Re: [Emc-users] hm2 IO

2009-05-30 Thread ejohnson
attachment: clip.jpgattachment: clip0001.jpg--
Register Now for Creativity and Technology (CaT), June 3rd, NYC. CaT 
is a gathering of tech-side developers  brand creativity professionals. Meet
the minds behind Google Creative Lab, Visual Complexity, Processing,  
iPhoneDevCamp as they present alongside digital heavyweights like Barbarian 
Group, R/GA,  Big Spaceship. http://p.sf.net/sfu/creativitycat-com ___
Emc-users mailing list
Emc-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/emc-users


[Emc-users] Latest hostmot2 drivers for m5i20

2008-09-15 Thread ejohnson
Sebastian, et al

I built the latest hostmot2 drivers from source. Unfortunately where the machine is does not have an internet connecton, nor are the development tools loaded on the target system, so I compiled on my development system. I then copied the files hm2_5i20.ko and hostmot2.ko to the target machine. This of course could be the problem, although it is very close to working.

They seem to load and run, as I noticed the stepper paramters (step length, etc.) were changed from floats to U32s. I patched the hal file for that and everything loads without error.

What I get is, the PWM gives some output, but does not seem to be changing (I don't have a scope connected) or at least doesn't change much. Additionally the two stepper motors don't seem to be working at all. The GPIO however seems to work fine.

Any ideas? Are there any other files I need to copy over to the target system?

Thanks,
Eric




-
This SF.Net email is sponsored by the Moblin Your Move Developer's challenge
Build the coolest Linux based applications with Moblin SDK  win great prizes
Grand prize is a trip for two to an Open Source event anywhere in the world
http://moblin-contest.org/redirect.php?banner_id=100url=/___
Emc-users mailing list
Emc-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/emc-users


Re: [Emc-users] Latest hostmot2 drivers for m5i20

2008-09-15 Thread ejohnson
Sebastian,

Here are the links for thehal file and ini file I am using. Any other config filesare unmodified from the defaults (core_stepper.hal, emc.nml, etc.).

HAL file:
http://pastebin.com/mfb07fa0

Ini file:
http://pastebin.com/m4a155f84

Is there a script for building the deb file or do you have a link on how to do it? I did a quick search on linuxcnc.org but did not see how to do that.

Thanks,
Eric

I would suggest you compile the CVS TRUNK to emc2.deb on your development machine and install the .deb on your CNC machine.Also, please pastebin the machine config files you're using.

-
This SF.Net email is sponsored by the Moblin Your Move Developer's challenge
Build the coolest Linux based applications with Moblin SDK  win great prizes
Grand prize is a trip for two to an Open Source event anywhere in the world
http://moblin-contest.org/redirect.php?banner_id=100url=/___
Emc-users mailing list
Emc-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/emc-users


Re: [Emc-users] Latest hostmot2 drivers for m5i20

2008-09-15 Thread ejohnson
Sebastian,
 If you're using the latest stepgen.c (top-of-tree TRUNK  2.2 branch), then you don't need the raw stuff starting on line 30. 

Ok, I will take them out. Looks like you're using the same stepper timings as I do on the steppers I test with - is that correct for your setup too? 

I used them for a start and they seem to work. Oops, and it looks like the u32 stepgen timing parameters are commented out, and the old float ones are being used. I'm not sure how those floats get read in, but they might all read as (u32)0, which could also cause your steppers to not work. 

Sorry about that, I went back to the old drivers because at least the stepper motors worked. As you can see, I have two sets of parameters, one for the old driver and one for the new driver. I just enable the set approrpiate for the current driver. I see you have the Axis 2 motor command hooked up to the LaserPower net and from there to pwmgen.00.value. Since you're not setting pwmgen.00.scale it'll default to 1.0. You should be seeing full PWM (100% duty cycle) when Axis 2 amp enable is on and Axis 2 motor position is +/- 1. Is that not happening? 

