I wrote the original stiffness v5 sheet that you mentioned in the first post a few years ago, plus later posted a v7 which may be of interest.
I've tried to replicate your results but cannot match them (don't know all the dimensions etc that you used). So I'll just make some general comments.
The 13um in Z is bending due to the spindle weight applied to the centre of the gantry. This seems fine to me.
The 30um in X sounds like you've only noted the 'tool deflection' number which is only the bit due to twisting. There will be a bit more overall deflection due to X bending to give you a final total deflection in X. This is starting to sound a bit high (30-40um is 0.03 - 0.04mm) for aluminium. However, this is at maximum extension so you can reduce this problem as you've stated by raising the job in Z.
Biggest thing to point out though is that you haven't run the analysis on the gantry sides (plates) in the Y direction. Depending on your exact numbers this is likely to be similar or greater than the X deflection and because it is just bending (you can ignore Y torsion) and there is no work-around in terms of raising the job in Z etc. This is because all the tool forces in Y go through the ballnut and this is at a fixed height above the X axis bearings. This is why Jazz keeps pointing you to the raised X axis designs for aluminium since this offset between ballnut and X axis bearings is basically zero. You are then free to beef up the bed/raised X axis sides to react this force.
I have a machine broadly similar to your posts so far and it will cut aluminium but I you can tell from the sound it makes and machine vibration that it is approaching the limit. If I was cutting aluminium more regularly then this machine would not do. For the occasional piece it is OK.
Finally, feel free to experiment with the spreadsheet but do just use it to get a feel for general comparisons and where to place effort and material. As a like-for-like comparison of different gauge, sections, and dimensions it will give you direction, but don't expect to get the quoted deflections out as there too many simplifications. So long as you use it as just another way of investigating the design (along with reviewing the build logs and the invaluable subjective opinions of experienced builders) then you will be OK.
Good luck with it and welcome.
Thanks for the info. The deflection was calculated only for one 100x50 5mm box section as in my first design, now there are 2 in an L shape.
This is meant to be a hobby machine so I honestly dont know how much aluminium will be cut on it but I do know I want to be able to cut so I will take the advice and modify the x sides to bring the y axis down.
Has anyone ever built the gantry with the carriages in a triangle formation? I am thinking if the 2 bottom cariages were 150mm apart and the top 2 were 250mm apart (from outside dimension) I could keep the same footprint and machining area?
As for rest of machine then like I say it will cut aluminium no problem and if it's only occasionally and your not wanting perfection on the finish then it will be fine. If your wanting to cut deeper and get high quality finish the machine won't be stiff enough.
Personally I'd just go for it and if you find your cutting more aluminium than anything then build a machine just for this purpose because while a machine for cutting wood isn't 100% for Ali the same applies in reverse to some degree. Can't beat purpose built machines and Jacks of all trades always fall short in some department.
The bed and workholding on any machine can be a pain and often under estimated how important it is. This is esp true on machines that are intended to cut multi materials.
Problem comes from how you hold it down and the type of work and how you machine the material. If your cutting or drilling thru then having a nice aluminium bed isn't much use because it doesn't leave a nice warm fuzzy feeling when you chop into it.! So you need some spoil piece anyway and often this needs surfacing each time you use it if any decent accurecy is required.
Now IME the best bed is a combination of both.? So nice stable material for Machine bed that is surfaced parallel to the cutter with either all of it or part covered with sacrificial spoil board. If bed is large enough like mine is then having half n half works well and saves swaping between and surfacing for jobs that need the accuracy of the base.
Last edited by JAZZCNC; 19-07-2014 at 02:18 PM.
Well ive started the redesign so no turning back now :-) and tbh I think I will mostly be machining metals more than plastics and wood so I may as well make it right. I have changed the z axis but that may well of mucked up my plans for the 4th axis as the spindle plate and rails come down now!
I will try and get the plan finished and post it up so you can all tell me why it wont work again!
Thanks for the help and advice
Here goes, My 4th design with base, The base is made mainly with aluminium 2" x 2" 1/8th thickness with some 1 3/4 bar inserted at the bolting points (not added in yet) I will also have adjustable feet, do you think it will be strong enough or does it need beefing up.
Also my redesigned gantry again! :-) there will be a 15mm plate across the back for the X stepper and pulleys to attach to but I thought I would post my plans before I start adding in all the details such as bolt holes etc.
As before, feedback good or bad appreciated.
Thinking more about the frame, I don't like it so maybe I will have to learn welding. My dad used to do it years ago and he's retired now so maybe I can get him onboard.
Where is the best (read cheapest!) place to get cut to length steel including mitre cuts or would I be better ordering full lengths and investing in one of these:
Its £100 but says it will cut steel, ally, wood and whatever.
Last edited by CharlieRam; 20-07-2014 at 09:52 PM. Reason: Steel frame maybe?
Another question, Do all these plates need to be 20mm, If I make them 15mm will it affect the machining ability substantially? I'm on a cost/weight cutting exercise!
350mm rear plate and bearing plates etc could be 15mm but wouldn't change the front plate.
Don't go chasing weight it will be your friend when cutting and if you need really high speeds with high velocity changes then put the correct motors on it and you won't have any trouble.
My X rails are 810mm apart, carriages are 250mm apart from outer dimension, Would I suffer from binding if I just had a central ballscrew or would you still recomend one on either side?
The Following User Says Thank You to CharlieRam For This Useful Post:
The normal convention for the Z axis is to have the rails on the front plate and the carriages/bearings on the back plate. This adds stiffness to the front plate which is where the axis most needs it. Also stick with 20mm plate if you can rather than 15mm. With regards to ballscrews two are way better than one, you undoubtedly would get racking with one central ballscrew at that width I would say.Neil...
BUILD LOG: Steel frame cnc router design/buildBy CraftyGeek in forum DIY Router Build LogsReplies: 110Last Post: 06-05-2015, 09:00 PM
By mart154 in forum DIY Router Build LogsReplies: 27Last Post: 19-09-2013, 04:48 PM
By mZeles in forum Projects, Jobs & RequestsReplies: 4Last Post: 27-12-2012, 07:17 PM
By Rogue in forum Gantry/Router Machines & BuildingReplies: 0Last Post: 12-11-2011, 04:02 PM
By MikeyC38 in forum Gantry/Router Machines & BuildingReplies: 12Last Post: 21-10-2011, 03:50 PM