Provided that the environment is not electrically noisy, it's not too difficult to measure milliohms.

Look up 'milliohm circuit' to find some analog designs.

One trick for your back pocket is that, with the addition of a controller, you might be able to use a low duty cycle so that the measuring current is only applied a tiny fraction of the time (in case needed). This assumes that the level isn't changing too quickly.

VE1BLL,
Thanks for the reply. Unfortunately I am restricted by my data acquisition equipment to measuring either a 0-10 vDC or a mili volt (thermocouple module) signal. I'm unable to measure resistance or current directly.

?? If you can measure 0-10VDC, then you can measure voltage, and you would use a simple voltage divider. That said, however, you need to determine how to keep the molten metal from freezing onto your probes. One possible option is to have them covered with something to which the metal will not stick permanently, except for certain spots along the length of the probes.

What is the temperature of the metal? Most materials increase resistance as temperature increases.

TTFN
FAQ731-376: Eng-Tips.com Forum Policies
[IMG http://tinyurl.com/7ofakss]
Need help writing a question or understanding a reply? forum1529: Translation Assistance for Engineers


Of course I can. I can do anything. I can do absolutely anything. I'm an expert!
There is a homework forum hosted by engineering.com: http://www.engineering.com/AskForum/aff/32.aspx

Random idea: Perhaps the resistance wire could be coiled around a ceramic form, so that you end up with perhaps about ten feet of resistance wire spanning the one foot overall dimension. Obviously the turns would have to be spaced slightly apart.

You can't weigh the crucible?

You have a third problem, in that whatever wire alloy you use can't be attacked (dissolved) by whatever liquid metal you have, and it needs to resist oxidation where it extends past the molten metal pool. It sounds like you need some very fine platinum wire.

Hi,

This is going to be used in a permanent die Al casting. one set of probes will be cast into each casting to measure the fill rate. This is for a trial to characterize die filling. The probes are disposable and the fill time is short so the erosion of the probes should not be a problem. I can calibrate for the degradation of the probes. I can't weight the crucible or die as it is not practical. I believe i have all of the method and experimental design well thought out. My problem is measuring the signal with such a small voltage drop.

Although my voltage module in the DAQ can read 0-10V if i only get 0.1V voltage drop over the 1 foot length of wire then the noise in the system and lack of resolution will compromise the quality of the data. Having a 5-10V voltage drop would significantly improve the data. The problem with using a voltage divider is that the get a reasonable voltage drop the other resistor in the circuit would have to be 1-6 ohm. The resistance of any wire used in the circuit might be larger than this. I have also never seen a resistor made with this resistance magnitude. I have limited electrical experience so i might be wrong about these last two points.

The question is how can i obtain a significant voltage drop (>4v) with only a variation of 6 ohms of resistance?

You need to amplify the signal into your data acquisition. I normally work with ICs, so I would make a circuit like 10V-fixedresistor-probewire-fixedresistor-0V, use the two fixed resistors so not a lot of current was running through the probe, connect a differential amplifier across the probe wire, and gain the voltage, by, say 10 to 100, into my data system.

You've not stated what kind of DAQ you have, but many of them have strain-gauge amplifiers available, and such an amplifer would work fine in this setup. You could get fancier, and work your probe into one leg of a Wheatstone bridge (which is what a strain-gauge is), and have a loop of wire that is the probe, and a loop of wire outside of the die that is a reference leg.

Most resistor wire have a thermal coefficient, so the resistance measurement will be a mixture of both level and temperature.

A typical DAQ has at least 16 bits of resolution, which means its voltage resolution is at least 152 uV, which is 600 times smaller than the voltage you're citing.

TTFN
FAQ731-376: Eng-Tips.com Forum Policies
[IMG http://tinyurl.com/7ofakss]
Need help writing a question or understanding a reply? forum1529: Translation Assistance for Engineers


Of course I can. I can do anything. I can do absolutely anything. I'm an expert!
There is a homework forum hosted by engineering.com: http://www.engineering.com/AskForum/aff/32.aspx

Thanks Comcokid. I'll look into that.

IRstuff, Yes you are correct, however the I suspect that the noise from the power supply and other areas will have an effect. However i may try it and see what i get.

Thanks all.

Noise can be filtered; that's an EE problem, for which you should probably hire one to do the job right from the get-go.

TTFN
FAQ731-376: Eng-Tips.com Forum Policies
[IMG http://tinyurl.com/7ofakss]
Need help writing a question or understanding a reply? forum1529: Translation Assistance for Engineers


Of course I can. I can do anything. I can do absolutely anything. I'm an expert!
There is a homework forum hosted by engineering.com: http://www.engineering.com/AskForum/aff/32.aspx

I suggest you look up Wheatstone bridge and strain gauge theory. Both involve accurately measuring very small changes in resistance.

How about a simple set of probes at the full line. Continuity between the probes will indicate a full die and can be used to stop the fill process.

Bill
--------------------
"Why not the best?"
Jimmy Carter

Bill's concept could be extended to employ an array of such continuity sensors, one contact for each mm of depth if you wish.

...and you could print an array of them on a glass plate using standard pc etch techniques.

https://www.emf-corp.com/copper-slides.html

Here's a simple amplifier that will do the job. This however is just the starting point. Environment factors will make it a lot more fun than a simple non-inverting amplifier with the correct gain. Number one problem is that most resistance wire has a high temperature coefficient. Presumably this is an indicator, not a precision instrument, so empirical calibration will be sufficient.

                                               +12V
              --------------------------o---------------------------
              |                         |
              |                         |
             .-.                        |
             | |                        |
             | | 100 Ohm 5W             |
             '-'                        |
              |                         |
              |                         |
              |                         |
              |                       |\   LM358
              o-----------------------|+\              Output
              |                       |  >- -------o---------
              |                 |--- -|-/          |
              |                 |     |/           |
              |                 |      |  ___      |
              |                 o------|-|___|------
              |                 |      |
             .-.               .-.     |   51K
             | |               | |     |
             | |  Probe        | |     |
             '-'               '-'     |
              | 0-12 Ohm        |      |
              |              7K5|      |
              |                 |      |
              |                 |      |
              |                 |      |
              |                 |      |          0V (Gnd)
              ------------------o------o--------------------------
(created by AACircuit v1.28.6 beta 04/19/05 www.tech-chat.de) 

David Stonier-Gibson
http://splatco.com