Right now we're just sharing pins for both x-axis drivers. This is obviously a low-tech approach, and EMC does allow you to configure multiple motors on a single axis (I believe that axis U is set up as the alternate X axis), but it works well enough as is.

Anytime. Looking forward to some baby pics.

Whoa! That's too close to my kid's grandparents! Say . . . Asheville is appropriately named for this simulation.

How did you wire the servo motor to the gecko.

Brown goes to: ??
White goes to: ??
Black goes to: ??

Ok, Brown can be left disconnected (it's motor ground), or can be hooked back into safety ground or neutral at your power supply.

White is Clock-wise, and should be connected to the ARM + terminal.

Black is Counter-Clock-wise and should be connected to the ARM - terminal.


And here's how to connect the encoders:

+5v to ENC + terminal
Channel A to Phase A terminal
Gnd to ENC - terminal
Channel B to Phase B terminal

about wiring:
For the motor power leads, they should at the very least be twisted together, and not allowed to run parallel with the encoder wires for any appreciable distance.
For the encoder wires, it is advisable to use sheilded cable, if possible, and either have it as ground (the sheild) and connect it to ground on both ends- or at the very least, connect it to signal ground (ENC -) on the gecko and use a seperate wire to connect the grounds. But Do Not connect it to the motor ground.

Thanks for the quick answer.

Hey Robin, thanks! I also had the same question on wiring the encoders. As I understand it, the gecko will provide +5v (up to 50mA) for the encoder to work? (at one time I was thinking I needed to provide an external 5v power source)

Also, while on the subject of wiring. Do you have a pin-out for the wiring on the DB25 cable when using EMC2? (haven't read into EMC2 enough yet to answer this question)

Cheers!

can you post the HAL file from your EMC setup ? (or list the settings) ?

Hopfully no need to reinvent the wheel

Thanks
Jan Erik

In EMC2 you can decide what the functions of the pins are going to be.
It might be that Lumenlab have chosen differently than the EMC suggestion

by default it is:
1 E stop out
2 X step
3 X direction
4 Y step
5 Y direction
6 Z step
7 Z directopn
8 A step
9 A direction

Hi Charlie, You need to supply 5vdc. The gecko only has enough juice for its error signal, not to drive the encoders. Note the 5vdc is connected to "common" on the amp (why they called it 'com' is beyond comprehension). The negative -5vdc is connected to pin 18 on the breakout (ground pin).

In EMC2, you can assign the pins as you like. We usually just stack up the drives on 1-2, 3-4 etc.

Ok, I've attached the applicable configuration files for the RoGR. These are preliminary, and will need to be changed. However, they are what we're using right now to run our RoGR.

I still suggest using the EMC stepconf wizard to set yours up, it is very easy.

1) name your configuration

2) choose your generic axis configuration (XYZ)

3) set your machine units (Note: everything we're using is configured to inches. For millimeters multiply appropriately)

4) On the next screen, configure your pins. At the moment, you will probably only be using Step and Direction for X, Y, and Z. how these are assigned is unimportant, as long as they correspond with your actual pin-outs. Leave all unused pins as "unused".

5) Configure the individual axes.
-With our current pairing of servo and encoder, it is 1200 steps per revolution.
-Microstepping = 1
-pulley/gear ratio should remain at 1 : 1
-leadscrew pitch = 6 revolutions / inch (4.233 mm / revolution)
-when setting table travel, make certain to leave a little buffer on the edges. Home can be any number within range of the table, but, is easiest if set to 0. Make home slightly away from the edges of maximum travel, and set the minimum distance less than zero (example, home location : 0.0; table travel : -0.25 to 26.0)
-It's important to test the axes. remember that (for reasons as of yet unknown) using the Jog buttons do not accurately reproduce the velocity (it will be about .6 the velocity speed), so it's important to perform actual runs to establish an acceptable maximum velocity and acceleration.

That's the jist of it.....

Ok, one important thing to mention about wiring, that seems to be somehow too obvious to remember, and yet too obscure for the new-comer to know......

Pin 18 on your parallel port break-out (sometimes it's actually labeled as "ground") needs to be tied in with the ground of your gecko driver power-supplies. Preferably with the ground of your 5vdc power supply (which, incidentally, should be tied to the ground of your motor power supply). Otherwise signal from your parallel port will not be able to flow into the gecko drivers.

