Spindle Encoder issue
- Lazarov
- Offline
- New Member
Less
More
- Posts: 4
- Thank you received: 0
06 Oct 2017 08:43 - 06 Oct 2017 12:30 #99980
by Lazarov
Spindle Encoder issue was created by Lazarov
Hello,
I want to share with you an experiment I made yesterday.
I've mentioned a problem with the spindle encoder of the lathe I've digitized (at revolutions greater than 400, the display is inconsistent). Example: spindle revolutions 1000rpm - screen display 200rpm.
The encoder is 1024p / rpm and has Phase A, Phase B and Index (I use only Phase A and Index).
Yesterday, I put a chip that plays the frequency divider and so I made a Phase A feedback to be 256p/rpm. Before the monrtry was made an oscilloscope test that is so (frequency to be reduced 4 times).
After installing the chip, the display on the Linuxcnc screen improved dramatically - at spindle revolutions of 1000rpm - a reading of about 1035. Everything great, but when running the same threading program (G76), the result differs from making threads before to mount the chip.
In both cases I set up in StetupConf wizarda, the new encoder resolution.
Does Linuxcnc do not take the Encoder Index (does it not pass through the extra chip at all)?
After the experiment I made I found that for G76 Linuxcnc takes Phase A, not Index.
If it takes Phase A, then why does it mix the thread? All I have done is to download the encoder resolution 4 times (256p / rpm).
I've attached a file with my Encoder configuration.
Thank you in advance.
Best regards:
Lazarov
I want to share with you an experiment I made yesterday.
I've mentioned a problem with the spindle encoder of the lathe I've digitized (at revolutions greater than 400, the display is inconsistent). Example: spindle revolutions 1000rpm - screen display 200rpm.
The encoder is 1024p / rpm and has Phase A, Phase B and Index (I use only Phase A and Index).
Yesterday, I put a chip that plays the frequency divider and so I made a Phase A feedback to be 256p/rpm. Before the monrtry was made an oscilloscope test that is so (frequency to be reduced 4 times).
After installing the chip, the display on the Linuxcnc screen improved dramatically - at spindle revolutions of 1000rpm - a reading of about 1035. Everything great, but when running the same threading program (G76), the result differs from making threads before to mount the chip.
In both cases I set up in StetupConf wizarda, the new encoder resolution.
Does Linuxcnc do not take the Encoder Index (does it not pass through the extra chip at all)?
After the experiment I made I found that for G76 Linuxcnc takes Phase A, not Index.
If it takes Phase A, then why does it mix the thread? All I have done is to download the encoder resolution 4 times (256p / rpm).
I've attached a file with my Encoder configuration.
Thank you in advance.
Best regards:
Lazarov
Last edit: 06 Oct 2017 12:30 by Lazarov.
Please Log in or Create an account to join the conversation.
- PCW
- Away
- Moderator
Less
More
- Posts: 18088
- Thank you received: 4879
06 Oct 2017 13:52 #100006
by PCW
Replied by PCW on topic Spindle Encoder issue
Its not surprising that LinuxCNC cannot count a 1024 PPR encoder with a parallel port
at anything above a few hundred RPM
Dividing A lowered the frequency to a number that LinuxCNC could count,
but its likely you would have to stretch the index pulse for it to be detected reliably
Like all spindle synchronized motion, G76 depends on index working
at anything above a few hundred RPM
Dividing A lowered the frequency to a number that LinuxCNC could count,
but its likely you would have to stretch the index pulse for it to be detected reliably
Like all spindle synchronized motion, G76 depends on index working
Please Log in or Create an account to join the conversation.
- andypugh
- Offline
- Moderator
Less
More
- Posts: 23193
- Thank you received: 4871
10 Oct 2017 22:02 #100167
by andypugh
Your position scale is 1024. Was it previously 4096?
How much does the HAL pin motion.spindle-revs increase by for one spindle revolution?
Replied by andypugh on topic Spindle Encoder issue
If it takes Phase A, then why does it mix the thread? All I have done is to download the encoder resolution 4 times (256p / rpm).
Your position scale is 1024. Was it previously 4096?
How much does the HAL pin motion.spindle-revs increase by for one spindle revolution?
Please Log in or Create an account to join the conversation.
- Lazarov
- Offline
- New Member
Less
More
- Posts: 4
- Thank you received: 0
18 Oct 2017 07:09 #100503
by Lazarov
Replied by Lazarov on topic Spindle Encoder issue
Hello, sorry for my delayed response.
The encoder has a standard resolution of 1024. In one case, I install a frequency divider and reduce the resolution 4 times (256) to get a correct screen display. In the other case it is standard with 1024.
The index signal does not change it both cases, but the same thread program gives different results.I can not understand what caused this.
I apply both files to my settings in both cases.
Thank you in advance.
Best regards,
Lazarov
The encoder has a standard resolution of 1024. In one case, I install a frequency divider and reduce the resolution 4 times (256) to get a correct screen display. In the other case it is standard with 1024.
The index signal does not change it both cases, but the same thread program gives different results.I can not understand what caused this.
I apply both files to my settings in both cases.
Thank you in advance.
Best regards,
Lazarov
Please Log in or Create an account to join the conversation.
Time to create page: 0.053 seconds