Need Help!- Feed rates out of whack

[chipwitch1]

I've just wasted an entire day trying to debug a program I wrote, only to discover the problem is with my feed rates. I just used a stop watch to time a 1 inch travel at f10. It should take 6 seconds, but it takes between 10 and 11! Spindle and Feed pots are both set to 100% but even when I use M49, my little test still takes 10 seconds. Is this normal? Is there something I can do to calibrate it?

The machine is a 1995 VMC15RT. Specifically, I'm trying to synch the spindle with the z-axis similar to a tapping cycle. The result I'm getting isn't even close to what it should be.

Any help or suggestions would be greatly appreciated.

[Neal]

Chipwitch1--
This is DC servo machine and it appears that the clock card (1020) needs adjusting. I suggest doing a 20" move with the M49 invoked. It should take 12.5 secs.

Neal

BTW, thanks for the kind words!! It is appreciated.

[chipwitch1]

12.5 seconds? I'm confused. How is that calculated? Do I need to set the feed to 96 ipm or is that something internal and automatic I'm unaware of?

Also, I'm noting that when I S150, I get 146 rpm output from the spindle. Is that within the machines tolerance?

This morning when I turned on the machine and ran the same program, I found the travel feeds, though still off, were nowhere near as bad as they were. This has me wondering if *I* was doing something wrong or if the machine hiccupped and reset when rebooted? We're talking the same program here. Strange.

Neal, as for the kind words? They are but a small token of my gratitude... and sincere.

[alex gitzel]

run this program( m49 g1 g8 g90/x-10.f100./x10./x0./m30)this should take 24-25.5 seconds on the movement timer on the screen.if not set pot R6 on the clock card i think ccw is slower.also enter command ME and see if the memory has less that 15% available if so it will have odd glitches and you should zero the memory and put in less programs.the clock card should be reseated.

[chipwitch1]

Thanks for the replies. It seems that there's more tolerance in axis travel than I expected. I see now that you can't simply set two different axis in motion and expect them synch very well. In my case, I'm trying to sync c axis with z axis. By setting the spindle speed and expecting the z axis travel to sync to it with the accuracy of a screw thread is asking for more than the machine was designed for. I now understand that cnc machines use small incremental moves, check for in position, then go to the next move in order to accomplish a single contour move or "rigid tap". My thinking was flawed.

I've just discovered how to index the spindle. By checking the progress (through the use of macros) of my tool path at the end of each 1/4 revolution to the current position of the z-axis, I achieve the results I was originally looking for.

[Neal]

Sorry, I left out the fact that the feed rate should be F100. That would give you the 12.5 seconds. The field servicemen frequently use this test to verify the clock card adjustments.

Neal