cancel
Showing results for 
Search instead for 
Did you mean: 

PSoC 5, 3 & 1 MCU

New Contributor II

I am using an ADC in a design and I am getting readings that are lower than I'd expect.  I set up a test pin (Pin 1), and it's reading properly on my multimeter.  There are no external components connected to the psoc board at the moment.  The test pin is reading 1020 mV which is consistent with the design to that pin (VRef through buffer to pin).  But the PSOC is giving me 972 mV.    And it doesn't seem to matter if I use the SAR or DelSig.  

   

 

   

I've used the DelSig before, and it read flawlessly.  The only major difference with that design and this is that I am using hardware muxes.  Could they be leaking current to ground?

   

 

   

Thanks!

   

 

   

jk

0 Likes
Reply
1 Solution
Esteemed Contributor II

In analog view you can see using the ohmmeter that there is a resistance of 550Ohm between the GPIO pin and the SAR input. Can this affect your measured voltage? the SAR input itself is not high impedance, the analog input pin is. ADC continuous mode is not valid when muxing the inputs. It is not advisable using delays in an interrupt handler. I would suggest to convert the current cycle and when appropriate switch the mux, so you get time to settle the voltages until the next cycle starts.

   

 

   

Bob

View solution in original post

0 Likes
Reply
5 Replies
Honored Contributor II

Didi you add a decoupling cap to the pin with Vref?

0 Likes
Reply
New Contributor II

I did, and it didn't help.  And when I test the point with the multimeter, it's giving me the right voltage.

0 Likes
Reply
Esteemed Contributor II

In analog view you can see using the ohmmeter that there is a resistance of 550Ohm between the GPIO pin and the SAR input. Can this affect your measured voltage? the SAR input itself is not high impedance, the analog input pin is. ADC continuous mode is not valid when muxing the inputs. It is not advisable using delays in an interrupt handler. I would suggest to convert the current cycle and when appropriate switch the mux, so you get time to settle the voltages until the next cycle starts.

   

 

   

Bob

View solution in original post

0 Likes
Reply
Valued Contributor II


It seems there is an inaccuracy in the processing result
Perhaps this comparison will help:

New Contributor II

That did the trick.  Thanks so much!!!

0 Likes
Reply