6040Z-S80 shakes and grinds

[TheGyrator]

The title pretty much says it all.
Chunking, rattling noises from all 3 motors and shaking/skipping as it moves.
It sounds like it's dying!
Polarity is correct and in terms of moving when controlling it manually or testing a file it's going in the directions it should just not very well so I've shut it down for fear of it self destructing.

I have the same blue box controller most people condemn running on Mach3.
Please help!

Similar Threads:

• Problem- Pro 4824 with the shakes
• Need Help!- SX3 Head shakes when entering material.
• HELP Lathe shakes while threading
• Stepper shakes when it normally stand still

[tgdavies]

Has it always done this, or did it previously work OK?

How does it behave if you move it manually via the knobs on the stepper motors, with the power off?

What are all your Mach3 motor settings?

[Justblair]

I had shakes and grinds on my 3040t X axis... I tried lubrication everywhere.. Turns out it was my controller. Replaced the stock with a gecko 540 and with a, little adjustment managed to get to silky smooth.

[TheGyrator]

Hi thanks for the quick response guys.
I think it's the dreaded blue box controller that has the lousy reputation.
It ran fine for a couple minutes this morning as I've witnessed other owners of the model then went into some sort of wicked voodoo dance.
I pulled the motor and that's what you see in the video.
For a split second it runs smooth then goes wonky.
http://tinypic.com/player.php?v=91bi...s=8#.VJdp6V4AM

If I move it around manually I get those 2-3 second bursts of smooth running.
I know it can be modded/repaired and I've got a fair bit of college behind me as an electrical engineer but, I see degree bearing engineers far more experienced and knowledgeable than myself do this all the time.
Buy something rather than repair or build the existent!
I must have a dozen less complicated gizmos to repair as it is.

[aarggh]

I think it's the dreaded blue box controller that has the lousy reputation..

More than likely!

Buy a Gecko G540 and never look back!

cheers, Ian

[TheGyrator]

Thanks but that's WAY out of my price range.
I'm looking at the typical Chinese controllers everyone has a love/hate relationship ranging from worked out of the box to I had to replace 3 devices and add filter caps.

I think $150 on a Chinese unit is going to be my limit which is probably 4x what it costs to manufacture, advertise and mail it to me but obviously as the end user what matters is it works.

[aarggh]

Thanks but that's WAY out of my price range.
I'm looking at the typical Chinese controllers everyone has a love/hate relationship ranging from worked out of the box to I had to replace 3 devices and add filter caps.

I think $150 on a Chinese unit is going to be my limit which is probably 4x what it costs to manufacture, advertise and mail it to me but obviously as the end user what matters is it works.

If your buying the all-in-one or the stand-alone YOOCNC clone boards, you'll have tremendous troubles, as the design simply doesn't work. The TB6560's have a very peculiar power up/down sequence that blows the drivers erratically. The design of almost all these boards also is sub-standard, on most the opto-couplers either don't work, or impede signals, the buffering and signal conditioning doesn't work, the extra 5VDC rail has been a source of issues, etc, etc.

Sometimes spending cheap ends up costing more, I'd hate to think how many people have bought these boards and then had to spend double again to get their machine running.

Been there, done that, I absolutely wouldn't do it again no matter how cheap it is, if it doesn't work, it just aint cheap. I'd recommend reading through the numerous posts before laying down cash anyway.

cheers, Ian

[TheGyrator]

Now I'm at that point where it seems Santa's elves fixed everything in the dead of night.
I had set the LPT to plug and play detection port 3 and "try not to use an interrupt" which just changed the pitch I would hear from the motors in their standby power state. It still went into earthquake mode so I messed with that a bit until tonight.
This time I changed it to those settings in safe mode, restarted and it's zinging around as it should.

I'm running XP on an old 2600+ AMD x64 athalon which is obviously a dinosaur.

I'm going to keep testing it to see how stable it is.
I definitely do not feel I am out of the woods yet, however, I do believe the blue box isn't to blame but in fact it's a windows process that's starting up and creating distortion on the LPT port. I have another PC with a bit better specs run by a dual core and was avoiding the sacrifice but what must be done must be done.
The current PC has a clean XP install without anything to bloat the system and the only aftermarket installs are Mach3, Mill Wizard, Eagle and Eagle's add on PCB-to-Gcode.
I'll try to post some relevant data after the holiday as I fear sooner or later it's going to go bonkers again.

[TheGyrator]

This is the first real step of progress.
I changed from ACPI user to a standard user in device manager.
It runs 100% off the keyboard and at just about any speed I set it to within reason(I triggered the sensor shutdown built into mach3 running too fast)

Then I load any of the included templates in mach3, post the g-code and well you see what happens at 1:14.
It bumps and grinds like it's in a sleazy night club making a pass at miley cyrus!
As of early 2014 Microsoft no longer offers XP support for updates, work arounds, patches etc.
I'm betting it's a porting issue with either the Mach3 driver or part of the Window's OS but either way Window's won't be patching it and I can't afford to buy a new version of Mach3.
It's probably a safe bet to assume the older version of Mach3 that came with the mill isn't licensed aside from being outdated.

I'm at the last of my savings as this is simply a home start up business venture based primarily on the year I've completed in college for electronic engineering, the mill and a variety of other tools and building supplies.
It's a difficult financial situation when I have no income to spare ATM.

[TheGyrator]

All problems 100% solved by changing the PC from an ACPI computer to a Standard computer in device manager under computer in Windows XP.

My problem when running G-code tests was due to my Steps Per Unit being set waaaaay too high.
It's now at mfr. manual specs and doing great!
For anyone that lands in the same boat I hope you find this thread rather than throwing away so much bread.

My next course of action is watching the Mach3 tutorials so I have a true understanding of what I'm doing vs the current which is understanding enough of it to fiddle around but I would surely destroy things if I tried to mill materials at this very moment.

[Dean Elliott]

In an effort to consolidate all of the 6040 and 6040z threads please post here: http://www.cnczone.com/forums/open-s...ml#post1790768

[aarggh]

In an effort to consolidate all of the 6040 and 6040z threads please post here: http://www.cnczone.com/forums/open-s...ml#post1790768

Why?

This is your link expanded:

"Forum - OpenSource CNC Design Center - Open Source Controller BoardsBuild Thread - CNC 6040 6040Z - China Made CNC Routers"

Can't see that working for the multitude and variance of posts related to 6040's, 3040's, 3020's, and the associated drivers, steppers, power supplies, etc.

cheers, Ian

[Jon.N.CNC]

Pretty much guarantee 90% of your problems are the tb chips in your drivers. Those things output 1.2a repeatable current max and that's the good ones the 15% minority that have genuine non counterfeit chips with heatsinks. The bad ones will output 0.3-0.5a repeatable current max before they overheat. For those that are confused they are rated on their peak current so if they boast 4.5a you may well get 4.5a for a millisecond or one step, then the current drops like a discharge of capacitor in a camera flash.

Best advice I can give you is waste little time on them and get some mosfet drivers. Tb drivers are not capable of smooth accurate motion.