LinkBack URL
About LinkBacks
Are the red part aluminum? If they are it seems a solid build. Not light duty at all. Maybe some offset on the gantry to make full use of the cutting area?
| | |
I have not had a chance to get my solidworks fix lately, so it was back to the lappy.
Here is an idea I have had running around my brain for a bit. It's a lighter duty 2x3-ish table made to be super simple and be easy to build with just fasteners. It will run on 1/2 ACME with CNCRouterParts nuts, can use 1 or 2 steppers on the X, and have a really low parts count. I think it could be assembled in an afternoon if the risers and 4 end caps were already cut.
It's incomplete obviously. Some 80/20 connecting the two main beams will be needed, and this might operate better with a carriage. Also the two risers will most certainly go thru a couple dozen revisions before it's all done.
Thoughts?
Are the red part aluminum? If they are it seems a solid build. Not light duty at all. Maybe some offset on the gantry to make full use of the cutting area?
I would extend the risers a few more inches but below the table a little so that you can join the two risers below the table with another piece of 8020.
This would help rigidity significantly and allow you to drive the X axis with a single stepper instead of two at the same time.
Just mount the X axis stepper on a bracket at the middle of the cross pieces you haven't added yet. This also simplifies the end cap design.
Gary
Crane, this looks a lot like the one I build in this thread. 80/20, THK, CNCRP
My build is a little larger but not by much, 38"x48". It works fine with one screw on the X-axis. 80/20 is great stuff to work with but more expensive than MDF for light duty. If you were to use .75" uprights and beef up the gantry piece by using 3060 or even 3030 then you can call it "Medium Duty", if there is such a rating.
You could also use the CNCRP parts such as bearing blocks/ stepper mounts as I did.
Thank You.
I agree that the gantry beam is entirely too small. It needs to be 3060 or somewhere in that neighborhood. Could not be in more agreement.
For various reasons I really like the dual leadscrew design. One motor could be used with a belt no problem too. I will work on this model a bit later tonight and make some more progress.
Alex
The THK or in your case you show HIWIN blocks are quite sturdy and with a 1545 bottom cross member racking is not a problem. I have not calculated the trade off in cost to the two methods but I would be curious of the benifets over 2 lead screws as oppsed to one.
Thank You.
I'd use 1545 for the gantry vs 3030 because its almost twice as strong in this application + cheaper and lighter weight if your going for that. I think 3060 is severe overkill for this unless you just want the weight.
Nice design by the way
Here she is at the moment. I have made a few modifications and tweaks, but it still has a long way to go.
Here are a couple things I think I am going to stick with:
* No cross member underneath. Don't think it's necessary.
* Still need L/R supports. Got a good idea for this, just have not got to it yet.
* Design still allows for 1 or 2 X motors. 1 motor can be used with a belt.
* Decided to go with a 3030 for the gantry, as well as made the under supports 3030. This makes things simple.
* Never been a fan of the 90* THK rails (yes, the model is HIWIN, but it easier then spending 4 hour just making bearing blocks), but after playing with it I am starting to like this.
Stay tuned.
The design continues to get revised. I have landed here for the end caps for, and am satisfied. They might get tweaked in the future, but for now I'm leaving them. I am still working on the risers. As it sits now there is too much wasted space from the risers/router offset. Most of my effort as of late has gone into the carriage.
Someone mentioned earlier about offsetting the gantry and bearing blocks- thank you. It was definitely needed.
Well, your drawings look awesome, and the machine looks like a modern race car.
However, you did come here for critique, right ?
A few things. The machine is small enough to not really need two screws. Add up the cost of that extra screw drive.... Just off the top, say a $40 motor, $40 bearings, $10 coupling, $125 Gecko, $150 Kerk screw and $50 Kerk nut.... You end up with quite the money in just that ONE aspect, and, you add the complication of using 2 software axis in the control. I'd rather stick that 4th axis money into an actual 4th rotary axis !
