My first CNC build (or is it a retrofit?)

[jamesbond007]

Hi ya'll, it feels so good to finally being able to post my first CNC build, It can be considered a retrofit, since I'm taking a Junkyard found TechnoIsel 3 axis table (small table but I think good for my first) its Model 027 with 500x540 table travel size.

I've reading and reading more about CNC for over 3 years and since I don't have an engineering background (except genetic) Its been a wild ride but I think I'm finally on the verge of accomplishing this machine to work, so I can later build one bigger.

My setup:

Gecko G540 with 425 oz/in steppers and a 25V 10A power Supply kit bought to Hubbard CNC (excelent service BTW)

Planning on use MAch3 (trial version first)

Limit switches I will set them in serires and home switches individually, will take all four inputs of the G540 but my Techno Gantry already has the microswitches on each end of each axis. May be I'll change this later if I get a touch probe.

I'm trying to wire everything with the same cables the table had with servos, those amphenol connectors are really nice.

I'll post the progess as it develops.

Here are some pictures of what I have so far:

[louieatienza]

HolyCow! I really need to know where these junkyards are! Great find!

[jamesbond007]

This is the cabinet getting ready to be wired, I will use the same amphenol connectors the gantry came with...

[jamesbond007]

Hi, I've been busy doing great progress to my build, but I have a problem, I don't know what is it but its like I'm losing steps, the X axis (the main gantry movement) gets kind of stuck for a second when doing simoultaneous movement with the Y axis, so the program (mach3) thinks that the position is right while its a couple fractions of an inch wrong.

I've been reading a lot about this, at first I tought it was the speed set wrong, I checked twice all the mechanics, ballscrew etc and looks fine, I exchanged X and Y axis, and know the Y is the one that fails, so I think its electonics or software related.

I have a Gecko G540 with keling 425oz-in stepper rated at 1.4 amps (bipolar series) the smaller resistor the kit came with was labeled 1.5K but reads 1.78K, my PSU is a 25V 250 W switching (looks like a PSU for a high end video card or CPU). Some other users say that you can lose steps because of bad grounding, or if the resistors are larger than the ones you need.

Here are some pictures of the problem as well as the machine almost finished.

Thanks for the advice.

[PaulRowntree]

Hi, I've been busy doing great progress to my build, but I have a problem, I don't know what is it but its like I'm losing steps, the X axis (the main gantry movement) gets kind of stuck for a second when doing simoultaneous movement with the Y axis, so the program (mach3) thinks that the position is right while its a couple fractions of an inch wrong.

I've been reading a lot about this, at first I tought it was the speed set wrong, I checked twice all the mechanics, ballscrew etc and looks fine, I exchanged X and Y axis, and know the Y is the one that fails, so I think its electonics or software related.

I have a Gecko G540 with keling 425oz-in stepper rated at 1.4 amps (bipolar series) the smaller resistor the kit came with was labeled 1.5K but reads 1.78K, my PSU is a 25V 250 W switching (looks like a PSU for a high end video card or CPU). Some other users say that you can lose steps because of bad grounding, or if the resistors are larger than the ones you need.

Here are some pictures of the problem as well as the machine almost finished.

Thanks for the advice.

The three problem spots all seem to occur during rapid moves, never while cutting. If this is true, perhaps the rapid motion speeds and accelerations are set too high, at least on that axis?

[jamesbond007]

Yes, they all happen whe doing the rapid moves, I haven't found any guide yet that says how to really adjust speed, I did some autoconfig of the axis from mach3 and it came out with a speed of 2500 mm/min which are 98.4 in/min, Is that too fast? The gantry brochure from TechnoIsel says it can go up to 400 in/min, of course not with steppers, acceleration is set to 49.9 mm/sec, step-pulse is set to 2 microsec. and dir-pulse I use 1 (Mach3 automatic config was 6 on step-pulse and 6 on dir-pulse)
I've tried lower speeds and have same problem, I did't try very slow speeds yet tough.

Thanks for the advice

[PaulRowntree]

Yes, they all happen whe doing the rapid moves, I haven't found any guide yet that says how to really adjust speed, I did some autoconfig of the axis from mach3 and it came out with a speed of 2500 mm/min which are 98.4 in/min, Is that too fast? The gantry brochure from TechnoIsel says it can go up to 400 in/min, of course not with steppers, acceleration is set to 49.9 mm/sec, step-pulse is set to 2 microsec. and dir-pulse I use 1 (Mach3 automatic config was 6 on step-pulse and 6 on dir-pulse)
I've tried lower speeds and have same problem, I did't try very slow speeds yet tough.

Thanks for the advice

"Too fast" depends on the machine, so it is empirical. The acceleration seems low to me (2"/s2) so your machine may never get to the rapids limit you have established in Mach3; does the speed shown on M3's front panel during rapids correspond to the speed limit you have imposed? If not, then you will only see the effect of changing speeds when you reduce it below what M3 is actually using on a particular move. Does this make sense?

The best way to find out is to keep cutting the speeds down by a factor of 2 or 4 and see if it improves the situation. Since you know that the problem is in rapids, you don't really have to be cutting to do these tests. That said, if the problem is only present when the spindle is running, it is probably a noise issue.

You are running your G540 on 25V I think; increasing to 32xsqrt(coil inductance) as per Gecko suggestions may give your steppers more oompf (I think the G540 is limited to 50 V max). I run 48 V.

I would increase the pulse lengths too, just to be more tolerant of parallel port weirdness.
Cheers!