I believe you've got me nailed there...

Thanks RWT and BC.

So. I need an external 5v power supply. The '+' side to the 'common' (pin 12) on the gecko (which is listed in the manual as '+5 VDC', but on the unit itself as 'common'...??) The '-' side goes to pin 18 on the breakout. Check. I was planning on using the 5v rail from a PC power supply for the encoder, so I guess that means I should be using the 470ohm resistor between pin 6&7 on the gecko?

I did have a feeling EMC2 would allow pin assignments, but didn't know if there would be a "LL standard" that you would be using with the preconfigured PC's. I'll take a look at the file you posted, and hopefully i'll get to do some soldering this weekend.... my mission is just to simply be able to turn the servo from EMC2, etc... and confirm wiring without burning up the damn drive! I've got a 1amp 24v power supply that might just do the trick for this mission...

I'm gonna have to do a schematic or something. Isn't the gecko driver power-supplies and the motor power supplies the same thing? Sounds like pin '1' on the gecko will join the 5v power supply on pin 18 of the DB25 connector. (should this also go back to the brown wire on the motor?)

god, i'm such a noob... i've got tons of learning to do and no time to read! ahh....

Nice work BTW. Love seeing the video's and hearing about all the cool stuff coming down the pipe... what to make first?? Maybe just cutting out some simple shapes for some speaker cabinets, or shelving... hey, gotta start somewhere!

Cheers!

No no, don't connect pin 18 to +V! Pin 18 is the ground. One of those "things" you'd never know is that the "com' terminal on the g320 is actually the +5V input. The negative side of the 5V goes to pin 18(ground). You must also tie the NEG of the amp's main V supply (-24v from your power supply) to pin 18. Make sense?

We've never used the resistor...not sure what that's about.

Ok, 24vdc at 1 amp (aka 24 watts) will be woefully under-powered for one of these servos- if you are actually making it perform work. I suggest something on the order of ~40vdc (though, more is better up to a point), and ~ 3 amps (per motor/driver)... Those aren't hard numbers, but they'll work well.

There will be a standard configuration, or at least a set of standard configurations down the line... However, for all intents and purposes- it makes little to no difference how the parallel port is connected to the drivers.

regarding wiring:
+5V (from your 5-volt power supply) should be connected to the "common" terminal of the gecko... again, why they decided to call it "common" is a mystery (when "common" is commonly used to refer to reference-ground).

-V (from your 5-volt power supply) should be connected to both -V on the motor power supply and pin 18 (the ground pin) of your parallel port.

Ok, this will probably be a little confusing.......

A FRANK TALK ABOUT GROUNDING

The brown wire (ground wire) for the servo only provides one real function: safety.
It could be argued that it helps block some EMI (electro-magnetic interference) when properly connected........ But I don't really buy that. The motor is, of course, a bunch of magnetic fields- which are not likely to be absorbed by passive shielding of this sort (which works ok for a different type of interference).
So....
Properly grounding the servos could maybe reduce some of the RFI (radio frequency interference)... And I'm pretty un-convinced of this.... But really, what the brown wire does - when connected to the safety ground of your power supply (functionally it doesn't matter which PSU..... but I'll get to this in a little bit) - is to provide a low-impedence path back to the neutral tap of your breaker-box... Meaning that if there were some catastrophic failure and power was being dumped into the frame of the motor (or the RoGR), YOU do not have to be the lowest-impedence conductor around- and hopefully the current draw will trip the breaker as well...
BUT, the chances of that kind of failure in the motor are pretty low (next to nothing really) and you could connect the chassis with the safety ground and acheive the same results. Let me say that you should DEFINITELY connect the chassis/frame to a safety ground (or some combination of motors), since the the chances of short circuiting else-where in the machine are pretty reasonable (power-tools + coolants + dust + error) - and as beautiful as it is, the RoGR should not be the last thing you see.

There is a nit-pickey argument about grounding........ Basically, that noise from things such as motors and the like can interfere with the operation of things such as encoders. This interference basically creeps up and down their shared ground/neutral (in the USA there's almost no difference between the two in a house's wiring). If this is truly a problem, a de-coupling capacitor may be used...... But that's another story, and probably pointless.

