[louieatienza]

I search for the book Solsylva 25x37 "Plans for Inexpensive CNC Router Tables by David K. Steele". It isn't present on a site of the author

It has been replaced by other plans. The nice thing however is that you do not have to buy all three plans in a book, you can just buy one. The book you refer to was sold as hard copy and there is no digital download. David's new plans are now all available for download; that is of course after you pay for them. He no longer does the 25 x 37 size, though it's pretty easy to extend the x axis on the 25 x 25 plans...

[louieatienza]

Due to a misunderstanding of Misumi's 30% discount special for new customers, I will be updating this machine!

I thought I would get the discount on my first order. Instead the 30% is a credit toward the NEXT order. If I knew this beforehand, I would have left out some parts for my new build, and placed the second order immediately after. Just as well, since I want to get all the parts in one shipment anyway.

So now I have some credit with Misumi which to purchase more extrusions! For this machine however, I'll probably go with their more economical systems, etiher EFS or HFS.

I will keep all the electronics, rails, and leadscrews. I havent decided whether to keep the raised rail configuration, since I want to open the ends to process longer pieces (getting a 3/0 x 6/8 door on the machine would be nice. I will also make the gantry beam longer, increasing the cutting envevlope from 25 x 34 to 34 x 34.

[louieatienza]

For those who haven't seen my other thread, I think I killed my xylotex box. The Y axis started stalling inexplicably, and my machine cut through a washer! The endmill seemed unharmed, but the workpiece was trash.

Blowing out the xylotex left a plume of smoke befitting of a 70's lounge. Afterwards the Y started behaving uncontrollably, stalling at slow speeds but sounding normal at rapids. I hope to purchase a replacement.

I was forced to use the G540 box purchased from DeepGroove1, since I have a potentially lucrative contract job riding on samples I need to make. This turned out to be a blessing in disguise, as I nearly doubled my rapids on the Y from 400 to 720 with teh accel set at 100, and the X from 200 to 570 with the accel set at 40. I do get the X to stall with speeds over 600 and accels over 50. The Y I believe is limited to 750 by the processor.

I shot a video to document the change:

GeckoDrive G540 Installed in my DIY CNC Router - YouTube

[louieatienza]

Every once in a while, between the ho-hum projects and commissions, I get a fun part to cut out. This is for a custom grille for an F-250. The grille is black, so the client will paint the bevel black to let the outline and raised areas stand out.

First pic: I used the Kyocera 1/8" single edge spiral-"O"-fllute carbide endmill to "pocket" my outline, and leave room for the bevel and clearance for cutting out the part. I don't have the SuperPID hooked up yet, so I used the slowest speed on the Craftsman router, and ran this at 45ipm and .032" doc.

Second Pic: I used a 1/16" 2-flute general purpose carbide endmill to "clean up" the first pass and get the profile of the lettering right. Since there is ample clearance for chips I was able to do each pass at .032" doc as well; 24ipm feedrate with the spindle at the same speed.

Third Pic: I used a 60 degree drill mill to bevel the edge of the lettering. To avoid cutting with the tip, I created a .020" offset and set the final depth .020" more than I wanted. I left about .063" flat on the lettering walls.

Fourth Pic: With the spoilboard surfaced, I can accurately set the depths of cut. As you can see, I keep the part on the table by cutting right to the Mic-6 protective film!

Fifth and Sixth Pics: Views of the finished part! Can't wait to see this on the truck...

[TrickyCNC]

looks great Louie

don't forget to post the pics of it mounted !

[louieatienza]

Thanks Rich. My client liked it so much that we're doing a hitch plate with the same logo...

Pic 1 shows the design, basically the same logo, beveled similarly (to create "prismatic" letter, milled out of a solid chunk of 1/2" aluminum (I believev it's 2000 series but not sure.)

Pic 2 shows the roughing toolpaths. This is OneCNC's high speed *constant engagement" strategy. By not "burying" the bit and adjusting the radial doc to the feed, high feedrates can be achieved. The overall depth is .18", and I roughed this out in 2 passes, 1/4" spiral-"O", .09" depth, with 12% rdoc, at 150ipm! Left .01" to finish.

Pic 3 is the finish toolpath (for the field). I like the look of the HSM toolpaths so I used it instead of the traditional offset or planar paths. I'm used a 1/8" spiral-"O". Reduced the feedrate to 40 to leave a nice finish.

Pic 4 shows the toolpaths for the bevel. I could've drawn it but it's actually faster to program it this way.

Pic 5 shows the cutout toolpath.

