Mesa 5i25 with plain G540, no 7i77
25 Apr 2013 19:12 #33188
by pws
Replied by pws on topic Mesa 5i25 with plain G540, no 7i77
Thanks for the response, Andy.
A little more educated today - so I have a specific request!
Where can I get an updated "g540x2.xml" - the one I have from the Mesa ZIp file is from 2/11/2013. Based on other threads and this one, it is clear that a newer one exists somewhere!
Thanks again!
Paul
A little more educated today - so I have a specific request!
Where can I get an updated "g540x2.xml" - the one I have from the Mesa ZIp file is from 2/11/2013. Based on other threads and this one, it is clear that a newer one exists somewhere!
Thanks again!
Paul
Please Log in or Create an account to join the conversation.
25 Apr 2013 19:15 #33189
by andypugh
There may have been updates to the 5i25 firmware.
Replied by andypugh on topic Mesa 5i25 with plain G540, no 7i77
Unless the pinout of the G540 drives has changed, I can't see any reason why the XML would change.Where can I get an updated "g540x2.xml" - the one I have from the Mesa ZIp file is from 2/11/2013. Based on other threads and this one, it is clear that a newer one exists somewhere!
There may have been updates to the 5i25 firmware.
Please Log in or Create an account to join the conversation.
25 Apr 2013 19:53 #33193
by pws
Replied by pws on topic Mesa 5i25 with plain G540, no 7i77
OK then - I will proceed with what I have.
I have created a "new" configuration with PNCConf (I found where g540x2.xml needed to be so it was "seen") and I plan to compare what it came up with vs. what I have created by combining info from my original parallel port version and information in this and other threads. I did a quick compare last night after the light dawned on what this was all about and things are close - but a couple of things (times) are very different so I need to digest what they mean and what looks appropriate for my hardware!
The original system (that is working and talking to the G540 through the Parallel port), was simply too old - tons of jitter and a bit slow! So, I decided to go to the other end of the spectum and build the latest and greatest - plenty of room for LinuxCNC to mature and grow! At the moment, it is a bit overkill - but it sure runs nice!
New Config looks like this:
System:
- ASUS F2A85-V PRO FM2 AMD A85X Motherboard
- AMD A10-5800K Trinity 3.8GHz (4.2GHz Turbo) Socket FM2 100W Quad-Core Desktop APU (Radeon 7000 GPU)
- G.SKILL Sniper Series 16GB (2 x 8GB) 240-Pin DDR3 SDRAM
- WD Velociraptor 300G 10KRPM Drive
- Mesa 6i25 Anything I/O
I have yet to plug it into the G540 yet - I want to make sure I get things right in my mind and on the system before I do that - but I am getting close and I believe they will mate this weekend!
The target is a Grizzly X2 Mini Mill. I have purchased a Large Table for it from LMS. The CNCFusion X and Y ball screws are mounted on the Large Base and are completely functional. Nothing done on the Y - will do that when I swap the bases. A axis is in the future but that channel and Y are working the motors fine.
Talk about a learning curve.....
I have created a "new" configuration with PNCConf (I found where g540x2.xml needed to be so it was "seen") and I plan to compare what it came up with vs. what I have created by combining info from my original parallel port version and information in this and other threads. I did a quick compare last night after the light dawned on what this was all about and things are close - but a couple of things (times) are very different so I need to digest what they mean and what looks appropriate for my hardware!
The original system (that is working and talking to the G540 through the Parallel port), was simply too old - tons of jitter and a bit slow! So, I decided to go to the other end of the spectum and build the latest and greatest - plenty of room for LinuxCNC to mature and grow! At the moment, it is a bit overkill - but it sure runs nice!
New Config looks like this:
System:
- ASUS F2A85-V PRO FM2 AMD A85X Motherboard
- AMD A10-5800K Trinity 3.8GHz (4.2GHz Turbo) Socket FM2 100W Quad-Core Desktop APU (Radeon 7000 GPU)
- G.SKILL Sniper Series 16GB (2 x 8GB) 240-Pin DDR3 SDRAM
- WD Velociraptor 300G 10KRPM Drive
- Mesa 6i25 Anything I/O
I have yet to plug it into the G540 yet - I want to make sure I get things right in my mind and on the system before I do that - but I am getting close and I believe they will mate this weekend!
The target is a Grizzly X2 Mini Mill. I have purchased a Large Table for it from LMS. The CNCFusion X and Y ball screws are mounted on the Large Base and are completely functional. Nothing done on the Y - will do that when I swap the bases. A axis is in the future but that channel and Y are working the motors fine.
