GME's New 80/20 CNC Build - My Design

[Arjay]

Not related to the build, but another area gem for electronic components is Vetco Electronics in Bellevue.

[GME]

Not related to the build, but another area gem for electronic components is Vetco Electronics in Bellevue.

Good to know. I've looked for electronics stores in the area, but hadn't found this one. Actually, Bellevue is a nasty drive from Gig Harbor, but it's good to have sources within driving distance in an "I need it now" situation.

Gary

[wmgeorge]

Nope my tracks are on 8 inch centers and they should have been 6 which would have been perfect. If your time is so limited, why not just use a $20 piece of MDF with added sides and
just "git er done"? A couple of C clamps (or bolts and nuts) to the cross braces, clamp down the aluminum do the holes first use them as hold downs while you do the edges. The real critical
part is the hole spacing and you've already made a prototype that fits perfect.

[GME]

Nope my tracks are on 8 inch centers and they should have been 6 which would have been perfect. If your time is so limited, why not just use a $20 piece of MDF with added sides and
just "git er done"? A couple of C clamps (or bolts and nuts) to the cross braces, clamp down the aluminum do the holes first use them as hold downs while you do the edges. The real critical
part is the hole spacing and you've already made a prototype that fits perfect.

Thanks, Bill. Now that you've corrected the record, I was probably keying in on what you wished the spacing was vs what it actually is.

I keep waffling on what I want to do next. Based on my testing on homing speeds, and the return I seen from greasing everything and adding the extrusions on the sides, the gussets don't seem like such a high priority. I'm anxious to start cutting stuff, which is why I doing the spoilboard started looking more attractive. The downside to putting on the spoilboard before cutting the gussets is the risk of damaging it. On the other hand, if I plan to cut aluminum in the future, I'll still face the risk of damage. I guess I'm in a awkward (indecisive?) stage in the build.

Maybe I'll just breakdown the aluminum plate with my track saw (really not looking forward to it) and decide then what to do next. It's all got to get done and I'm anxious to get it there.

It all has to get done. The only question is in what order.

Off to the shop now to breakdown the aluminum. I'll take a picture or two to show how I set it up.

Gary

[mactec54]

The going is slow right now. My wife has a never-ending honey-do list and seems intent on keeping me from finishing the project.

I have made a little progress, though.

I've settled on how I'm going to do my spoilboard, created it in CAD and created the toolpaths for the base sheet. The spoilboard base is 18mm Baltic Birch - 60" x 60". To that I add my t-track spaced 4" OC. 1/2" MDF between each t-track, glued to the base sheet. Finally, 3/4" MDF cut to overlap the slot edges of of the t-track and glued to the 1/2" MDF. By overlapping the t-track slot edges, mechanical fastening of the t-track is unnecessary (I'll add a couple of screws in each piece to keep it from moving around). The Orange Aluminum t-tracks make the process easy, because they are 1/2" high. I don't need to cut rabbets (rebates for those not from the U.S.) in any of the pieces.

Bill, I recall that you have your t-tracks spaced at 6" OC. Since I cut directly on my spoilboard (versus using sacrificial pieces between the spoilboard and the stock), 4" OC will be more versatile for my purposes. I thought about variable spacing, but in the end, decided to go with fixed.

UPDATE ON RESONANCE ISSUE

I did some more fooling around with my homing speed. Using 5 ipm as my reference point, I incrementally increased and decreased the speed to see what would happen, i.e., at what point does the resonance start, and go away. Turns out that the unacceptable resonance was present at only 2 speeds - 5 ipm and 6 ipm. At 3 ipm, 4 ipm and 7 ipm everything was within an acceptable range. I decided to take ger21's suggestion and contact the folks at Gecko. However, before doing so, I tried to anticipate what diagnostics they might suggest and go ahead and perform them. I probably missed one or two things, but that's to be expected. I don't know what I don't know, if that makes sense.

One other thing I did was go back and regrease all of the bearing blocks and R & P parts. As expected, the blocks didn't take much additional grease, but I thought they might take some, given that I've been moving axes around and some of the grease may have settled in.

