Measuring Resistivity in a sample of A356

[esp42089]

I'm trying to measure the resistivity of a sample of A356 to back out the thermal conductivity. I do not have anything sensitive enough to measure the micro-ohm resistance (my sample is 0.5" x 0.25" x 3").

If I want to use an oscilloscope or multimeter to measure the voltage drop, I should need in the neighborhood of 1000A for a rough 10mV drop. Could I use a car battery to provide this colossal current?

Thanks for any help.

 

[f95toli]

I have a feeling that this is one of those "if you have to ask..." questions.
Measuring micro-ohms resistances is very tricky, at least if you want reasonable accuracy.

What kind of equipment do you have available?
And, more importantly, what kind of accuracy do you need?
I am assuming you will use a 4-point setup, but have you considered factors like thermoelectric effects etc?

 

[Carl Pugh]

The computer says that A356 is aluminum casting alloy.
Why don't you just get a long thin piece where you can get a reasonable voltage measurement with a reasonable current?

 

[esp42089]

I will be using a 4 point method to measure this and I have several multimeters and an oscilloscope on hand to play with.

I am trying to use tensile bars that we already have made instead of tooling a whole new bar for resistivity experimentation. If this proves to be a reasonably consistent measurement, then I will retool and put a "resistivity" bar in with the tensiles that is longer and thinner.

Thermoelectric effects are going to be ignored at first and if the sample gets hot too fast then I will start figuring how to mitigate that.

My plan is to measure our A356 sample and then a 6061 sample, and use the 6061 which has known data to linearly scale my results for A356. This is not terribly accurate, only little better than a back of the napkin sort of calculation to see what we get. I plan on using the Wiedermann-Franz law to back my thermal conductivity from the resistivity.

 

[sophiecentaur]

Measuring very low resistance is sometimes done with a bridge circuit. This can ,at least, reduce the effects of contact resistance, which may dominate in any simple V/I measurement. Also, I seem to remember seeing a bridge circuit working with AC and transformers, used for actually measuring very low contact resistances. Does this ring a bell with anyone?

 

[AlephZero]

Is there any reason why you really want to measure this, rather than just look up the data?
http://www.springerlink.com/content/h535l6u243112827/
http://www.engineersedge.com/properties_of_metals.htm

The first one also describes the experimental setup they used...

 

[esp42089]

I think it may help you out to know I am an industrial engineer working for a foundry.

My main reason for this effort is that I need to be able to compare our metal to industry standards. Especially since we aren't using straight A356.2 ingots in our batches anymore, we are recycling old castings and we want to check that it falls within certification boundaries.

EDIT: I've looked into bridge circuits and it looks like a modified wheatstone bridge such as the kelvin double bridge may be of great assistance in the actual measuring of the resistance. It seems I will still need a large current source to get these measurements to be accurate as resistors in each branch should be proportionate, so we come back to the issue of detecting micro-ohms requires a tremendous current to develop an appreciable voltage drop.

 

[Carl Pugh]

This has to be a fairly common measurement. Have you done any research to see how other engineers make this measurement?
If you google microvolt meter there are many hits. Is it practical to use a reasonable current and a microvolt meter?
When you load a car battery, the voltage decreases. This will be a problem if you are making accurate measurements.
Lower currents will also give less problems with the sample heating.

 

[Carl Pugh]

After thinking it over, here are some other comments.
If you are working for a foundry then the measurements should be done according to some standard.
In the USA the "American National Standards Institute" (ANSI) is usually the standards that everything meets. You should get a copy of the appropriate standard for making this measurement and follow it.

 

[f95toli]

I will be using a 4 point method to measure this and I have several multimeters and an oscilloscope on hand to play with.

The problem is that this kind of measurement requires equipment that is in a different league compared to what you would use in "normal" everyday measurements, a normal multimeter would be pretty much useless unless you put a good pre-amp in front of it (and to be honest I am not sure that would work very well either).
I'd say you'll either need a specialized instrument (micro-ohm meter) or you have to make your own setup using a few different instruments (say a good current source and a nV-meter etc). You should probably be looking at equipment from Keithley or a similar company.


If you are going to do this routinely it would probably make sense to spend a few thousand dollars on a good quality micro-ohm meter (my guess would be that you can get one for say $4000 or so).

 

[sophiecentaur]

I think it may help you out to know I am an industrial engineer working for a foundry.

My main reason for this effort is that I need to be able to compare our metal to industry standards. Especially since we aren't using straight A356.2 ingots in our batches anymore, we are recycling old castings and we want to check that it falls within certification boundaries.

EDIT: I've looked into bridge circuits and it looks like a modified wheatstone bridge such as the kelvin double bridge may be of great assistance in the actual measuring of the resistance. It seems I will still need a large current source to get these measurements to be accurate as resistors in each branch should be proportionate, so we come back to the issue of detecting micro-ohms requires a tremendous current to develop an appreciable voltage drop.

I wonder whether a totally different approach could help. What about looking at the effect of surface resistance on the absorption of em waves? The Q of a cavity resonator is affected by the surface resistance. A clean surface (freshly cut) of the metal to test could be inserted into a cavity and the resulting bandwidth of the cavity would tell you the resistivity.

OR one could look at the Isquared R heating of a sample, calorimetrically. It would be easy to produce wires of a measured CSA and a high current.

I guess the first thing to know would be what percentage difference in resistivity you are actually looking to measure.

 

[esp42089]

Thank you for the replies.

So, this isn't a standard measurement, in fact no one does it. What we are looking to do is get some data on our metal without having to invest in expensive equipment. We want to eventually get a spectrometer and an micro-ohm meter but it isn't in the budget, so here I am being told to think up some backyard ways of measuring micro-scale properties.

This should be in the ballpark of 44 micro-ohms/ sq. cm. going by the tables. So I am prolly trying to see differences on the 1 micro-ohm or 0.1 micro-ohm scale. I agree, this isn't really possible without more sophisticated equipment. I'm trying to come up with a reasonably inexpensive way to show my boss this is unreasonable.

 

[Carl Pugh]

Google A356 properties.
The first one that comes up gives all the standards for A356.
Ask you boss which standard he wishes to meet.
Trying to devise your own standard is just wasting your time.

You are probably looking for "Electrical Conductivity, % of IACS at 20 deg C (68 deg F)"

Good Luck
Carl