[vbgraves]

I'm working with Ahren to replace some fasteners related to the motor tensioners, but in the meantime I decided to attach my router mount. There are instructions on the CNCRP web site, but for these PRO machines there's a slight problem. The instructions say to loosely thread the t-nuts onto the cap screws, then slide the nuts into the end of the extrusion, which is what is done on most of the other extrusion attachments and works very well. However, since the Z-axis on the PRO machines is a ball-screw assembly, it has end plates on both ends of the extrusion, so the slots aren't open on the ends. And if the t-nuts are on the screws, they won't roll into the slots directly.

Rather than disassemble the Z-axis, which I don't recommend, I just removed it from the machine and laid it on the ground. I removed the t-nuts from the screws, rolled them into the extrusion slots and then easily attached the router mount. Reattached the assembly to the gantry, and I was done. I'll figure out the final vertical position of the router mount later, but I assume the lowest position is the normal location.

I'm using a PC 7518 router on my system, and as you can see the mount is pretty large!


Attachment 196994 Attachment 196996 Attachment 196998

[ahren]

The instructions need to be updated, but the roll-in t-nuts currently supplied with the router mount kits are easy to install on the ballscrew z axis. As the name implies, they can just "roll in" to the slot, and do not need to be installed from the end. An allen wrench can be handy to help maneuver them into place in the slot. The z axis does not need to be removed from the machine to do this, although the nuts slide around less in a horizontal position.

Ahren
CNCRouterParts

[vbgraves]

Yes the nuts can be put in the slots, where they will fall to the bottom. I didn't want to have to get clamps, magnets, wrenches, or another set of hands in order to hold the router mount in place while trying to get a couple of fasteners installed. Removing it seemed to be much easier to me. Anyone reading this before they get this far in their build, I highly recommend installing your router mount prior to installing your Z axis.

[seesoe]

vbgraves, i put the first 2 roll in nuts and let them fall to the bottom. i then screwed the first 2 top screws into that. after that i moved the mount up and put the other 2 roll in nuts and let them fall to the bottom and then lowered the mount to those nuts. i didn't have any issues with any z axis assembly, it went together smoothly. what i did have issues with was the gantry sled, the back 4 adjustment bolts. being that the backplate is painted, the slots are smaller and the plate moves while you adjust it, so i would recommend a clamp or scrape the slots clean for easier and precise adjustment.

[vbgraves]

Update on progress. After some fastener issues with the motor tensioners was resolved I got them installed.

Attachment 198382

Also got most of the cable track installed and ran the motor cables.

Attachment 198384 Attachment 198386 Attachment 198388

The issue came when mounting the control box, which is the CNCRP DIY NEMA34 box. It's fairly large, and I wanted to maximize the space under the box to make it easier to install the cables and not put any pressure on the connectors. Because of my table design, the box wouldn't really fit underneath the LVL and still leave adequate space underneath. So I mounted it as high on the table side as I could, leaving ~1 inch between the bottom of the V-con adjusting screw and the top of the control box. This issue is due to the fact that the PRO V-con system extends below the surface of the table top, which reduces the amount of space available for the control box.

The other thing I was wanting to do is mount the control box towards the back of the table, which would leave the front to mount my computer. Since I plan on working primarily from the front of the machine for material loading and machine setup, I wanted the standard keyboard and monitor cables to reach the computer.

Attachment 198390

I was hoping this configuration would work with the standard length motor cables, but it did not. So I moved the control box to the front of the machine, which is probably where most people put it. I guess I could have swapped the 12ft cables for 20ft cables, but it seemed cheaper to just buy some longer USB and monitor cables.

You'll also note that by mounting my control box close to the V-con rails, the cable track tray mounting may need to be modified. I'm looking at this now.

[vbgraves]

Just wanted to mention that last night I reached a milestone and got motion out of the machine! I'm running LinuxCNC and already had the software loaded and ready to go, as I had tested my control box with the motors several months ago. So it was a matter of configuring LinuxCNC with the correct motor, drive, and gearing information. Ahren sent me the details, I entered them into the Step Config Wizard, and it worked! Still need to clean up the wiring and mount the side cable track, but this was a good feeling! Will spend some more time on it after the holiday weekend.

