0
\$\begingroup\$

Background: I am designing a test board to measure the voltages from multiple point of the target board (DUT). I am using 2 pair of mux to toggle between the pairs. One mux (mux506ipwr from TI) is toggling between positive side, and the output will go thru an opamp buffer before going to relay and eventually go to positive side of DMM; another mux (DG408 from Vishay) is toggling between different ground, and output will go straight to relay and the relay will go to the negative side of the DMM.

Circuit as below. Measurement circuit using 2 MUX

Problem: When I use the DG408 to toggle to GND, it will have around ~18mV higher than actual GND. I got this measurement by measuring the output pin of the DG408 (pin8; D) to the actual GND on the DUT, I got ~18mV on my multimeter. So when the positive side is giving low signal which is ~2mV, my DMM ended up reading -0.016V. Although I still can adjust the test limit to accept this, I am curious why it behaves in such way. Plus, this reading is not useful - my measurement result does not really tell the actual reading, leaving my test not reliable.

Some add on:

  1. This was not observed earlier; until one point (after running thru 200+ DUT with the same setup) due to some reason I have re-aligned the test jig and it started to behave in such way. I have tried to replace the DG408, MUX506 and even the OPAMP (OPA2991), the negative voltage still remains. I am quite certain that the alignment is still correct, as I am still able to run the full test with expected readings, except the negative voltage when measuring low signal.

Link to datasheet: MUX506 DG408 OPA2991

This is the first time I am using mux to perform measurement. Is the concept wrong? Appreciate if anyone could give me some insights on this. Please let me know if more info is needed.

\$\endgroup\$
3
  • \$\begingroup\$ Please measure the voltage on the actual DG508 chip pins from pin 8 to pin 12 and report back. \$\endgroup\$ Commented Jan 10 at 6:24
  • \$\begingroup\$ @Spehro'speff'Pefhany Thank you for pointing out. I have measured, DG408 pin 8 and pin 12 have 0.0mV voltage different during the measurement step. Strange enough, the voltage difference is come from pin 12 and DUT GND; which is unexpected since DUT GND is directly connected to pin 12 thru pogo pin and wire connector and PCB trace, with no other component in between. \$\endgroup\$ Commented Jan 13 at 3:03
  • \$\begingroup\$ To add on, the pin 12 of DG408 is having positive 20mV against DUT GND, makes me wonder where the voltage is coming from... \$\endgroup\$ Commented Jan 13 at 3:10

1 Answer 1

Reset to default
1
\$\begingroup\$

Thank you for confirming the 0.0mV potential between DG508 pins 8 and 12.

So it appears there is enough current passing through the PCB trace to result in a 20mV voltage drop.

Changing the layout so a different reference point is used can change this. It may also be an indicator that the layout could be better (fatter current-carrying traces or a ground plane might be called for) but there will always be be some voltage drop where current is flowing through PCB traces.

(If there was a voltage across the switch that would indicate an unexpected current through the switch)

\$\endgroup\$
4
  • \$\begingroup\$ Thank you for your command. However, the pin 12 is instead, 20mV higher than the DUT itself, so I dont think this is due to voltage drop. Actually, I was expecting almost no current flow on that path due to the OPAMP buffer at the back (U5.1 in the schematic), but I could be wrong on this. \$\endgroup\$ Commented Jan 13 at 5:37
  • \$\begingroup\$ If it is energized, where is the relay coil current flowing? \$\endgroup\$ Commented Jan 13 at 6:05
  • \$\begingroup\$ I am using a ULN2003 to sink current (Relay36 net) when microcontroller send a signal. ULN2003 is connected to Relay+/- as well. I just removed the relay and measure again and got the same result, pin 12 is having ~18 - 20mV higher than DUT. \$\endgroup\$ Commented Jan 13 at 6:37
  • \$\begingroup\$ Current must be flowing through that ground to generate that offset. \$\endgroup\$ Commented May 24 at 19:43

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Start asking to get answers

Find the answer to your question by asking.

Ask question

Explore related questions

See similar questions with these tags.