On Apr 24, 8:05*am, wrote:

So, could you do the usual electronics fault finding thing of using
the diode check function on the output lead of the alternator? - (then
reverse, repeat, see difference or none) even cheap DMM have them now
- (BTW - was yours a SABTRONICS kit?)?

Andrew VK3BFA.-

The alternator windings look like shorts to the meter and complicate
the diode measurements. If you can break such loops in the circuit it
should work, but the most informative way to check resistance and
forward drop is to apply a controlled current, ie a power supply and
some limiting resistance, and measure the drop with a voltmeter. This
will let you check devices with forward drops above the open-circuit
Ohms voltage of a multimeter (which you can check with another meter).

If you measure the voltage at several widely separated currents like
0.1, 1, 10, 100mA and plot them you can see resistance as the slope of
the line, Volts per Amp, and junction drop as the voltage projected to
zero current, in other words a junction with series resistance. If the
current is an Amp or so you can measure the resistance of motor
windings this way pretty accurately.

If you haven't acquired a heap of surplus power resistors, try light
bulbs.

I think that's all correct. It's been a long day. I used a variable-
speed Bridgeport with a DRO for the very first time.

The DVM is based on Intersil evaluation boards and high precision
resistors salvaged from old projects. At the time I was a lab tech
building the creations of a brilliant Ph.D who had previously designed
some of Keithley's fanciest equipment. The home-made meter and a
wirewrapped 8080 computer got me a test engineer position.

A recession killed off that small company after lawsuits by worried
competitors had seriously weakened it.

Jim Wilkins

"David Lesher" wrote: (CLIP) But Leo seems to think proportional ==
linearly proportional.
I disagree but decided it wasn't worth arguing about.
^^^^^^^^^^^^^^^^^^^^^
From Your Dictionary.com: "Math.: "having the same or a constant ratio"

I've always been taught to put a test light between the negative
lead and the negative post. Do your same fuse study. If the
light is burning, something is on.

--
______________________________
Keep the whole world singing . . . .
DanG (remove the sevens)




"Karl Townsend" wrote in message
anews.com...
I need the truck to haul home my latest toy...

I've had to leave the battery disconnected cause the juice gets
sucked down. So, today I put the ohm meter between the + post
and ground. Got 180 ohm. Pulled one fuse at a time hoping to
find a faulty circuit. No joy, same reading the whole time.

Any suggestions on how to find what's draining my battery?

Karl

"DanG" wroteI've always been taught to put a test light between the negative
lead and the negative post. Do your same fuse study. If the light is
burning, something is on.
^^^^^^^^^^^^^^^^^^^^
It's not an electrical reason--it has to do with safety. Having the ground
cable dangling in the engine compartment poses no risk of a short (to
ground.) However, I consider an ammeter much better than a light bulb.
It's going to be in series with whatever is draining the battery, which
could cut down the current to a level that doesn't light the bulb. If you
do use a light bulb, use the lowest wattage you can find. Possibly even a
6v bulb in a 12v system.

On Apr 25, 3:55*pm, "Leo Lichtman" wrote:
"DanG" wroteI've always been taught to put a test light between the negative lead and the negative post. *Do your same fuse study. *If the light is
burning, something is on.

^^^^^^^^^^^^^^^^^^^^
It's not an electrical reason--it has to do with safety. *Having the ground
cable dangling in the engine compartment poses no risk of a short (to
ground.) *However, I consider an ammeter much better than a light bulb.
It's going to be in series with whatever is draining the battery, which
could cut down the current to a level that doesn't light the bulb. *If you
do use a light bulb, use the lowest wattage you can find. *Possibly even a
6v bulb in a 12v system.

You connect both in series and the bulb protects the meter from a
short or heavy load. The 10A range on the $4 HF meter reads down to
10mA, so it and a headlight bulb should be enough.

jsw

On Thu, 23 Apr 2009 11:18:17 -0500, "Karl Townsend"
wrote:

Follow up. JOY! Got it out with the torch. melted AL on the alternator I had
it so hot.

Now I really really hope this is the cause of the current drain.

