Homing & Limit Switch UNSOLVED MYSTERY! <solved>

All things related to the Centroid Acorn CNC Controller

Moderator: cnckeith

suntravel
Posts: 1967
Joined: Thu Sep 23, 2021 3:49 pm
Acorn CNC Controller: Yes
Allin1DC CNC Controller: No
Oak CNC controller: No
CNC Control System Serial Number: 6433DB0446C1-08115074
DC3IOB: No
CNC12: Yes
CNC11: No
CPU10 or CPU7: No
Location: Germany

Re: Homing & Limit Switch UNSOLVED MYSTERY! <solved>

Post by suntravel »

You can be sure there is no software problem with Acorn I never had a false axis movement....

Uwe
RJS100
Posts: 380
Joined: Tue Oct 20, 2020 8:41 pm
Acorn CNC Controller: Yes
Allin1DC CNC Controller: No
Oak CNC controller: No
CNC Control System Serial Number: none
DC3IOB: No
CNC12: Yes
CNC11: No
CPU10 or CPU7: No
Location: Arizona

Re: Homing & Limit Switch UNSOLVED MYSTERY! <solved>

Post by RJS100 »

I didn't mean a coding error. I meant a parameter, setup or operator error. Something is causing the z axis to overshoot to some extent the proximity sensor. I may still re-wire these sensors, so each goes to a separate input.
tblough
Posts: 3089
Joined: Tue Mar 22, 2016 10:03 am
Acorn CNC Controller: Yes
Allin1DC CNC Controller: Yes
Oak CNC controller: Yes
CNC Control System Serial Number: 100505
100327
102696
103432
7804732B977B-0624192192
DC3IOB: No
CNC12: Yes
CNC11: No
CPU10 or CPU7: No
Location: Boston, MA
Contact:

Re: Homing & Limit Switch UNSOLVED MYSTERY! <solved>

Post by tblough »

It will ALWAYS home to the same place because homing is not based on steps only a switch. RETURNING to the home position is based on steps, so if steps are off it will not return to the same position.

There are no parameters or configuration settings that will randomly work one way and someway different a little while later.
Cheers,

Tom
Confidence is the feeling you have before you fully understand the situation.
I have CDO. It's like OCD, but the letters are where they should be.
RJS100
Posts: 380
Joined: Tue Oct 20, 2020 8:41 pm
Acorn CNC Controller: Yes
Allin1DC CNC Controller: No
Oak CNC controller: No
CNC Control System Serial Number: none
DC3IOB: No
CNC12: Yes
CNC11: No
CPU10 or CPU7: No
Location: Arizona

Re: Homing & Limit Switch UNSOLVED MYSTERY! <solved>

Post by RJS100 »

Thanks Tom. I am going to try to recreate this problem again taking better note. I will also try to load up the z-axis an see if I can make it loose steps while returning to the home position. Thanks for your thoughts.
RJS100
Posts: 380
Joined: Tue Oct 20, 2020 8:41 pm
Acorn CNC Controller: Yes
Allin1DC CNC Controller: No
Oak CNC controller: No
CNC Control System Serial Number: none
DC3IOB: No
CNC12: Yes
CNC11: No
CPU10 or CPU7: No
Location: Arizona

Re: Homing & Limit Switch UNSOLVED MYSTERY! <solved>

Post by RJS100 »

WOW... This is really frustrating. I recreated the problem and noted the process:

1. I shut down CNC12 and Acorn, restarted and homed the machine. I carefully marked with blue tape the position of the flag with respect to the proximity sensor when the machine completed the homing cycle (z-axis). See image "Original Home Position".

2. I repeated step one four times, re-homing the machine each time. The machine returned precisely to the same position. Each time I homed the machine, I jogged away from and back the home position and each time the machine returned to the same position. In addition to jogging the machine, I also issued G0 G28 and again the machine returned multiple times correctly.

3. I repeated step one again, this time pulling with almost all my weight down on the z-axis to try to overload it. I repeated this process of homing the machine and loading the z-axis simultaneously and each time the machine returned to the same place shown on image "Original Home Position". I could not seem to make it fail no matter what I did.

