Part location wire EDM

Hello,
When it comes to wire EDM, I am a total newb. I wanted to know how you guys locate the parts origin. I do have a ram type EDM which has an electronic edgefinder and also a rotating shaft that I can mount an indicator. Do you use the wire as an edgefinder? How accuratly can you locate an edge?
Thanks

On a sinker, you locate just like you would on a mill with an edgefinder. Touch off the electrode until it makes contact (CNC's have automated functions for this), set an origin point from there, and go. More common these days is to use a conductive probe to locate, then set your offsets from there. Since most graphite machining is done on machining centers these days, the datum points for all of the electrodes are the same and can be easily calculated.

On a wire, the process is pretty much the same. With wire feed on, the axis is moved until the wire makes contact. Again CNC's have automated pickups for this. Divide by 1/2 the diameter of the wire for zero, just like a standard mill edgefinder.

As far as accuracy, if the electrodes and workpeice have good finishes and are perfectly clean, accuracy is .0001 or better in most cases.

e723,

What kind of a machine are you using and how old is it? In the auto edgefinder on the newer wire's they automatically compensate for the wire diameter so if the program is loaded and ready the cnc already knows what kind of wire you are using. There is a lot of older machines where you have to manually add or subtract 1/2 the wire dia to get your location as MitsTech said. Locations on a lot of machines can be found within 1 or 2 tenths most of the time with good contact. Newer machines can find edges within 20-40 millionths most of the time and the readings will repeat when performing the operation a second time. On my older machines when I use the auto cycle to find an edge I find that the machine tends to overdrive the wire by a tenth or two to get the contact it needs for a reading. So, if you had .010 wire I would move .0048 to reset my zero instead of .005. This usually would get me very, very close. Exceptional accuracy can be had by using an oscilloscope to detect the touch point as it is much more sensitive than the machine is.

TMD

MetalDoctor, what kind of machine is it and how old is it? Have you ever replaced power feed or feedback cables on it? If that stuff's all fairly new, chances are there is a parameter or potentiometer you can adjust for the sensativity. Also cleanliness of your connections, condition of power feeders, and diamond guides can all make a difference in pickup accuracy.You are correct, and oscilloscope or even a good DVOM can get a pickup dialed in very nicely

Not to hi-jack this thread, but how are you setting up a scope to edge find? I've seen you mention it one time before and am curious. What are the setting on the scope and what do the wave forms look like? I would asume you are connection one lead to the upper or lower contact and the other to ground. Thanks Mark

Pretty much! When the wire makes contact you will see a voltage drop on the scope.

Space,

When I edge find with the scope, I put my machine into continuity check mode, wire threads and runs and I will jog the table over until I see he wave form become distorted. I just checked to see what it was so I can better describe to you:

In continuity mode I have essentially a positive running square wave about 15vpp and 20 usec pulse width. A very stable, synced image and when I begin to touch the part the waveform becomes distorted more and more the further I move into the part and gain on full continuity. At full continuity I have a dead short and the signal is gone. I stop and take my reading when I first see the signal scratch the part.

TMD

Mits,

The machine I do this with is 12 years old. I have never replaced the feed cables, but do revolve the contacts when needed. I have only had to replace the contacts once in 12 years. My machine has about 33,000 hrs on it. (AGIE)

TMD

TheMetalDoctor,

I am using a late 90's Ram EDM. I wish I had a wire EDM. I am working on getting one by the end of the year. You will see a lot more of me on this forum

Just a follow up question. I got how you would find the X & Y cordinates for a particular work origin. Let say you have to cut a taper in your block. How would you locate the top of the part. Would you cut undersize, then measure, offset, cut again, like cutting countersinks on a mill, or somehow locate the top of the part and cut in one shot.

Thanks!

You don't need to locate the top of the part. You set your Z-height to the flush cup from the table with a gauge block. Then, since the diamond guides are actually a longer distance away from the part than the flush cups you have to do some type of measurement of the U-V movement relative to the actual angle of the wire. On a Mits this is done with the same tool you do a wire alignment with. You tell the machine the height of your Z axis flush cup, the desired angle, and hit start. It will kick out the U-axis a set amount, then move the whole tilted wire to the test fixture, which is really a contact based probe with two precision ground flats on it. Here's a crappy pic of one...
Without getting too into detail because I can't figure a way to explain it easily, it basicly figures out the actual angle and distance needed to tilt the wire to set the wire at whatever the programmed angle is.

As far as your workpeice is concerned, you set a Z1, Z2, and Z5 value. They represent the bottom, top, and center height of the taper (not in that order). Once the machine knows how far to tilt the wire, where the workpeice is, and the height of the workpeice, tapers can generally be cut extremely accurately.

Sorry if my explaination sucks, I just woke up It's really pretty simple when you see it. I should also note, this is how Mitsubishi does it, other companies may have slightly different terms or methods but the end result is pretty much the same.

My Agie works similarly to what MitsTech explains:

Very basically -- I create geometry for the profile at the bottom (typically, though not necessarily the large end) of the taper, and geometry for the upper end of the taper. I tell the machine how far above the zero reference plane the bottom profile is, and how tall the part is (distance between the 2 "profiles" I have created).

The machine "knows" where the bottom wire guide is (it's a fixed position that never changes), and knows where the top wire guide is by absolute position of the Z axis (upper support assembly).

Since the machine knows where both guides are relative to the Zero plane, and it also knows where the upper and lower profiles of the tapered part are relative to that same zero plane... the machine can "figure out" where to position the wire guides to correctly cut the taper.

A slight bit of an over-simplification, but hopefully that makes sense.

Bottom line: if the position of the bottom of the part and height of the part are precisely measured, I've found that size and position on a tapered cut are usually right on the money first shot. (easily within tenths).

----

Just thought of something else that you might be asking: the way to reference position of a tapered cut within the part? My above explanation was more about controlling size/geometry of the tapered cut... but not about positioning that tapered cut in X/Y within the part.

On my machine, there is an "entry point" that defines the beginning of the operation - (this is typically done in the CAM package where the program is created). That entry point may be on the part, or it may be out in space off the side of the part (or obviously it could be in space in the center of a cavity within the part, or even more likely it could simply be in the center of a vertical "start hole" for threading the wire). That all depends on what the tapered cut looks like.

The entry point is nothing more than a start point for the cut where the wire is (can be) vertical. Once again... a bit of a "loose" explanation, but basically if the entire cut is tapered, then the entry point is off the cut somewhere and the machine begins moving the upper head immediately so that upon reaching the imaginary surface of the tapered geometry... the wire is at the correct angle.

Whew... that was a long sentence, but hope it makes sense.