I would simply run one screw and one motor underneath the center of the machine. You CAN build a gantry that will NOT rack easy enough. Dual screw or dual sided drives are perhaps much more necessary on gantries that are not exactly rigid, or machines that are just too wide to expect a single drive method to do the job effectively.
But the design change is a bit deeper. Now that you have SAVED $400 on not using all that extra stuff, call a Mic6 supplier and buy yourself a piece of 1/2" thick aluminum tooling plate for the table top. It will be less than $400, and you will end up with a MUCH better machine when done. As it stands in the picture, your machine has no "backbone".... if that is only a piece of wood for a table top, your whole machine really depends on whatever cabinet you set it on to keep it square, true and straight.
Therein lies the advantage of that tooling plate. The plate combined with the extrusions backing it up will give that machine a real backbone. It will "carry itself" a lot better than the current design. On top of that, your bearings will last for years and years because they no longer act as a SHELF for all the dust that is going to fly off the spindle IF you mount them under the table edge.
So back to the single screw....
Now that you have a solid table top (and you'll love mic6's specs for twist and straightness), put a few short legs under those extrusions so there is space under them and whatever you are setting this machine on. Use a 1/2" thick piece of tooling plate to run just under them from side to side.... Make this plate as wide as your gantry uprights and let your gantry uprights sit ON that plate. You will NOT have any "racking" if you are pushing in the middle of a flat plate with that width of machine. If you were not using true linear components, this would be a different story, but true linear is too tight to "rack".
Now, about that Gantry extrusion. I've seen this tried by a friend. He had, has an awful time fully pulling the linear rails to the extrusion without binding. The two locations, one on top, the other on the side seems to make it even more difficult to get it straight. Thats because while extrusions are wonderful, they are not as straight as linear rail CAN be when it is mounted to something like...... tooling plate ! In my opinion, and I bet my friends, We would suggest using some 5/8" - 3/4" thick tooling plate across the back of the gantry with the rails screwed in parallel along the front or back, depending on what you can devise for screw location.
I realize that I suggest some major differences. But, I usually stay silent, but all too often, I do see people "spend the money" on the real stuff (true linear rails and bearings), but then introduce design issues that do not support the accuracy that true linear bearings are purchased to provide......
Many would do well to get a price on Mic6 tooling plate to use for a rigid top. When you call a supplier (I use Howard Precision Metals), you really need to ask about "drops" or "cutoffs". They all have various sized pieces already cut some size for less money than if you just ask them to whack a chunk out of a new sheet. If I were to guess off the top of my head, I'd think something in a drop around 36" square and 1/2" thick might run about $300. WORTH EVERY NICKLE if you use it as the straight and true baseline for linear components. You also eliminate "bounce" and "ring" as well as add MASS, something a CNC machine can not get enough of.
But.... I really do think you did a bang up set of drawings there. I have attached a photo of a small engraver I built some years ago. You should be able to see that the linear rails are right under the outer table edge. This machine is small (12" x 15" x 5") but is rigid as you can get in its final form. I don't think I spent more than $200 on tooling plate for it. Again, worth every nickle because it is very real, straight and true.
Chris
Chris L
There are certainly advantages for using a single lead screw (cost being the main one) but I think I am going to stick with the dual screw in this case. There is enough about them that I really like, and its a major enough part of this particular design that warrants it stay. If I was going to use a single screw design (which I still probably wouldn't) my design would probably be fairly different from the get-go.
The Gecko G540 has 4 outputs, having them configured for an A axis really is not that difficult- I could probably set it back up from scratch in 10 minutes. No worries about complexities there, cause it's really not.
The first goal of this was to see how easy 80/20 could make a project like this. Next is just the need to be constantly creating stuff. I think there is a void between high end machines and your typical DIY home build. I would like to bridge that gap with this one and have a professional level build within the realm of your average Joe. If a guy were to have some of the parts water-jet cut and then be able to bolt those parts onto extrusions- there is value in that.
Posting Permissions
| | |
