Sounds good to me. I think I'll go ahead and drill one of them out. Maybe.

Hey guys. I wanted to say I've had great success tonight with my micRo. I got my new (refurbed dell - my other 2 old comps bit the dust) up and running -- although Axis crashes the computer...I'm going to look into it, but the mini interface seems to be more what I want anyway. Once I got that all running I tested my wiring with an ATX power supply I hot wired (green wire to ground)...I wouldn't really recommend that as a everyday thing, but in a bind it works and from my old computers dieing I had some laying around.

So, using that to test, I found my Z and Y axes were working great. My X were not wired proper. Some intimate time with my meter and I figured out where I'd gone wrong (wiring was just botched at the RJ45 jacks). So, rewiring those got me squared away and I now have all 3 axis operating via the computer. Although, I recommend when you put it together...try to have it near zeroed. I had the dang X-axis all the way back and through the mini interface I could could 0.625" before having to re-zero and back it forward more--over 10" that takes a while. I'm sure there is a better way to go about that but oh well.

I've got a dry erase pen and a white board I'm going to use for testing, but I want to get some g-code written that will work with the underpowered 12v supply (it doesn't give the x motors enough power to go faster than 9, and then Z and Y are good to about 16. But with the 24v supply I am sure it won't have any problem moving along.

This has definitely been worth the wait. I have my Harbor Freight spindle decked out with some ABEC 9's and I'll see about getting some bits ordered this weekend, so they'll be ready when the holder gets in.

Too bad my brother has a basketball game in the morning or I'd pull an all nighter down there working on it.

Here's a quick cell phone pic of we're I've been living the last few days. Notice the netbook, server, emc2 box, micro, radio, switch, tv, dvd, beer fridge aaaaaand the hanging spindled tucked up out of the way. No wonder I'm home alone on a Friday night =/

Click to view attachment

You're not alone, you've got a bot. Using your wall for target practice?

Please don't drill the thread out of a gantry blok...

man ma chine ma chine ma chine ma chine machine machine maaachine !!!!!!!!!!!!!!!!!

Tour de France

So the only thing I am confused buy are these jobbies:



I can not see them anywhere on the Micro pictures, and am not entirely sure what they are for.

I'm sure BC will chime in, but my understanding was that those were part of the initial China order along with the screws and ways. The design change since then made those obsolete. The UMHW blocks in the kit already support the ways, so those are basically extras. Somewhere back in the initial drawings I think you can find them (maybe in the first 10 pages of this thread). If I remember the micRo started out with design based on the Rogr, before it was optimized out.

Just to make everyone else jealous, I took a picture.



Just waiting for a couple of parts and my robot will be ALIVE!

Yeah, those shaft supports are definitely part of the old design... We might take them out of the boxes, just so no one else is left wondering though.....

Anyways, some advances..... Last week had some time to sit down with the EMC integrator manual, and now have the RoGR working off of two dedicated sets of direction/step pins out of the parallel port for the X axis. I haven't taken the time to actually measure the performance, but, it seems to run a lot smoother along that axis now (and at higher speeds). So, the problem is the same for enabling this on the micRo, and I'll be posting directions to allow this shortly... The two catches are: once you make the changes, editing your configuration with stepconf will nix it... And that Axis, for some as-of-yet unknown reason loads into "joint mode", so needs to be changed to "world mode" before any operations are performed (world mode sets the bot in the XYZ coordinate system we all know and love)...

Ok, here are updated .ini and .hal files....

Again, opening these with stepconf will erase important data. Also, due to an unknown cause, Axis loads automatically into "joint mode". You need to be in "world mode" before any operations are performed. Change this by going to the View menu, and clicking "world mode".

Here is the pin mapping (axis "U" is the clone for the X axis, that is, it should go to the second X-axis driver and motor):

X step (pin 2)
X dir (pin 3)
Y step (pin 4)
Y dir (pin 5)
Z step (pin 6)
Z dir (pin 7)
U step (pin 8)
U dir (pin 9)

Production problems resolved, shipping begins again!

Awesome.

Can we get an update on the spindle order if you have heard anything more?

Hi Narwhal, Yes I just got an email from them. The spindles will ship in ~10 days.

While your giving out good news about updates, any news on the base?

We should have those ready within 14 days if the supplier delivers the pieces on time. The big hang-up now is the new drivers which have a longer lead time, as they are produced for us in Taiwan. I'll inform as to when as soon as I get a reply...

Sweetness. I know my order is a ways down on the list, but it is exciting to keep up on the process, especially with all of the wonderful thoughts of what I can do with this running through my head. You guys are outstanding and it is great to have regular updates on what is happening, unlike many places where the best a person can do is just kinda sit and wait and at most get an answer of "We don't have it yet, please wait" without any valuable information. You guys have a customer service system that should be the standard for pretty much every company.

Oh, one more question. Did you get any information concerning the shank size and specs for the bits to be used with the upgraded spindle?

If I ordered the upgrade package, is it possible to have the micro sent now as if I hadn't? I ask because while I do want the upgraded drivers and other hardware components, I'd like to start messing with the steppers and everything else.

