Machsupport Forum
Mach Discussion => General Mach Discussion => Topic started by: dsnellen on May 13, 2012, 09:22:03 AM
-
I posted this on CNCzone but no response. I have been using Mach 3 for several years, no problems. On Thursday this past week for some reason C-4 Safety Charge Pump would trip off several times. Of course, the router turned off and the motors stopped but Mach 3 would continue to run clicking off the X,Y & Z counters. Mach 3 responded to the ESC and acted like all was well. This is the first time ever this has occurred. The machine is properly grounded and since it is a new rebuild, the limit switches are not yet installed or deactivated. What would cause Mach 3 to drop the 12.5Kmz signal and yet continue to run? And what do I do to fix it? Thanks.
Dave
-
Hi Dave,
Are you certain that the 12.5kHz output from Mach has been stopped or could it be that the C4 is just not recognizing it for some reason ??
Tweakie.
-
Thanks for the reply. I can't say and don't have the equipment/knowledge to test the idea. The C4 has been working well for about a year. As soon as I hit the switch, it comes back on. And M3 has been working well for 3+ years now. It only happened that one day. I wonder if a stray electrical current would have tripped the C4 or spooked M3. Our electrical lines is not clean electric. Don't know if the C4 is sensitive to voltage flux. Sometimes turning on the 220v vac will cause the M3 computer to lock up. No clue why that would happen as they are two totally different circuits.
I went through the grounding thing last year with Al the Man. Every piece of aluminum and steel is grounded back to the control box, the wall and the computer. I will be back in the shop on Friday and see what happens.
Thanks
Dave
-
What kernel are you running?
Hood
-
I can let you know on Friday when I am back at the machine. Thanks. Dave
-
Hood - My kernel speed is 25,000Hz and my M3 version is R3.042.029. Thanks Dave
-
Ok that should be fine, it was just when you mentioned some instability I was thinking it may have been you were using a high kernel as that can lock up some computers.
Hood
-
Thanks for the reply. I have been running this configuration for about a year with no problems. I ran today for about two hours fine. No problems. Then I set up the next job and the charge pump flip off twice in about 5 minutes. M3 continued to run. Made restart a bitch. The second time I quit for the day. Still wondering if it's a power problem, a computer problem, a M3 problem or a charge pump problem. I am at the point of taking the charge pump out of the system. Would hate to do that. Any suggestion would be appreciated. Thanks Dave
-
You really need to find out if the problem is the breakout board or Mach and really the only way to do that would be with a scope I would think. Maybe you know someone that has one that you could borrow if you dont have one yourself.
Hood
-
probably the c4 board. crappy solder or connection. i had 1 give me problems.
-
Thanks. I can't think of anyone around with a scope. And it would take a long time of watching (or recording) the data to see which one misses the step.
Thanks tripleback. That is my solution. They are not that expense and can be swapped out quickly. Over the past couple of days I have made some expense firewood which costs far more than a C4. I have already ordered some digital sensors so I can restart with accuracy. See http://www.cnczone.com/forums/open_source_cnc_machine_designs/101878-electronic_home_switches_made_easy.html for details.
Thanks for the replies. Dave
-
You can use an earphone to detect the 12.5khz signal but detecting dropouts or intermittent errors is out.
If it fails for some reason such as a software bugger-up, will it stop altogether or recover itself?
-
Thanks BluePinnacle. It is easier to replace than troubleshoot. However, I revisited the M3 XP_optimization page and rechecked everything. All that I did was a disk cleanup, a defragment and reboot. It has ran perfectly for the past three days. Go figure. It dawned on me that since I have a tape on the X axis, I could simply write down the X stating zero position to within 1/64". And on Y, a piece of blue tape across the track at the zero position. Its eyeball accuracy but usually close enough for woodworking. Still waiting on the magnetic digital switch to arrive.