline Bruce Abbott

  • Frequent Contributor
  • **
  • Posts: 627
  • Country: nz
    • Bruce Abbott's R/C Models and Electronics
Re: How to measure a lower current with a shunt from 0A to 1A
« Reply #12 on: February 23, 2018, 07:41:22 pm »
The lowest readings are essentially in the noise when you take into account ADC error, especially if the design does not take into account ground current. Super sampling will not fix that.
When oversampling to increase resolution, noise is good. You want at least 1 LSB of noise.


 

Online iMo

  • Super Contributor
  • ***
  • Posts: 4675
  • Country: nr
  • It's important to try new things..
Re: How to measure a lower current with a shunt from 0A to 1A
« Reply #13 on: February 23, 2018, 09:07:51 pm »
If 15samples/sec is enough for you the TI's http://www.ti.com/lit/ds/symlink/ads1110.pdf is your friend.
Wire the differential inputs to the shunt (common GND at the lower-side) and you may get up to 8uV resolution..
 

Offline Rick Law

  • Super Contributor
  • ***
  • Posts: 3423
  • Country: us
Re: How to measure a lower current with a shunt from 0A to 1A
« Reply #14 on: February 23, 2018, 11:56:16 pm »
... he needs 1mA to 1A. With a 10 bit ADC (ie, an arduino), he will just barely meet spec, so at the very least, he needs something higher than a 10 bit ADC.  Also, this would be a very good design to look over the concept of ground current. It is crucial in these circuits.

May be using an ADS1115 16bit 4channel.

Use the first 2 channel as differential measuring low-side with a .1ohm, and use one of the other two channels to measure V-out.  Since ADS1115 goes to only 5.5V max, a voltage divider would be needed to go all the way to 20V.
 

Offline ats3788Topic starter

  • Contributor
  • Posts: 26
Re: How to measure a lower current with a shunt from 0A to 1A
« Reply #15 on: April 09, 2018, 11:16:52 am »
Thank you for all the replies.
I got this week my INA226 and it looks pretty well.
 

Offline ogden

  • Super Contributor
  • ***
  • Posts: 3731
  • Country: lv
Re: How to measure a lower current with a shunt from 0A to 1A
« Reply #16 on: April 09, 2018, 11:29:52 am »
... he needs 1mA to 1A. With a 10 bit ADC (ie, an arduino), he will just barely meet spec, so at the very least, he needs something higher than a 10 bit ADC.  Also, this would be a very good design to look over the concept of ground current. It is crucial in these circuits.

May be using an ADS1115 16bit 4channel.

Use the first 2 channel as differential measuring low-side with a .1ohm, and use one of the other two channels to measure V-out.  Since ADS1115 goes to only 5.5V max, a voltage divider would be needed to go all the way to 20V.

Other option is to add 2nd current shunt monitor (to same shunt) with, let's say, 8x higher gain. As easy as that.
 

Offline SiliconWizard

  • Super Contributor
  • ***
  • Posts: 14306
  • Country: fr
Re: How to measure a lower current with a shunt from 0A to 1A
« Reply #17 on: April 09, 2018, 12:23:13 pm »
INA series current sense amplifiers like you're already using are actually great. I'm using the INA226 in several designs. In one of them, using a 3mOhm shunt, I can measure up to a bit above 27A, in sub-1mA increments.

I also used the INA226 and it's great (in my case with a 10 mOhm shunt, to measure currents up to 8 A).

What we don't know is whether the author needs a digital or analog output, whether it should be high-side or low-side, the required bandwith...
 


Share me

Digg  Facebook  SlashDot  Delicious  Technorati  Twitter  Google  Yahoo
Smf