Messages

7301

General Mach Discussion / Re: Fanuac Yellow top motors w/...

« on: December 21, 2016, 02:43:52 AM »

Hi,
new drivers are expensive, I thought you were trying to minimise cost by retaining the existing motors and drives.

If so search until you find the documentation that describes what you've got. I'm not sure but I believe Fanuc have an 'input'
board that attaches the drive proper and that board is what determines the required input signals.

One possibility to be aware of is that some of these drives are not meant to be used as position control, they're intended as
speed controlled devices ie spindles.

If you have to replace/repair the drives/motors you may be advised to ditch Fanuc, good but really pricey. Have a look on Ebay
or Ali express, Delta servos and drives are readily available around the $600-700 for 1hp with 10000ppr incremental encoders
and around $1000 for absolute encoders, that's the motor AND drive, good value and they have a steady reputation. They're a bit
like an 'asshole'...'everyones got one!'. There are cheaper makes on Ali Express as well but for me at least unknown reputation.

I of course don't know your boss, I don't know what makes my boss tick either... but I know what makes him explode!

Craig

7302

General Mach Discussion / Re: Fanuac Yellow top motors w/...

« on: December 20, 2016, 01:12:28 PM »

Hi,
the documentation of the drives you've got should tell you how to configure it so that it can accept step/direction inputs.
You need that documentation.

Craig

7303

General Mach Discussion / Re: Fanuac Yellow top motors w/...

« on: December 20, 2016, 12:40:33 PM »

Hi mrich0908,
that sounds pretty hopeful.

The 'digital' part refers to the input of the  driver. In M3/SS terms that means two digital signals, one controlling direction and the other to step in the
given direction. Older style drivers had one analogue input, a varying voltage between -10 to +10V and the servo would drive to match that, say extreme
left at -10V and extreme right at 10V.

The AC part of the description refers to the fact that the driver synthesises three AC voltages to Apply to the windings of the motor to make it move
and the encoder tells it if its done enuf or whether it needs to make more AC to match your position instruction.

What you have described sounds like it could be controlled by a SmoothStepper. Would need to see some document which proves it, ie 'with jumper X in
position Y the input control method is step/direction...' or least something that would prove it to your boss.

Craig

7304

General Mach Discussion / Re: No communication between computer and controller box.

« on: December 20, 2016, 12:21:29 PM »

Hi captainleeward,
just had another thought, can you beg, borrow, steal, or worse buy another BOB, just a plain one with no built in drivers but one with LEDs
on the outputs? It would confirm whether M3/PP is producing pulses.

At the moment you can't tell. It is possible to probe the PP with a multimeter but one slip with a probe and you risk blowing the TTL inside
the PP. I try to avoid probing the PP whether with multimeter or oscilloscope unless I have insulated contacts that fit perfectly, replacing
chips on a PC mother board is a hit and miss proposition at best!

Craig

7305

General Mach Discussion / Re: Fanuac Yellow top motors w/...

« on: December 20, 2016, 05:34:40 AM »

Hi mrich0908,
don't know anything about Fanuc except that they're frighteningly expensive and the byword in quality.

When you say that the servos/drives work, how do you know? What did you use to test them?

Given their apparent age I'm guessing that they might be analogue input type drives in which case a SmoothStepper
(being pulse step/dir) wont work. I know Hood uses CSlabs gear and they do an analogue version/addons as do other
manufacturers but they are all much more expensive than a SmoothStepper.

Do some research and find out what the drives require as input for position control.

Craig

7306

General Mach Discussion / Re: Xendoll micro machines

« on: December 20, 2016, 05:21:53 AM »

Hi Fred,
never heard of them before. Had a look at their website.

Is the price advantage such that you'd consider a Chinese 'unknown' over a well known/recognised/respected brand?
Only a quick search of this forum and you will find many-a-buyer disappointed with their purchase from such sources.

Craig

7307

General Mach Discussion / Re: No communication between computer and controller box.

« on: December 20, 2016, 05:09:52 AM »

Hi captainleeward,
try bumping up the step and direction pulse widths, 5us if you can't make them longer. Maybe the controller misses very short pulses.

Craig

7308

Mach4 General Discussion / Re: Does Mach4 have "Formula Axis Correction" like Mach3?

« on: December 20, 2016, 01:06:13 AM »

Hi,
while not 'formula axis correction' there is a very useful software utility I use for making circuit boards called Autoleveller.

The blank PCBs are not exactly flat no matter how hard you try by skimming the table, vacuum hold down, you name it.
Autoleveller, theres a freeware and subscriber version, probe the PCB blank on a mesh pattern and then applies corections
to the Z axis of your etching/engraving Gcode from you circuit board program.

With Autoleveller I can successfully isolate route a circuit board with 0.035mm copper layer (standard 1oz board) by taking only
a 0.05mm cut. Never been able to achieve precision depth cuts like that. Now making SMT boards with 0.2mm traces 0.2mm apart.

Probing a big area of MDF would present a few challenges and you would have to pre-process your G code before running the job
but otherwise its indentical to making a circuit board.

Craig

7309

General Mach Discussion / Re: No communication between computer and controller box.

« on: December 19, 2016, 04:43:08 AM »

Hi captainleewared,
got me stumped, given that all motors etc from the same supplier and presumably marked correctly....

The only way I can think of to check the BOB/driver combo is to disconnect the PP cable and feed pulses into
the step pin on each motor. Do you have a signal generator that can produce a squarewave at a few hundred Hz?

Some scopes have a 5V 1kHz squarewave for probe calibration, could at a pinch use that. Whatever you try don't feed the input
pins more than 5V. A 555 timer (output limited to 5V) would work to.

Craig

7310

Mach4 General Discussion / Re: mcCntlGetGcodeFileName ...Crashes M4

« on: December 16, 2016, 10:07:53 PM »

Hi All,
have some working code. Concentrating on the two macros M40 and M41. As DTG pointed out the names are case sensitive. AL and AE call the macros
in uppercase ergo I named them uppercase.

Found best combination of reliability/simplicity to use three persistent registers:
iRegs0/"ALcode" for the file/pathname of the G-code probe job,
iRegs0/"AEdata" for the file/pathname of the raw probe file and,
iRegs0/"ALtol" for your choice of permitted tolerance.

These will need to be set up prior to use. The write to the file/path registers occurs in M40 you will note.

While there are a few error traps in the code which will hopefully give you a clue if the code fails there is little/nothing in the way of recovery yet.

Craig
Also had hoped that the macros could be used as standalone processing jobs but have had little luck yet. As it stands they work if they are called
from the AL/AE generated G-code probe job and work from the editor but less successfully if called from the MDI line.
This code is tested by hand editing probe files and am happy that it works but be warned anyone who uses it are my guinea pigs. Please let me
know if you have problems.

Need to point out also that this is my first effort at Lua coding. If there are inconsistencies/mistakes they represent my newbie status.
Have just started looking at the string pattern/capture functions and believe I could at some future date make stronger and more readable code.