View Full Version : Opinions/Advice on my 800 x600 design
07-07-2015, 09:37 PM
Hi moving on from my first MDF fixed gantry build, I'm building again. Really upgrading. I'm learning a lot about some of the limitations of my MDF build an this time I want to build a moving gantry machine so I cut larger workpieces. I plan to use 15mm plate aluminium throughout. I think that should be strong enough? I already have a chinese 2.2kw spindle which I just love. Also plan to use SBR20 fully supported rails and RM1605 ballscrews. This is a big financial investment for me so I really want to get this right. Here's some pics; I'd be grateful for your opinions.
If you look close at pic2 you will see that I'm planning to be able to adjust the height and parallelism of the gantry. I think this could be useful in the future if I need to accomodate larger items under the gantry. Do you thing this would have significant impact on rigidity? Maybe I should use 6 bolts instead of 3?
I already have nema23s and the electronics from my old MDF machine. I've read some good stuff on chinese ballscrews and they seem viable. If they're anything like the spindle, they'll be a winner. Even if they dont last, they dont cost too much anyway and I might go with the pricey ones then.
Finally for those who use sketchup here's s link to the design and parts layout.
08-07-2015, 04:33 AM
Looks great I like the supported rails although what are you planning on fixing them too. It would need to be something quite sturdy and how would it be connected to the main bed of the router.
The other concern I had was with the single central leadscrew of the moving gantry. I think it's pretty likely to lead to serious racking i.e. where there is lateral flex in the gantry. I built one like this and found it quite flexible mine was wood but I still think it applies to your build. I ended up using a pulley system to stabilize the gantry which works well but was an afterthought and was tricky to get right needed a seriously tight steel cable. It seems now most people just use two leadscrews with twin slaved steppers. Given the pretty high spec construction of your build it seems worth the extra costs Good luck with your build.
08-07-2015, 08:03 AM
If you look close at pic2 you will see that I'm planning to be able to adjust the height and parallelism of the gantry. I think this could be useful in the future if I need to accomodate larger items under the gantry. Do you thing this would have significant impact on rigidity? Maybe I should use 6 bolts instead of 3? Dave I don't normally quote on system design but I do see a few things here and it also depends on what you want to cut on the machine.
I would not split the gantry with the two ends as it is very weak as it is already if the budget allows use 20mm end plates and put a couple of box sections across also have you thought about how you would put the bolts in to the bearing blocks and rails etc.
20mm linear rails are far better than the type you are using and they can be got from China with the screws.
With a small machine like this and designed correctly I think you would be alright with one central screw. Keep up the good work.:welcoming:
08-07-2015, 11:54 AM
Thanks for the feedback folks.
shoeswith: Sorry but its not clear from the pics. I already have a VERY heavy and sturdy bench. Its got 10mm plate steel sitting on top of 75mm of hardwood joist. You could run a railway over this thing. The plan is to drill and tap the rails from underneath so the rails and the bed are permanently fixed to the bench. I guess that I can use a 'suck it and see' approach to whether I need dual ballscrews. If it turns out I do It won't be a huge deal.
Clive: One of my constraints is that I already have a 2000mm x 1000mm piece of 15mm alu plate. I could double it on the gantry sides, I certainly have enough of the plate to do that. I certainly take your point about the horizontal box sections.
The supported rails in the design are going to be 20mm. Do you mean a different type of rail would be better? This machine will be for 3d profiling/ cutting wood, cutting aluminium and *possibly* drilling mild steel. Its all hobby stuff. I don't need micron precision, I just need it to be better than what I got. e.g I did a complex profiling op last weekend. The machine whirred off for about 30 minutes after I finished and sent the machine back to zero, the cutter was close to 3mm away from where it began. I'm putting this down to backlash and flexure etc. That's what I'm looking to avoid with this machine.
Also my existing machine, being a fixed gantry only allows me access to around 400mm x 300 mm on the table. I found this to be too limiting. I'm delighted I went with the MDF first and would recommend any first time builder to do so as your mistakes wont be too costly. Imagine if I had cut up that 2m x 1m 15mm plate and then realised I'd made some serious mistakes!
Thanks again for the feedback. I'll certainly start a build log for this one.
08-07-2015, 01:08 PM
Well if you think your bench is flat and true go with it but you certainly won't get any accuracy at all doing that. These are the type of rails
good luck with the build.