For testing purposes, I comment all that out andjust useLaserPower linked to the pwmgen.00.power. Then set it manually through halcmd. Since I do not have a scope, I use the laser power to indicate whether the PWM is changing. Basically it comes on at a fairly low power, and does not seem to change.

When I measured the PWM with a meter, I got about the full 3.3 V, but thought that was just the meter. It is new, and I have not tried it on PWMs before. On many meters you can get a good estimate of the duty cycle through the voltage readings, but this one seems to show only 0 and 3.3V. However if it really were 3.3, then the output of the laser should be significantly higher.  One thing that would help debugging this is to set things up so you think you *should* be seeing a strong pwm dutycycle, and then run (with halcmd) "setp hm2_5i20.0.raw.dump-state 1" and "show all", and pastebin the output of those two commands and also the dmesg after them. 

I will try to post that shortly. This would be much easier of your CNC machine was on the net, but I understand sometimes that's a hassle to do 

Unfortunately the computer box is panel mounted to the machine, so a real pain to disconnect and move to where there is an internet connection.

Regards,
Eric

-
This SF.Net email is sponsored by the Moblin Your Move Developer's challenge
Build the coolest Linux based applications with Moblin SDK  win great prizes
Grand prize is a trip for two to an Open Source event anywhere in the world
http://moblin-contest.org/redirect.php?banner_id=100url=/___
Emc-users mailing list
Emc-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/emc-users


Re: [Emc-users] Latest hostmot2 drivers for m5i20

2008-09-15 Thread ejohnson
Sebastian,

I am using an unmodified, up-to-date CVS checkout from source, then built from source and copied the binaries I thought had changed.

While it got lost in the exchange of emails, I was really most concerned about the stepper motors not working, and figured the problem with the pwm might be related. If I set the power level to 50%, or almost any other non-zero value, my meter reads about 3.3 volts. I know that can't be right because at 3.3 volts or 98% duty cycle, the laser should be full on and it is at more like 10%.

If I use theoriginal m5i20 driver, the laser (pwm) responds correctly, so the wiring sure seems to be correct.

BTW, what is the frequency of the PWM? The laser accepts 5-25Khz.

Here are the things posted to pastebin requested in the prior email:

Show all:
http://pastebin.com/m6ab8b073

Dmesg:
http://pastebin.com/m1478cde5

Regards,
Eric


Hm, I didnt think I'd broken stepgen again... It works fine on my development system. We'll have to debug this one too.Are you using an unmodified, up-to-date CVS checkout? Are you on TRUNK or 2.2?Wait, so is the problem with the pwmgen or with whatever it is that takes the PWM signal as input?If you set pwmgen.value to 0.5, do you read 1.65 V with your voltmeter?

-
This SF.Net email is sponsored by the Moblin Your Move Developer's challenge
Build the coolest Linux based applications with Moblin SDK  win great prizes
Grand prize is a trip for two to an Open Source event anywhere in the world
http://moblin-contest.org/redirect.php?banner_id=100url=/___
Emc-users mailing list
Emc-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/emc-users


[Emc-users] Debounce component syntax for multiple groups

2008-04-11 Thread ejohnson
Hi all,

The documentation for the debounce component shows the syntax as:
loadrt debounce [cfg=size[,size...]]

Trying various permutations, the  only syntax  I got to work so far was
(for example):
loadrt debounce cfg=3 2

I checked the various example configurations, but did not find one that
defined more than one group of debounce components.

The first question is am I doing something wrong, and if not, what is
the appropriate fix? Should the documentation be changed to match the
implementation, or should the component's command parsing be changed to
match the documentation?

I don't mind giving a go at fixing it, if that is the appropriate thing
to do.

Thanks,
Eric




-
This SF.net email is sponsored by the 2008 JavaOne(SM) Conference 
Don't miss this year's exciting event. There's still time to save $100. 
Use priority code J8TL2D2. 
http://ad.doubleclick.net/clk;198757673;13503038;p?http://java.sun.com/javaone
___
Emc-users mailing list
Emc-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/emc-users