Dirks 4x8 CNC Router Parts Build

[dirktheeng]

Got the spindle moving today!

[dirktheeng]

Got the spindle moving in Mach 3... got a couple bugs to work out yet.

[dirktheeng]

Got the dust shoe installed today... did a mini review of it.

[dirktheeng]

small discussion about how I plan to get the dust from the source to the collector.

[dirktheeng]

Made a prototype tram system for the hose... worked ok, not awesome.

[dirktheeng]

All,

Based on some suggestions of friends at buildlog.net - CNC Laser Buildlogs, I made some changes and got the hose system working much better... Check it out:

[dirktheeng]

The DC will help collapse the uber flexible hose so that the loops don't form and it has a natural spring return.

[dirktheeng]

All,

got the new limit switches installed from Ahren at CNC Router parts. Here's some initial testing:

[dirktheeng]

All,

Started up Mach 3, had another problem with my limit switches and also got my first real cut file tested.

[dirktheeng]

So this ends the rapid video posting. I started this buildlog over at buildlog.net - CNC Laser Buildlogs because my first machine was a CNC laser built last year. I met a ton of really good people there and they really helped me push the envelope of what I ever thought I could do. I decided to move over here because this is a much larger cnc website and there are a LOT more people that do cnc routing/milling here. Bart Dring over at buildlog.net - CNC Laser Buildlogs has built a community around cnc lasers and 3d plastic extrusion printers. He's a brilliant guy if you have the chance to interact with him. I think his bigest ever contribution to DIY cnc was the makerslide which enables us to do so many cnc designs at really low cost.

Anyhow, this video catches you up to where I am now. All the past posts was stuff I had done over the last 4-5 weeks. I am close to getting running now, but I still have a lot to go and a lot to learn. I really value this community and rely on all of you to share and learn together.

Without further delay, here's a short discussion of potential problems I am having now that I could use a hand with.



As a matter of record, you can find my cnc laser buildlog at:
Constructing Janus

You can also find the buildlog of this cnc machine started at:
Dirks 4x8 CNC Router Build at buildlog.net

Also, you can find all of these videos and more at my youtube site:
dirktheeng

[dirktheeng]

All,

As you have seen I have been having some trouble with the inductive limit switches. I was hoping to hear from some of you about your experience with them. I have a feeling that the switches I have are not going to cut it. I'm waiting to make my final decision until I get my dial caliper with magnetic base to do some repeatability studies. However, I have a feeling that these switches will drift with temperature. They may be fine for limit aplications, but homing with them is a different story. I think I need a contact switch to home accurately. What do you think?

If you are in favor of a different solution, could you please give some details and make an argument as to why I should do things that way?

[dirktheeng]

All,

I got my first cuts this week!

Check it out!

[dirktheeng]

All,

I did a bunch of measurements to characterize and attempt to correct for any error in the z axis alignment. check it out:



Here are the results:

[lx007]

All,

...However, I have a feeling that these switches will drift with temperature. They may be fine for limit aplications, but homing with them is a different story. I think I need a contact switch to home accurately. What do you think?

If you are in favor of a different solution, could you please give some details and make an argument as to why I should do things that way?

I built a CNC Router using mostly CNC Router Parts and I installed Sunx (now Panasonic) GL-N12F inductive proximity sensors for homing. I get very accurate and repeatable homing at any temp. (They're rated for 14 - 131 F). (These have been replaced with the GX-F models, which claim to have the "Industry No. 1 in stable sensing and temperature characteristics").

I bought a box of 10 on eBay for about $50. I found the new GX-F8A for $26 at SUNX Sensors USA, Largest Stocking SUNX Distributor in the USA, you may find them cheaper.

I have found a lot of really good Industrial quality Inductive Proximity Sensors (NPN/NO) that would likely work even better than these, just sharing my results.

[LeeWay]

Excellent thread. I really enjoyed the video blog.
Kept wanting to post some ideas, but you had already remedied most of the things by the next video.

One solution that I thought of for your hose was shower curtain hangers with rollers on them. They are cheap and work really well in such application.




