Milliohms

Thread Starter

Thender21

Joined Jun 7, 2015
27
i work as an aitomotive tech and we generally use voltage drops for testing rather than resistance readings.

There are some advantages. Volt meters are cheap. They are only accurate to 1/10th of an ohm however and so they are not very useful on low current circuits.

A starter cable will not show a tenth of an ohm but still have snough resistance to fail to operate. .01 ohms * 200 amps = 20 volt drop on a twelve volt system? 10.5 when cranking due to battery internal resistance?

In other cases we are plagued by bad connections that wiggle or broken wires in harnesses. Spread terminals all of which cause intermittent issues that are plague to diagnose.

Sometimes a conductivity enhancer like stabilant 22 helps.

I am womderinf if a milliohm meter may be useful in identifying poor circuit connections for sensors or evaluating the quality of wiring from end to end.

More information is better. .1 ohms is zero information.

Yeah there are complications but the very most complicated and difficult tests we do are with intermittent resistance issues.

Ideally I would want a meter that could output a voltage proportional to the resistance measurement to a scope.

This would enabl many resistance based transducers to be scoped as well. Such as thermocouples.

Something like one third of cars lemon lawed after repeat repair attempts are electrical problems.

Thanks very much for advice ome way or the other.
-Andrew
 

WBahn

Joined Mar 31, 2012
29,979
I can't tell what it is you are looking for or asking about.

Most low-ohm measurements are made using a four-wire technique, often known as a Kelvin connection.
 

Thread Starter

Thender21

Joined Jun 7, 2015
27
This is a good example

http://www.brakeandfrontend.com/tech-tip-so-what-s-this-fretting-corrosion-stuff-anyway/

Problems with a throttle by wire throttle pedal position sensor are often caused be these types of issues.

Small resistances confuse many things intermittenly where fixed resistances are switched into circuits and the net coltage drop indicates the switch combinations for example.

Some applications have low tolerance for anhthing but perfect connections. I wonder if milliohm meters could shed insight. They seem expensive.
 

AnalogKid

Joined Aug 1, 2013
10,987
Since measuring a voltage across a resistance is equivalent to measuring a current, what about a clamp-on hall effect current probe for your voltmeter or scope?

ak
 

wayneh

Joined Sep 9, 2010
17,496
I'm not sure there is one answer to your problem. Looking at the ∆V across a starter cable or a ground strap during cranking is very different than finding a corroded wire in the harness to the computer. The difference is the current involved and the potential for damage - I wouldn't feel comfortable probing computer pins with +12V or ground, for instance, for fear of doing damage.

Maybe a variable, constant current supply would be useful.
 

ian field

Joined Oct 27, 2012
6,536
i work as an aitomotive tech and we generally use voltage drops for testing rather than resistance readings.

There are some advantages. Volt meters are cheap. They are only accurate to 1/10th of an ohm however and so they are not very useful on low current circuits.

A starter cable will not show a tenth of an ohm but still have snough resistance to fail to operate. .01 ohms * 200 amps = 20 volt drop on a twelve volt system? 10.5 when cranking due to battery internal resistance?

In other cases we are plagued by bad connections that wiggle or broken wires in harnesses. Spread terminals all of which cause intermittent issues that are plague to diagnose.
When I made my living repairing VGA monitors, I got a lot on the bench with intermittent colour because the wires in the cable had fatigued close to the plug or chassis strain relief gland. It was pretty much impossible to detect the faults with a continuity tester or DMM, so I rigged up a 12V 9Ah SLA battery and a H4 headlamp bulb - it not only finished off the intermittent wires, but most of the ones that were about to fail very soon.
 

Dr.killjoy

Joined Apr 28, 2013
1,196
When I made my living repairing VGA monitors, I got a lot on the bench with intermittent colour because the wires in the cable had fatigued close to the plug or chassis strain relief gland. It was pretty much impossible to detect the faults with a continuity tester or DMM, so I rigged up a 12V 9Ah SLA battery and a H4 headlamp bulb - it not only finished off the intermittent wires, but most of the ones that were about to fail very soon.
I forgot lets burn some more crap up to figure out the problems and then replace lol ...Nothing like the smell of burning wires in the morning...Awesome times.
 

ian field

Joined Oct 27, 2012
6,536
I forgot lets burn some more crap up to figure out the problems and then replace lol ...Nothing like the smell of burning wires in the morning...Awesome times.
The cable became noticeably warm if I left the current flowing for a few minutes, but none ever burned.