I also have a use for the easy drivers after I replace them with the upgrades.

I'm 100% willing to pay whatever extra costs this would incur (double shipping, additional $$ towards the drivers, etc.). Thanks.

Just checked my email and surprise, there was an email answering my questions about various chuck options. Currently they only produce the 4mm chuck. That makes 1/8" the standard Imperial bit size, though 4mm is bigger so if you can find them, buy them. Naturally, any bit size smaller than 4mm can be used, as it is a drill chuck, not a collet chuck.



Sure, as long as all of the parts are ready to ship...we are still catching up, but we are making good progress now. We pulled an all-nighter last night which free'd up 20 robots to ship. Big thanks to Robin for remaking all of those gantry bloks I wasted; a week's worth of big time effort on his part; and he didn't even tease me about it! He may not be human...

So basically a "Dremel" bit is going to be our best bet for the Micro to begin with. That is kinda awesome since I have a buttload of those lying around the house and they are sold in abundance in my area. Now to find the best ones to use, lol.

I think something more like this: http://www.amazon.com/4Fl-SE-Carbide-End-Mill/dp/B000I6I5NK would be in order.

That's the ticket.

Ok. so pardon the stupid questions since I am new to this, but I want to make sure I understand correctly. You stated that smaller bits like the one below would work as well? I am thinking this would be great for working with fine detail, but are there any concerns that it would break under the stresses of the Micro?

http://www.amazon.com/16-4Fl-Carbide-End-M..._bxgy_hi_text_c

These guys also have their own site, not just Amazon, with a lot more stuff. They are in Brooklyn NY right next to a subway stop (the L for those who care) and when I am in the city they are real convienient for getting stuff right away and real nice people to deal with.

If anyone's having trouble sourcing these screws I may have a suggestion. They look like the same kind of screws that are used in car heater/fan/aircon boxes to hold the plastic case together. Nearly every car made since at least 1990 would have these type of screws in it somewhere. So if you can't find a place to buy any, take a trip to you local auto wreckers and have a look there.

Some other sources of 1/8" shank bits (in the US)....

http://www.miniatureendmill.com

http://www.microcutusa.com/fractional.php

http://www.harveytool.com/products/index.php?category_id=2

Is there a wiki for the micRo already? Should we have one to house all of the nuggets of wisdom from the forum?

Check out Rutland Tool they sell everything! (eBay)

Although the web site kinda sucks, Enco is another source for all kinds of machining stuff, including a bunch of miniature end mills.

Shars (AKA Discount Machine on eBay is also good for cheap/decent quality stuff, their precision toolmaker's vises are favorites.

A carbide bit like that is incredibly hard but very brittle too. I would guess you could probably snap it off with the micRo, but you would really have to be trying. As long as your speeds and feeds were appropriate that bit would be a good general purpose bit, with the ability to take some relatively large cuts and yet still leave a nice finish.

Do they ship in ~10 days from China, or will you have them in stock and be able to ship them out to us then?

My experience with any purchase is to assume the longer time frame. That way you are not left wanting unless something disastrous happens.

4mm end mills you say? Excellent, now I can start on some CAM for my micRo (while I wait that is).

It seems (from my breif checking) that metric end mills cost more then their Imperial brethren in the US as far as I can tell. Seems that 1/8" end mill shanks may be the ideal from a cost perspective.

awesome idea!

None of your questions have been "stupid" at all. If the spindle is up-to-par (remember, we are all testing it together!); and truly runs in ABEC 7's, which can hold about .002" tolerance; then we can start to see some benefit. But to answer your question exactly...you could run any bit size smaller than 4mm, so long as you can set the speed and feed appropriate to your material and rate of removal. If the bit breaks, you went too fast.


Good question!

As Brainchild says, if you snap the bit, you moved it too fast.

When using any kind of rotary tooling that is moving sideways, you have to spin it fast enough to cut the material as it comes into contact with the cutting edge(s) but slow enough that it doesn't simply bend and or snap that little cutter. Yes, micRo has lots of power, but you really won't be using that if your cutting tool is a tiny carbide mill. You need the power so you can move the cutter to where it is cutting fast, not while cutting.

Of course, lots of power means that you can use a big wood cutting bit and lots more speed without issues, too, or a cutting blade on vinyl, etc. But for milling steel, no, you don't need it.

I am sure it will take some experimentation on my part when I get it and have it up and running. I will be using it mostly with pine boards, or plywood, at least to begin with,

I appreciate all of the support, this is my first time dealing with anything CNC.

BC, do you have a suggestion as to lube for the ways and the lead screws? And is there a different lube for the X axis rails and bearings then the y and z?

I got my tax refund! Can I order the upgrades even though my order already shipped?

I'd use a "pocket oiler" with PTFE (non filamentous) for all axes:

http://www.amessupply.com/products1.cfm?ai...amp;fid=1305011

In the future, a pocket oiler will come with each micRo, with custom formulation and logo printing.



