ad623an in. amp. constant output!

P

player1

New Member

• Mar 1, 2019

• #1

Hello Guys,
I am trying to amplify a load cell output signal using ad623 amplifier. I am using a 5kg load cell and its excitation voltage is the same power source of the in. amp. itself.
I've made a simulation using Multisim and the output was great. the problem now is in the actual circuit, whatever I do, the output of the ad623 always is constant even if a 0V is applied to its input. any ideas to solve this problem?

 

Attachments

• ad623.JPG

  52.2 KB · Views: 199

crutschow

Well-Known Member

Most Helpful Member

• Mar 1, 2019

• #2

Is one of the inputs grounded in the real circuit, as shown in the schematic?

 

P

player1

New Member

• Mar 1, 2019

• #3

no, should it?

 

P

player1

New Member

• Mar 1, 2019

• #4

there was no change when I tried to ground it

 

P

Pommie

Well-Known Member

Most Helpful Member

• Mar 2, 2019

• #5

I've had a lot of success using the HX711 chip. Banggood sell a 5kg load cell with the amp.

Mike.

 

crutschow

Well-Known Member

Most Helpful Member

• Mar 2, 2019

• #6

player1 said:

no, should it?

Click to expand...

Yes, the inputs needs a ground path for the input bias currents.
Show the exact schematic of the circuit you have built.

 

P

player1

New Member

• Mar 2, 2019

• #7

Here is the exact schematic

 

Attachments

• loadCell-schematic.JPG

  33.4 KB · Views: 244

J

jjw

Member

• Mar 2, 2019

• #8

What is the voltage from load cell?
If it is negative ( + lower than -) AD623 can't handle it with single supply.
The gain with 100 ohm resistor is 1000.
Is it too much?

 

J

jjw

Member

• Mar 2, 2019

• #9

crutschow said:

Yes, the inputs needs a ground path for the input bias currents.
Show the exact schematic of the circuit you have built.

Click to expand...

The input bias currents come from load cell?
If the minus input is grounded, the load cell bridge will be unbalanced and the plus input is about 0.5*Vsupply.

 

crutschow

Well-Known Member

Most Helpful Member

• Mar 2, 2019

• #10

jjw said:

The input bias currents come from load cell?

Click to expand...

Yes, the load cell can provide that, so you don't need to ground one of the inputs.
What's the impedance and output voltage from the load cell?

 

P

player1

New Member

• Mar 3, 2019

• #11

the impedance = 990 Ohm
the output voltage = 1 mV/V (from the datasheet), I tried to measure it using a multimeter but it is always = 0 while applying a force on it. is that mean the problem in the load cell itself?

 

crutschow

Well-Known Member

Most Helpful Member

• Mar 3, 2019

• #12

I mean what is the output voltage to ground from the load cell, not the differential voltage..
Is that zero volts?

 

P

player1

New Member

• Mar 3, 2019

• #13

the voltage from the green or white wire of the load cell to the ground is also 0v

 

rjenkinsgb

Well-Known Member

Most Helpful Member

• Mar 4, 2019

• #14

player1 said:

the voltage from the green or white wire of the load cell to the ground is also 0v

Click to expand...

That sounds like the load cell is faulty or has burned out due to too high a supply voltage.

A full bridge "load cell" is just four strain gauge elements wired in a bridge configuration:
ie. **broken link removed**

The outputs should be at half supply, with no applied load.

The diagonally opposite pairs are configured so one diagonal is under compression and the other diagonal is under tension when load is applied to the cell.

That results in the two outputs being changed in opposite directions by some amount depending on the applied load, so giving a differential output voltage.