If a wire in the cable has fatigued and only a couple of strands still intact, I'm going to get a warranty return that takes my time away from paying jobs - its far better to detect any failing wires while its on the bench where I can do something about it.
 

Thread Starter

Thender21

Joined Jun 7, 2015
27
The schematic drawing of that tool is not exactly in a format I am used to but it looks as though it generates a regulated current of 10 ma with the device under test in series

Then the voltage drop can be measured across the DUT to indicate resistance.

The tool would have to supply a voltage and regulate the output current despite variances in the battery voltage and adaptively to the DUT which could have a widely varying resistance.


Using a fixed current across a fixed resistance reveals the resistance via voltage a la ohms law?

Did i understand? Can i make one?
 

Thread Starter

Thender21

Joined Jun 7, 2015
27
The cable became noticeably warm if I left the current flowing for a few minutes, but none ever burned.

If a wire in the cable has fatigued and only a couple of strands still intact, I'm going to get a warranty return that takes my time away from paying jobs - its far better to detect any failing wires while its on the bench where I can do something about it.
I had a disagreement with another tech today because we were working on a car that had a small electrical fire that discolored the left fender. And we were replacing a fuse holder there that melted. The fuse is secured to eyelets via two small screws.

When I replaced it I noticed one screw was hot and the other wasn't. Only makes sense if there is more resistance there since there is no variation in cooling.

120 fahrenheit on that screw. 100 on the other. .2 volt drop at 20 amps. 4watt loss?

Replaced fuse and cleaned terminals no further issue there.

More electrical and wiring problems down the line but not my monkey. I would have liked to make a correct repair and determine the cause of the fire. It was near a gasoline vapor emissions line.
 

Thread Starter

Thender21

Joined Jun 7, 2015
27
Say what?! --- A current of 200A through a resistance of 10 milliohms produces a drop of 2V
Moreover, drops greater than the supplied EMF are not possible in pure resistive and/or pure DC circuits...

Best regards
HP
Typo/mental retardation.

In terms of raw energy in wattage, 2V * 200A = 400 watts!! Yikes.
 

#12

Joined Nov 30, 2010
18,224
Using a fixed current across a fixed resistance reveals the resistance via voltage a la ohms law?

Did i understand? Can i make one?
Yes, you understood. This is an excellent circuit for ignoring its own battery variations and revealing small resistances by forcing a very stable current through the suspected components. Whether you can make one? I don't see why not. It's only 3 parts and a battery! (The LED is optional) and a sample circuit board is pictured! If you can't do that, maybe Tracecom will sell you one.
 

Thread Starter

Thender21

Joined Jun 7, 2015
27
I forgot lets burn some more crap up to figure out the problems and then replace lol ...Nothing like the smell of burning wires in the morning...Awesome times.
I had a wire that could have been a fusible link it was broken and needed replacement. It was not documented what it was but it was weird and connected directly to B+.

In the end I put it in series with a jumper wire and a jump box that has an on off switch.

Flipped it and let the smoke out like crazy.

Not a fusible link. Safe to repair with regular wire. Poor electrical work by the OEM, the wire is surrounded by grounds and could shafe easily.
 

#12

Joined Nov 30, 2010
18,224
Not a fusible link. Safe to repair with regular wire. Poor electrical work by the OEM, the wire is surrounded by grounds and could chafe easily.
This is not the place to wax poetic about automotive repairs, they are against the rules of the website. As long as you were working on how to do measurements, you were safe. Say the word, "car" and the moderators will shut this down.
 

Thread Starter

Thender21

Joined Jun 7, 2015
27
This is not the place to wax poetic about automotive repairs, they are against the rules of the website. As long as you were working on how to do measurements, you were safe. Say the word, "car" and the moderators will shut this down.
I see

I dont understand yet but I see how it is.
 

#12

Joined Nov 30, 2010
18,224
No reason, just policy. :D

Seriously, that's the rules. You were supposed to read the Terms of Service when you signed up and that is where it says, "No Automotive"! This isn't a Democracy, it's a privately owned website. The owner doesn't want automotive conversations, and that's all there is to it. He who pays the bills makes the rules.
 
