No announcement yet.

Slightly OT, repair of a Makino wire EDM that eats wire.. need ideas.

  • Filter
  • Time
  • Show
Clear All
new posts

  • J Tiers
    I verified a bit but not too much by looking at the current before the wire started sparking, but the water was hitting the work.

    Got about 2% of spark current. Looked reasonable.

    The Makino info says water conductivity has to be at least a minimum, but not too high.........
    Last edited by J Tiers; 02-18-2015, 02:27 PM.

    Leave a comment:

  • toolmaker76
    It has been more than 20 years since I have run a wire edm machine, ran them for 5 years up to that point. I looked up your model for a picture, and it seems that everything is controlled at the computer. Agie had outboard controls for the power supply. In any case here are some of the things I remember.

    Might ask a couple questions first. Have there been several different workpieces run? I have seen some bad batches of steel. What kind of finish are you getting? Have you used different spools of wire? We switched suppliers going after a cheaper price, had such a terrible time with the cheaper wire that we went back to the original supplier. Wire quality is important, and I have also seen wire spools that were contaminated, with similar results as yours.

    What kind of finish are you getting out of what you have been running?

    Eliminating steel and wire, there are a couple of things might be worth a look. In your wire drive, there are bearings inside the pulleys. not really talking about next to the contacts but out where the wire is driven from the spool to the work. How are the bearings? The bearings have to be in good shape, it doesn't take much of a "stick" or rough spot in your drive to cause similar problems. Where I was working we had three machines that we tried to run 24 hours a day, 5 days a week. Seems like we had to do some bearing work every two or three years.

    Incidentally, found that bearings were extremely expensive on our Swiss machines, so set about matching them up with an American supplier. As it turned out, they were American made, but by the time we got them they had been exported and then imported again. Bought the exact bearing locally for 1/3 the price.

    Another unusual problem that comes to mind was the wire harness that powered your wire contacts (the carbides and such). It was a heavy power cable similar to a battery cable, and it was exposed to all the water and nastiness right in the work area, they would corrode. Submerged machines would go before the flush machines, but like the bearings seemed like it was 2-3 years on submerged, maybe 3-4 years on the flush machines. After changing a few of them, it did not seem like it would be a stretch to make your own cables if they weren't available from the manufacturer.

    There are some other goofy problems that can come into play, I don't know how the Makino works but remember the Agie's pretty well. On the Agie, your resin bottle is bypassed until it is signal by the machine that it is needed, in fact it would error out if the conductance of the water was too high. If the machine is not getting the signal that the resin is needed, you can have new resin but not be actually filtering the water. There will be a probe that may need cleaning. If the water conductivity is too high you may experience the problems you are having. I can't think off hand of a way to verify your water conductivity, though.

    Hope this helps.

    Leave a comment:

  • J Tiers
    They replaced the two carbide contact blocks, the sapphire guides, and the other wire contact points, rollers, etc. Wire looks good up until the work, looks very pitted and eroded after it exits. It breaks at or in the work, not elsewhere.

    The wire shows up with abruptly necked-down areas at the break. Not gradual necking, as with an elongation (tension) break, but a sharp hourglass shape, as if the wire were either eroded by a heavy spark right there, OR somehow locally heated so that it was weak and stretched there. There are other such spots in the exiting wire near the breaks, and the exiting wire looks more unevenly eroded than I'd expect.

    They are running the wire pretty fast, so wire speed *shouldn't* be a problem.....

    A sudden heating sounds more like a hot spot caused by running up to an edge, or possibly by particles getting hung up. But since it was seen well into a cut (at least 80 thou in when I saw it) there should not have been any edges sticking out.

    That's what suggested current variance as a problem. But I am beginning to suspect it is a symptom more than a cause.
    Last edited by J Tiers; 02-18-2015, 09:15 AM.

    Leave a comment:

  • George Bulliss
    Can't say as to what the red sparks mean, as my experience is with fully submerged EDMs. Just about any decrease in performance will show itself in broken wire, requiring you dial back on power, speed, or both. With water that is becoming ionized, you will definitely get wire breaks. If they recently changed the resin filters (not just the regular filters) maybe things aren't flowing through as it should?