[louieatienza]

Pic 1: First thing to do was surface the plate. To reduce setup time I just used the same 1/4" spiral-"O" that I would use for the roughing. These have a raker edge on the bottom which leave a nice surface.

Pic 2: I really wanted to see how much I could push my machine. 150ipm cutting is usually no problem for wood, but for aluminum it's a different story. The Craftsman router was up to the task, only getting slightly warm. It looks like Christmas in my basement, with all the aluminum chips!

Pic 3: For the bevel I just used a 60 degree drill mill. I ran it slow to minimize chatter, though my setup is not the most ridgid. I took light cuts with higher spindle speed and lower feedrate.

Pic 4: When your aluminum cuts look nice and shiny, you're doing well! I used ZEP45 lube, spread with a brush. Not a lot is needed.

Pics 5-7: The finished product. I'll leave it to the client to do all the finish work. Not a lot of cleanup to do, maybe feather out some sharp edges. Polishing would be an option as well. All in all a fun (and paying) job!

[louieatienza]

I am almost done on the machine rebuild. To make things more modular, I have decided to incorporate t-slot extrusions into this build. THis should make any future upgrades/repairs easier.

My big beef with my current machine is that I currently cannot remove or work on one axis without disturbing another. On my redesign, I can remove the whole gantry assembly from the table without affecting those linear components, and I can also remove the carriage from the gantry with little fuss.

Another issue I faced was the work envelope versus the machine footprint. By moving the rails below the table, I was able to add another 5" or so of Y axis cutting area, while at the same time, decrease the table footprint from 37" wide to 31" wide. I also increased the carriage travel to 33" from 27"; this will allow me to 'lead in' with larger form tools like rail an stile bits, panel bits, etc.

The increased gantry weight necessitated me to abandon the belt synched dual leadscrew system in favor of a dual stepper configuration. I couldn't do this with the 3-axis Xylotex box I had, but is easier with the G540. To save costs for now, I will use timing belts and pulleys with a 1:1 ratio to drive the screws, and will upgrade to direct drive and flex couplers when the budget allows. I will also incorporate t-slot extrusions for the gantry legs which will allow me to adjust the bridge height as well.

All the connecting plates are made of 3/4" black phenolic that I have on hand. As time goes I may replace some of the parts with aluminum, though I like the look of phenolic. I used 1/4" aluminum for the stepper plates.

I just finished my last batch of auto parts for one of my clients, and started diassembling my machine and assembling the new frame. I mght make a new table of Unistrut as well.

Hope to have the brown truck drop off the rest of the stuff I need. I'll take pics tonight to show what I've done so far...

[louieatienza]

A little late, but here are some progress pics...

Pic 1 shows a general view of the machine. Reused are the ROLLON bearings and rails. Instead of the phenolic frame, I used 4545 profile for most of the machine, with 4590 at the center of the table and for the gantry legs. I did use phenolic for all connecting plates.

Pic 2 shows the X axis stepper assembly. This was done to use what I had on hand, but will swap these out for couplers, and new plates, when the time comes.

Pic 3 shows the Y axis stepper assembly. On both X and Y screws, there is a 1/8" brass spacer between the pulley/coupler and bearing. On the opposite end is another spacer and shaft collar. There is a vvery slight preload applied to the screws.

Pic 4 shows the Z axis assembly. To keep a low profile, the leadnut and screw are "set in" to the carriage plate, thus keeping the unit as compact as possible. I abandoned the round rail in favor of these ROLLON bearings and rails as well.

Pic 5 shows the Y-Z bearing saddle.

Pic 6 shows the carriage assembly in close up, with its relatively small overhang. I also attempted as much as possible to have the leadscrews in the same plane as the centerline of the bearings/rails...

I hope to have everything operational in the next couple days, save for the brackets on order. I will also utilize cable chain to keep the wiring neat.

[Falcon69]

Hey Louie,

My Nephew really loves anything that has to do with Monster. I had missed his birthday a couple weeks ago. What would you charge me to make me one of those in post 187?

[louieatienza]

Hey Louie,

My Nephew really loves anything that has to do with Monster. I had missed his birthday a couple weeks ago. What would you charge me to make me one of those in post 187?

PM sent...

[louieatienza]

I'm back up and running! This time only took me about 3 days, working on the machine a few hours each day. Having all the new parts fabricated beforehand made it so much easier, as well as using the extrusions. I still need more t-nuts to bolt the spoilboard down, but those are on the way. I'm using some gussets I had on-hand for now to hold the gantry beam until I either buy them or make them. I will also make gussets for the front of the gantry legs as well, and a new dust shoe.

These are better pics which I took with a new camera!

