|
Topic: Mach 3 conversion
|
| Author |
Post |
|
FA_Marsh
8 posts
Popping In
|
15-01-2009 00:29
I am relatively new to the profiler, but was thinking of trying to convert to mach 3 control to give me 3 Axis machining capabilities. Have many tried and been successful before me? I am aware of bennys conversion, but I am also intrested by other expriences
many thanks
|
|
Bartman
60 posts
Frequent Visitor
|
15-01-2009 01:27
The 2.5amp kit from RoutoutCNC http://www.routoutcnc.com/lpkits.html works a treat with Mach or EMC, and the three driver boards, para breakout board plus a psu cap and a 160VA transformer fit neatly into the existing profiler psu box. I chose these controllers as they have opto isolators on board. Instructions are clear and Mike @ Routout gives excellent support. Alternatively, you can get in touch with Kevin at DamenCNC in Holland http://www.damencnc.com. Again. Kevin is excellent on info and support. Additionally, Kevin has good prices on Kress motors and does an excellent Kress replacement mounting bracket.
Either way you can salvage your investment from the Profiler fiasco and get the Profiler machine running with industry standard interface and software that does what it should do.
I attach a pic of my bash-bash get it up and running temp construction 
Hth.
ps. The transformer and cap are from www.cpc.co.uk(please note you cannot use the original toroid as it's output voltage is too high for the Mach controller boards). Also, I chose a larger toroid as I intend to fit larger stepper motors.
FF01565 toroid 160va 2x18v @ £20.40 + vat
CA04832 cap 22000uF 35v @ £3.98 + vat
CA00780 cap clip 35mm(pack 5) £1.57 + vat
The bridge rectifier was from my bits box but any bolt mounted 20A one will do.
Post edited by Bartman
on 15-01-2009 01:28 Post edited by Bartman
on 15-01-2009 01:31
|
|
vibo
52 posts
Frequent Visitor
|
15-01-2009 10:51
Hello Fa_Marsh,
Not happy with the hard and software delivered with the Profiler I also started to look to alternatives. Many ways to mill...
I did implement the Benny-card, works without any problem I can tell you after almost a year experience with it.
I tried Mach3, worked fine, but I was limited to about 500 lines of G-code so I had to buy it. But this is Windows based software and not ideal for the timing. So I hesitated to spent my money yet...
At that moment I found EMC2, linux based milling software. http://www.linuxcnc.org/ . It is very reliable, and it was relatively easy to install it.
Its Open source, comes completely installed in an Ubuntu package.
Because Linux nowadays looks very similar to control as Windows or OSX, it is not so difficult to get into it. A picture:
http://www.elektor.nl/forum/forum/actuele-projecten/profiler-en-linux.494755.lynkx.
My other suggestion is http://www.usbcnc.com/. The hardware on the card does all the timing work.
DamenCnc als changed from Mach3 to the hard/software solution of usbcnc.
And DamenCnC is indeed a great on his info and support.
If you want help for a Linux experience I can help you!
Pieter
|
|
FA_Marsh
8 posts
Popping In
|
16-01-2009 00:41
Hello Vibo and Bartman,
Thank you for your prompt advise and direction. I will look over the possibilities you have suggested this week end and report back progress. I am tempted to recreate the mach interface with the charge pump safety interlock on pin 1 of the printer port. I t seems not so difficult, but I am pre-judging your replies. Any way I look forward to being able to use the profiler more easy in the near future
Many thanks
FA_Marsh
|
|
vibo
52 posts
Frequent Visitor
|
16-01-2009 09:27
Hi FA_Marsh
With EMC2 I implemented this charge-pump-safety, and auto-switch on/of of my kress and vacuumcleaner. This makes live much easier!
I can recommend switching to a setup to work directly with G-code (Mach3, EMC2 etc.) this opens the way to work with many more software-applications instantly.
The goal is to get an instrument that works straightforward from design to milling and without all the conversion-hassl.
I bought myself Cut2D and Cut3D from vectric to create the G-code. Very nice for hobbyists, but it has some disadvantages for more complex tasks...