4. I jogged the machine to machine coordinates X15.9932, Y-4.6860 and Z0 and set this as my G59 location. The location I plan on measuring all future tool offsets from. I jogged up and down and once again and the machine returned to the ("Original Home Position"). Then I measured my master tool. I then measured another tool. I then jogged all the way to the top. Then BAM (as Emeril Lagasse would say), the machine returned about .375" too high. See image "After Jogging". For fun I jogged the z axis down about 5 inches and issued G0 G28 and the machine returned to this .375" too high location again. As you can see from the screen shot, the machine was indicating machine coordinates of Z=0 when it was actually .375" too high.

I welcome your thoughts... Am I just doing something wrong here?

Thanks... Richard
Attachments
Screen Shot.jpg
After Jogging.jpg
Original Home Position.jpg
RJS100
Posts: 380
Joined: Tue Oct 20, 2020 8:41 pm
Acorn CNC Controller: Yes
Allin1DC CNC Controller: No
Oak CNC controller: No
CNC Control System Serial Number: none
DC3IOB: No
CNC12: Yes
CNC11: No
CPU10 or CPU7: No
Location: Arizona

Re: Homing & Limit Switch UNSOLVED MYSTERY! <solved>

Post by RJS100 »

I forgot to mention. I also overloaded the z axis simultaneously when issuing G0 G28 and it continued to return to the correct "Original Home Position" shown in the image.
RJS100
Posts: 380
Joined: Tue Oct 20, 2020 8:41 pm
Acorn CNC Controller: Yes
Allin1DC CNC Controller: No
Oak CNC controller: No
CNC Control System Serial Number: none
DC3IOB: No
CNC12: Yes
CNC11: No
CPU10 or CPU7: No
Location: Arizona

Re: Homing & Limit Switch UNSOLVED MYSTERY! <solved>

Post by RJS100 »

I started thinking about this issue and realized that each time I measure the master tool length followed by measuring a tool offset, the machine no longer returns to the correct z home position. I realized that I have a power drawbar and thought maybe I am getting EMI from the drawbar. I cycled the drawbar a bunch of times and could not create any issues.

I then power cycled CNC12 and Acorn and the machine returned to the correct Z-home (Image "Original Home Position"). I repeated the procedure of re-measuring the master tool length followed by measuring a tool offset and again, when I jogged the machine to the top of the Z-axis, it no longer returned to correct position, but exactly the same .375" beyond the proximity sensor as posted previously. See image "After Jogging".

I have no past experience here, but I cannot imagine if this issue is caused by EMI or lost steps that it would continue to overrun the proximity sensor by the same .375 inches each time. I also looked at the x and y axis and noticed that both of these axes stopped about .125 short of where they stop after homing the machine.

The observations I have made here:
-power cycle CNC12 and Acorn - the machine returns to the correct home position (X,Y and Z axis).
-Set the master tool height and measure a tool, the flag for the Z-axis passes the proximity sensor by .375" beyond it's previous home position.
-x and y axis no longer stop at the correct position either (I will have to track these positions closer).

Any idea as to what can cause this.

Thanks... Richard
Attachments
After Jogging.jpg
Original Home Position.jpg
suntravel
Posts: 1967
Joined: Thu Sep 23, 2021 3:49 pm
Acorn CNC Controller: Yes
Allin1DC CNC Controller: No
Oak CNC controller: No
CNC Control System Serial Number: 6433DB0446C1-08115074
DC3IOB: No
CNC12: Yes
CNC11: No
CPU10 or CPU7: No
Location: Germany

Re: Homing & Limit Switch UNSOLVED MYSTERY! <solved>

Post by suntravel »

So after measuring the master tool Z0 in machine coordinates is .375" higher?

Uwe
Nigelo
Posts: 368
Joined: Tue Dec 11, 2018 4:03 pm
Acorn CNC Controller: Yes
Allin1DC CNC Controller: No
Oak CNC controller: No
CNC Control System Serial Number: F045DA80C8B8-0905181172
E415F6F70BC3-0318203049
98F07B91FC6B-0123236802
DC3IOB: No
CNC12: Yes
CNC11: No
CPU10 or CPU7: No
Location: UK

Re: Homing & Limit Switch UNSOLVED MYSTERY! <solved>

Post by Nigelo »

