Another KX1 problem

[Mick Knights]

Hi. All.
I posted a few weeks back about a problem I was experiencing where a proven program seemed to stall during the cycle especially when cutting in the X- direction. I eventually traced that one back to the coupling clamping screw on the end of the ball screw which was just tight enough to drive in the X+ direction but allowed the coupling to slip when the torque was reversed.
Having solved that one I've now got a problem with the Z axis. The Z zero and T01 offset positions are correct, as since this condition appeared yesterday I've re-set both several times, but on the first Z drive move instead of the cutter traveling to the Z zero face on the work piece in order to pick up its offset it drives to Z-2.5mm. below the Z zero face. This 2.5 is an exact distance because when the Z zero in the program is changed to a Z 2.5 value the cutter will just cut a whisper of material. In order to run a program I now have to program 2.5mm of the total depth as Z+. I don't use CAD/CAM as I'm an old school Gcode man but I do use the wizards on a regular basis, which will become a problem if this condition persists as I'll have to edit all the Z moves. Any suggestions would be welcomed. Bearing in mind I've had my machine for thirteen years with over 2,800 machining hours on the clock the situation I'm experiencing is not likely to be due to setting. Thanks.

 

[ddmckee54]

Programs/programming don't normally change without something telling them to change. I have seen it happen, but it was determined that problem was caused by a nearby lightning strike. Jan./Feb. weather in the UK doesn't seem likely to generate those type of conditions.

How are you getting your initial Z zero position, any limit switches involved? If a limit switch loses part of its' actuating lever, it will still actuate, just in a different position than expected. When the programming suddenly starts behaving differently than expected, I always look for mechanical causes first.

You DID say the machine is 13 years old, it's a teenager now, and by definition unpredictable - I wish you good luck for the next 7 years.

Don

 

[Carbuilder]

Sorry I can't help you, but I do have a KX1. I think it might be that last one that Little Machine Shop sold about 4 years ago. Really nice little machine, but not much support available on the net for it. I've added a high speed spindle since I do a lot of work with small cutters.

Anyways, I'm mentioning this in case there is anything I can check on my machine that may help you at some point.

It is currently 6 hours into a camshaft, with hopefully another hour or 2 left without issue.

Rick

 

[awake]

Hmmm ... 2.5mm and cutting just a whisker ... any chance some sort of imperial measurement of .1" got in there somehow? The whisker of cut could be the additional .04mm, perhaps ... ???

I don't know enough to know if this is even a possibility, so I apologize if this is totally missing the mark.

 

[Mick Knights]

Hi. All.
Thanks for your input. I did get to the bottom of the problem the other day, which is good, but it does present a new question. I remembered what the instructor who taught me the black art back in the eighties told me, which was the control is logical and so will only do what its told to. As the distance the cutter moves to below z zero is constant the control must be doing what its been told. My control has what's called a gauge block distance which is used to set the tool offsets, this is something I've never seen the point in and so haven't used in the thirteen years I've had the machine. Sure enough, in the gauge block window was the value of 2.5mm. which of course was being added to the T01 offset value. Everything now is back to normal, except, the gauge block value has to be manually entered and if I didn't do it how did it get there?

 

[ddmckee54]

Not even the Shadow knows what evils lurk in the hearts of programmers, or what can happen at 3:00AM when they are running on Mountain Dew and no sleep, trying to meet their schedule.

There could be more than just you setting the gage block value. I'm not gonna call it a bug, maybe it's just an "undocumented extra feature"? Just make a note of it and hang it somplace to remind youself how you fixed to this time, just in case it happens again.

 

[awake]

As a former programmer, I can tell you exactly what it was. We called this sort of thing "operator error." It must be, since of course WE didn't do it ...

 

[ddmckee54]

As a former programmer myself, those issues came under the headiing of "undocumented extra features" for us. Being the good guys that we are we didn't charge any extra for them...

 

[awake]

Indeed - we always tried to assure the customer that this was a FEATURE, not a BUG!

 

[Asm109]

If you can't fix it. FEATURE IT!!