Hello Guest it is April 29, 2024, 09:11:50 AM

Show Posts

This section allows you to view all posts made by this member. Note that you can only see posts made in areas you currently have access to.


Messages - jallitt

Pages: « 1 2 3 4 5 »
31
General Mach Discussion / Re: Help!!!!!
« on: April 01, 2008, 05:18:33 PM »
I just stepped through the hole pocketing in single-blk mode - you can see the toolpath move to the right (with a G03) before each arc is cut. Maybe try slowing down engage-feed in VM. Does it skip steps with aircuts or only when cutting?


32
General Mach Discussion / Re: Help!!!!!
« on: April 01, 2008, 05:10:32 AM »
is it skipping steps in one axis in the G00's? If you turn off "use rapids" in the feedrate dialog in VM it'll do G01's at the selected feedrate instead. Have you checked your motor tuning to make sure it can still run at top speed in that axis?

33
General Mach Discussion / Re: MaxNC and mach3
« on: March 16, 2008, 07:37:04 AM »
Actually MaxNC OL uses 12 outputs from the parallel port - 4 for each axis which are amplified directly into the stepper phases. The maxNC driver (if you can call it that) has 3 7400 chips, 12 power transistors, 12 resistors and the power supply - it's definitely not doing any sequencing itself :)

Mach has a special "maxNC wave mode" which cycles the correct bit patterns on the output to turn the motors so my question about theoretically supporting full step and half step is valid.



34
General Mach Discussion / Re: MaxNC and mach3
« on: March 15, 2008, 02:23:43 AM »
So... rearranged the wiring and all 3 motors are spinning nicely.

Now I do have a hypothetical/theoretical question. I realise MaxNC support isn't a priority and we're lucky to have it work at all :) but since all the MaxNC driver circuit does is turn the 4 outputs  for each stepper into power for the 4 phases then wouldn't it be possible to run exactly the same hardware in full-step (for more torque), wave and half step modes from Mach since all that requires is modifying the bit patterns that are output? I'm guessing full-step would run hotter (although you could switch to wave when the motor wasn't turning)

--

Jon

35
General Mach Discussion / MaxNC and mach3
« on: March 14, 2008, 10:34:04 PM »
This isn't really a question - just thinking out loud.

Just picked up an old MaxNC OL controller, a win95 PC  :(  with MaxNC loaded and 3 steppers. This is going to be my budget lathe controller but the last thing I need is another PC so I'm hoping to get it working with Mach3 and share the PC I have on my router.

Plugged everything into the win95 PC and all three axii jog (albeit slowly) just to make sure it's all working.

Plugged it into an XP PC to test with mach3, set MaxNC-10 wave mode and the motors twitch but don't spin. I'm guessing I need to check the motor phases are wired exactly right. MaxNC seems to let you wire up the motors anyway you want, then has a sequence checker program so you can cycle through all the possible combinations until the motors turn smoothly.

I also grabbed the MAXNC pinout diagram from the FTP site so I'll put a scope on the LPT port to triple check the 12 output phases are in the right places.

--

Jon


36
I picked up a second-hand chinese CNC router/engraver a couple of months ago. Cast iron frame, 500x300mm working area, german ballscrews, no software and a manual in chinese... Finally got it all working (after much trial and error) but wasn't happy with the quality of cutting I was getting from the onboard RS232/HPGL controller. If I didn't have a Roland MDX 20 I might have accepted the 5mm/sec feedfate I needed to get a clean finish but the little roland machine makes really nice renshape parts at 15mm/sec (assuming I don't go too deep and stall the spindle). The "new" chinese machine seemed like it should be able to do better and faster. After pulling the covers off and doing some research it seemed like a fairly simple project to convert it to run under Mach3 - and it was.

I used the existing XY motors, drives, optointerrupter home switches and power supply. Z axis was driven directly from the controller board so I needed a driver (from http://www.oceancontrols.com.au) and I also bought a new motor for Z.  Breakout board was also from ocean controls. Local electronics supplier Jaycar had a parallel port 8 way relay board kitset so I'm using that to control the spindle and to cut off power from the steppers. Added a second parallel port to an old PC and got the router cutting last weekend. I also managed to get the breakout board and relay board mounted inside the machine. I still need to tidy up the wiring but it's all working. Solid 3 meter per minute in XY and about half that in Z (the spindle motor is pretty heavy so unless I can find a way to spring load or counterweight it that's about as fast as it's going to go). I also hooked up a touchplate made from 1/16" thick printed circuit board, a contour shuttlexpress and, the PC used to run a security camera and it has a video capture card, so I mounted an IR camera so I can remotely monitor the job - can't really see what's going on but at least I can tell when it's finished cutting a pass.

Ran some test cuts (to compare to a part I made on the Roland) and it's running pretty smooth. Also, since I could rough much more aggressively and cut without worrying about stalling the spindle I got the part cut in 2 passes in about an hour rather than 4 passes in 6 hours. I have another test I want to run which will really tell me if it's working reliably since the toolpath has thousands of rapids and z lifts.

There are some pics of the machine here:

http://www.b26354.co.nz/CNC

Now I keep looking at my Seig C6 lathe and thinking about retrofitting it :) If I can find some really cheap stepper drivers it might almost be worth converting the Roland to run with Mach3 (and since I have an aftermarket 4th axis for it I could run simultaneous 4 axis)

37
something like this - but if your output from mach is somehow turning itself on it's probably not going to help... This only inhibits the spindle if Mach isn't running or the PC reboots.


38
not sure if this is any help but I wired my spindle through 2 relays in series throught the NO contact of one relay to the NC contact of the other. Which means the only way the spindle can start is if output 1 is on and output 2 is off (which should only happen when Mach is in control). If I yank the parallel cable out (or the PC restarts when the milll is connected) all the relay ouptus float high, all the relays activate and the NC contact breaks so the spindle can't run. Since I have more relay outputs than I'll ever need I'm also running the stepper power supply through 2 relays the same way so they're only powered up when the enable output is on. I'll probably install a charge-pump at some stage so this is a temporary solution.

39
General Mach Discussion / Re: New PlugIn just about done..
« on: March 04, 2008, 08:59:30 PM »
So - just pluged in a shuttleExpress and downloaded the plugin and everything seems to be working fine (once I figured out which buttons were which) - but I'm wondering if there's any way to speed up the middle wheel. The outer ring has a nice range of speeds but, working in mm, I seem to need to spin the middle wheel around several times before I can hear a stepper stepping. Does it go 25.4 times slower that it should working in mm :)


-- edit nevermind - found the answer on page 3 of this thread :)

40
Poppabear, thanks for the screen set. That's exactly what I was looking for as an example of custom buttons with scripts attached and a way to clear the tool from the workpiece for inspection (or vacuuming out cuttings)

I've modified the button scripts so that it remembers X, Y and Z position so when you resume it goes to the stored XY position, starts the spindle, goes down to the prev Z and restarts cutting (so it bypasses the preparational move dialog).

I want to do some more testing and get the zoffsets right to get Z to goto a safe machine coordinate (rather than an offset from 0) which should be easy enough but if anyone is interested I can attach the scripts.

Pages: « 1 2 3 4 5 »