I welded my own base for my Avid Pro 4x8, but I was still getting some flex. I improved it a bit by adding some stiffeners to the bottom, and it helped a ton. Here’s a video on the process, including how I measured the movement before and after to see if it actually improved anything:
Looks very solid. I beefed my standard Avid frame up with several 1" super refined MDF panels mounted to the extrusions. It dramatically cut the vibrations down but I did not do the detailed measurements like you did. I may try to replicate your measurement setup to see how close I come to your excellent results.
Awesome – I’ve heard of some people doing that with MDF and getting a lot more stability. Definitely report back about the movement you get - I’m super curious!
Nice video as usual! I’m surprised at the amount of flex you were getting. Your base is now pretty heavy and large. Hope you don’t need to move that machine any time soon.
@corbin, I agree with @gordo there, that seems oddly large. Is there any chance there are adjustable feet on the bottom of this and the whole structure is moving/translating on industrial thread fitup/slop rather than flexing? I wonder if this could be check by putting an indicator on the top and bottom of the leg at the same time and reproducing the flexure test you did in the video. Then go back and look at a video capture of the new test so we could tell if both the top and bottom are moving at the same time and that this isn’t flexure?
Just thinking out loud there, maybe I’m going down the wrong path…
That is a good idea! I did do this test early on when I noticed the problem, but I didn’t record it on video. I saw twice the amount of flex at the top of the table compared to the bottom, so that told me it was the table flexing and not movement of the table itself.
It would still be an interesting test to do just to see if everything moves in sync, or if the table still has some flex at one point or the other.
I haven’t done any real projects on the CNC since I welded up the additional supports; the real test will be to see if I get better cuts. On some specific projects, like my spoons, I would see chatter marks from y-movement – I’m hoping that will be gone, but we’ll see!
Here are a couple. All panels are 1" super refined MDF which is much more dense and heavier than standard MDF. There is a full length spoilboard top, a middle and lower shelf running to the full height vertical board. Also 8" wide stingers are on the bottom perimeter. The back spoilboard piece is removable to gain access to the vertical board. I use the bottom shelf for lumber storage adding a lot to the total weight. I have no objectionable movement even on rapid moves up to 900 IPM with a 50 acceleration set. I don’t do any laser work so it may not be good enough for that but it is fine for routing.
Thanks for sharing those. I wish the 4896 model had those intermediate extrusions halfway down the legs as well, it looks like ithat plus your additional mdf would help rigidity quite a bit.
I hadn’t really considered this until seeing posts from a few others, (I had nearly forgotten). Not for this reason, but a couple years after buying the machine from CNCRouterParts/AVIDcnc, I went back and bought additional extrusion material. With this material I added a lower stringer to the middle set of legs and then added a number of 80/20 cross members between them and finally tied them in with 3/4in MDF. In addition there is a good amount of mass (steel stock) stored on those shelves. Not huge, but definitly adding some inertial mass. Without considering it from that perspective, I’ve likely been getting a small boost in rigidity from this for the past ~3yrs or so. The image below are of a few of the additional extremely simple gussets I added between the new cross-members/bottom shelves and upright legs.