Karl

Sorry I've not been tracking this thread. For future reference: a
thick-walled "impact" 1/2" drive 9/16" good quality chrome-moly
6-point socket should not have split even with a 3-foot cheater -- and
no bolt with 9/16" head exists that can resist that much torque.

Given enough force (long enough cheater) it should have either turned
the bolt or sheared it in torsion. After splitting a few Craftsman
sockets I started buying good ones. Oddly enough, some really good
ones are quite inexpensive: Wright at Fleet Farm. They're not
pretty (black, not shiney) but they're strong and well made.

As you discovered, red to orange heat always works but minimizing
collateral damage can be an issue.

On Fri, 24 Apr 2009 05:05:07 -0700 (PDT), wrote:



And I would humbly disagree re ohmmeters vs ammeter's - its all to
easy to fry an ammeter, the 10a range on most DMM's is rated for a few
seconds only, then the copper wire shunt disengages itself....besides,
10a is leakage current in a vehicle...and DMM's are now cheap, throw
away items, but a decent ammeter is still expensive....

Andrew VK3BFA.

Replacing fuses in a DMM used in ammeter mode can get old after a
while. Been there, done that.

For those inclined to tinker a bit, as I think Andrew might be, a
solution is to make some shunts of known (low) resistance. Connect
shunt in series with load, measure the voltage across the shunt,
calculate the current. Weston shunts of yesteryear were 50 millivolt
shunts. 100 amps at 50 millivolts is 5 watts, no big deal but not
negligable either. The shunt can be made of convenient material --
iron, steel, stainless, brass, whatever. Nichrome and carbon are
readily-available materials with fairly low temperature coefficients.
Constantin and manganin are much lower but not readily available.
Just make it big enough so it doesn't get spit-sizzle hot and
fuggedaboudit. Don't need 1% accuracy to be useful, right? My shunts
are made of copper, stainless, brass and steel. They work for me.

I calibrate/trim the shunts by measuring both current and voltage with
a controllable bench supply so as not to cremate the ammeter. Once
calibrated, the shunt and a voltmeter become my robust ammeter.

Don Foreman wrote:

(...)

I calibrate/trim the shunts by measuring both current and voltage with
a controllable bench supply so as not to cremate the ammeter. Once
calibrated, the shunt and a voltmeter become my robust ammeter.

Also, for those super low-impedance current measuring requirements,
you can use a Hall Effect current sensor to great effect.

I used some of these:
http://www.allegromicro.com/en/Produ...0756/index.asp

... to measure current in a situation where a 2 milliohm (0.002 Ω)
shunt was too much resistance for loop stability in a power
supply.
The sensors require a separate power supply, but they work a treat.

--Winston


--

Don't *faff*, dear.

On Apr 26, 2:44 pm, Don Foreman wrote:
On Fri, 24 Apr 2009 05:05:07 -0700 (PDT), wrote:



Replacing fuses in a DMM used in ammeter mode can get old after a
while. Been there, done that.

Especially if its a Fluke with UL listed fuses at $10 each...(in Oz,
anyway)
Which is ridiculous - you can fry a $10 DMM and just curse about it.
Same with over voltage protection in Flukes - had to fix mine twice
now, not cheap. Again, the $10 DMM would be cheaper/easier.
But I paid approx $200 for the Fluke, years ago, its my favourite
meter (no logic to that, of course)

For those inclined to tinker a bit, as I think Andrew might be, a
solution is to make some shunts of known (low) resistance.

Yep. Done that - gets a bit dodgy with high current and very low
resistance shunts - the R of the terminals etc etc. Very rarely do I
attempt to measure high currents anyway, the bench power supplies have
in built 30A meters - and the clouds of smoke are usually a dead give
away that something is wrong......(bugger)....
But its a good idea Don, thank you - I do have a few WW2 vintage round
Bakelite cased 100ma meters, so having one set up like you describe as
a dedicated high current meter would be a good idea....

Winston - thank you for the link, what a truly amazing device that IC
is....wonder if its available here in less than tube quantities...

Andrew VK3BFA.