Talk about a learning curve.....
Please Log in or Create an account to join the conversation.
12 May 2013 23:00 - 12 May 2013 23:00 #33955
by pws
Replied by pws on topic Mesa 5i25 with plain G540, no 7i77
SUCCESS!!
Based on Andy's comments here and PCW's comments here and on another thread, I was finally able to grasp some of the more subtle concepts and I now have a very nicely working X and Y on this new system. HUGE improvements all around with the new system and some of the strange errors/messages I was getting are all gone!
Next step is to marry the new table with the X2 and build up the Z axis. Yahoo!!
Thanks to all!
Paul
Based on Andy's comments here and PCW's comments here and on another thread, I was finally able to grasp some of the more subtle concepts and I now have a very nicely working X and Y on this new system. HUGE improvements all around with the new system and some of the strange errors/messages I was getting are all gone!
Next step is to marry the new table with the X2 and build up the Z axis. Yahoo!!
Thanks to all!
Paul
Last edit: 12 May 2013 23:00 by pws.
Please Log in or Create an account to join the conversation.
- iointerrupt
- Offline
- New Member
Less
More
- Posts: 1
- Thank you received: 1
21 Oct 2013 08:50 #40117
by iointerrupt
Replied by iointerrupt on topic Mesa 5i25 with plain G540, no 7i77
That sample file is nearly perfect, though I found I needed just one more tweak to get the charge pump working:
Changed this line:
setp hm2_[HOSTMOT2](BOARD).0.stepgen.04.velocity-cmd 0.25
To:
setp hm2_[HOSTMOT2](BOARD).0.stepgen.04.velocity-cmd 1.0
The 0.25 value was giving me a 2.5kHz square wave, but the G540 doesn't turn on unless it's 10kHz. Bumping it up to 1.0 fixed it.
Attached are my working configuration files with all the fixes. Do note the velocities and accelerations are insane and would re-doing for anyone using them.
Changed this line:
setp hm2_[HOSTMOT2](BOARD).0.stepgen.04.velocity-cmd 0.25
To:
setp hm2_[HOSTMOT2](BOARD).0.stepgen.04.velocity-cmd 1.0
The 0.25 value was giving me a 2.5kHz square wave, but the G540 doesn't turn on unless it's 10kHz. Bumping it up to 1.0 fixed it.
Attached are my working configuration files with all the fixes. Do note the velocities and accelerations are insane and would re-doing for anyone using them.
The following user(s) said Thank You: BobS
Please Log in or Create an account to join the conversation.
06 Nov 2014 08:32 #52841
by jaybobc
Replied by jaybobc on topic Mesa 5i25 with plain G540, no 7i77
I know this thread is almost a year old, but I'm pulling my hair out on an issue I'm having and don't know where to turn. I'm new to linuxcnc so If the answer is painfully obvious, pleas take it easy on me. I have a Mesa 5i25 and a Gecko G540 driver. I've flashed the 5i25 with the G540X2 firmware. Through research I was able to get the charge pump and e-stop working properly. I'm running a XYYZ setup with the two y steppers driving the gantry. I'm only using one pair of home/limit switches for each axis. I've used gantrykins in my configuration and everything appeared to be working correctly. All axis home properly. When I go into world mode, I can jog all three axis properly and the two y joints appear to be synced (no racking or popping). However, when I try to run a GCODE program The two y joints appear to be out of sync. The gantry will rack and the belt will "pop" on the pulley. When visually watching the Steppers, it appears the main one (joint 1) moves in "pulses" and the other one (joint 3) moves in a smooth motion. I've looked at the gantry-hs example and tried to use the gantrykins code out of it (as it's a little different from mine), but the results were the same. I've attached my configs to this posting. Can someone please take a look to see if you can find anything wrong. Any help would be greatly appreciated.
Please Log in or Create an account to join the conversation.
06 Nov 2014 08:43 #52842
by andypugh
This seems very strange. I find it hard to see how jogging might work and G-code not work.
Are you jogging at the same speed as the G-code rapids?
In Axis, if you are keyboard-jogging, you can do a rapid-jog by holding the shift key.
Replied by andypugh on topic Mesa 5i25 with plain G540, no 7i77
When I go into world mode, I can jog all three axis properly and the two y joints appear to be synced (no racking or popping). However, when I try to run a GCODE program The two y joints appear to be out of sync. The gantry will rack and the belt will "pop" on the pulley. When visually watching the Steppers, it appears the main one (joint 1) moves in "pulses" and the other one (joint 3) moves in a smooth motion.
This seems very strange. I find it hard to see how jogging might work and G-code not work.
