Need Help!- Probing---Need electronic guru

[ozzie34231]

Hi All,
I've been working on a setup for probing on my lathe; screenset, VB scripts, etc.
It seened to me that for the probe there are 3 basic ways to go about it, an insulated cylindrical probe in a chuck, an insulated master tool, or an insulated QC tool post/holder.
The third is the really practical way to go if it can be done in a way that doesn't sacrifice rigidity.
Circuit board material is fairly thin, has good compressive strength, and cheap.
So I've insulated the post by placing a layer under it, a layer under the nut, and wrapping the stud with two layers of electrical tape.
Now comes my little problem:
My digital multi-meter shows total insulation between the post and all other parts of the lathe.
If I touch the probe wire to the post, there is no lighting of the screen LED.
Any place else I touch shows ground, (lighted LED).
If I run any one of my scripts from my screenset and use the probe wire to end the probe by touching the machine, all is well; the scripts conclude as they should.
Final test! I clip the probe wire to the insulated tool post and start a probe.
THE PROBE TRIPS OUT IN A SECOND. THE LED FLASHES, THE SCRIPT CONCLUDES.
I think I've gone over everything a hundred times, I have even tried a shielded probe wire, thinking it is noise when the servo starts. I've changed the insulation scheme, nothing helps.
I need HELP!
Ozzie

[Eurisko]

Ozzie,

I am not an electronics guru, but am willing to offer my 2 cents...

You'll have to post more details of the circuit to get a better answer, but here goes...

If you're using any CMOS gates, don't let the inputs float. Tie them to either supply rail through a resistor. A static charge can cause the outputs to change state.

Another thought is that the probe might be too sensitive to vibration. A slight movement of the probe (or machine) might be causing the probe to trigger.

[ozzie34231]

Well the electronics are as simple as possible, and by the way, the exact same setup I use on my mill, same computer, control box, same all. And there it all works. The breakout board is a direct connect to the parallel port cable, nothing on it but traces from the posts to the wires.
Ozzie

[Eurisko]

Ozzie,

Just a few questions to make sure that I understand your setup.

Your lathe is grounded to the computer's logic ground ?
The probe wire is directly connected to one of the parallel port input pins?
When the probe wire touches any metal on the lathe, the input pin is forced to logic ground?

[Al_The_Man]

In spite of the PC P.S. common being at ground potential, it is not a good idea to use ground as a conductor, normally probes should have two conductors, one for the trigger input and the other should go back to the P.S. common.
I know many hobby probes use the ground method, but industrial types always use a two conductor system.
Whether this is the reason for the glitch is debatable and would have to be verified.
Al.

[ozzie34231]

Eurisko, yes to all.

[ozzie34231]

Well with my limited knowledge of such things I found a fix. I put a 100 ohm resistor in the probe wire, and now it works, though the testing is limited so far.

Okay that's a solution, but I'd still like to know what problem the is.
Ozzie

[florins]

One 0.1 uF capacitor installed in parallel to the input (between the probe wire and the ground) might help in case you have a lot of noise generated by the drivers. There is no fast fix for this, but with such probes you should really look into protecting the input of the BOB / parallel port card, as the high static voltages generated can easy damage the ports.
The attached schematic might help, sometimes the diodes are included already in the circuitry of the BOB, but adding the extra protection will not produce any harm. The 1k resistor can be a bit on the hight side if the inputs are pure TTL, but for the modern circuits is just fine. Otherwise, you can reduce it to 330 ohms.

All the best,
Florin

[ozzie34231]

Thanks for the thoughts Florins.
I'd still like someone to suggest why the insulated toolpost act differently from a plastic handled screwdriver stuck in the chuck?????????
Ozzie

[florins]

...why the insulated toolpost act differently from a plastic handled screwdriver stuck in the chuck?...

Just a guess, as you didn't provide too much data: bigger capacitance between the toolpost and the machine ground, proximity of the cable to servo/stepper cables, etc.

Florin

[ozzie34231]

Since I'm electronically disadvantaged I don't know what data to supply. I don't understand "capacitance between the toolpost and the machine ground". Can you explain in oldslowmind to me.
I stuck the resistor in there because it seemed to me the post was too sensitive and I thought a resistor would calm it down; not through any insight.
I did try various arrangements of the probe wire relative to the servos and servo wires; no effect!
Ozzie

[Al_The_Man]

#1, the toolpost is on a travelling carriage and not rigidly attached to the rest of the lathe.
If I were to look for the source of the problem, the first thing I would do is take a 5v common conductor from carriage back to the PC common, even if it is only to the P.P. common.
We do not know how you have wired the machine grounding and bonding etc.
This can have a large bearing on how your circuit behaves, especially if you are depending on common return through the frame of the machine.
Al.