Where I actually saw a reduction in the resonance was with adding more grease to the R & P parts. That surprised me some. I thought I had lubed properly when I installed the parts. The post re-lube told me I hadn't. Don't get me wrong. There was still resonance, but the degree changed. Before lubing, there was enough to cause a 21mm wrench to move around. After, the wrench didn't move at all - no rattling noise, nothing. However, a pair of reading glasses and mechanical pencil still moved around at 5 ipm and 6 ipm. So, I am left to conclude that the lubing mitigated the effects of the resonance, but did make the resonance go away. Progress.

I detailed what I did in email to Gecko support. I suspect I will have a conversation with the folks at Gecko, but I wanted to make sure I communicated everything I've done first, so I didn't miss something if I started the dialog with a conversation. I just sent the email last evening, so I expect hear back in the next couple of days.

I think I've worked out how to cut the diamond plate braces. Rather than use a sacrificial piece of MDF, I think I will fasten a piece of HDPE to the spoilboard and tape plastic to it to keep the coolant from reaching the MDF spoilboard surface. I plan to cut some threaded holes in the HDPE for holding the diamond plate down, but I won't cut the threaded holes all the way though. That way, the coolant can't make its way around the threads and onto the MDF. If I do it right, I should be able to reuse the HDPE for future projects.

I am still toying with using WD40, instead of my water/soluble oil mix for cutting. I'd rather use with water/oil mix, but I'm reluctant to risk my spoilboard by exposing it to water. There is still a risk with WD40, though. If any of it gets on the spoilboard, it can later transfer to wood parts, which will create problems with finishing. nlancaster and I had some dialogue on this thread about cutting aluminum. I was pleasantly surprised to learn that whatever is used to cut aluminum tends to remain fairly localized to the cutting area. If I spray with a can of WD40 (vs using my mister), I can more easily minimize the amount of WD40 that's needed to do the job.

More to follow.

Gary

To Protect the spoil board put down a plastic sheet or pallet cling rap of some kind

[GME]

To Protect the spoil board put down a plastic sheet or pallet cling rap of some kind

Thanks. That was the plan.

Gary

[GME]

Moving forward.



This is a picture of my setup for breaking down the diamond plate. You can't see it, but there are two pieces of 16" x 48" x 2" thick white rigid Styrofoam insulating board under the plate. Using it as a sacrificial base layer is great way to save some wear and tear on saw blades. The stuff lasts and lasts. I've used it for years.

The saw is a festool TS 75 plunge cut track saw fitted with an Oslun non-ferrous/plastics blade. The blue thing to the left of the saw is a precision guide from TSO Products for cutting perpendicular to the reference edge. It allows one to easily straighten and edge, and saves having to measure from the upper and lower edges and having to align the rail to 2 marks. With the guide, just measure the width of cut from the near edge, mark it, and set up on that mark.

I used a stick lubricating wax on the blade teeth before each cut. It reduces wear, gives a smoother cut, and adds are margin of safety. The wax reduces cutting temperatures and discourages the aluminum from trying fuse to the blade. The saw cut this stuff like butter. I never cut aluminum with a saw without the wax. I probably wax more than I need to, but I'd rather err on the side of too much.




Here's the stack of cut plates ready for milling. I only need 8, but the larger plate I cut yielded 9. That give me on extra to use for adjusting feeds & speeds. I cut the pieces a little oversize to leave some material for the profile cut.

As I mentioned in a previous post, the cut plates will be 12" x 15", with 2 sides cut on a diagonal. The legs under my CNC are 3" square profile.. 12" x 15" gives me the same number of fixing screws on both the leg and the horizontal brace at equal spacing.

I'm still waffling on whether to use a temporary spoilboard to cut the gussets, or wait until I get the permanent spoilboard installed. Stiffness is an issue. A temporary spoilboard will be 3/4" thick, unless I glue multiple pieces together. The permanent spoilboard will be a combination of 18mm Baltic Birch, 1/2" MDF and 3/4" MDF, or nearly 2" in total. The 3 layers will be glued together, so should be quite stiff. I'm leaning toward finishing the permanent spoilboard before cutting the aluminum.

I took a scrap of MDF and spot sprayed with some WD40 to see what it looks like. As ger21 advised, and as I expected, it doesn't cause swelling. It just darkens the color some. It also seems to retain its structural integrity. Even if I use the permanent spoilboard, I'll still be using a piece of sacrificial MDF (and plastic sheeting and aluminum foil duct tape to protect the spoilboard). I'm thinking I may drill some holes in the sacrificial to insert metal alignment dowels. They will be removed before running the profile cut. That should speed up setup time for each piece.

