Continue to Site

Welcome to our site!

Electro Tech is an online community (with over 170,000 members) who enjoy talking about and building electronic circuits, projects and gadgets. To participate you need to register. Registration is free. Click here to register now.

  • Welcome to our site! Electro Tech is an online community (with over 170,000 members) who enjoy talking about and building electronic circuits, projects and gadgets. To participate you need to register. Registration is free. Click here to register now.

Current Sense resistor

Status
Not open for further replies.

Mosaic

Well-Known Member
Hi I am ganging 4 of these https://www.jameco.com/Jameco/Products/ProdDS/659956.pdf to create a 0.1 ohm current sense (20W) resistor. I have an actual precision current sense to compare it against. My concern is I am using a 7800Hz PWM load and these wirewound resistors may introduce some reactance that will throw off the sense results.

I was wondering if anyone has experience in using wirewound loads in a PWM train?
 
Given the low resistance values, I wouldn't expect the inductive reactance to be significant. The resistance wire will likely be wound with sufficient spacing between turns, that the inductance likely will be very close to the same total length of a straight piece of wire. (In fact, if the winding pitch is chosen correctly, the inductance can be lower than a straight piece of wire.) If you have them located close together, then you can also cross connect them so that for adjacent resistors, the current is flowing in opposite directions, and the mutual inductance effect will partially cancel the self inductance.
However, you'll still need to measure them to make sure they're suitable. If you get noticeable voltage spikes across them, you could look at using an RC snubber, tuned for your switching frequency.
 
Hi I am ganging 4 of these https://www.jameco.com/Jameco/Products/ProdDS/659956.pdf to create a 0.1 ohm current sense (20W) resistor.
Just a quick note;
Normally I would not use more than 1 resister for a current sense application, especially if it needs precision.
The tolerances are additive so with four 5% resisters in a network the overall tolerance is now 20%

OK this was all wrong, nevermind. I dont know what I was thinking.
Infact using multiple resistors you can get closer to your target resistance.
 
Last edited:
Actually, not it's not. If the resistors are all the same value, the overall tolerance is still the same, regardless of whether it's a series or a parallel arrangement.
 
Actually, if errors in resistor values are independent, averaging four values would produce 50% less error than a single resistor taken alone. So, 4 resistors are likely to be more accurate than one.

If values are interdependent, such as if they all biased the same way (e.g. because it is the same batch, or because they all are in the same environment), the error will be the same for either 4 resistors or one, but that's the worst case.
 
Just a quick note;
Normally I would not use more than 1 resister for a current sense application, especially if it needs precision.
The tolerances are additive so with four 5% resisters in a network the overall tolerance is now 20%

OK this was all wrong, nevermind. I dont know what I was thinking.
Infact using multiple resistors you can get closer to your target resistance.
See my corrected post above.
 
I have used a similar resistance, actually just one resistor of similar value, it was wirewound, and it was part of a 40khz switcher supply.
It worked well.
Connect them up and measure the inductance, that'll give you an idea of whats going to happen, the reistance is probably going to appear higher with reactance, which allthough undesireable at least it'll create a safe condition, rather than the circuit see the current lower than it actually is.
Another thing to look out for if the reactance is high, if the power switch or fet switches very quickly you'll get nasty spikes on the drain/emitter, and the resistor will create a resonant loop with its own capacitance.
 
Hi,

Yes it's true that the tolerance of four resistors if each is totally random will produce a value that is more precise than any one resistor. If they are not random however than that's not the case.
The best bet is to measure the resulting resistance or just compare it to a known more precise resistance.

The inductance of the wire wound resistors should be small for a frequency of only around 8kHz, but since inductance increases with series connected inductors and decreases with parallel connected inductors the best connection arrangement is probably to connect the multiple resistors in parallel. For four resistors this means the resistance of each resistor should be four times as high as the required final value. The inductance will then be one quarter of the inductance of any one resistor.
 
Check the spec for the resistors, some are half wound one way, and half wound the other to combat inductance.
 
Status
Not open for further replies.

Latest threads

New Articles From Microcontroller Tips

Back
Top