Finally, Here is mine

[RedLabel]

Decided to build one about 1.5-2 years ago, started with jgro but then I changed my mind and built this thing instead.

I'd love to hear suggestions.

BTW, Is there a simple guide for setting up Turbocnc or emc2 to use with xylotex board. I can learn it myself but I want to get this thing running as fast as possible.


Thanks

[harryn]

I like your build. Always amazing what people come up with.\

Maybe you can help me understand how the work mounts on the unit ?

Do you have any plans on how to keep dust away from your drives ?

[RedLabel]

Thanks for the reply harry. I've mounted dremel tool and the cutting table here are some more photos:

[RedLabel]

BTW the the machine jogs very well. However I'm having difficulty in calibration of the machine.

One of the problems is that when I increase the speed I've noticed lockdowns(noise comes from stepper but no movement), when it happens I have to leave the button and push it again. Decreased the acceleration and it helped some but did not solve the problem entirely. I think it is about the pulse shape but I have no idea how to change that in EMC2.

Another thing is scaling. Well I don't know how to calculate(or what is) encoder counts, i can see that this is a counter's value when the stepper turns one round but this is not clear (the logic behind the program is clear for me but there is a lot of ambiguity in the labeling, ex: cycle period refers to period of which cycle?). I read the manual but a lot of things is not explained. So I still could not adjust the lengths correctly? If there is any help I'll be very happy.

[RedLabel]

Need some help here.

Adjusted the machine, using mach3 demo and generating the codes from Mastercam X I have great difficulty in getting clear shapes.

The lines are perfect on each direction(linear and the axises are perpendicular).

But when I tried to cut a square I saw that the 3 edges were nice but the starting edge did not connect properly(You can see in the pictures).

When tried to cut a circle the result was a disaster, the starting point is again not aligned to the finishing point, moreover at 180degrees there is a bad defect too. Its like shifted two half circles. Again I attached a picture about the result.

Broke a nice dremel tip because of an interesting error causing the z axis to move into the table after an operation without any reason.

I hope I can get some help.]

Thanks

[walter]

It could be number of things but the most important is that your gantry might be too heavy for your components.

[RedLabel]

Originally Posted by walter It could be number of things but the most important is that your gantry might be too heavy for your components.

hmm, logical explanation, observed no visible backlash though. I'll check that.

What are the other possibilities? Could the crappy Mastercam postprocessor cause the problem?

Thanks...

[rrrevels]

Generally with that type of error you have either a backlash problem or are losing steps. Run each axis back and forth independently and check with an indicator to make sure that the axis is returning to the same spot each time. Also make sure the computer is able to process the steps quickly enough especially on cirlces. I had a similar problem using an old laptop. It worked fairly well but I would have the little glitches on the circles. When I changed computers and started using Mach3 the glitches went away.

Russ

[aminear]

Are you using any thrust bearings on your leadscrews?
Are you using any backlash device on your leadnut?

Your picture looks a lot like backlash.

[RedLabel]

Actually I was first using emc2 then switched to mach3. There is a huge difference in terms of stepper movement. While using EMC2 at 20ipm there was lockdowns now each axis can jog around 30ipm without any problem.

Well I tried to measure the backlash. When I moved each axis independently I see that there is no measurable backlash(each axis is moved back and forth 4 times and it came to the initial point).

The interesting thing is when I jogged axises jointly(that is I was rapidly changing directions and axises, that was like a random walk) the problem occured. There was a considerable amount of error. The machine was 1-2mm away from the initial point.

What do you think the problem can be?

BTW the current machine has AMD 3000+ venice processor OC to 2.4ghz and running very stable, also it has 1gb of memory which is I think enough for mach3. The signal waveform is quite nice having very very little jitter.

[rrrevels]

The only thing I can think of at the moment is to try the Mach test program, I can't remember the name of it at the moment, to see if there is something running on the computer taking clock cycles away from Mach. My main computer at the house won't run Mach at all with all the stuff I have on it. My shop computer is a clean install with nothing else to interfere. Take a look at the Mach Yahoo group there is a tremendous amount of information there.

http://groups.yahoo.com/group/mach1mach2cnc/messages

[ger21]

What kind of PC are you using? How fast?