McMaster-Carr

No need now though.
Very nice machine. I am doing a 5' by 5' plasma using CNC Router parts Nema 23 version.
My first router project had a similar skate bearing truck design, but I used rolled ball screws on it.
Those bearings held up very well over the years, so I knew I wanted what Ahren was selling.
I was not disappointed at all.

I chose to go with a different type of limit switch.
Roller Arm Type AC Limit Switch for CNC Mill Laser Plasma Me 8108 Brand New | eBay




I'm not quite as far along as you are yet. Just need to find a bit more of the precious stuff. If I could buy time, I'd order it by the drum.

I hope you get it sorted out.
I recently built a Mendell Prusa and it came with some nice optical sensors.
They seemed to be very accurate, but a little on the dainty side for a cutting machine. A chunk of saw dust might set it off.
I'll keep watching and I'll check out your other threads. Keep it up. You'll get there.

[dirktheeng]

All,

I decided to go to a different type of limit switch. Since everything is grounded on the machine, I can use a direct electrical contact switch. That is, a conductive element touching another conductive element to close an electrical path. The same thing we do when we use a touch pad with a bit to zero the z height.

I did a lot of searching to find a reasonably cheap solution that would be highly accurate and accomadate limited over travel. I came up with a Pogo Pin which is used in the circuit board testing world. It should be very accurate, relatively immune to dust (though my dust collection is good as long as my bit fits in the dust shoe). The only possible concern I have is a piece of dust coming between the contact plate and the pin, but I am using a rounded point for the contact and should be ok. They have pins with sharp points as well and I may use those if I find I have a problem here. BTW, the dust you see on the machine now is the result of poor collection when I tried the 2.75" surfacing bit. I didn't get that all cleaned up apparently. When I use any other bit, the machine doesn't even produce the smell of wood.

[dirktheeng]

All,

I got one of the limit switches put together today and tested it out. It works good and is pretty hearty. I purposely crashed the mill to see if it would break and it didn't. I have to to a repeatability study yet, but I think it will do very very well. I turned the kernel speed up to get Mach 3 to respond more quickly and that really helped get the axis stopped when I crashed it. I think I am going to switch the bracket around so that it zero's when the single bumper is compressed a little to protect the homing switch a bit more.

Anyhow, here's the test:

[dirktheeng]

All,

Today my dad and I finished up the homing/limit switches. I have to say that I would challenge anybody to come up with a more precise homing system for less money than I spent. I have about $15 of materials into the system and ended up with a very good homing system (not counting the extra that I had in the system because I couldn't buy a small enough quantity). The only thing to be mindful of is heavy dusting. I don't have that problem if my dust shoe and dust collector are working correctly. I don't even get the smell of wood dust (which is a good way to test for very small, invisible wood dust) when I run the system with mdf with a smallish bit.

Anyhow, I did a repeatability study and basically my dial indicator is not accurate enough to really assess the homing error. That means that it is in the sub 0.00025" region.

Check it out:



I also got the gantry squared up and it has about the same precision (which is expected), but my setup for measuring that is a bit shaky. I need a way to lock the spindle a bit better. Anybody have any suggestions?

[dirktheeng]

All,

Today my dad and I worked on making a few shop jigs for us to use. We made a bowl gouge sharpening jig and a bunch of push sticks for our saws and such. I tried my hand at double sided machining and it didn't go so well. I used pins for alignment and somehow I got my pin gcode mixed in with my cutout gcode and I ended up machining stainless steel... I was actually surprised that the bit didn't break up as soon as it touched the steel... it actually cut about half an inch of the 3/8" pin away before the bit broke.

I think I should make some plastic locating pins instead of steel ones or use wood dowels so I reduce the risk of breaking bits this way again. I try so hard to be vigilant about how I set stuff up, but I made a lot of mistakes today... some were rookie stuff others were just stupid oversites. I'm going to make a check list to go through every time I start to mill something.

[cjsamples]

Just wanted to say I am enjoying the videos. I have had my share of unexpected issues as well. I thought for sure I could stop a code and adjust something, only to find out that didn't work so well........
Glad it was only some plywood.

Keep up the posting as I am wanting another machine, not sure what yet, but just another one.

Chris