G-luck!
Pieter
|
|
Sudar
220 posts
Power User
|
16-01-2009 11:31
|
|
FA_Marsh
8 posts
Popping In
|
17-01-2009 12:03
Hi All,
And thank you for the further input. Last night I set to and hand built a basic Mach 3 opto isolated interface, I managed to get the motors moving and then stopped. I will continue later today. I can post a photo and circuits for anyone interested.
I do however have a further question.
Should the charge pump circuit stop all movement or only the spindle rotation?
Many thanks
FA_Marsh
|
|
FA_Marsh
8 posts
Popping In
|
20-01-2009 10:23
Hi All,
The hand wired prototype is now all working to my satisfaction. I felt it best to disable all movement in the absence of the charge pump signal. I will try to upload a picture of it now. The next job is to enter the circuit into protel and publish it as a pdf. then I would be open to suggestions for improvements!
Many Thanks
FA_Marsh
|
|
vibo
52 posts
Frequent Visitor
|
20-01-2009 12:46
Gratulations if it works!
Looks nice in this prototyping state.
Can I consider this as your own version of the Benny approach, controlled through the parallel port?
Indeed the chargepump switch disables all movement. If not you would distroy a lot when only the mill stops, but its movement continues. I wouldn't mill, just crash itself and your cutters in your workpiece.
P.
|
|
Sudar
220 posts
Power User
|
20-01-2009 19:29
Hi FA_Marsh
Wow realy Cool
Is it possible to draw it on Eagle or Target? Then the most of us will be able to mil it with our Profiler 
(German proverb: What was first: the chicken or the egg? )
Regards Lukacs
|
|
FA_Marsh
8 posts
Popping In
|
21-01-2009 13:45
Hi All,
I guess in answer to vibo's question, yes you could consider this a rehash of bennys board. I have only seen a photo of his board.
In the end I used the three chip enable lines to the L6208s to disable the drive electronics on no charge pump.
This is the moment when I expose myself to constructive criticism by posting the schematic as pdf.
I chose to use the parallel port with active low signals to minimise the connections to the PC electronics, no 5 volts required to drive PC side electronics, however I am still open to criticism because the home switchs and E-stop drive are directly connected to the PC as is the PC_GND and machine GND.
I also chose to use a small processor to manage the charge pump logic.
I am unsure at this stage whether to include a further input for a probe and or to include an extra opto isolator to give a 4th axis on the expansion connect pins PC_P8, PC_P9
I have noticed that the motors 'whine' alters while driving and wonder if this is because the XP windows system interfers with the constant pulse stream. Does this affect work quality? Does the linux system exhibit the same symptoms?
Is this why people have graduated to a serial or usb pulse generator design? hmm... a whole new oportunity!
Enough from me, I am looking forward to some feedback
FA_Marsh
 08010_iss1.pdf
|
|
BoeingBrown
689 posts
Power User
|
21-01-2009 13:55
Below the schematics of the "Benny board".
He's famous in the mean time !
It was free to use (he wrote to me), but it's not available in Eagle file type format.
Probably usefull for comparison ?
Hessel
p.s. Visuals are better after downloading and opening it again.
Post edited by BoeingBrown
on 21-01-2009 13:56
|
|
vibo
52 posts
Frequent Visitor
|
21-01-2009 15:39
I have noticed that the motors 'whine' alters while driving and wonder if this is because the XP windows system interfers with the constant pulse stream. Does this affect work quality? Does the linux system exhibit the same symptoms?
Hi FA-Marsch, I am using the Linux EMC2 system. I don't know exactly what you mean with the altering whine effect. Is it altering while you travel at constant speed? If it is really important could you record the sound so I can check it?
My linux system is very stable and I have not noticed any problem with timing ever. (I believe we're the only two Benny-isch adapts on this forum).