08-07-2015, 01:53 PM
As you've designed it you'll not be able to get to the bolts for the X axis linear bearings, they'll clash with the gantry uprights the same is true possibly on the Y axis where the Z axis rail mounting bolts might go. I had to re-design mine a few times because of this, it looked good until I realised I couldn't put it together. There's still a few things I'd do differently next time (yes there will be a next time!!) which I've found out from experience in building this one!! Your Z axis looks to be relying on the linear rails for strength? Why not use a nice slab of the 15mm plate you've got? The rails we'd normally suggest to go on the front plate of the Z axis as they add a bit more strength.
I'd also suggest getting 1610 ballscrews for the X and Y axis as they'll give you better speed, if you connect to the ball screw via pulleys and belts you'll a) reduce any possible resonance getting back to the stepper and also give yourself flexibility to cut wood at a decent speed and also gear down if you need the accuracy.
As Clive says, if you can stretch to profiled linear rails they you'll be upping the machine to a whole new ball game, they're quite cheap from China now, I got mine second hand off eBay from Korea. One thing you'll need to make absolutly sure of, as Clive says, is that your table is totally flat, i.e filling and scraping using a straight edge as reference to get a really flat surface, especially if you go with the profiled linear rails.
08-07-2015, 07:13 PM
from all what you say and having in mind you have steel top and 2000x1000x15 aluminum plate at hand, if i were you i would start again with some parts of the drawing.
For example make gantry back plate one whole piece, not 2 pieces that will bend like crazy, use the stiffness calculator found here in the forum to confirm and strengthen further if needed. Same with Z. Maybe read #8 here (http://www.mycncuk.com/threads/8778-Driving-the-gantry-on-Z-axis?p=70040#post70040) where i explain much better Z design with the same effort
Generally speaking better look all the similar builds as there are much better designed machines from aluminum plates and similar rails.
About the rails, i assume you know already and they told you - square rails are the way to go.
09-07-2015, 11:50 PM
Thanks Neil and Boyan.
. (see i learned the dot trick:friendly_wink:)
Really good spot on the x axis bearing bolt access Neil. I hadnt noticed it.
I can certainly take into account most of the advice kindly offered. I will certainly double the gantry sides and make the back plate a solid piece as recommended. I'll still have a head of 15mm Alum over, so I'll probably use it to double up wherever it will add extra strength. The bed supports being an obvious one to me. I'm sort of stuck with the supported rail for the mo as I already have some, found them locally at a price too good to ignore. However a bit of redesign, so I can easily replace them with profiled rails at a later stage.
I also take the point about the flatness of the base. The steel plate it sits on is certainly as flat as a straight edge can tell, however before I lay the rails, I'll give that bit extra care with a stone to remove any possible high spots and then clean it to an inch of its life.
As for the adjustable height gantry, I think I'll drop that idea. I'll adjust the height in the design until the gantry clears my milling vice with a bit to spare for protruding stock.
Boyan, I studied your comments regarding z-axis vibration and strength. That sound of the whole machine shuddering when you go down into aluminium is exactly the sound my existing MDF machine makes:) So in short if I widen the spacing between the bearings (vertically) and reduce the overhang as much as possible so that the spindle is as close to center as possible? Maybe again double that main block on the Z axis. Do I understand correctly? As you can see from my images, I already have one of those 80mm chinese spindle mounts from my existing machine, so maybe get a second one so that the spindle is supported top and bottom? I don't need to be concerned about weight as once this is in place it ain't ever moving:) However, I don't want overkill weight to affect the + travel of the z-axis. I don't think an extra aluminium block will make too much difference?
Clive S? I'm interested in your comment regarding accuracy? Can I ask what you consider to be accurate? I would consider my MDF machine to be reasonable accurate, insofar as it cuts to the right dimension as far as my calipers can tell. However where it loses out is if I do a longish complex run, it won't return to exactly the same origin, so clearly it isn't all that accurate afterall. For what I need I don't need micron accuracly but I would like reliability and repeatability.
Thanks again for all the input it is genuinely appreciated.
10-07-2015, 12:03 AM
no extra weight from one bracket is nothing for the z
On the design i like the rails move together with the plate adding stiffness. But in some cases its not necessary. In your case i believe you could benefit. On my design all depends on the HD long bearing block and type of rails. Search here in the forum there were very good design with rails like yours.
Powered by vBulletin® Version 4.2.3 Copyright © 2016 vBulletin Solutions, Inc. All rights reserved.