On Sun, 26 Apr 2009 01:37:46 -0700 (PDT), the renowned
wrote:

On Apr 26, 2:44 pm, Don Foreman wrote:
On Fri, 24 Apr 2009 05:05:07 -0700 (PDT), wrote:



Replacing fuses in a DMM used in ammeter mode can get old after a
while. Been there, done that.

Especially if its a Fluke with UL listed fuses at $10 each...(in Oz,
anyway)
Which is ridiculous - you can fry a $10 DMM and just curse about it.

The fuse is there for SAFETY. Your $10 DMM probably has an unfused
shunt. Or a crappy fuse that is short enough that 240VAC can arc from
end to end if the fault current is high enough. Fine for electronics
and 12V, but not for an industrial control panel. How much is your
vision worth? Those aids for even partially blind folks are VERY
expensive.


Best regards,
Spehro Pefhany
--
"it's the network..." "The Journey is the reward"
Info for manufacturers: http://www.trexon.com
Embedded software/hardware/analog Info for designers: http://www.speff.com

On Apr 26, 12:44*am, Don Foreman
wrote:

For those inclined to tinker a bit, as I think Andrew might be, a
solution is to make some shunts of known (low) resistance....

I calibrate/trim the shunts by measuring both current and voltage with
a controllable bench supply so as not to cremate the ammeter. Once
calibrated, the shunt and a voltmeter become my robust ammeter. *

I just made one to test. 15" of 0.030" galvanized steel wire dropped
125mV at 1.25A. You could put it in plastic water pipe with terninal
screws through the wall near the ends, covered with tape. I would
crimp solderless terminals on the shunt and connecting leads so they
don't work loose.

When you experiment with your own wire, the length to use is between
the inner two connections and it doesn't matter whether they are the
voltmeter or the current. The reason is a good student exercise. Hint
- what's the I*R drop in the voltmeter leads?

The original HF probe tips plug into the curled edges of female 0.250"
Faston terminals.
http://www.aircraftmech.com/pics/faston.gif
You might have to open or close them slightly, and insulate them with
tape or heatshrink. I'd make the shunt-to-Faston wires short so the
partly exposed meter probes are close to the plastic pipe, maybe even
taped to it. If you've ever shorted a wrench across a car battery you
understand these precautions.

The $4 HF meter resolves 0.1mV on the 200mV range, ignore the decimal
point and take the entire displayed reading as Milliamps.

I have at least four of those HF meters for fault monitoring like
this, as well as good meters for serious measurements. The HF ones are
accurate to +- 1 or 2 counts, the usual standard for DVMs, but the
good ones have more digits.

Jim Wilkins

wrote:

(..)

Winston - thank you for the link, what a truly amazing device that IC
is....wonder if its available here in less than tube quantities...

Can you buy from any of these vendors, Andrew?
http://service.stkcheck.com/Default.aspx

I see they are available in single quantities from my second -
favorite vendor:
http://search.digikey.com/scripts/Dk...CS756&x=19&y=9

Digikey will sell in single unit quantities for U$ 5.18.
They only have 1273 of them left though.

--Winston

--

Don't *faff*, dear.

On Apr 26, 8:51*am, Jim Wilkins wrote:
...
I just made one to test. 15" of 0.030" galvanized steel wire dropped
125mV at 1.25A. ...

Jim Wilkins

That steel wire stayed shiny while charging a battery at 5A.
Galvanizing dulls when it nears its melting point.

Winston wrote:
wrote:

(..)

Winston - thank you for the link, what a truly amazing device that IC
is....wonder if its available here in less than tube quantities...

Can you buy from any of these vendors, Andrew?
http://service.stkcheck.com/Default.aspx

Well, that didn't work at all!
Try this:
http://www.allegromicro.com/en/Produ...0756/index.asp
Scroll down and click on 'Check Stock'.

That should work better.

--Winston



--

Don't *faff*, dear.

On Apr 26, 11:11*am, Winston wrote:
wrote:

(..)

Winston - thank you for the link, what a truly amazing device that IC
is....wonder if its available here in less than tube quantities...