Are you jogging at the same speed as the G-code rapids?
In Axis, if you are keyboard-jogging, you can do a rapid-jog by holding the shift key.
Please Log in or Create an account to join the conversation.
06 Nov 2014 09:05 - 06 Nov 2014 09:12 #52845
by PCW
Replied by PCW on topic Mesa 5i25 with plain G540, no 7i77
To check whether this is a commanded position problem or lower level (hardware/timing etc) problem
can you plot:
axis.1.motor-pos-cmd
axis.3 motor-pos-cmd
with a simple Y move?
(do MDI moves work?)
Also the ferror settings are huge (20 mills at stop to 1 inch at full speed),
this may mask a stepgen issue. I would set them for say .002 and .005
can you plot:
axis.1.motor-pos-cmd
axis.3 motor-pos-cmd
with a simple Y move?
(do MDI moves work?)
Also the ferror settings are huge (20 mills at stop to 1 inch at full speed),
this may mask a stepgen issue. I would set them for say .002 and .005
Last edit: 06 Nov 2014 09:12 by PCW.
Please Log in or Create an account to join the conversation.
07 Nov 2014 08:40 #52887
by jaybobc
Replied by jaybobc on topic Mesa 5i25 with plain G540, no 7i77
I'll attempt to answer your questions (the best I can anyway):
Andy:
I agree it's a strange problem to have. I wouldn't have expected it either. When I first saw it, I figured it should be a simple case of two configurations, of some sort that didn't quite match, but I've been over it with a fine tooth comb and can't find anything like that.
I have my jog speed set at 60ipm (the max on my slider). When I tried the rapid jog by jogging with the shift key depressed, as you suggested. It appears to be the same (smooth moving & and no racking or popping).
PCW:
I was experiencing some joint following errors and changed the ferror & min_ferror setting (still not fully understanding them) to stop the errors. I have now changed them to your suggested settings and jogging still works fine (no following errors), but still get the racking and popping when running the gcode program.
I don't fully understand what you mean by plot:
axis.1.motor-pos-cmd
axis.3 motor-pos-cmd
with a simple Y move?
I'm assuming we're talking about using halscope here. I set channel 1 to axis.1.motor-pos-cmd and channel 2 to axis.3 motor-pos-cmd and when I jog the y axis, I get two parallel lines (no spikes or anything). Is this what you're asking me to do?
If I issue a G0 y10 command, on the MDI tab, it quickly sends the y 10 inches away from the home position and if I issue the G0 y0 command it quickly returns. All happens smoothly with no racking/popping. If that's not what you're asking for, please let me know as I'm a newbie here.
Also if you need me to make a short video or something, please let me know and I'll try to get that together for you.
Thanks,
Jay C.
Andy:
I agree it's a strange problem to have. I wouldn't have expected it either. When I first saw it, I figured it should be a simple case of two configurations, of some sort that didn't quite match, but I've been over it with a fine tooth comb and can't find anything like that.
I have my jog speed set at 60ipm (the max on my slider). When I tried the rapid jog by jogging with the shift key depressed, as you suggested. It appears to be the same (smooth moving & and no racking or popping).
PCW:
I was experiencing some joint following errors and changed the ferror & min_ferror setting (still not fully understanding them) to stop the errors. I have now changed them to your suggested settings and jogging still works fine (no following errors), but still get the racking and popping when running the gcode program.
I don't fully understand what you mean by plot:
axis.1.motor-pos-cmd
axis.3 motor-pos-cmd
with a simple Y move?
I'm assuming we're talking about using halscope here. I set channel 1 to axis.1.motor-pos-cmd and channel 2 to axis.3 motor-pos-cmd and when I jog the y axis, I get two parallel lines (no spikes or anything). Is this what you're asking me to do?
If I issue a G0 y10 command, on the MDI tab, it quickly sends the y 10 inches away from the home position and if I issue the G0 y0 command it quickly returns. All happens smoothly with no racking/popping. If that's not what you're asking for, please let me know as I'm a newbie here.
Also if you need me to make a short video or something, please let me know and I'll try to get that together for you.
Thanks,
Jay C.
Please Log in or Create an account to join the conversation.
07 Nov 2014 08:45 #52888
by andypugh
Replied by andypugh on topic Mesa 5i25 with plain G540, no 7i77
Can you jog two axes at a time? (some, but not all, combinations of cursor keys move X an Y together).
I am wondering if you are running short of PSU current when moving multiple axes.
I am wondering if you are running short of PSU current when moving multiple axes.
Please Log in or Create an account to join the conversation.
Time to create page: 0.102 seconds