Gary

[GME]

Update on the resonance issue when homing at 5 ipm.

Marcus at Gecko asked me to records some video during the Y axis homing at 5 ipm. I just ran the gantry to the rear and started the homing sequence. At 5 ipm, it takes awhile to reach the front. I had plenty of recording time. In conjunction with the video, Marcus was interested in the sounds during the homing exercise.

After sending the videos to Marcus, he replied that he thought I have a mechanical resonance issue. Looks like I'm coming back full circle.

Marcus asked me to remove the steppers and run them at 5 ipm on my bench and record the sights and sounds. I'll be doing that over the weekend.

I've done some reading on eliminating/reducing mechanical resonance. I don't have an engineering background. I understand some of what I've read; some I don't.

One thing that stood out for me was that adding stiffness can increase the resonance frequency. I don't know what that means in day-to-day practical terms, but it sounds like adding rigidity could take a problem and make it worse. Then again, could it mean that if you get the resonant frequency high enough, it could crease to be an issue. It's hell when you read things and don't fully understand the material.

I also looked at vibration dampening. That part I understand, but what's best? There are pads, like those made by an outfit named Sorbothane. They look good, but jacking up an already put together and meticulously leveled machine to insert pads under the feet doesn't sound like something to look forward to. Apparently, adding lead shot or sand to the extrusion cavities can work wonders, but that route would call for a complete tear down and reconstruction. I'm not sure I have that in me right now. What a horrible prospect. I've read a little on filling the cavities with polyurethane foam. That has promise, but I'd either have to tear down and rebuild, or drill holes all over the place. Pretty much all the fixes, except pads, would really go better after a complete tear down.

Since the problem is centered on a relatively small IPM band, on way of addressing the issue is to simply avoid that speed band. Not my first choice, but maybe the most practical one at the end of the day.

More to follow.

Gary

[peteeng]

Hello Gary - Every part of the machine is vibrating. It's just one part or maybe two parts are vibrating in sympathy at that speed. A stepper motor is a great resonator as it steps along. If you have a long steel rod, even a long screw driver put it to your ear ( or jaw) and place the other end on the motor and then on the mount and start to work away from the motor. using the rod as a stethoscope you will be able to hear the various machine parts singing. You should be able to hear the major vibration and work towards it. It will be something that's thin and cantilevered, or eared . Something that can vibrate freely. The tubing and major structural parts will vibrate at a lower (I think) frequency than what you are getting. But I'm sure using the rod, you will track down the culprit. Then you need to clamp it, damp it, or change it somehow so it sings at a different vib. Peter

[wmgeorge]

Since the problem is centered on a relatively small IPM band, on way of addressing the issue is to simply avoid that speed band. Not my first choice, but maybe the most practical one at the end of the day.


So the exact same setup was on your old machine and no issues? I can not ever think of a time when I needed to cut at exactly 5 ipm.

[ger21]

Adding the multiple layer spoilboard may help dampen the frame enough to get rid of it.

[pippin88]

Ways of reducing resonance:
1. Raise the natural frequency above the operating range
- increase stiffness
- decrease mass (without compromising stiffness)
2. Increase damping
- passive - largely material dependent. There are good charts around. Cast iron better than steel. Epoxy granite quite good for damping.
- active damping - largely outside the realms of a diy machine. Though "stepper dampers" were popular on DIY machines some years ago.
3. Avoid the frequencies when operating.

Some steppers drivers have systems to reduce stepper induced resonance. These seem to have largely supplanted the diy mechanical stepper damper solutions.

[GME]

Since the problem is centered on a relatively small IPM band, on way of addressing the issue is to simply avoid that speed band. Not my first choice, but maybe the most practical one at the end of the day.


So the exact same setup was on your old machine and no issues? I can not ever think of a time when I needed to cut at exactly 5 ipm.

Thanks, Bill. My current setup isn't even close to my old one. I am using extrusions that are twice the size of the old one, The gantry mounts and bracing are completely different. Linear rails vs steel plates and skate bearings. My old one was mounted on wooden torsion box cabinets which had 12 hockey puck leveling feet, vs my current leg setup, The pucks and cabinet construction probably provided vibration dampening. About the only thing that's the same as my old one is reusing the Steppers and R&P drive system. I don't believe I can draw any valid by conclusions by comparing old and new. Very different machines.