[vbgraves]

seesoe, good suggestion on the router mount. Obviously that didn't occur to me when I was trying to mount mine. Sounds like you're ahead of me in your build. Do you have a build log I missed somewhere?

[vbgraves]

Never mind, I found it. Looks good so far!

[seesoe]

vbgraves, im building a pro 48x96 machine, i just finished the bulk of the assembly last night, i still have to mount the motors and racks and then wire up my panel
http://www.cnczone.com/forums/cnc_ro...build_log.html
i'll going to write and update later tonight on my progress and where im at

your machine is coming along nicely, i like you how your base come together. i wanted to use lvl as well but my local supply only had it in 24' x 12" and couldn't cut it down haha.

[vbgraves]

Been a while since I've posted, so here's an update. I've been waiting for the newer version of the brackets that the side cable tray rests on. The original version didn't work for my table setup, and CNCRP shipped them last week, so I expect them tomorrow. So for now I just have a mess of cables on the side of the machine.



While I've been waiting, I cleaned up the cabling for the gantry and router. I also installed my KentCNC dust shoe. I haven't run my 4" PVC dust collection system to the router yet, so I temporarily will try to use my shopvac. It really doesn't move enough air through the 4" hose, but it is better than nothing.

Attachment 199708 Attachment 199710

I'm controlling my router with a SuperPID. The cable entering the top of the router is the light sensor that serves as the tachometer feedback. The second small cable below the router mount is the thermocouple. It would be preferable to be closer to a bearing, but I didn't want to cut another hole in the router's top casing to get it inside. Maybe later. I just wanted to get it installed for now.

Attachment 199712 Attachment 199714

I also installed the CNCRP limit switches but have not connected them yet. Will wait till I clean up the other wiring. Finally, I worked on my spoil board. I cut it to size on my table saw and generated some g-code to machine counterbored thru holes that line up with the cross members. My concept was to use t-nuts in the cross member slots and hold the spoil board down with some button head cap screws. To do this, I machined the holes in the spoil board and positioned the holes over the slots. I had some hobby paint and a small brush from painting the top bearing of my router for the SPID sensor, and I used them to put a small paint dot in the bottom of the slots where the holes were located. I then removed the spoil board and placed the t-nuts over the paint spots, reinstalled the spoil board, and threaded in the screws. Worked great.

Attachment 199720

So now I've actually machined some MDF under computer control! I haven't tried to tram the spindle yet and am looking for a magnetic base to do that with. Hopefully this week I can get things cleaned up and finish the basic machine construction.

Van

[vbgraves]

I made some progress this week by installing the rail side cable track, which made it easier to start doing some testing. I started getting random e-stops during both jog and programmed movements. Checked the connections but found nothing amiss. I looked through the PMDX-126 documentation and followed their grounding recommendations as best as I could, but nothing changed. Finally tried adding a capacitor to the e-stop terminals on the PMDX board, and so far that seems to have solved the problem. I was able to make a video of the machine under program control. Feed speed is 400 ipm.

CNCRP PRO4848 warmup cycle - YouTube

So I reached another milestone!

[vbgraves]

I thought I'd give a more detailed description of my machine setup, since it seems to be "non-standard".

In the CNCRP documentation and in most build logs I've read through, the assumed setup has the gantry moving in the X direction, so the two gantry motors are X and X'. Motion along the gantry is then Y and Z is vertical.

I prefer to work from the front of my machine, facing the router, and I wanted the machine movements and material loading to match the coordinate system in my CAM software, that is +X is to the right and +Y is towards the rear of the machine. So that's how I configured the axes in LinuxCNC: I have a single X motor and two Y motors, Y and Y'. For those who might want to configure LinuxCNC for this setup, here's my parallel port setup from StepConfig. Note that I haven't set up any limit or home switches yet.

