Re: [CAD_CAM_EDM_DRO] Re: Driving unmarked steppers

----- Original Message -----
From: <ccs@...>
To: <CAD_CAM_EDM_DRO@yahoogroups.com>
Sent: Friday, February 15, 2002 8:02 PM
Subject: Re: [CAD_CAM_EDM_DRO] Re: Driving unmarked steppers


>
> > One might ask "what is a good current source" ? Well a good
> > lab supply can do that but so can a norton equivalent... a battery and a
> > resistor.
>
> Nitpick, but a voltage source with a series resistor is the THEVENIN
> equivelent of a current source with a shunt resistor, which is the
> NORTON equivelent of the voltage source with series resistor.

That is correct.
>
> A battery is neither a voltage source nor a current source - rather it
> gets modeled as your choice of these equivelents with an internal
> resistor

Also true... it is an approximation. But let's see how good it is.
Let's say you want to measure some small resistance between .1 ohm and 0
ohm. Let's say you have an inexpensive (perhaps fet) voltmeter that can read
100
millivolts. We will have to push 1 amp to get that voltage
across the .1 ohm. I^2R will be .1w which is not too bad as far as heating
up a motor winding.

For that 1 amp let's use a car battery (12.8v or so) and
a 12.8 ohm (or so) resistor in series. I^2 R for that is
12.8 Watts. It can get a little warm but not too bad.
A cheap wirewound resistor can have a TCR (temperature
coefficient of resistance) of well under 100 PPM/degree
centigrade. I just ordered some $4 10 ohm ones that are
+/- 20 PPM. If it gets to boiling water temperature it will
have changed by less than .7 percent. I just use a coil
of fine nichrome wire sometimes and it is big enough to not warm up much due
to its surface area. Pure copper has a much higher TCR but if it stays cool
to the touch it will be ok.
It needs to stay within a few degrees for 1% .

So what we want is a constant current. OK with a .1 ohm test
load hooked up total series resistance is 12.8 + .1 or 12.9.
Current is 12.8/12.9 or .992 amps. Now with a short instead
of .1 ohm the current is just the 12.8/12.8 or one amp. The error of the
current source approximation is less than one
percent over that range!

Our DVM will now read one milliohm per millivolt.

What I am getting at is that you can measure low resistances
with just some wire or a resistor and a cheap DVM certainly within a few
percent even with battery loading and such. Much much better than trying to
read it with a regular ohm meter. Just flexing the meter leads often changes
the resistance more than the figures I discussed. The key is the
4 wire kelvin system.



The production rig I mentioned had its resistors in a temperature controlled
bath and analyzed the voltage
division in software. It easily measured .001 ohm and
stayed in calibration well. In fact it is still being used to
measure my thermistor and conductor polymer inks.

Leslie Watts
L M Watts Furniture
Tiger, Georgia USA
http://www.rabun.net/~leswatts/wattsfurniturewp.html