Hi all! I finished a video on how I built my steel base table for my Pro 4896 machine. I’m sharing the info in case anyone else wants to do the same thing.
The Fusion 360 model for the legs is here: Table Model – feel free to download it and tweak it for your particular setup!
Video:
Great video! I did something similar years back and may post that sometime. Course I don’t have any video. Stuff moves when you weld it. Sometimes bolts might be better?
Awesome! Do you have any pictures you can post?
Yeah, I am thinking of adding some more braces on some other feedback; I still get some shaking movement, particularly on the y-axis, and some people on Facebook think I can eliminate this by adding a lower brace and a 45 or two. I might just go ahead and tack weld it and not do a full bead, to help avoid movement. Or bolt it!
Yes I have pictures, however they aren’t digital. Will have to experiment with taking a picture of a picture.
I had a welder fabricate a table like that. I put jack screws around the perimeter. They are great for easily leveling the table. I have braces on it. It only moves when the gantry stops after manual rapid move. Never does when gcode is running.
Cool - did you do any tuning on the acceleration curves? Mine does still get more movement than I’d like…I’m going to add some 45s and another bottom brace. The screw feet I got are cheap ~$9 ones from McMaster Carr
No, I haven’t tuned the acceleration curve. How do you do that? The inertia from the gantry does bug me. A graceful slowdown would be nice.
I bolted the welded table to the floor, which my or may not have eliminated some flexing.
Jim did a good video on it:
I haven’t done it yet myself, but reducing the acceleration can definitely clean up chatter. His video is a bit long; I might try to make a shorter version when I end up doing it.
I went straight to the steel table, so I don’t have a direct comparison. I know a lot of people don’t have problems with the aluminum table, and beef it up by adding on MDF sides. I was pretty close to doing that, but thought it’d be better to weld up my own.
I do see some chatter in parts, particularly when the y-axis does a short fast move. I’m planning on adding a bottom lengthwise piece and some 45s, based on the suggestions of some other people.
I also plan on doing a before-and-after comparison of the mod; I’ll measure deflection on the table top when doing a pre-canned GCODE move using a dial test indicator, and then compare it to what I get after stiffening it up. I roughly do see up to .010" of movement at the top of the machine.
I built a steel frame similar to this three years ago when I first got my pro 4-8.
I have a couple of notes or comments just for reference to anyone considering this.
I ended up doing some similar reasoning as Corbin as far as hanging the cable track supports, etc. I am sure there are some improvements to the way I did mine, Hopefully there is some helpful information for anyone thinking of building there own frame, table and they can improve upon mine.
** I will try and dig up some pictures.
I did the structural welding at a professional welding shop I had access to. This is a picture of when I brought my table home from the shop. I had to finish up small details like putting the control box rack on and the bases for the feet before taking it to be powder coated.
I looked through my pics and don’t really have much of the build. This pic does show the overall structure though.
Here are a couple of pictures I took this morning of the overall configuration and showing the rails installed directly to the frame. You can also see in the pics where I had the control boxes mounted and then had to move them for better access when I moved my machine to a different location.
Awesome! That looks like some nice clean welding you did there! I did end up getting a bit more racking than I liked on the y-axis, and I added a few more supports. I forgot to follow up on this thread. I did make a video on it:
Maybe a compelling argument for cold connections of the gross assembly, and utilize welding to build components, such as brackets. Strategic use of cold connections would also maintain a primary feature of AVID systems that make them a consideration for so many: portability.
Corbin,
I have to start out by letting you know how much I appreciate your interaction in the community. I have gained an incredible amount of knowledge and experience through the videos and chats of people like yourself. I need to do a better job of giving back and contributing for sure, I am hoping to improve on my contributions going forward.
I watched your video on the mods you made. I had just a few thoughts on that, remember with these thoughts, I am not a machinist with extensive machining background, I am just an engineer/craftsman who makes a lot of things : ), Obviously, the stiffer the machine, the better for accuracy, but that all depends on what you are making and to what extent motion effects the things you are making. For me, the accuracy of my machine is as good as I will need it for what I make, one motivation for a stiffer machine for me might be to improve surface finish when cutting, at this point, I think I am satisfied in that respect as well, , I know of other things that are affecting the surface finish that vary and are less controllable. … As far as what I could tell from your machine, it seems like you have majorly improved it and I would definitely be satisfied. one thing to keep in mind, based on your test setup, you may have been measuring 10 thou movement, but the relative movement at that part/bit might not have been nearly that. does that make sense?, I am probably not describing it very well, but the gist of what I am saying is that if your machine head, gears, cross member, uprights, etc were infinitely stiff, you could still potentially measure much larger displacements at the end of the table where you were measuring, but in fact you may have had zero relative displacement where the tool meets the part. Just a thought.
One other idea which is pretty simple to implement is to bolt your machine down to the floor. When I first set my machine up, I drilled some anchor bolts into the slab underneath the machine and used two small cables and turnbuckles to pull the machine down tight to the floor. This essentially stiffened the machine up. I had some feet on the machine at first that were no good and the machine would actually “walk” after a lot of use, I have since become a better programmer for acceleration and have better feet, so that is not an issue anymore, but securing the machine to the floor was my first attempt to eliminate that movement, and it in turn stiffened up my machine.
Thanks again for sharing so much of your experiences. I really appreciate it and am positive many many others do as well.
rg
Thanks rg! Yeah, what you said definitely makes sense. I was seeing chatter on my parts, and I eventually realized it was shake on the y-movement when making 180 degree turns (ie: back and forth). The extra bracing did help. I think your 45’s that you used from the get-go were a good idea, and add a lot of strength to the y-axis.
I also like that you put the linear rails directly on the steel; I’ve heard that the aluminum extrusions will tend to “ping” a bit, and not absorb as much movement as it would when right on steel.
I’m not looking forward to moving my machine! The base weighs a ton (almost literally).
