Author Topic: Differential Amplifier did'nt work  (Read 1322 times)

0 Members and 1 Guest are viewing this topic.

Offline Eldi4Topic starter

  • Contributor
  • Posts: 41
  • Country: id
Differential Amplifier did'nt work
« on: June 30, 2019, 03:15:50 pm »
Hi,
I have some differential amplifier for hi-side current sensing that did'nt work, the output is always sitting on 3.8V whatever the condition is (no current/current flow), the one that did'nt work is the bat and load side (same symptoms), the diagram is on the attachment, i used 10k/220k for 22 gain for both amp (bat and load) instead of what is showed on the diagram, the amp is powered with 5V single supply, i have tried to replace the IC with other LM358, but the problem remains the same.
However, i had differential amplifier with 47 gain (1k/47k) that did work, it powered with 5V too, the diagram is attached too.

What probably is the problem?

NOTE : The bat diff amp and load diff amp is on different IC, not like what is showed on diagram.
« Last Edit: June 30, 2019, 03:18:59 pm by Eldi4 »
 

Offline capt bullshot

  • Super Contributor
  • ***
  • Posts: 3033
  • Country: de
    • Mostly useless stuff, but nice to have: wunderkis.de
Re: Differential Amplifier did'nt work
« Reply #1 on: June 30, 2019, 03:27:08 pm »
What probably is the problem?

Your circuit places an input voltage at the LM358 that is way above its supply voltage.
You must observe the allowed input voltage range (differential and common mode) of the amplifier, having a 5V supply doesn't cover 12V common mode voltage.
Safety devices hinder evolution
 

Offline Eldi4Topic starter

  • Contributor
  • Posts: 41
  • Country: id
Re: Differential Amplifier did'nt work
« Reply #2 on: June 30, 2019, 03:32:03 pm »
What probably is the problem?

Your circuit places an input voltage at the LM358 that is way above its supply voltage.
You must observe the allowed input voltage range (differential and common mode) of the amplifier, having a 5V supply doesn't cover 12V common mode voltage.

So, using 12V supply should fix this problem then?

My PV voltage is even 30V, but the current sensing was seem to be working when the voltage was dragged down to 0.7V because of diode short circuit, but no idea when the voltage is 10V+ (havent tried it yet).

How can i solve problem for those diff amps?
« Last Edit: June 30, 2019, 03:34:32 pm by Eldi4 »
 

Offline capt bullshot

  • Super Contributor
  • ***
  • Posts: 3033
  • Country: de
    • Mostly useless stuff, but nice to have: wunderkis.de
Re: Differential Amplifier did'nt work
« Reply #3 on: June 30, 2019, 03:48:43 pm »
So, using 12V supply should fix this problem then?
No, not in general. Observe the specifications of the amplifier and do the calculation of your circuits common mode voltage. Most simple and cheap amplifiers require some headroom between the supply voltage and the input voltages. Look for an rail-to-rail input if you want to supply the amplifier from a voltage that is equal or greater than your max. common mode voltage.

Quote
My PV voltage is even 30V, but the current sensing was seem to be working when the voltage was dragged down to 0.7V because of diode short circuit, but no idea when the voltage is 10V+ (havent tried it yet).
The circuit is the same, so it will behave the same.

Quote
How can i solve problem for those diff amps?
- There are special "current sense" amplifiers that are designed to work under such conditions - ready to buy from the major suppliers (TI, ADI).
- You can build your own current sense amplifier using some tricks, you might be able to find such circuits e.g. in the datasheet of the OPA333.
- Add noise gain to your circuit (this will increase common mode voltage for the cost of reduced accuracy and increased noise).
- If applicable, use low side sensing

Safety devices hinder evolution
 
The following users thanked this post: Eldi4

Offline Eldi4Topic starter

  • Contributor
  • Posts: 41
  • Country: id
Re: Differential Amplifier did'nt work
« Reply #4 on: June 30, 2019, 03:58:18 pm »
Thanks for the explanation!
I will go for low-side sensing, the quick and costless way...
 


Share me

Digg  Facebook  SlashDot  Delicious  Technorati  Twitter  Google  Yahoo
Smf