Can you buy from any of these vendors, Andrew?http://service.stkcheck.com/Default.aspx

I see they are available in single quantities from my second -
favorite vendor:http://search.digikey.com/scripts/Dk...ng=en&site=US&...

Digikey will sell in single unit quantities for U$ 5.18.
They only have 1273 of them left though.

--Winston

Notice that the noise level equals about a quarter of an amp. and the
total error can be 5% of 50A for the X050 over the Commercial
temperature range, so you need a way to calibrate or at least zero it.
The simple way to use it would be to connect a 1 - 5K pot across the
supply ( 3 x AA cells) and meter between the device output and the pot
wiper. Use the pot to zero the meter when current = 0. I would add a
shorting switch to zero the meter reading without disconnecting the
circuit.

That ought to be enough to check a truck electrical system.

Sears sells a DC clamp-on meter which may use that device or one
similar. I tried one in their parking lot on a cold day and found it
was unusable because the reading drifted so fast, 300 mA in less than
a minute as it cooled down.

Jim Wilkins

Jim Wilkins wrote:

(...)

Notice that the noise level equals about a quarter of an amp. and the
total error can be 5% of 50A for the X050 over the Commercial
temperature range, so you need a way to calibrate or at least zero it.

The noise is A.C. that A.M.s the output voltage, yes?
It can be filtered and / or averaged out.

The simple way to use it would be to connect a 1 - 5K pot across the
supply ( 3 x AA cells) and meter between the device output and the pot
wiper. Use the pot to zero the meter when current = 0. I would add a
shorting switch to zero the meter reading without disconnecting the
circuit.

Sounds like D.C. offset compensation. That is a good thing because
the offset spec is +- 30 mV.

That ought to be enough to check a truck electrical system.

For those otherwise impossible low-impedance current measuring
tasks, an accuracy of +- 7.5% from -40 to +125 C was
sufficient for my purposes. Readings from these devices
agreed within 2% with a good quality ammeter during my cross -
checks so I figured that was good 'nuff.

Clearly a naked DMM is more than sufficient for truck
electrical system testing, without these fancy devices.

I haven't seen a car battery yet that breaks into open loop
oscillation when presented with a load impedance 10% higher
than normal.

--Winston

--

Don't *faff*, dear.

On Apr 26, 1:20*pm, Winston wrote:
Jim Wilkins wrote:

(...)

Notice that the noise level equals about a quarter of an amp. and the
total error can be 5% of 50A for the X050 over the Commercial
temperature range, so you need a way to calibrate or at least zero it.

The noise is A.C. that A.M.s the output voltage, yes?
It can be filtered and / or averaged out.

They don't give the frequency range, it could include drift. If you
use an op amp to filter and level-shift the output, be careful that
the op amp doesn't rectify the noise into a false DC offset.

The simple way to use it would be to connect a 1 - 5K pot across the
supply ( 3 x AA cells) and meter between the device output and the pot
wiper. Use the pot to zero the meter when current = 0. I would add a
shorting switch to zero the meter reading without disconnecting the
circuit.

Sounds like D.C. offset compensation. That is a good thing because
the offset spec is +- 30 mV.

That ought to be enough to check a truck electrical system.

For those otherwise impossible low-impedance current measuring
tasks, an accuracy of +- 7.5% from -40 to +125 C was
sufficient for my purposes. *Readings from these devices
agreed within 2% with a good quality ammeter during my cross -
checks so I figured that was good 'nuff.

Clearly a naked DMM is more than sufficient for truck
electrical system testing, without these fancy devices.

I haven't seen a car battery yet that breaks into open loop
oscillation when presented with a load impedance 10% higher
than normal.

--Winston

I'm so used to having the right electronic test equipment that it's
hard to think up simple alternatives sometimes. This morning I checked
the DC output of a genny using a 300A Weston shunt to measure the
current.

jsw

Jim Wilkins wrote:

This morning I checked the DC output of a genny using a
300A Weston shunt to measure the current.

Sounds like a "git 'er done" attitude to me.
That's the way to go.

--Winston


--

Don't *faff*, dear.