2
\$\begingroup\$

In my design I want to measure the current going through a load with a shunt resistor. Based on some preliminary calculations, I've selected WSBS85181L000JK34 from Vishay. Being my first time using such high power shunt, I'm confused about the mounting system: does it need to be soldered on the PCB or screwed down with a nut and a bolt on opposite sides of the PCB?

Also, that resistor has a notch on one side. Should it face the up or down to the PCB? My guess is that the middle part shouldn't touch the PCB, which is where the current sense lines would go.

For clarity, the max load to be measured will be 5A @ 5V (30W), so I selected a 36W (1mOhm) resistor to have some margin.

Thanks

EDIT: Thanks to @brhans and @John D for highlighting my conceptual mistake. When choosing a shunt power rating one must think at the power dissipated by the shunt, not the load, so in this case I have 5A going through a 1mOhm shunt, which gives a power loss of P=I^2*R -> 25mW, so a anything higher than 30mW is prefectly suited for the job, certainly not a 36W shunt resistor.

\$\endgroup\$
8
  • 2
    \$\begingroup\$ For the current levels these shunts are aimed at, they would most likely be bolted between segments of bus bars, not mounted on a PCB. Though I suppose it's possible to surface mount them if the PCB traces are maybe enhanced with bus bars. These look like way overkill to measure 5A. \$\endgroup\$
    – John D
    Commented Nov 9, 2022 at 23:24
  • 5
    \$\begingroup\$ You've conflated the power dissipated in your load with the power dissipated by the shunt resistor. At 5A, your 1mOhm resistor is only going to dissipate 25mW. The load voltage is irrelevant to the shunt resistor. \$\endgroup\$
    – brhans
    Commented Nov 9, 2022 at 23:33
  • 1
    \$\begingroup\$ That shunt you've chosen is good for almost 190A btw - more than a little overkill for your application where something the size of a 1206 would probably work just fine. \$\endgroup\$
    – brhans
    Commented Nov 9, 2022 at 23:42
  • 1
    \$\begingroup\$ If you can afford to lose 50 mV, a 10 mOhm would be a better choice. It will dissipate 0.25 W at 5 A. Unless you need high accuracy, a 10 mOhm, 1W SMD should work. \$\endgroup\$
    – Mattman944
    Commented Nov 9, 2022 at 23:52
  • \$\begingroup\$ @JohnD where do you attach the sense wires for Kelvin sensing? \$\endgroup\$
    – Criticizing Israel not allowed
    Commented Nov 10, 2022 at 3:06

1 Answer 1

Reset to default
1
\$\begingroup\$

Looks like its a bolt, but with standoffs. They also wire the sense lines from underneath, but I would guess that this would be up to the designer and what works best for them.

enter image description here enter image description here Source: https://www.vishay.com/en/product/30134/tab/product-videos/

\$\endgroup\$
1
  • 1
    \$\begingroup\$ I'd guess the standoffs are to allow for better convection cooling. It probably only needs them if you're planning on pushing a reasonable amount of current through the shunt. \$\endgroup\$
    – brhans
    Commented Nov 10, 2022 at 1:43

Your Answer

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

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