1
\$\begingroup\$

I just received a brand new Brymen 869s and I'm concerned it may be broken as it's giving a strange reading.

When measuring a 100k 0603 resistor which is soldered into a PCB (forming part of a larger circuit) I get the following readings:

  • Direction away from J1: 3.29MΩ
  • Direction towards J1: 2.02MΩ

I took some the same readings using a different multimeter, a Proster VC99 and got these readings:

  • Direction away from J1: 100kΩ
  • Direction towards J1: 8.66kΩ

At J1 there is a 10kHz 2Vpp sine wave.

Why would the Brymen show a value above 100kΩ whereas my cheapo Proster VC99 does not? Surely there's no way that the resistance can be larger than 100k in this circuit.

Here's the signal processing part of the circuit with the 100k resistor marked by the blue arrow. There are 4 similar circuits in the whole PCB since this is a 4 x voltage controlled amplifier circuit. enter image description here

There's also a power supply. enter image description here

|improve this question
\$\endgroup\$
  • 2
    \$\begingroup\$ Did you disconnect one side of the resistor before taking any measurements? If you did not then any result is unreliable... \$\endgroup\$ – Solar Mike Mar 26 at 15:54
  • 1
    \$\begingroup\$ We would be speculating on the cause without seeing the surrounding circuitry. I suspect protection diodes may be at least a part of it. Also, I suspect your cheapo meter is somewhat more generous with its sense current (this can be verified by using each meter to measure the voltage across a stand-alone resistor while its resistance is being measured with the other). Are you completely powered down when measuring, caps drained and everything? \$\endgroup\$ – Cristobol Polychronopolis Mar 26 at 16:06
  • 1
    \$\begingroup\$ @Dave Tweed Very often charged capacitors may hold enough voltage to cheat an ohmmeter even after minutes from power off. It's not uncommon to measure more than resistor itself even on a good one \$\endgroup\$ – carloc Mar 26 at 17:48
  • 1
    \$\begingroup\$ @carloc: Yes, that's possible, but I don't see any significant capacitors in the schematic. And even when that happens, you usually get vastly different measurements in different directions. \$\endgroup\$ – Dave Tweed♦ Mar 26 at 18:00
  • 2
    \$\begingroup\$ Thank you! The comments about charged capacitors triggered me to see if any current would flow if I shorted the 100k resistor (by using the mA range on my Brymen). Sure enough there was current flow for a few seconds. The readings were still off, however, until I realised that my signal generator plugged into J1 was still generating a 2V sine wave. Unplugging this as well and both meters now show the same reading. Turns out the Brymen was trying to tell me something. I feel somewhat stupid, but I also learned quite a few things. Thanks very much for your help. \$\endgroup\$ – donturner Mar 26 at 18:35
2
\$\begingroup\$

resistor which is soldered into a PCB (forming part of a larger circuit)

DMMs measure resistance by applying small current and measuring the resulting voltage. Then assuming that the component is pure passive resistor, the DMM calculates the value in accord with Ohm's Law.

The issue is that when a resistor is a part of a larger circuit, the resulting net contains semiconductor-kind elements and possibly active sources as not fully discharged capacitors. Semiconductor have non-linear dependence on current direction and applied voltage. This would explain one difference. The higher than nominal resistance could be shown if there are active sources that confuse DMM algorithm.

The circuit in question is a 4-channel variable-gain amplifier. It has power supply part that is not shown on schematics:

enter image description here

This hidden part must have bypass capacitors.

Now, the V2164SZ IC likely has clamping diodes as ESD protection. While the DMM usually applies small probing voltage, 0.2- 0.3V, high impedance measurements require higher voltage. For example, my Fluke 179 applies up to 3.8V in autorange mode. So it is possible that the probing voltage from DMM charges these bypass caps through clamping diodes. Try different directions for different time and different DMMs, and the readings can be anything.

Therefore, don't trust any in-circuit measurement of components without disconnecting at least one end.

|improve this answer
\$\endgroup\$

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.