Prototrak Plus Retrofit with Mesa 7i77 & 5i25 cards

More
22 Feb 2017 15:34 - 22 Feb 2017 15:49 #88499 by new2linux
Many, Many Thanks to andypugh, Todd Zuercher, PCW, tommylight, Bevins and lakeweb for all you help!!

I have the encoders show/register motion, believe turning the correct direction, and ruffly the correct distance moved. I
can't get them to move under power. With the AXIS screen open, the axes homed, press "Run" and get "linear move on line 11 would exceed joint 2's positive limit"

In the Hal Meter there are several Pins and Signals that will reflect motion when servo is turned by hand if you select thru the list.

I have reviewed the work from the past few days, and believe the 7i77 card is ok, lights on with green heart beat blinking. I am not show where to look and or if there is some recommended reading I am ok with that. Where do I attach multi-meter when using "setp" command, at the DRV 0, TB5?

Many thanks, to all in advance, for any help or words of wisdom that may aid me in my effort.
Attachments:
Last edit: 22 Feb 2017 15:49 by new2linux. Reason: typo, add question

Please Log in or Create an account to join the conversation.

More
22 Feb 2017 16:55 #88503 by lakeweb
First, when you start LinuxCNC are your DROs zeroed? If not, delete the .var file. Start with all your positions in the middle and then set some safe numbers in the INI, like:
MIN_LIMIT = -4.0
MAX_LIMIT = 4.0
For each of your axis's.

Don't run the sample, you don't want the machine to take off until you are ready. You are not even close until you tune this thing. You can start with commands like:
G1F3X.5
G1F3X0
Nice slow and short jogs that you can toggle between. And if all is right at these speeds it probably won't fault. Then you can use the HAL scope, speed up and lengthen your moves and start to tune.

Best, Dan.

Best, Dan.
The following user(s) said Thank You: new2linux

Please Log in or Create an account to join the conversation.

More
22 Feb 2017 19:22 #88510 by new2linux
Many thanks Dan for your thoughts, I will not installed these on the machine until I can get some motion going. I can appreciate the suggesting though, it could get ugly quick.

Please Log in or Create an account to join the conversation.

More
24 Feb 2017 13:08 #88580 by andypugh

I have the encoders show/register motion, believe turning the correct direction, and ruffly the correct distance moved. I
can't get them to move under power. With the AXIS screen open, the axes homed, press "Run" and get "linear move on line 11 would exceed joint 2's positive limit"


If you look in the Axis preview you will see a dotted red box which shows the axis limits.
Your G-code file is probably shown overlapping these limits.

With a fresh setup the origin of the working coordinate system is the same as the machine coordinate system. Most g-code files have a G-code origin somewhere in the part.

If you jog your machine so that the tool position corresponds to the origin of the G-code file, then use the touch-off button to set X, Y and Z to zero you will see the G-code preview move inside the axis limits box, and the code will run.
The following user(s) said Thank You: new2linux

Please Log in or Create an account to join the conversation.

More
24 Feb 2017 15:17 #88585 by new2linux
andypugh and lakeweb, Many, many thanks! for your help, I naturally have been trying the sample g code that comes with the program, and after rereading lakeweb post saved the file a new, edited it with Pluma (text editor), select and delete, 1st tried to save all the beginning of the sample file, added the "G1F3.5" as well as"G1F3x0" where the original "g" code started, then deleted all but.

this is all there is
#<depth>=1.0
#<scale>=.50
G1F3.5
G1F3x0

the error says:
R i j k words all missing for arc
The LinuxCNC logo in the sample, even though the g code is edited is still seen on the screen, even with the code as follows below:

The machine origin appears to be in the red dotted line, if you look at x or y views the origin looks to be on the very edge of the dotted red line, but still no motion. The servos are not on the machine. I plan to reread the prior posts in this thread and see if I can see any voltage change in the DRV 0 as I try the "setp" in halrun, I needed away to hold the meter probes to the TB5 contacts while cycling thru "setp". I fear that the one or both of the cards is bad, not even jog (in set up) has moved the servos.

Many thanks to all that have helped, if you have suggest or comment please do so.

Please Log in or Create an account to join the conversation.

More
24 Feb 2017 16:53 #88588 by lakeweb
HI new2linux,
Don't try to run any program until you have the machine working. When I said use:
g1f3x.5
g1f3x0

I meant them to be entered into the MDI tab. Once entered you can just double click between them when you want to move. The sample has a lot of baggage you just don't need. At that, you should read about any g-code commands you would edit into a run.
G1 documentation
And when you start your machined needs to be at x:0 y:0 z:0. It should be if delete the .var file.

As far as your hardware goes, it is unlikely that there is anything wrong with the 7i77. You would almost have to use a scope as the drive voltage will come and go pretty fast when LinuxCNC starts and then faults. Use clipped leads, don't try to do three handed work. Add pig tails to the output.

As far as testing the servo amps, this may be frowned upon but...
Get yourself a really dead AAA and solder pig tails to that. Like .5 volts. Then you can touch it to the unplugged drive input of the amp and the motor should move. I consider this a fairly save test because as soon as you remove the battery the motor should stop. .5 volts is not very fast, like 5ipm on my machine.

Best, Dan.
The following user(s) said Thank You: new2linux

Please Log in or Create an account to join the conversation.

More
25 Feb 2017 00:07 #88611 by new2linux
lakeweb, many thanks! I have a DC power supply that is variable current and voltage that is settable, so not to over power, I will try to get that out and try your approach.

many thanks, for all the help and suggestions!

Please Log in or Create an account to join the conversation.

More
25 Feb 2017 16:29 #88618 by lakeweb
Hi new2linux,
The input draws no meaningful current. If you can't move the motor with a battery, something is wrong. Last thing I would do is grab a bigger hammer.

Best, Dan.
The following user(s) said Thank You: new2linux

Please Log in or Create an account to join the conversation.

More
27 Feb 2017 15:38 - 27 Feb 2017 15:45 #88675 by new2linux
Many thanks to andypugh and lakeweb! I have reviewed and preformed the prior several posts and do not get any motion from servos, I have applied .5 v to DRV 0, I have setp to see if I could get the multi-meter to show any motion in the needle, have not seen any. I have reviewed the files (.ini &.hal) for the ".var file" to set to 0.0, could not find any that looked close.
I am of the opinion that one or both cards 5i25 or 7i77 are bad, I could use a second opinion and or additional reading/testing to insure the decision is correct.

Many thanks to all that have helped, I have reviewed thoroughly my prior posts and wonder how I can best help myself.

Edit: Prior to using the cards 5i25 or 7i77 I tested the Ptrak in its original conf, I'm shore it ran then.
Attachments:
Last edit: 27 Feb 2017 15:45 by new2linux. Reason: clearify

Please Log in or Create an account to join the conversation.

More
27 Feb 2017 16:00 #88676 by andypugh
I have never used a 7i77. It is entirely possible that there is an "enable" pin that needs to be set true (in HAL) to enable the voltage outputs.
...
In fact, looking at your pin list
hm2_5i25.0.7i77.0.1.analogena
You need to issue:
setp hm2_5i25.0.7i77.0.1.analogena 1
to turn on the analogue outputs.

This does not explain why the battery does not move the drives, but this is again likely to be an enable input (hardware this time) that needs to be set.
The following user(s) said Thank You: new2linux

Please Log in or Create an account to join the conversation.

Moderators: piasdom
Time to create page: 0.283 seconds
Powered by Kunena Forum