As has been noted previously, about the only time I should get into the 5 ipm range when cutting is during movements like direction changes. It may or may not be enough to be an issue to cause problems, but if I can reasonably solve the resonance issue, I plan to do it. I don't pretend to know much about resonance (I'm learning), but a few things are predictable. Fasteners will be more prone to loosening, and moving parts will probably wear out faster. One could probably conjure up a long list of potential list of horribles, but I doubt the problem is severe enough to warrant it.. Still, I owe it to myself to make this machine as good as I reasonably can.

Gary

[GME]

Hello Gary - Every part of the machine is vibrating. It's just one part or maybe two parts are vibrating in sympathy at that speed. A stepper motor is a great resonator as it steps along. If you have a long steel rod, even a long screw driver put it to your ear ( or jaw) and place the other end on the motor and then on the mount and start to work away from the motor. using the rod as a stethoscope you will be able to hear the various machine parts singing. You should be able to hear the major vibration and work towards it. It will be something that's thin and cantilevered, or eared . Something that can vibrate freely. The tubing and major structural parts will vibrate at a lower (I think) frequency than what you are getting. But I'm sure using the rod, you will track down the culprit. Then you need to clamp it, damp it, or change it somehow so it sings at a different vib. Peter

Thanks, Peter. I would never have come up the the "stethoscope" test method for tracking down the the "singing" part(s). I appreciate the suggestion and will try it. I just happen to have a 5/8" x 3' long solid steel rod in my shop. One thing may be an issue. I have some hearing loss and wear hearing aids. Hopefully, if the parts sing at a frequency I don't hear so well, I can still "feel" differences in the vibrations through the rod. I'll just have to see how it goes. I report out on the results.

Thanks again,

Gary

[GME]

Adding the multiple layer spoilboard may help dampen the frame enough to get rid of it.

Thank you for weighing in, Ger. From some of the reading I've done, it sounds like both plywood and MDF have vibration dampening properties. Most of what's written is so far over my head that I'm not able to really appreciate how much dampening to expect. I'm encouraged by both plywood and MDF being used for speaker boxes because of their dampening properties. Anyway, I hope you're right and the spoilboard will be enough.

Gary

[GME]

Ways of reducing resonance:
1. Raise the natural frequency above the operating range
- increase stiffness
- decrease mass (without compromising stiffness)
2. Increase damping
- passive - largely material dependent. There are good charts around. Cast iron better than steel. Epoxy granite quite good for damping.
- active damping - largely outside the realms of a diy machine. Though "stepper dampers" were popular on DIY machines some years ago.
3. Avoid the frequencies when operating.

Some steppers drivers have systems to reduce stepper induced resonance. These seem to have largely supplanted the diy mechanical stepper damper solutions.

Thank you, pippin. Useful information.

I'm not sure how I could reduce mass, so that probably isn't an option. However, increasing stiffness is. I still have the aluminum gussets to add, will increase rigidity. My concern is that I might raise the frequency to the more normal operating range and make things worse. On the plus side, if the gussets exacerbate the problem, it will be easy enough to remove them. So far, adding the horizontal braces on the side did reduce the resonance at 5 ipm without moving up to higher speed area. That's encouraging.

As ger21 mentioned, the spoilboard may add enough dampening to solve the problem. One can only hope.

ger21 earlier mentioned a stepper drive that was reputed to be smoother. Before that, I was unaware that drives were being sold with that enhancement. I could go that way, but would have to rebuild my electronics box to make the transition. On my current box, I cut out part of the top of my box and added a 15" x 4.25" x 1" high heat sink for my Geckos. The heat sink is open to ambient air. Although I have a fan in my box, it's exposure to ambient air that makes the arrangement work. It's a method I used when making a semi automated brewing control where I was using 40 amp SSRs, which generate way more heat than the stepper drives. The heat sink on the brewing system only gets a little warm while powering a 5500W (about 23 amp) heating elements, sometimes 2 elements going at a time.. Anyway, I'd need a new box and new design to go with different drives. If necessary, I'll do it.

Thanks again, pippin,

Gary

[wmgeorge]

Gary, I guess I should have clarified my statement above. Your using the Exact same control system right down to the stepper motors, drives and settings. You never had an Issue on
your old machine using the above, so its not the settings its your new machine construction. I hope that clears things up for you.