Attachment 200752

Here are the three axis configurations. These values correspond to the NEMA34 motors and gearing along with the 5mm Z axis ball screw. The max velocities and accelerations are my initial values. I'll probably change the X and Y velocities at some point but for now they are adequate.

Attachment 200754 Attachment 200756 Attachment 200758

At this point I haven't done much cutting, so I haven't checked any measurements with a caliper or indicator, but I believe these to be the correct nominal values from the hardware.

[FoxCNC1]

Very nice !! I am waiting for my prokit as well.
I have been thinking hard on whether I want to get the nema 34 motors or stay with the nema 23... decisions, decisions.

Why did you go w the N43?

[vbgraves]

Profoxcg, I created a thread several months ago when I was thinking about a CNCRP machine that gives some of the thought process (http://www.cnczone.com/forums/open_s...g_crp4848.html), but for me the basic reason was expandability - the N23 G540 solution only accommodates 3-axis machines. I had/have visions of someday adding a 4th-axis rotary indexer. With the DIY N34 electronics kit, all I need to do is add another Gecko drive module to the enclosure, wire it up and configure it in software. That's not possible with the G540 in a straightforward fashion, at least not as far as I know.

Also, I decided I'd rather wait till I can buy / build the machine I think I'll keep rather than try to upgrade or replace the first one later (a "build your second machine first" mentality, which is another reason I waited till the PRO machines were available). Right now, as I'm just getting started, the N34 system is overkill, and it may turn out to always be that way. But ask me again in 6 months or a year, and I hope to be able to say I'm using the additional performance of these motors, especially if I replace my PC router with a spindle.

Going N34 is a $1k+ adder, and I thought long and hard about it too. Good luck with your decision!

[Dylwad]

N34 is definitely overkill, not that it's a bad thing other than cost. IMO more suited to a machine that needs to process full sheets, full time, full depth of cut, with a beefy spindle.

Adding a separate stepper drive to the 540 isn't too hard for a 4th axis, not impossible by any means. Easier than the n34 route, but it all depends on how many I/O's your using on the 540, otherwise a break out board or card with second parallel port is needed.

Sent from tapatalk

[FoxCNC1]

Profoxcg, I created a thread several months ago when I was thinking about a CNCRP machine that gives some of the thought process (http://www.cnczone.com/forums/open_s...g_crp4848.html), but for me the basic reason was expandability - the N23 G540 solution only accommodates 3-axis machines. I had/have visions of someday adding a 4th-axis rotary indexer. With the DIY N34 electronics kit, all I need to do is add another Gecko drive module to the enclosure, wire it up and configure it in software. That's not possible with the G540 in a straightforward fashion, at least not as far as I know.

Also, I decided I'd rather wait till I can buy / build the machine I think I'll keep rather than try to upgrade or replace the first one later (a "build your second machine first" mentality, which is another reason I waited till the PRO machines were available). Right now, as I'm just getting started, the N34 system is overkill, and it may turn out to always be that way. But ask me again in 6 months or a year, and I hope to be able to say I'm using the additional performance of these motors, especially if I replace my PC router with a spindle.

Going N34 is a $1k+ adder, and I thought long and hard about it too. Good luck with your decision!

Thank you for the link and your response.
I also agree with the idea of not having to "buy twice" and have thought about it hard before reading your post and linked thread.
I already have a machine with N23 motors. It works great and I thought I could use those motors to get me going, but I ultimately want to be able to throw anything I want at the machine. I too will buy the DIY kit. (btw what gauge wire were you using for the connections from the motors to the drives etc)

In short yes its more money - but it would even more money from do the upgrade once you had a N23 setup or even just upgrading the motor mount and R&P. Although I already upgraded my order I am still having some second thoughts but its all done and paid for. Maybe on the spindle gantry and Z should be Nema34. The other axis, because it driven by two motors could be Nema23. Either way, just like in a car its better to have the power than now.