Sure, but your order shipped so we'll need to work it out....just write to support@lumenlab.com and Kellie will take care of you.

Sweet, just picked a couple up.

Also while I appreciate that my question about spindle shipping dates was a good one, I was wondering if you had an idea of the answer :-p

heres one of those uninformed questions
how / where is the pinout or diagram on how to wire the controllers to the parallel port ? or is adjusted in emc2 to tell it what pin is controlling what servo ?

Yep, it is configures in emc2.

Read up on Stepconf :
http://www.distinctperspectives.com/emc2installsetup
http://www.linuxcnc.org/docs/2.2/html/config_stepconf.html

Robin has posted the micRoVdot1B and micRoVdot1C config files.

In B it goes as follows:
Pin
2 X Step
3 X Dir
4 Y Step
5 Y Dir
6 Z Step
7 Z Dir

In C it is the same as B, except you split the doubles you have connected in pins 2 and 3, to a cloned X axis, U.

8 U Step
9 U Dir

You can check the ini config file. On your EMC box go to the console and try this:

less microVdot1B.ini | grep pin

That will just display the pertinent lines to pins.

ok 1 more question i see you can also configure limit switches would any normaly open leaf switch work for this ? i remeber them being mentioned before but not in detail

Brings up a noob question I had for micRo or RoGR. There are switch options for homing on closure or backing off after closure, etc. What is the BEST type of switch for these applications (and the best option for homing on approach or backing off, etc.) and can one actually depend up them for a very close tolerance (a few thousandths of an inch) home or start position?

When you guys are cutting parts do you fasten a jig in the RoGR and from then on just clamp in blocks and run a program with no manual touching off or homing - that is, does RoGR always know where he is even if you take his juice away?

I just got a call from Ames Supply saying that they were out of that product and were unsure as to when they were getting more. Will any PTFE lube work?

On some other 3-axis motion instrumentation, we always used IR sensors (photointerrupters) for high precision repeatability, plus they are no-contact devices. All you would need is a small tab attached to the axis in question that breaks the beam in the home position.

Something like: Sharp GP1A50HR

I believe they will ship from China in about 7 days, then a few days to get here.




We don't use them due to the lowered precision. We have scribes indexed in the couplings. Before every run, the axes are indexed to the scribes. When I say "run", I mean everytime the machine is powered-up, not between every piece.



http://www.amazon.com/Super-Lube-Synthetic...4/dp/B000UKUHXK

Regarding limit switches:
Limit switches mainly function as a failsafe against bad coding and/or setup and out-of-control drivers. In my experience they are not necessary as long as the machine is sent to a known home position (more on that in a second), and the machine extents are known. EMC will not automatically run a program that violates the machine extents... Stepper drivers (that I know about) in particular only clone out the step pulses, and can't generate their own, so the chances of having a run-away driver are next to 0... Mostly, limit switches are there to prevent operator error, so, again, if properly set up there is little chance of ever needing them. However, I'd recommend that you stick around your machine while its running, other errors are more common (and usually because of improper g-code, that is, not accounting for jigs/fixture clearances, Z-axis clearance paths, wrong plunge depths, wrong feed rates, and the occassional broken bit..) and are usually obvious if you're in ear-shot, but would almost never (in my experience) have been prevented by a limit switch.

Regarding Homing, home, and home switches:
Most mechanical home switches have quite a bit of tolerance (on the order of .010" +/- .005") that would be unacceptable in most cases. Not to mention that they need to be kept clean or sealed to IP64+ standards (seems obvious, I know..). Capacitive proximity sensors are more resilient (but are sensitive to metallic chip), and also have positional accuracies that are slightly worse than mechanical switches. Optical switches, though I don't have much experience with them, are the most sensitive to chip/debris, but apparently can maintain pretty good tolerances when properly set up and maintained.... I have to discourage the use of magnetic switches altogether (though they're popular amongst inexpensive CNC machines), because they have poor positional tolerance, and are sensitive to metallic debris.

But all of that is besides the point. What we've found to be the best method is to use indexing marks on the couplings. Things will shift and move, and your machine will lose steps, our servo drivers will occassionally cause the motors to jump as they start, but as long as the coupling/screw doesn't slip, you'll have a reliable index mark that will maintain your home position. In the case of the RoGR, we can maintain home within under ~.00045", making it the most accurate part of the machine aside from the encoders.

So best practices in this case means:
Add a few lines to the end of each program you run to return your machine to 0,0,0.
If there's an error, stop the machine via the EMC/Axis E-stop button, so that you can maintain your current position. Instead of killing power to your drivers.
Verify that your machine is in the 0,0,0 position when you start your machine.
If it is slightly out of position, load EMC, move to establish the correct 0,0,0, and then restart emc to set this as 0,0,0 inside of EMC...

Future improvements:
Encoding directly off of the leadscrew.
New servo drivers that will share the encoder with EMC, so that even when power to the drivers is off, position can be known and maintained.
Including a physical E-stop button that will disable the amps/drivers through EMC (the new stepper drivers allow for enable/disable commands that can be linked to the EMC e-stop BTW).