Google Groups no longer supports new Usenet posts or subscriptions. Historical content remains viewable.
Dismiss

shunt/ current sense resistor question

34 views
Skip to first unread message

Jezebels_couz

unread,
Dec 1, 2020, 9:41:44 PM12/1/20
to
In the process of finishing a 0-12 VDC @ 0-10 A switching power supply.
Fairly new at this and wanted to ask about the shunt resistor. It will
be in the ground line. Plans mentioned two options: either a custom
shunt made from constantan wire or similar, or simply two 5W 0.1 ohm
power resistors in parallel. Since I wanted to keep costs down, I am
going with the power resistors, but I have a concern: I will be running
12 gauge wire for the output lines, but when the ground wire reaches the
shunt, even the two resistors in parallel combined are far less than 12
gauge. Should I be concerned, or is the tiny resistor/ 12 gauge wire
distance (basically just for the 12 ga wire to resistor wire connection)
irrelevant to heating and losses?

Johann Klammer

unread,
Dec 2, 2020, 2:47:18 AM12/2/20
to
On 12/02/2020 03:41 AM, Jezebels_couz wrote:
> In the process of finishing a 0-12 VDC @ 0-10 A switching power supply. Fairly new at this and wanted to ask about the shunt resistor. It will be in the ground line. Plans mentioned two options: either a custom shunt made from constantan wire or similar, or simply two 5W 0.1 ohm power resistors in parallel. Since I wanted to keep costs down, I am going with the power resistors, but I have a concern: I will be running 12 gauge wire for the output lines, but when the ground wire reaches the shunt, even the two resistors in parallel combined are far less than 12 gauge. Should I be concerned, or is the tiny resistor/ 12 gauge wire distance (basically just for the 12 ga wire to resistor wire connection) irrelevant to heating and losses?
If it's not too long, the heat might dissipate through the adjoining traces.
(if they didn't lie about the resistor wattage, that is)

Bob Engelhardt

unread,
Dec 2, 2020, 8:11:29 AM12/2/20
to
12ga wire is 1.6R per 1000 feet*. If your connection is 6", its
resistance is .0008R - <2% of your shunt & negligible.

* - https://www.engineeringtoolbox.com/copper-wire-d_1429.html

Ralph Mowery

unread,
Dec 2, 2020, 9:51:08 AM12/2/20
to
In article <rq6ut4$jh7$2...@dont-email.me>, jb...@three.net says...
The small wire of the resistors should handle it just fine. The main
reason for # 12 wire on the circuit is for the voltage drop.

Jezebels_couz

unread,
Dec 2, 2020, 12:45:02 PM12/2/20
to
Here's a diagram of what I was trying to explain. I wasn't sure I was
very clear, so hopefully the diagram will better relate:

https://i.imgur.com/pYSRxiX.jpg

Jezebels_couz

unread,
Dec 2, 2020, 12:45:36 PM12/2/20
to

Jezebels_couz

unread,
Dec 2, 2020, 12:45:56 PM12/2/20
to

Ralph Mowery

unread,
Dec 2, 2020, 2:47:12 PM12/2/20
to
In article <rq8jsg$ne0$3...@dont-email.me>, jb...@three.net says...
>
>
> Here's a diagram of what I was trying to explain. I wasn't sure I was
> very clear, so hopefully the diagram will better relate:
>
> https://i.imgur.com/pYSRxiX.jpg
>
>

I understand what you are going to do.

To put your mind at ease Astron that makes power supplies has a 20 amp
one that has two pass transistors. That means that each one would be
good for 10 amps. They use only one .1 ohm 10 watt per transistor.

I have larger ones that do the same thing and they have worked for years
with no problems.

Each of your resistors will need to be less than 3 watts at a 10 amp
current if a quick calculation is correct.


Jezebels_couz

unread,
Dec 2, 2020, 2:57:16 PM12/2/20
to
Ok, just wanted to be sure as high current would still have to squeeze
through those small areas and I wanted to minimize heating and loss. In
my mind, it sort of defeated the purpose of using 12 gauge wire
everywhere else to have a bottleneck at the end, but if all is well then
no longer concerned. There will be two 5W resistors in parallel, so
good to go on wattage. Thanks again.


Transition Zone

unread,
Dec 5, 2020, 8:59:59 AM12/5/20
to
Also, most current-sensing transformers don't need that high voltage (of shunt resistors) and can isolate any especially sensitive circuitry that might be there.
0 new messages