Edge finding with a multimeter

[digiex-chris]

I don't have an edge finder, and didn't want to wait to get one to start my project. I discovered this a while ago when I was using a Taig, and it worked well today.

Take some thin uniform paper (thermal receipt paper works well), and place it between the jaws of your milling vise to electrically insulate your workpiece from the vise. It doesn't seem to affect accuracy that I can see, and seems to slightly increase holding friction. Place your multimeter into continuity beeping mode (so it will beep when you touch the probes together). Check that you've got continuity from wherever you're clamping the one probe to your tool (put one probe on the tool, the other where you're planning on clamping the probe). Replace your cutter with a dowel of a known diameter.

Now you can place one probe on the workpiece, and one probe wherever on the machine (where you'd tested before), and feed in the dowel till it beeps. Back it off, and feed it in in smaller and smaller increments, soon you'll find the point where your dials will show 1 division away it won't beep, and if you feed it in 1 division, it will beep. You've found the edge within the accuracy of your dial. I suspect you can get as accurate as you want, just make sure you keep any metal dust or shavings off of the dowel and the part or it'll get continuity through that.

I also use the same technique when I'm switching between a roughing tool and a finishing tool on my lathe, by wrapping the finishing tool bit in paper. I use a thin shim to keep the clamp bolts from breaking through the paper.

 

[mklotz]

Be careful of oil film on either the dowel or the part it is touching. I'm speculating but I would think that if you have to press the parts together hard enough to get metal-to-metal contact in the presence of oil, you might get a false edge indication.

 

[digiex-chris]

Oil tends to be non-conductive, so it shouldn't interfere. At least, it hasn't been a problem for me, but I'm usually going for 0.001" or so. Automotive points ignition systems arn't affected by oil at all. Worth experimenting with though. I tend to make sure things are extra clean using this method anyway though, because chips and dust will cause trouble.

 

[mklotz]

If there's a slight oil layer you may need to push the two pieces of metal together with extra force to get metal-to-metal contact. In effect you need to push the oil aside. This extra force might lead to erroneous edge detection.

Automotive points have plenty of voltage behind them to punch through thin oil layers plus they're sprung tightly against each other. It's not comparable to sensing a few thousandths.

Try a few experiments and report on your results. I'll be happy to be proved wrong.

 

[Mosey]

Oil tends to be non-conductive, so it shouldn't interfere. At least, it hasn't been a problem for me, but I'm usually going for 0.001" or so. Automotive points ignition systems arn't affected by oil at all. Worth experimenting with though. I tend to make sure things are extra clean using this method anyway though, because chips and dust will cause trouble.

Gee, I've had problems for decades with points and condensor ignitions because of oil on the contacts insulating the sparks. How come?

 

[blighty]

what an excellent idea!

this is the same way a cnc touch probe does it (well one way) but instead of a beep it makes the circuit that the software picks up, records the DRO and then does a number of other things you may want do.

i made one for my cnc, but as i don't have a BOB i would get false triggers from time to time.

the only problem i can see with this, is the work may be sitting on parallels but you could go the other way i.e isolate the dowel. the probe i made was 20mm od of delrin with a 10mm hole up the middle. then there was turned bit of aluminum stuck out if the 10mm hole with a 6mm cylinder turned on the end of it. turned out to be more accurate then the collets i was using.

either way its probably just as good as using a cigarette paper.

as to the oil....... just give it a wipe.

 

[n4zou]

ENCO sells an electronic edge finder for $26.20. When the edge finder contacts a conductive metal surface it lights up.
http://www.fvfowler.com/pdf/2304/2304_481.pdf

 

[blighty]

ENCO sells an electronic edge finder for $26.20. When the edge finder contacts a conductive metal surface it lights up.
http://www.fvfowler.com/pdf/2304/2304_481.pdf

and that's the other way ;D same as digitizing probes, but with out the ruby ball.

 

[Mosey]

Look for an electric edgefinder that has the ability to pop out of position if you run it into the work, cause the cheap ones I had were damaged by hitting the work and then bent. There are some with this spring-loaded safety, but they are pricey.

 

[fcheslop]

Whats wrong with the old ciggy paper :big:I must be to low tech :big:
Nice idea
best wishes frazer

 

[digiex-chris]

The nearest place that sold cig paper is 60km away and it was near midnight

 

[mu38&Bg#]

I use the multimeter in the CNC mill I built, but only when centering a round object. And the table is isolated from the spindle so I don't have to worry about it.

I have this link bookmarked. http://www.einfach-cnc.de/kantentaster.html I just have to find a ball type edgefinder with a small shank. I've only found 3/4" shank.

 

[kvom]

Look for an electric edgefinder that has the ability to pop out of position if you run it into the work, cause the cheap ones I had were damaged by hitting the work and then bent. There are some with this spring-loaded safety, but they are pricey.

I have a bent one too.

 

[Omnimill]

ENCO sells an electronic edge finder for $26.20. When the edge finder contacts a conductive metal surface it lights up.
http://www.fvfowler.com/pdf/2304/2304_481.pdf

Nice!

 

[digiex-chris]

Wouldn't oil have the opposite effect on the offsetting type? It would make it break sooner leaving the cutter futher from where you want it?