GadgetSmith must be talking about this from the Gecko g340 manual (I'm guessing it's the same for the g320)...

A short update of my project

Click to view attachment
Painted and the x-ways mounted with the linear blocks in place.

Click to view attachment
The X-Y plates testmounted

More info

Click to view attachment
Screws & Nuts next, progress on those are in Lumenlabs hands


Click to view attachment

Backplate of the Z-axis facemilled on both sides.

Haha. Didn't make that clear did I! Yes, '-' side of 5v power supply to pin 18. (this is why I like schematics/pictures to explain things rather than my incoherant writing style)

You mention the NEG of the amp power supply being -24v, which brings up a question I have. Can I use an old PC ATX power supply and connect the +12v (20a) rail to the gecko +18 to 80v input, and the -12v (1.0a) rail to the gecko 'power ground' to get a 24v DC power supply, good for 1a? Is it possible to connect two PC ATX power supplies in series by connecting the ground from PS1 (12v rail ground) to the +12v rail of PS2? (I've read conflicting reports, but most seem to indicate that this would 'blow up' the PS)



Yea, for normal operation I was considering the above option (if possible), but for testing purposes my 24v 1.1a power supply from my scanner should more than suffice to simply get the servo to turn on the bench, no?

Realistically, i'm thinking of just getting a 48v 10a (or so) power supply off ebay (or such), but that will wait, for now i'll just use what I have lying around.


Yea, that was the part in the manual I was referencing. I bought a 470ohm resistor while at Radio Shack in th event i'd need it, just not sure of it's purpose... I'll have a look on the gecko yahoo group to figure it out.

I have a question about the Screw Pass Through Hole in the Frame Column. It seems awful close to the Rod Support Hole, there is like 1/4" maybe less material between the holes.

This doesnt seem right to me.. is there any reason why they holes couldnt be spaced a little farther apart?
The screw is 3/4" so im leaning towards a 7/8 hole for nice clearance, but that would leave even less material between the holes.

How much clearance do you have between the top Rod support hole and the Screw passthroughs?

Also how much play do your holes have, All my bolts are 1/2" and my holes are 1/2" but there is no play, its a tight fit. Do i want to go a little larger in the holes, say 9/16" for play? (adjustment?)

One last thing.. can we see a picture of how the motors are mounted to the frame? Rturner mentioned a common shared hole with the rod support, is there a picture of the machine from the motor side? and one of the motor support bracket?

Thanks guys!

Interesting to see the rectangular gantry bloks....the final design has a piece of broke sheet over the trailing 1/2; this creates the rigid "box" gantry. So far, I can not stop the machine to finish this.


I only said that because you said that was your power supply. Sure, the motor will turn on the 'bench'....as per long-ago-video demonstrated; the ol'skoool servos will turn on a AAA battery.


Oh, the pain.

You could prolly rotate the shaft support.
Click to view attachment

Hmmm... I guess you might end up drilling another hole for the motor that would be just as close anyway?

Really its quite simple a question.. you guys have been answering everyone elses questions.. why not mine?

Yeah, other than moving the lead screw up 3/8" I dont think that should be a problem but not sure.. i just wanted to make sure these things are right before i drill.

Ive been holding off on drilling for like 3 days now so i can get an answer.. guess i'll be waiting some more...

Jan, maybe you can answer me, how far apart are your Top Rod support hole and your Lead screw pass through hole? and how large did you make the screw pass through hole?

And should i have play in my bolt holes?

Simple questions no?

Thanks to anyone that can answer my questions!

Well i just went through this entire thread again and found a picture of the x motor mount.

Click to view attachment

I looks like the screw hole and mount hole are close together but not crazy, i also realized that the lead screw is turned down at the ends so that its smaller than 3/4" when it passes through the frame.

So say i stay to LL Spec, Robin said to make the hole 3/4 or larger, but really 3/4 should be enough right?

So just 1 question now.. should i drill my frame bolt holes larger than my bolts (1/2" bolts, 9/16" hole?) for adjustment/play? or is it not necessary?