The only disadvantage of my Linux setup is that the programs to design my project and create the final toolpaths are Windows based, so I always have to use a second computer or make the system dual boot...
regards, Pieter
|
|
FA_Marsh
8 posts
Popping In
|
31-01-2009 19:57
Hi all,
It has been a few days since I last posted, work intervened! The PCB layout has been finished and I will send the board off on monday. If there is still any interest I will post the final circuit diagram, gerbers etc here monday evening. It seems to me that there is remarkably little activity here for a project that kept the manuafacturer in so much difficulties for so long. Even if all the interest is not positive.
If the people dont like the profiler then it would seem to me an ideal spring board for further development.
Hope some one is still listening
FA_Marsh
|
|
vibo
52 posts
Frequent Visitor
|
01-02-2009 22:26
Hi FA_Marsh,
Yes, I'm very interested in your final design.
And I think it's good to show the Profiler audience that it is not that difficult to upgrade the hardware (thanks to people like you who are able to make and share these nice solutions!).
I will put a link on the Dutch forum too...
Btw, did you read my pdf on how I converted mine, I completely forgot to mention it before. There is a little about the setup of Mach3 in it.
http://www.oreid.nl/profiler/instructions-parallel-interface.pdf
Regards, Pieter
This was the original post...
http://www.elektor.nl/forum/forum/actuele-projecten/g-code-via-parallelle-poort-3d-met-de-profiler.431887.lynkx
|
|
FA_Marsh
8 posts
Popping In
|
02-02-2009 23:05
Hi,
Here are the final schematics and the gerber plots as promised. Now I have to wait for the boards. You will see that I work almost exclusively in SMT. If anybody wants the protel file I might have to send it directly, 16.5 meg
In the mean time I might look at a serial port interface for mach 3
Thanks for the encouragement Pieter, I am not holding any copyright over the interface designs and will happilly supply the source and hex files for the processor handling the charge pump etc. i hope someone enjoys it. Nothing worse than designing something nobody wants!
FA_Marsh
08010_schem.pdf Top.pdf Bottom.pdf
|
|
vibo
52 posts
Frequent Visitor
|
03-02-2009 12:27
Hi FA_Marsh,
Thanks for sharing your work!
Looking at your schematcs I notice a pic18F1320. Does this one have to be programmed?
I've put I link on the dutch forum...
Pieter
|
|
Bartman
60 posts
Frequent Visitor
|
03-02-2009 16:58
Good to see people beavering away at their own solutions to the Profiler fiasco.
Perhaps Elektor could come up with their own Mach interface and stepper driver project?
|
|
FA_Marsh
8 posts
Popping In
|
05-02-2009 00:37
Good evening,
The PIC18f1320 needs programming I will post the C program(very simple) and the hex file as soon as the pcb comes back. I changed some pins and it needs a tweak compared to the hand wired prototype. the 6 pin connector has been wired to match the pickit 2 programmer, extremely cheap microchip dev tool. If anybody is in a rush I can mod the program sooner but I wont have checked the software in the final board.
thanks for the encouragement
Pieter, I did see your original article, which prompted me to start experimenting!!
In reply to bartman, It shouldn't be too difficult to design the 3 axis drive electronics. the problem would be getting everyone to agree on the specs, total drive current etc. Having said that I have had the existing motors stall on my profiler, so maybe I am tempted, anyone attempted to change the motors out there??
FA_Marsh
|
|
Bartman
60 posts
Frequent Visitor
|
05-02-2009 01:16
I find with the RoutoutCNC 2.5A driver board that it takes an awful lot of effort to push against the Z axis driving up(which given that it uses the same thread pitch and nut as the x & y axis would give a similar resulting drive force) and would suggest that you either have a misalignment between the lead screw and nut causing binding, a bent lead screw, or the original Colinbus driver is not up to the job(can't do a comparison as my original Colinbus controller went in the trash months ago). You would also be advised to replace the non flexy couplers that Colinbus supplies with something like this http://cgi.ebay.co.uk/3pc-0-25-x0-25-HBL-shaft-coupler-for-servo-step-motor_W0QQitemZ260357176589QQihZ016QQcategoryZ633QQcmdZViewItemQQ_trksidZp1713.m153.l1262 or similar from DamenCNC.
Post edited by Bartman
on 05-02-2009 01:19
|
Informative articles
Password forgotten?