[GME]

Hello Gary - Every part of the machine is vibrating. It's just one part or maybe two parts are vibrating in sympathy at that speed. A stepper motor is a great resonator as it steps along. If you have a long steel rod, even a long screw driver put it to your ear ( or jaw) and place the other end on the motor and then on the mount and start to work away from the motor. using the rod as a stethoscope you will be able to hear the various machine parts singing. You should be able to hear the major vibration and work towards it. It will be something that's thin and cantilevered, or eared . Something that can vibrate freely. The tubing and major structural parts will vibrate at a lower (I think) frequency than what you are getting. But I'm sure using the rod, you will track down the culprit. Then you need to clamp it, damp it, or change it somehow so it sings at a different vib. Peter

I tried the "Stethoscope" thing. I could hear/feel some differences, but am not sure what do with what I learned.

As Peter suggested, I started with the motor. As expected, the greatest amount of resonance was in the stepper itself - it sang the loudest. The R & P mount came i second, but didn't sing quite as loudly as the stepper. I moved to the gantry mounts. A little less noise than the steppers, but not by a lot. Close to the song of the R & P mount. The gantry extrusion sang close to the same as gantry mount, Maybe a little louder than the gantry mount, but no more than the R & P mount. The difference were fairly subtle. The frame and legs were pretty much consistent everywhere. I could hear them, but they sang quieter by a large margin. The linear rails were about the same as the frame.

Conclusions

-The steppers are setting up the resonance.
-The extrusion frame members are dampening the resonance.
-The gantry extrusion is singing close to the same as the stepper

I am sure that filling the gantry cavities with something like lead shot would have a significant, if not total, dampening effect on the extrusion itself. Unfortunately, lead shot would add around 200# (quick & dirty volume estimate) to the gantry, which is a lot. The gantry beam deflection would only be about .0014" according to the 80/20 toolkit, which is amazingly low. 50# of sand gets deflection down to .0008" Either one would require taking the gantry off the machine and machining caps on the ends to contain it. I'm not excited about the prospect, but it's doable. I read that some folks have considered expanding urethane foam, but there's not much written about it. I haven't read about anyone actually doing it. Even if foam would adequately dampen the gantry, finding a rig to fill to the center of a 74" beam could be problematic. In normal home construction applications, installers don't need such a long reach. I've seen some guns with 2 foot long nozzles that use the canned stuff, but it's questionable whether 2' is long enough to fill to the center of my 74" gantry beam.

I guess I'm back to thinking about 1. how to reduce the stepper resonance, since that's where it's most prevalent, or 2. avoiding the speed producing the resonance as much as possible.

Changing out the stepper drives may solve or reduce the problem, but at the cost of new drives and a new control box. Servos, like the Clearpath would probably solve the problem, but they are outside my budget right now.

Avoiding the 5-6 ipm speed as much as possible will probably be the way to go for now.


It will be interesting to see how the steppers behave when I pull the motors to make the videos for Gecko. I'm expecting them to vibrate around on the bench when testing, but we'll see. I'll really be stumped if they don't. Depending upon how they act, I may take another shot at tuning the 203Vs off the machine and see what happens.

Gary

[GME]

Gary, I guess I should have clarified my statement above. Your using the Exact same control system right down to the stepper motors, drives and settings. You never had an Issue on
your old machine using the above, so its not the settings its your new machine construction. I hope that clears things up for you.

I understand. Thanks for the clarification, Bill.

Yup, no changes with the control system from old to new. The problem isn't there. Unless I'm missing something, the steppers are setting up the resonance. At 5 ipm the steppers themselves exhibit the greatest amount of vibration.

Gary

[davida1234]

-The steppers are setting up the resonance.

50# of sand gets deflection down to .0008" Either one would require taking the gantry off the machine and machining caps on the ends to contain it.

I read that some folks have considered expanding urethane foam

Probably.

Could also be wear on the gear bushing. Maybe conical wear towards the ends which lets the gear "cant" but difficult to feel as the middle portion is still tight.

In terms of sand, you could drill small (1/4 or so) holes and funnel the sand without gantry disassembly. I would expect it will dampen the resonance nicely.

The expanding foam will definitely not do anything at all. It doesn't have the mass, the structural strength or the adhesion to have any effect.