Pic 1 shows the machine, with the spoilboard cut but not bolted down.

Pic 2 shows the dual X stepper configuration.

Pic 3 shows the Z carriage assembly.

Pic 4 shows the rear of the gantry assembly, including the gantry gussets, Y leadnut between the extrusions, and steppermount.

Pic 5 shows the Y stepper assembly.

Pic 6 shows another full view.

With the dual X leadscrew assembly, I can now jog the X axis at over 750ipm, and I can achievev a blended (X,Y) speed of ovver 1000ipm! As far as the extrusions are concerned, I'm really amazed how stiff these things really are! Assembly was almost effortless, and adjustments just as easy. I'm hoping the extra ridgidity translates into better quality cuts.

Video coming soon....

[louieatienza]

And the video...

DIY CNC Rebuild using 8020 Extrusions! 1000IPM Rapids - YouTube

[louieatienza]

And the video...

DIY CNC Rebuild using 8020 Extrusions! 1000IPM Rapids - YouTube

Almost a bad experience today... After installing cable chain, cleaning up the wiring, and making a temporary router nount, I started the machine up. The X axis started to stall. I couldn't even get it to move without the motors going out of synch and making bad noises. I thought I fried my G540!

Turns out those damn Molex connectors were getting hot. I checked and could see the arcing. I ended up cutting them off, splicing and soldering the wires, and protecting the splice with shrink-tubing. While I was at it, I did the same to the Y axis motor. The Z seems fine for now but I may change that over soon; I can't afford for that to happen in the middle of a job.

I'm beginning to think my xylotex box is not fried, will have to pull it out and check.

Does anyone know of a more reliable 4-wire connector system I can use?

[Phife]

I used XLR Microphone connectors for my servos a few years back and they are stil working well, and look great. On my new machine some DIN connectors came with the electronics enclosure and Nema24 back covers, they also seem to work well. Although the Molex should have been able to handle it.. strange that they were getting hot, perhaps they were not properly crimped?

[CarveOne]

I would not be surprised if the Xylotex died from an intermittent connector. It has no short or open circuit protections. Mine blew an output when I snagged a motor cable on my first build a few years ago and one of the wires pulled partially out of the connector. It apparently only takes microseconds to destroy the output device. Cost $70 back then to get it repaired.

(Insert plug for a new G540 here )

[louieatienza]

I used XLR Microphone connectors for my servos a few years back and they are stil working well, and look great. On my new machine some DIN connectors came with the electronics enclosure and Nema24 back covers, they also seem to work well. Although the Molex should have been able to handle it.. strange that they were getting hot, perhaps they were not properly crimped?

Thanks for that idea... was thinking maybe s-video connectors too. I have those DIN connectors as well but was saving them for the aluminum router....

I would not be surprised if the Xylotex died from an intermittent connector. It has no short or open circuit protections. Mine blew an output when I snagged a motor cable on my first build a few years ago and one of the wires pulled partially out of the connector. It apparently only takes microseconds to destroy the output device. Cost $70 back then to get it repaired.

(Insert plug for a new G540 here )

I had both the intermittent connector AND excessive dust in the xylotex box. Since the rest of the box is still good I will reuse it for another project. Just need a new xylotex board...

[Phife]

Thanks for that idea... was thinking maybe s-video connectors too. I have those DIN connectors as well but was saving them for the aluminum router....

I would avoid using svideo connectors because they do not lock in place and could get pulled out, also they probably arent rated to handle the voltage and current..

[harryn]

Your rebuild project is great - it is more like a brand new machine using some of the old machine's parts. Thanks for the video - very helpful.

I kind of got lost in how the final gear ratio works out. Am I right that with 8 start screws and your belt setup, you are somewhere in the 200 - 300 full steps per inch range ?

[louieatienza]

Your rebuild project is great - it is more like a brand new machine using some of the old machine's parts. Thanks for the video - very helpful.

I kind of got lost in how the final gear ratio works out. Am I right that with 8 start screws and your belt setup, you are somewhere in the 200 - 300 full steps per inch range ?

Thanks Harry. It does 'feel' like a new machine! There are in fact less machined parts in this build since using the extrusions.

The 8-start screws have a 1tpi effective travel. Coupled with the 1:1 belt drive, that gives me 200 steps/in. I find the dumpstercnc AB nuts are more than adequate at keeping repeatability, and I've never had a part ruined from missed steps during a cut (unless I ran into a fixture, clamp, or the frame of the old machine.

I'll have pics later of my cable chain setup. I got by fine for years just hanging the wires, but thought it'd be nice to not have to worry about them anymore!