    Flushing will cause wire breaks too, and is the main cause in most cases. Especially important in non-submerged machines. If they are running the same jobs at the same settings but getting more breaks, perhaps the flushing nozzles have worn and need replacing?

    Wire EDMs are pretty dirty machines and there may be some areas where the wire is being abraded. On the ones I operated (Mitsubishi) there was a carbide guide that would wear and need rotation or replacement; it could cause wire breaks if the wire had worn a groove. The carbide guides and the flush cups are wear items and we got less than a year out of them.

    Good luck!

    Leave a comment:

  • J Tiers
    Supposedly, recently.

    That was the suggestion I got elsewhere as well.

    The question is, what SORT of decrease in performance? They specifically break wire at low currents, possibly associated with the red sparks I mentioned. At least the number of red sparks increases as the likelihood of breakage goes up.

    Leave a comment:

  • George Bulliss
    When was the last time they changed the filters for the water? The resin filters don't last forever and a gradual decrease in performance is the result.

    Leave a comment:

  • Slightly OT, repair of a Makino wire EDM that eats wire.. need ideas.

    This machine belongs to a vendor of ours. We, being the electronics and motor control folks, apparently were the folks of choice to ask to look at it, and, being the only shop-type guy, I got the job....

    I am not any sort of expert with wire EDM. I am looking for ideas. Makino cannot fix it, it's an EC-7050, and too old for their guys to know about. They cost money, didn't fix it, and had no good ideas, is what I was told....

    The problem with it is that it has begun to break wire easily. It's a 16A machine, but can't be run much above 3A or it eats wire. It may go as much as a minute at 5A before it breaks the wire. I have watched this, and done some investigating.

    What I know:

    The machine is clean and many of the wire handling parts are new. The weighted dancer is stable, no jerking.

    At 4A it may break wire, at 3A it will not. At 5A it does very soon. It used to be fine, but the allowable current has decreased over time.

    Voltage is in recommended range. Capacitor values selected are in recommended range. Wire speed is, if anything , high, and is not a limit.

    Movement max speed is in normal range.

    used wire seems reasonable, although it looks a bit more unevenly worn than I expected. There are thin spots, and where broken, it has abrupt thinning that looks almost like a shear pin, coned down to a thin spot. Could be tension elongation, or not.

    breakage was at top, bottom and in middle of work, as far as I could see.

    White sparks are continuous, even in apparent number and rate, with no obvious stoppages or changes that might indicate an overshoot on table movement, etc.

    White sparking seemed to be continuous and even in nature right up to the wire breaking... I cannot be certain whether the sparks stopped first or the wire broke first, of course. But my impression was that it worked right up to the time of breakage.

    Water flow appeared good and effective, but might not be all it should be. However they state that they have always run it the way I saw, without problems until more recently.

    I had them back out of the cut, and re-enter it after a breakage, and with the wire in the cut, powered up, and with water flow, but not yet sparking, I saw a little current flow, which was quite steady. I expected that, and the current was quite low, maybe 2% of max.

    The current appears, when using a Fluke 87V on DC, to be rather more variable than I thought. Might be a form of "aliasing" due to the meter sampling slowly, but when set for 4A peak current, I saw peaks on the meter that were indicating over 5A, and minimums under 3A. I was measuring across the test points that are on the current sense resistor.

    Do not have an oscilloscope observation of it, will get one next time I go there.

    One thing I DID see, was that at 3A there were occasional red spurts that looked very much like grinding sparks, with hot red particles shot out of the water flow.

    At 4A there were more of them, and at 5A they were quite regular. I would say they increased more than the current flow appeared to, out of proportion to the current increase.

    So, that's basically what I know, and I am looking for ideas.

    It has been suggested that it could be due to particles, either from insufficient water cleaning capability recirculating them, or possibly from water flow not clearing them from the cut. I suppose that might give a different spark which might affect the wire somehow. I am not at all sure how that would work.

    It has also been suggested that the current is noisy because the current control is not working right, and it gives bursts of heavy current that may burn the wire. Dunno about that...
    Last edited by J Tiers; 02-17-2015, 09:41 PM.