Tool Change

[DaveCVI]

Another thought:
Is there a large amount of debounce set for Mach?
Debounce is actually set as a integer number; the number is the number of time slots before mach will recognize that the probe input is active. So if you have a debounce of 10,000 (for example), you are telling mach that the probe trigger has to be active for 10000*40us = 0.4 seconds.. that means that the spindle will keep traveling down for almost 1/2 a second after the plate touches and only then will mach try to ramp down the the Z motion.

Dave

[LGreen]

Dave,

Besides Mach3, the only thing on the computer is Adobe reader, an antivirus program and a little utility called "Mach Pendant", it's an iPad-iPhone app that works wirelessly using the computer IP address so obviously the computer is online. I could do without those additional programs, disconnect the computer from the Internet and disable the antivirus program and try it that way. Probably be better anyway. I'll try that tomorrow and report back to you.By the way thank you so much for working on this. I appreciate it.
Lou

[LGreen]

I didn't see the second post before I responded to the first. I have never changed the de-bounce since Mach was installed. I had no idea what it was or what it did. The debounce interval is set to zero, the index debounce is also zero. Well I'm on the general configuration page there is a shuttle wheel setting that is set to 0.25 seconds. I also have no idea what that is. Do you see anything in these settings that could benefit from being different?
Lou

[LGreen]

So here is what I've found,
In the Mach standard mill user manual, paragraph,
8.5.1 Common Probing Operation Parameters;
8.5.1.4 Probe Fast FR (Feed Rate) Initial Feed rate to use for all probing operations; I had entered 50 instead of 5 in the entry box. After correcting that, the machine settled down and went back to its normal operation. Feeling much better now and maybe a little wiser, as well as quite embarrassed.
Lou

[DaveCVI]

Hi,
Go ahead and forget the embarrassment aspect - we all learn as we go along.
Debouce is a filter that mach can apply to the probe input signal. Essentially it is a way to ignore very short pulses so that they do not cause mach to think the input is triggered, while still letting mach react to inputs that last a longer time. This can be effective in making mach ignore noise or switch contact bounce pulses (Switches do not always close one time as the contact are brought together). The more the debounce used, the longer the pulse has to be present before mach will recognize the input signal. This is essentially a software fix to a hardware electrical problem.

If you system works with 0 debounce that is great - count your blessings. Many systems require a little bit of debounce and people tend to use it (as it;s easy to set) rather than track down and fix electrical noise.

Re the feed rate of 5 - that's about as slow as I've ever run a probe. SLow feed rates can exacerbate contact bounce issues - so if you start getting false triggers at that rate, you might want to raise the feed rate some. Section 8.11.1 of the mill user manual has more info on this.

Dave

[LGreen]

Thanks Dave,
In truth there is still a little bit of a hard touch on the touch plate, so the slower the feed the softer the touch. My concern is for smaller bits such as PCB Bits breaking under the pressure. I haven't tried any yet and my concern might be unwarranted. I would definitely like a higher feed rate and have been wondering if there could be such a thing as a spring-loaded touch plate. It would be easy to make, firm enough to ensure a good solid contact and soft enough to give way when need be. Just an old man's musings.
Thanks a million
Lou

[DaveCVI]

Ah, the old "if only it were spring loaded" thought....
Before you go heading down that path please read this thread:
http://www.calypsoventures.com/forums/v ... etter#p734

Dave

[LGreen]

here comes the old dizzy feeling again

[venom51]

Hi,
Go ahead and forget the embarrassment aspect - we all learn as we go along.
Debouce is a filter that mach can apply to the probe input signal. Essentially it is a way to ignore very short pulses so that they do not cause mach to think the input is triggered, while still letting mach react to inputs that last a longer time. This can be effective in making mach ignore noise or switch contact bounce pulses (Switches do not always close one time as the contact are brought together). The more the debounce used, the longer the pulse has to be present before mach will recognize the input signal. This is essentially a software fix to a hardware electrical problem.

If you system works with 0 debounce that is great - count your blessings. Many systems require a little bit of debounce and people tend to use it (as it;s easy to set) rather than track down and fix electrical noise.

Re the feed rate of 5 - that's about as slow as I've ever run a probe. SLow feed rates can exacerbate contact bounce issues - so if you start getting false triggers at that rate, you might want to raise the feed rate some. Section 8.11.1 of the mill user manual has more info on this.

Dave

Just one more thing to mention that I have found out about the debounce setting a Mach. If you are using the PID loop the debounce will also get in the way of reading the tach signal. This leads to a run away of the motor RPM as the signal from the tach is ignored once the spindle is above a certain speed.

[DaveCVI]

Just one more thing to mention that I have found out about the debounce setting a Mach. If you are using the PID loop the debounce will also get in the way of reading the tach signal. This leads to a run away of the motor RPM as the signal from the tach is ignored once the spindle is above a certain speed.

Oh, that is not good. I didn't know about that one.. thanks for sharing that information.

Dave