Oh, on a side note.. I found a company that sells AD-DC power supplies. They sell a 48V and a 60V and put out 40A. The minimum purchase is 5 units at a price of $1200USD each.

Would something like this be good for the RoGR? If so would anyone be interested in a group buy?

The pain comment wasn't directed at you, it was just me joking about the pain of locating and drilling those holes. I was going to let Robin answer your question. He only has internet at the shop, so I guess he'll answer you tonight if he has a chance.


That's way more amps than you need, and about $1000 too expensive. I'm ordering power supplies now, I can get you 48v at 12a for $150, or 72v at 12a for $225.

Awesome thanks BC!

Hey phife, the bore-through for passing the screw, should be done for the inner diameter for the coupling you'll be using. We are currently using a 14mm coupling on the screw side. The idea is that you can fit a small bearing block onto the inside of the frame to help with thrusting and lateral stability. Admittedly, this doesn't leave tons of material there, but, then again, not a whole lot is needed.


FOR THE MOTOR X-AXIS MOTOR MOUNTS and SHAFT SUPPORTS:
1/2" holes leaves plenty of room for adjustment since the way supports use M8 bolts. But really, if you can maintain your precision, and verify the squareness/parallelness of your steel square tube, then you could theoretically have 8MM mounting holes (though I'd still recommend some adjustment, since the frame will invariably need some).

FOR THE FRAME:
We specify 1/2" bolts, but really you could use any larger bolt. I would absolutely recommend leaving some adjustment room, since your frame will almost definitely need adjustment. The rigidity of the frame will be made via the tightening of the bolts, not by the bolts resistance to shearing forces.



Attached are a few pictures of how the shaft support, lead-screw, bearing block, and motor mount go together....

I have fallen prey to Brain's unique sense of humor MANY times so don't feel bad. He's really a pussycat, sweetheart, etc. so don't take nuttin' personal-like. Unless he does mean it . . . then - WATCH OUT!!!

When can we expect the tramming tutorial and full plans for everything and the extra two axes and auto-feed kit and extra z-travel plans and . . .

Seriously, you guys are making fantastic progress from my point of view. But this is Head Hang Fire talking.

Those pics are like HD pROn - every blemish shows.



Gonna email my brother-in-law RIGHT NOW to see if he's ordered my gantry frame steel yet!!! <insert bit-chomping noise here>

Got the spindle chuck. Everything is great but I have doubts that the motor is balanced enough to be THE spindle motor....we are talking about holding better than .001" tolerance.

Yep, that end of the robot was re-designed about 10 times...I'll be sure to make the next one pretty, just for you.

Man that looks like a Unimat chuck. Where's it from?

Robin claims he's chattier w/drink but that red plastic cup does appear to be a prosthetic extension of his arm so I'm not so sure his claims are true.

China, certainly. Oh BTW Dan, might wanna watch vid #2, heheh...

http://www.lumenlab.com/forums/index.php?s...st&p=301888

PS, Thanks for the metal!

Oh yea....on the frame.

I noticed RW didn't mention that all of those 'critical bolts' (such as way supports) are supported in the tube by sections of steel tubing. This is to keep the framing rigid when bolted at high torque. In the near future, frames will be offered with welded studs, internal supports etc. (Another robot, sigh...)

np

Need more? Just PM me the rough dims.

PS: Dur-otch-koe not Dur-atch-koe.

Added a couple necessities to the RoGR inventory. Expect a lot more as my sample orders arrive over the next two weeks! Touchscreens, 19" rack enclosures etc. Main things for today are "robot wire" which is 16g wire to carry power to the motors; whether Nema 34 stepper (4 wire) or RoGR servo (2 wire). This wire is special in that it is conjoined in a parallel fashion, and is made for motion applications:

Click to view attachment

Also these RJ45 breakouts work not just for micRo, but also for cleaning up the critical RoGR encoder wiring:

Click to view attachment

There is plenty of room in the encoder box to create a ~3/4" hole to accept the encoder housing and elegantly breakout via screw lugs to shielded 14' Cat5e cables. I added these items to the inventory, so if you'd like a RoGR wiring kit, just contact me!

