Instrumentation INA125 VREF 10v

Andres2007

Hello,
I'm using this amp to connect to a load cell. I use Vref 10v pin from INA to the load cell. When I test the voltage without the load cell it has the correct voltage 9,998 v, but when I insert the load cell the voltage at pin 4 Vref down to 7,02. It is normal?? or should I get 9.998??

Thanks

dre

Hi,
The Vref circuit does not supply much amperage, so I am not surprised that there was a voltage drop. The bridge supply has to be routed through a transistor such as per figure 4 on page 12 of the INA125 datasheet. Measure the current through the bridge. You may be surprised. Or, you can calculate the current knowing the resistance of the gages comprising the bridge. Turns out that the total resistance of the bridge is equal to the resistance of the gage itself.

Others may have a better solution than using the TIP29C called out in figure 4 if you don't have a TIP.

cheers, David

Andres2007

Hello David,
thanks for your reply. I have used the TIP29C and the circuit works fine fien now. I get at the input of bridge 9,996 vref voltage. I have used this voltage as input of a voltage divisor with 2 resistors 1K, and the output of the divisor directly connected to the Vreference input of the PIC microcontroller to get 5 v as Vref. By the moment it works fine. I must work still to stabilize the analog read got by the PIC. Should I use capacitors in the input of analog read of PIC?

dre

Hi,

I am not an electronic engineer so I can't comment on using capacitors in this manner.
You can also 'filter' the input in the microprocessor using software. However, most knowledgeable will tell you to filter using software only as a last resort.

Software filtering can take on many forms: One form is a running average form: Take, say, 3/4 of the existing average and add 1/4 of the new reading. There are better Pbasic coding than I show. Look into the BS1 manual or applications [I don't remember the exact location.].

For an excellent discussion on 'data smoothing' see:

http://www.emesystems.com/BS2math5.htm

cheers, David

dre

Hi,

Under the Parallax web site I could not find the Stamp 1 ' Basic Stamp Starter Kit' documentation I referred to earlier. It is on page 33 of the hardcopy however. Here is the information:

"---- performs a weighted average by adding 1/4 of the current measurement {b2} to 3/4 of a previous {average} measurement {b4}. This has the effect of smoothing out noise."
Here is the subroutine [written in Stamp1 Pbasic]. [Remember that only the Stamp1 has a 'pot' command.

Test:
pot 0,100,b2 ' take a pot reading [b2]
let b2 = b2 /4 + b4 ' Make b2 = (0.25*b2) + b4
let b4 = b2*3/4 ' Make b4 = 0.75 * b2
return

Congratulations on the success of reading the bridge. A bit of trivia: 'Murphy's Law came about because of a strain gage wheatstone bridge wiring error.' :-(

cheers, David

Andres2007

:-) Murphy's Law is most allways a law

Thanks for the link and your help. I will try to filter noise via software using this kind of method. As you say it should be allways the last resort, but I have not other option by now.

Andres