I notice from the most recent report that your Automatic Homing Sequence is set to to 1 for all axes. I've no idea if this interferes with homing or not but surely should be corrected e.g. Axis 3 (Z) = 1 / Axis 2 (Y) = 2 / Axis 1 (X) = 3 / Axis 4 if used and with switch = 4. See Wizard's Homing and Travel section

It appears you are still using inputs for all axes set to Home rather than HomeLimit as Uwe pointed out back on Page 3 of this thread. This won't help with your missing steps but is the reason you keep sailing past the Z sensor and even causing damage on one occasion. FYI, when set to HomeLimit the sensor will trip thus preventing the axis passing its stop position under all circumstances, machine homed or not / missing steps or not i.e. an important failsafe. It will also allow you to use a bit more travel, if you need it, by positioning the sensor nearer to the physical Axis limit. I realise that HomeLimitAll doesn't appear to exist in v4.8 so you would need to use a separate HomeLimit input for each Axis. See Wizard's Input Definitions section

Earlier in this thread you seemed to doubt the repeatability of the sensors used. If this is the case, replace them with known good items. You could then avoid using a reference tool, if desired, by adding 2 to your Param 3 value (currently set to 4). As suggested earlier by Uwe, this would require all tools to be remeasured as the tool offsets will then be taken from Z home rather than compared with the ref tool.

Finally, you have several MPU resend requests in your log which need addressing and points to your missing steps

Hope this helps
Hope this helps
Nigel

"You can lead a horse to water but you cannot force it to drink"
RJS100
Posts: 380
Joined: Tue Oct 20, 2020 8:41 pm
Acorn CNC Controller: Yes
Allin1DC CNC Controller: No
Oak CNC controller: No
CNC Control System Serial Number: none
DC3IOB: No
CNC12: Yes
CNC11: No
CPU10 or CPU7: No
Location: Arizona

Re: Homing & Limit Switch UNSOLVED MYSTERY! <solved>

Post by RJS100 »

Hello Nigelo,

Thanks for taking the time to review this post so carefully. Very much appreciated. I am trying to implement your suggestions but have the following questions/comments:

1. With respect to the homing sequence. I am confused here. Please see attached screen shot from Wizard as I thought I set them up exactly as you suggested. Why is the report showing otherwise. Am I missing something?

2. Currently, I am using "HomeAll" because the proximity sensors are wired in series, and I have not rewired them YET. See attached input definition screen shot. It is still an option in v4.8. I did not rewire them as I was hoping this was not the cause of the problem. Shawn wires in series with great results. I am a bit perplexed at the fact that the problem is repeatable after a tool is measured and suddenly the axis rises to a new position .375" higher. That said, your explanation of fact that HomeLimit will stop action under all circumstances is new to me. For this reason alone, I think I will rewire now.

3. I initially doubted the repeatability of the sensor, but after marking the sensor with tape and seeing that each time it overran the sensor by the same .375", I am thinking there is another problem here. In either case I do have an extra sensor and when I rewire them, I will replace the current sensor.

4. With respect to Parameter 3, I believe Uwe suggested that it could be set to 4 or 6. I set it to "4" as I would prefer to measure my tool offsets relative to a master tool. Is this the correct setting for measuring relative to a master tool?

5. Can you help me understand what MPU resend requests means and how to address this?


Uwe, to answer your question "So after measuring the master tool Z0 in machine coordinates is .375" higher?" Let me clarify:

-First I started the machine for the first time of the day. It stopped at the location on the proximity sensor shown as "Original Home Position" see previous post. The machine coordinates were z=0. Note, it ALWAYS stops at this location each time I home the machine.
-I then jogged to the top of the z axis. Machine coordinates indicated z=0, and the machine returned to the same exact position noted by the tape on the proximity sensor.
-I measured the height offset for tool #2, hit F10 to save the position.
-Jogged to top of the z-axis or issued G0 G28.... BAM... the machine returns to a position .375" higher than before (see image "After Jogging" in previous post). The machine coordinates indicate z=0 at this time.

In essence, after I measure a tool, the z-axis datum shifts up .375" AND the x and y axis no longer return to their correct home positions. This does not appear to happen when I measure the height of the master tool.

Thanks.... Richard
Attachments
Input Definition Screen Shot.jpg
Homing Sequence Screen Shot.jpg
Post Reply