I should have added further to the above post, but I've been too tired to think straight. The cables are not regular Cat5e cables; those would not last long as they have solid conductors, no truss and no shield. The cables I stock are a variant called STP trussed patch cables, which have a woven fiberglass truss, a full shield to protect the sensitive encoder data, and use stranded wire. Solid wire would break quickly in motion applications. I also never showed the other end of the equation, which is the 12 port RJ45 patch bay, providing breakouts for two 6-axes machines. This would become the front of your enclosure containing your amps and the LPT breakout:

Click to view attachment

You need a type 110 punch down tool to wire the terminal blocks, but I provide that with the RoGR wiring kit. The tool also includes a cat5 stripper (meOw) for stripping the solid conductor cat5 cable used to wire from the encoder I/O to the amps, the LPT, and so forth:

Click to view attachment

Next up is the rack to hold the DIY/Retail 19" touchscreen LCD, the computer and the I/O rack. This one allows the touchscreen to be angled however you need: UPDATE, in stock in one week: dumbell toggles, metal casters, all steel:

Click to view attachment


And we need a proper computer / DIY case, I stock these empty or ready to run at a fraction of the cost:

Click to view attachment

The cases/computers have filters (very important), locks on the critical functions, essential IO on the front panel, heavy duty construction and they look sexy, too.

all that stuff looks great BC! very impressive! I may have to pick up those goodies, it would sure make the RoGR look Pro!

Thanks Phife, Though I think I should split this off into a another standalone product known as a "machine automation control" hmmm MAC, that sounds good.

The product can be used to control any number of robots, whether RoGR, lathe (even micRo, well maybe an army of them). It can also be used industrially to replace expensive, proprietary hardware (Rockwell, Hass, Fanuc).

Will these be selling for around what the OptiPlex GX270 is going for as of this posting? So nice. So many goodies popping up around here.

I take a mac with fries and a coke

How are the precision rods mounted so all 4 are parelel?
do you have some sort of screw adjustment afterward?
If it is drilled through the steel squair tubing how do you know the steel is streight or the drill is streight enough?
How about the x and y axis of if the rod ends are in the right place?

I'll take a crack at this, maybe BC will comment when he gets time.


All the Rods are mounted to the square frame via the Rod supports, Rod support adjustment is done by making a slightly larger holes in the frame so the support can move a little. I used 8mm bolts and drilled a 3/8" (about 9.5mm) so I get about .75mm adjustment in all directions.

I think BC mentioned that they were going to use some sort of plate that will make adjustments and setup/tramming easier for later RoGR models.

As for the square steel, its not hugely important but I was careful to make sure my ends were cut square and I choose the squarest edge so that the 2 faces of the tube that touch another tube would be square. I only had 1 or 2 square edges on my tubes, the other were out slightly.

And I also referenced all the tubes from the same end and triple checked all my measurements before center punching.

The critical parts are the X Gantry support blocks and the Z carriage, they must be made perfectly to maintain high precision.

Got a couple pics of the rack. I have 15 of these in stock...you can see how easy and useful it is to use this rack. At 45 degrees for sitting down, and adjustable for standing:

Click to view attachment

There are standard 19" rack spaces beneath for the MCU and breakouts:

Click to view attachment

I'd have the touchscreen recessed, but I can't get any time on the RoGR due to micRo:

Click to view attachment

RoBIN pretends to use it:

Click to view attachment

BTW; my "test" for the rack was to ride it around the shop, crashing into things, for about 5 minutes. It is fine.

So what are your ideas for a screw adjustable x & y axis for rod and screw mounts? I was looking at mcmaster catalog and couldn't find anything that would work. But I work with electronics more than mechanical stuff. I like squair tubing but I only have a drill press so you can see how attaching to squair tubing a screw on adjustable squair tubing mount for a rod or screw would be awsome.

I only have a drill press and my holes turned out fine, If I were to make another RoGR, i would Probably to waht BC said he was gonna do, and make the rod and screw supports attach to a milled alum. block and the block was attached to the frame. I cannot do that now because i have no space, the X Rod fits between the colums perfectly, Either the rod needs to get smaller, or the Long frame parts need to get longer.

So true. This is the "pain" part. (BTW, kick ass on your drills).

how much for the pc rack ? i want one of those for the living room
flip the table 180 and put the wheels under the couch
the utra lazy touchscreen home theater control /webbrowser umpc