Low Cost Electronic Edge Finder

By Rick Sparber with help from Alan Condit

I discovered that I could make an accurate yet low cost edge finder by using a few of the functions in the DRO to compensate for run-out. I've verified its accuracy with my mechanical edge finder and they are within a thousandth.

[ShumaTech] Direct-Coupled Edge Finder (from Shumatech Website)

"The design for the direct-coupled edge finder works by replacing the light bulb with two wires that connect directly to the edge finder, one wire to the negative battery terminal and the other to the probe. A general-purpose NPN transistor translates the 1.5V signal from the battery to a 5V signal as required by the DRO-350. When the probe touches the object, the switch representing the probe is closed which will bias the NPN transistor on via the 100K base resistor. The NPN transistor is in a common emitter configuration with a 10K collector resistor."

Scott Schumate’s Original Schematic

Comment on the above circuit

The battery, BT1, is actually within the DRO and is the positive voltage fed to the scales. You just have to touch the resistor, R1, to any metal on the mill to get Q1 to saturate. A minor point, but BT1 is drawn upside down. The voltage going to the probe is +1.5v but as drawn it is –1.5v.

 

I modified the edge finder interface circuit a little. It was too sensitive to noise plus had no ESD protection. The changes consist of lowering the base resistor R1 to 47K, adding a base-emitter resistor of 220K, adding a diode between base and emitter (cathode to base), and finally, I added a piezoelectric audible beeper across the pull up resistor. I sometimes just want the edge-finder to tell me it has hit a reference and not set a zero. When I hear the beep, I know I've arrived. It is also helpful in preventing excessive pressure on the edge-finder since the beep occurs the instant of touchdown.

 

 

Note: eventually I grew tired of the follow design's shortcomings and just modified an Enco electronic edge finder. The modification consisted of removing the battery and bulb. Then a wire was run from the tip, through the body, and out one of the holes. RGS 02/04/2007

 

The actual edge finder is described next. Here are my rough plans and a picture of the machined unit.

 

I made my prototype EF out of aluminum and the "product" from drill rod. The tip was turned to .314 thou diameter and the body was cut from 3/4" stock.

The EF is made from 2 machined parts plus a 3/4" long piece of surgical tubing. Start by locating a piece of tubing with an ID of around 3/8". Then try fitting various size drill shanks into it until you find a good snug fit. Measure the OD of the tubing with the drill inside and record that value along with the drill size.

Machine a 1 1/4" long cylinder the same diameter as the drill's diameter. Drill and tap the end 6-32. Call this part the tip.

Then take a second piece of round stock at least 0.1" larger in diameter than the OD of the tubing and about 2 1/2" long. Select a drill that is as close as possible to, but not larger than, the OD of the tubing and drill a hole in this round stock to a depth of 1". Cross drill a 1/8" diameter hole then enters at the bottom of this large hole. The wire will exit via this cross drilled hole. If you like, turn the top 1/2" to a diameter of 3/8", the next 1/2" to a diameter of 1/2", and the next 1/2" to a diameter of 5/8". In this way you don't have to change collets to use the EF. Call this part the body.

Run an insulated wire through the cross drilled hole. Strip the insulation off the end, tin it, and wrap it around a 6-32 screw. Screw the 6-32 into the tapped hole. Then slide the tubing down over this tip until it is even with the tapped end. Next soap up the tubing and slide it into the larger hole in the body. As you feed the tip and tubing down the hole, carefully pull the wire out the cross drilled hole. When none of the tubing is visible, stop pushing.

You now have an EF with the tip insulated from the body. More than likely, you also have a very large run-out. So the next step is to identify this run-out and compensate for it with the DRO.

Put the EF's body in a collet and mount it in the spindle of the mill. Using a dial test indicator pressing on the side of the insulated cylinder, find the point of maximum run-out. Draw a vertical line at this position on the body with a marker. Rotate the spindle 180 degrees and you should find the point of minimum run-out. Place a dot at that point on the body.

Measure the diameter of the tip, preferably with a micrometer and record it.

Connect the EF through the single transistor interface circuit described earlier to the DRO and enable the aux input.

There is a calibration procedure, which is done only occasionally, and a normal use procedure which is done every time.

Calibration:

1. Set the DRO to zero the axis perpendicular to the reference surface.

2. With the EF positioned so the line faces the reference surface, move the EF until it touches and the DRO zeros. At the instant of zeroing, you should hear a beep.

3. Back the EF away from this surface and rotate the spindle 180 degrees.

4. Crank the EF into the reference surface until you hear the circuit beep again.

5. Hit Function 1 and then the axis you are moving. Subtract the radius of the tip from this number. Store 2 times this result without the negative sign as a tool offset.

Normal Use:

1. Rotate the EF so the line on the side of the body faces the reference surface

2. Set the DRO for EF zero

3. Move the EF till it contacts reference surface and zeros that axis.

4. Hit Function 6 and then the tool offset number for the EF

5. Select the "cutter" face opposite the reference surface (for example, if the left side of the EF is touching the reference surface, select the right side)

6. Raise the EF above the reference surface and feed in to 0

7. Hit Function 6 and then select tool 0 (to turn off offset)

8. Set ABS 0 for this axis

9. The center of the spindle should now be lined up with the reference surface plane

 

I know this seems like a lot of steps but with a little practice it is

tolerable.

Rick Sparber

rgsparber@AOL.com

04/07/2006