FW: Making Low Ohm High Current Sense Resistors
Willie Hunt
-----Original Message-----
From: Willie Hunt
Sent: Wed 1/15/2003 6:41 PM
To: Marty Goodman
Cc:
Subject: RE: [BC] Making Low Ohm High Current Sense Resistors
Actually some steel wire is much better, because it has much higher resistivity and some alloys have OK temp. co. Anyway, yes, resistors can be made of wire, and I've done that myself many times, but if your a try to mearure current of a lamp, or a battery charger, etc, it isn't worth the aggravation of having such a poor and wandering calibration.
Which reminds me of the last time I had to make a resistor out of wire, which was just 2 weeks ago. I flew out to TN with all my new super power strobes, and one of the power pack's DC to DC convertor didn't work. Upon looking at it, the 4x 0.12 ohm 2% 2 watt non-inductive metal film resistors that sense the current on the power FET all broke in 2 pieces right down the center. Most likely the baggage gorillas at the airport did it! This would have been a real problem to fix on the trip, but I had brought the PCB from one of the 10 cheapo (like $2) Polaroid storbe that I bought surpus. On the PCB is a steel wire loop with a clear sleeve that is used to reduce the peak current for the quench tube when this tube short the 1000uF 330V cap to reduce flash power to the main Xenon flash tube. Anyway, I took the wire off, measured it at 0.06 ohm using the power supply as a 1 amp source and a volt meter. I needed 0.03 ohms, so I cur it in half and installed it, and it worked.
This was a good thing, because I wouldn't have been able to take these pictures without fixing the power pack:
http://www.cs.indiana.edu/~willie/willie_adv03.html
Willie
-----Original Message-----
From: Marty Goodman [mailto:martyg...@worldnet.att.net]
Sent: Wed 1/15/2003 6:03 PM
To: Willie Hunt
Cc:
Subject: Re: [BC] Making Low Ohm High Current Sense Resistors
OK! Sorry. I'll use Nichrome next time!
Seriously... I now do use quality low value high power resistors these days.
I was just trying to demonstrate the general concept of approximation based
on textbook data, then fine tuning by successive real world adjustment.
It's amazing the number of "precision" things an amateur can devise using
the approach, without buying "professinal" calibrated equipment.
I conceed that copper can get quite hot, tho.
---marty
----- Original Message -----
From: "Willie Hunt" <wil...@surefire.com>
To: <martyg...@worldnet.att.net>; <kur...@mtco.com>; "Bikecurrent"
Sent: Wednesday, January 15, 2003 4:56 PM
Subject: RE: [BC] Making Low Ohm High Current Sense Resistors
> Marty! don't do this! Copper wire has a horrible tempurature coefficient
so the resistor value will change radically with power applied. I seem to
recall it's doubles at 200C compared with 25C. There are plenty of
tempurature stable low value power resistor available from Digikey or Mouser
in single quantities.
>
> Willie
>
> -----Original Message-----
> From: Marty Goodman [mailto:martyg...@worldnet.att.net]
> Sent: Wed 1/15/2003 11:27 AM
> To: kur...@mtco.com; Bikecurrent
> Cc:
> Subject: [BC] Making Low Ohm High Current Sense Resistors
>
>
>
> Steve Kurt asked of Bruce Ingle:
>
> "How do you estimate the wire resistance? Just use a nominal figure for
> resistance of 30ga wire (35.65 ohms per 100m, according to one table I
> found)? Apply a fixed voltage, and measure current thru it? Or do you
> have a proper milli-ohm meter?
> One drawback of using a length of wire as a sense resistor is the
> inductance of the wire. Shouldn't be much of an issue for a linear
> regulator, but it might(?) be a factor for a switcher."
>
> In my limited experience with this, I take a wire gauge table and look up
> the right wire gauge and length to use to get a resistance that's
> approximately what I'm shooting for out of a length of wire in the range
of
> 3 to 12 inches. In any case, I pick a wire gauge that is able to handle
> the currents at issue. I cut it to a bit more than the calculated length.
> And hook up my wires on each side of this custom current sensing resistor,
> then MEASURE the current it has set! At that point, I determine what
> percentage above or below my desired current this first approximation of
the
> resistor is causing to be produced, and alter the points of contact to the
> wire accordingly to fine tune the value of the sense resistor, then
measure
> how the system behaves again. It doesn't take me a lot of time to wind up
> with something that causes the right current level to be selected.
>
> This is ridiculous for mass production, of course, but works quite nicely
> for one or two of a kind personal projects.
>
> Interesting concern about the effect of a few curles of wire in a sensing
> resistor. My GUESS... I invite others who know there stuff better to
> comment here... is that SINCE the sense resistor is typically right at the
> output of the voltage regulation circuit, dealing primarily with pretty
much
> flat DC, its inductance won't be a problem, even in switching regulator
> designs. But again... I could be off base here, and invite others to
> correct me if this SPECULATION is in error.
>
> ---marty
>
> --
> To subscribe to this list send a BLANK email to
> To unsubscribe from this list send a BLANK email to
>
> ==^================================================================
> This email was sent to: wil...@surefire.com
>
> EASY UNSUBSCRIBE click here: http://topica.com/u/?aVxiBc.a48bh7.d2lsbGll
> Or send an email to: BikeCurrent...@topica.com
>
> TOPICA - Start your own email discussion group. FREE!
> http://www.topica.com/partner/tag02/create/index2.html
> ==^================================================================
>
>
>
>