Kinda like the idea. There is something called a tone milliohmmeter that might help you. It's not quantitative,, but it might respond to wiggling.

A milliohmmeter typically requires force and sense leads and can be 4 or 5 wire. Think of the wires as force and sense and you can look up Kelvin measurements. There are Kelvin clips that essentially make two connections to one point.

I understand your frustration because some of the sophisticated methods sometimes fail to detect a "wiggle break" wit Ethernet cables.

"Thinking outloud", I'm wundering if you had something like a 10 mV current limited source applied to your wire/connector combo and then attached it to a feedback ammeter (Current to voltage converter) that has very low voltage drop (<1 mV possible) and can measure extremely small currents.

Forcing current and measuring voltage works too.

I'd probably start using the calculated resistance of 10' of 18 AWG and a typical contact resistance.

I thermal camera (pricey) might help you as well.
 

Thread Starter

Thender21

Joined Jun 7, 2015
27
Kinda like the idea. There is something called a tone milliohmmeter that might help you. It's not quantitative,, but it might respond to wiggling.

That may be viable. Whats implrtant is detecting anomalies.

I had one the other day a PWM solenoid with a standard resistance of about five ohms. It set a code for incorrect current flow. I measured 249 ohms from the connector inside the car. But the transaxle had been replaced and this wiring may have been crushed at that time.

A heat shield that covers the wiring from radiant exhaust hest was absent. Had to drive and get gas by which time the whole fault was gone completely.

Have to wonder if heat could have induced that much resistance into the wiring even if it was near the exhaust it seems unlikely.

Via analysis of a specific combination of circuit faults I deduced that elsewhere there is a bad splice within the main wiring harness that goes OPEN completely and I cant inspect it because it is in a 2 inch thick harness behind the dash.

It has opened a few times and although not our fault we take the blame. So I am in a tough spot of trying to guarantee repair on a car with ghosts.

A milliohmmeter typically requires force and sense leads and can be 4 or 5 wire. Think of the wires as force and sense and you can look up Kelvin measurements. There are Kelvin clips that essentially make two connections to one point.

Now I understand that clearly. I will look into it as I have planned. The cheapest milliohm meter is 2 wire, not compatible with a dvom or scope and $100

I understand your frustration because some of the sophisticated methods sometimes fail to detect a "wiggle break" wit Ethernet cables.

"Thinking outloud", I'm wundering if you had something like a 10 mV current limited source applied to your wire/connector combo and then attached it to a feedback ammeter (Current to voltage converter) that has very low voltage drop (<1 mV possible) and can measure extremely small currents.

It is not practical to do inline current measurements because we would usually have to break a wire open. We can open a single wire at a connector and insert a meter there but that normally opens all other wires as many as 18-26 that would need to be jumped to comsider it non invasive.

What we do is use a low amp clamp with a resolution of at best 1ma that outputs a voltage with adjustable scaling. Then scope the voltage while the circuit is loaded and perform wiggle tests etc.

A simpler approach is a dummy load of a lightbulb and buzzer for visual and audible feedback but you just have to get lucky.

Wires break near connectors inside the insulation sometimes from repetitive strain. Other times oxidized terminals rub and vibrate inducing intermittent conditions.

Sometimes terminals spread or back out of connectors causing poor comtact and intermittents. And when these happen an avalanche lf other problems may occur, due to for example a loss of communication between networked modules, in some designs power distribution is controlled via serial data requests.

Now you see it, now you dont. Often long enough to cause a malfunction but with no recorded disgnostic data. Any invasive test may erase all evidence or clues.

Forcing current and measuring voltage works too.

I'd probably start using the calculated resistance of 10' of 18 AWG and a typical contact resistance.

Wire gauge and length and current load if fixed could be used as a standard if you can measure fine voltage drops on a regulated low current. Our wiring is a mess though so only rough estimates of length are possible.

I thermal camera (pricey) might help you as well.
We are waiting for prices to come down. Some guys see potential but we have to keep things simple. See attached photos.


Thank you (and everyone else) for your helpful replies.

May the moderators have mercy on me I didnt have a minute to read the rules at the time...

I take my work seriously... If it goes out the door it is Not coming back if I can help it. Not unless it is back to bring Cookies!!!
 

Attachments

Top