N34 is definitely overkill, not that it's a bad thing other than cost. IMO more suited to a machine that needs to process full sheets, full time, full depth of cut, with a beefy spindle.

Adding a separate stepper drive to the 540 isn't too hard for a 4th axis, not impossible by any means. Easier than the n34 route, but it all depends on how many I/O's your using on the 540, otherwise a break out board or card with second parallel port is needed.

Sent from tapatalk

The whole "overkill" thought has had me thinking very hard about this. This will not only add power, but also add weight to the machine etc.
I do see how beefy the pro machine is, and I thought it would be good to have strong motors that can move the cnc with ease. I am sure the Nema 23 can, but the idea of the bigger motor just makes me think that it would be that much better.

corrent me if I am wrong, but I recently learned that you can use the G203 to drive nema 23's, and if I understand correctly, I could even use Nema 23 motors for the X and Y, and get a Nema 43 for the Z axis - in other words mix and match motors.

[Dylwad]

If the pro kit uses a 5mm pitch ballscrew for the Z the nema 23 might be a better match than the 34, just due to the different torque curves, usually the lower inductance smaller motor will have more torque at higher rpms than the larger motor, but there are too many factors going on to know for sure.

I built a machine years ago with 640oz 34's and G201's running at 65 volts, 1/2x8 2 start acme, it worked great and had plenty of torque and performance, my latest machine uses 381 oz 23's at 50 volts and G250's, 1/2x10 5 start acme and has better performance for about half the cost of the older system, machine weights are similar, both use only 3 motors.

If you took nema 23 on a r&p geared 3:1 and 34's on the same r&p geared 2:1, the top speed and torque might be very similar, but the 23's will do it with finer resolution.

So it's not cut and dry all of the time, bigger motors will sometimes mean better performance but not always. The problem with mixing different motors is they will require different voltages to get the best performance most of the time, one power supply is much easier than two.

I know cnc routerparts has it pretty well figured out in regards to the electronics they sell and recommend, IMO it's best to run the same motors all around just for the ease of keeping everything the same voltage, and the ability to swap components to track down issues in the future.

Sent from tapatalk

[FoxCNC1]

If the pro kit uses a 5mm pitch ballscrew for the Z the nema 23 might be a better match than the 34, just due to the different torque curves, usually the lower inductance smaller motor will have more torque at higher rpms than the larger motor, but there are too many factors going on to know for sure.

I built a machine years ago with 640oz 34's and G201's running at 65 volts, 1/2x8 2 start acme, it worked great and had plenty of torque and performance, my latest machine uses 381 oz 23's at 50 volts and G250's, 1/2x10 5 start acme and has better performance for about half the cost of the older system, machine weights are similar, both use only 3 motors.

If you took nema 23 on a r&p geared 3:1 and 34's on the same r&p geared 2:1, the top speed and torque might be very similar, but the 23's will do it with finer resolution.

So it's not cut and dry all of the time, bigger motors will sometimes mean better performance but not always. The problem with mixing different motors is they will require different voltages to get the best performance most of the time, one power supply is much easier than two.

I know cnc routerparts has it pretty well figured out in regards to the electronics they sell and recommend, IMO it's best to run the same motors all around just for the ease of keeping everything the same voltage, and the ability to swap components to track down issues in the future.

Sent from tapatalk

Did I understand you said that the nema 23 have better resolution than the Nema 34? (I understand this as better accuracy or the ability to cut smaller items with more precision?)

[Dylwad]

Did I understand you said that the nema 23 have better resolution than the Nema 34? (I understand this as better accuracy or the ability to cut smaller items with more precision?)

When the 23 is geared 3:1 and the 34 is geared 2:1, yes. Otherwise they are both 1.8°per step motors.

Sent from tapatalk

[FoxCNC1]

this came from CNC Router Parts' site:
PRO Rack and Pinion Drive, Nema 34 | CNCRouterParts

"The new (gear) spindles also now feature a 64 tooth pulley, offering higher resolution and incredible low end torque via a 3.2:1 reduction."

So it sound like they are taking care of that what do you think?