View Full Version : Matals ID via the Seebeck effect

Forrest Addy
05-22-2009, 04:45 AM
Here's something for the serious experimenter to monkey with.

A piece of apparatus long used in the machine shop QA section for material ID and verification is the AcroMag. The AcroMag is a heated thermocouple probe connected with a millivolt meter. The arrangement is a room temp copper plate grounding stage on which the unknown material is placed and a heated temperature controlled copper probe. Both stage and probe are connected to a multi range millivolt meter to sense the voltage developed between copper stage and the hot tip of the probe with the unknown sandwiched between.

All metals have distinctive thermo electric response when contacted by a heated probe of dissimilar metal. The response is different in varying degrees depending on the alloy constituents of the material under test. The material is identified by its thermoelectric response - that is, the thermoelectric generated voltage when the hot probe contacts the unknown sample. This works best and most reliably when known samples of materials can be compared to the unknown. When the generated millivoltage is equal between known specimen and the unknown to be ideentified the material may be presumed identical. This is not a perfect process but it's quick and convenient - and safe to use once its limitations are learned and kept firmly in mind.

In the home shop. the apparatus can be assembled easily from common shop equipment specifically a piece of copper plate, a millivolt meter, connecting wiring, and a temperature controlled soldering iron whose guts have been tweaked so that the soldering tip is electrically isolated from other circuits. The one manufactured part will be a copper tip to replace the original soldering tip. The millivolt meter is connected to the gounding plate and the heated bare copper tip.

The thermal regulation of the soldering iron will have to be pretty good; I strongly suggest one that is digitally control;ed. The iron is not run at soldering temperatures. AcroMag units uses 257 degrees F (125 degrees C). I suggest this figure be used. It seems to be the highest temp you can raise clean copper to before it films over with oxide.

Look up "Seebeck effect" in Google for further info.

05-22-2009, 08:50 AM
This may work if the temperature of the probe and temperature of the metal specimen under test can be accurately controlled or measured. This would be affected by the quality of the contact which would make the apparatus difficult to build. I'm not sure how much potential differences there would be for different alloys. I suspect that the differences may be small enough that it will be very difficult to measure.

Forrest Addy
05-22-2009, 10:38 AM
Rotate, I've used the AcuMag apparatus over many years. It's bone simple and reliable for matching thermo electric properties of unknown metals to specomens of known metals. The numerical reading of the meter is also helpful if a reliable chart exists.

The key to the whole thing would be reliable comparisaon specimens. If 4130 is desired and the unknown looks likely then the 4140 specimen is selected. The probe needs to be touched to the unknown and the speciment for a moment. If the reading match within a narrow range then you have a good start for a positive ID. You cannot witlessly rely on this apparatus. It's tradings have to be supported by other tests and observations and your native wit. If this test is followed by other simple tests (vissual exam, spark test, magnet test, hardness test, file test, etc, an very reliable comparison can be made and if the specimen is indeed the right stuff then the unknown is a certain match to is.

If the probles temperature is not reliable then good identifications ate still possible by operating in comparator mode using repeat reading of the known specimen technique. You're looking for a match within a narrow range. My remembered experience is that reliable distinctions can be made between 1018 and 1040 1430 and 4140, Water hardend drill rod and oil hardening, M2 HSS frpm M42. valve bronze from bearing bronze, 5000 series aluminum from 6000 , and so-on. Modered digitally controlled soldering its are accurare to a very few degrees. I wouldn't reccommen one not havng thermostatic control. That said, a massy solid copper probe heated in a light oil bath comes to mind. An electric fondue pot and a candy thermometer are likely ancillary items for this approach.

The reading in millivolts doesn't mean much except provide a sensitive and high resolution means of differentiation between the many alloys. If one were patient and had access to many alloys whose analysis was accurately known one could make up a chart giving millivolt readings for the various metals in the specimen collection. However you still need the specimens to rule out the odd erratic reading.

Needless to say the points of contact between copper ground, item under test, and proble have to be clean of rust, paint, plating, and goo.

05-22-2009, 12:40 PM
My buddy and I happened to stumble into a large SS fume hood last year while away at school. That is exactly how we identified the material as 316 SS. We had access to some very accurate electronics from the Cloud and Aersol Sciences research center and it worked well. We knew it was SS and 304 or 316 seemed reasonable. We took along a coupon of each material and compared them.

(not that we actually needed to know - this was more of an "excercise" because we were bored one weekend)

05-23-2009, 06:32 PM
material ID and verification
AcroMag; Forrest, that took me down memory lane. In my other life I used to run our companies X-Ray Micro Analytical Lab. Among other things we did a lot of Tig welding repair of jet engine blades and vanes. These components ran in the ‘hot’ section of the engine and were either Cobalt or Nickel base super alloys. In the old days they relied on the vendors chemical certifications of the different weld wires they supplied. Occasionally there were problems and I had to do a full quantitive analysis using the electron beam microprobe to see what was what. This was time consuming and a real PIA. Finally the welding department got an AcroMag with a set of standards and started checking all the incoming weld rod themselves. What a relief. Once in awhile I still had to do some work for them, but that little instrument sure saved me a ton of work.