[dragon_sa]

Hi
I just received my CNC machine (see attached picture) and have been going over it, I have discovered that from the side of the bed to the middle going across the bed, that the bed is not flat and dips down to 1mm-1.5 mm lower than the sides, the manufacturer said that I should use the vacuum bed and the dip in the middle will not be a problem, this doesn't ring true to me, so would like some advice on this issue from some cnc gurus weather or not the bed is meant to be like this for vacuum bed setup, or meant to be dead flat all the way across.
A when I am using clamps I have found I now need to pack the work up level.

Cheers

[routalot]

If you are only cutting panels from sheet materials,the suppliers advice is not really wrong.It still doesn't explain why they couldn't do better with their assembly.Is this your first experience with a CNC router?I would expect that your first job should be to place a spoilboard on the machine and skim it flat;by which time you need to be absolutely certain that the gantry rails are absolutely flat and not humped or dished because any variation there will be echoed in the face of the spoilboard.Is there access to the underside of the machine to allow for remedial steps?

If you ever engage in 3D machining and the job isn't supported on a reliable datum surface you have a good probability that the sagging will allow the stock to distort and while the trajectory of the toolpath will be as commanded by the program,the shape of the project sitting on a flat surface subsequently may not be as intended.

[dragon_sa]

If you are only cutting panels from sheet materials,the suppliers advice is not really wrong.It still doesn't explain why they couldn't do better with their assembly.Is this your first experience with a CNC router?I would expect that your first job should be to place a spoilboard on the machine and skim it flat;by which time you need to be absolutely certain that the gantry rails are absolutely flat and not humped or dished because any variation there will be echoed in the face of the spoilboard.Is there access to the underside of the machine to allow for remedial steps?

If you ever engage in 3D machining and the job isn't supported on a reliable datum surface you have a good probability that the sagging will allow the stock to distort and while the trajectory of the toolpath will be as commanded by the program,the shape of the project sitting on a flat surface subsequently may not be as intended.

HI..yes this is my first CNC router, the sides sit true and level, yes I can get under the machine, 3 of the side panels come off, should I just go through and shim the center rail at the fixing points of the bed?

[ger21]

Does the tool follow the curvature of the table? If not, then you'll need to shim and surface the table so it's flat. If not, you will get uneven cuts.

[routalot]

In the OP's position I would have no hesitation in shimming the table and while perfection is desirable,very minor discrepancies will be taken out when the spoilboard is skimmed.I would also let the supplier know that he has room for improvement in his manufacturing and that you hope not to discover any other problems.It looks like a promising machine in the image file-good luck with it.

[he1957]

As above. Ideally the table should be dead flat and level to the spindle across its travel surface. That dip of 1.5mm is a lot. If that surface is a rubber based compound it may also compress a little under vacuum. If you are using composite materials such as sign sheet (aluminium bonded to a rubber center), the aluminium thickness is about 0.3 mm so you can guess what results you'll get. Engraving will cause annoying problems with inconsistent cutting depths of only a few poofteenths of a mm.

As suggested, advise the manufacturer their tolerances could be a lot better and try shimming the table to get it as flat as possible. Also look at the construction of the vacuum chamber - perhaps there is minimal support between the top and bottom (guessing it is some sort of box arrangement) so shimming from underneath may still result in some downward deflection of the upper surface under vacuum.

As mentioned a spoil-board (like MDF) sealed around the outer edges and then surfaced flat should allow vacuum to pass through. You can cut slots to access your T-slots rails.

Edit:

After spindle tramming and table shimming, consider planing the rubber base surface flat (which is not unexpected before first use and is a one-off operation) and before adding a spoil-board :-)

[ZASto]

On these chineese machines, there is no box arrangement of a vacuum chamber. Vacuum hoses are directly connected to suction ports on a table.
Tabletop is laid on a crossbeams made with square profiles and in the middle of the worktable these should be supported from ground. Afterwards the table should be shimmed.
Don't ask me how I know that

[dragon_sa]

I should be able to shim it reasonably easily unless its glued down, as you can see in the cross struts under the machine, so I will give that a go...the bed does appear to be a very hard rubber, its all 1 piece and been routed out on the machine, but I dont think it was done with the machine.

The vacuum bed runs with 2 x 2.2kw vacuum pumps, vacuum ports are routed into the lower part of the bed as you can see in the pics also, the bed is broken into 4 vacuum sections which I can turn any of on and off, they also supplied rubbers to go into the grooves of the bed for isolating sections.

I wondered how the vacuum bed worked with a spoil board, what is the recommended thickness mdf sheet for that?
Do i use the rubbers in the grooves with the spoil board?

Does mach3 have a surfacing function or do I need to setup a job in my cad/cam software?

I dont have a surfacing tool, what would be a good one to get without spending hundreds of dollars?

The cuts I have done so far are not too bad as I work out feed rates etc..

The machine has delta servos and drivers and runs of an ethernet controller and uses a 3kw spindle.

Really appreciate everyones help and feedback its been great thank you.

[dragon_sa]

On these chineese machines, there is no box arrangement of a vacuum chamber. Vacuum hoses are directly connected to suction ports on a table.
Tabletop is laid on a crossbeams made with square profiles and in the middle of the worktable these should be supported from ground. Afterwards the table should be shimmed.
Don't ask me how I know that

You are correct...but unfortunately the square bars cannot be supported from the ground as the spindle rail bar would hit any supports that went to the bed

[routalot]

I would choose a 12mm MDF spoilboard with a flat surface to mount it on.Sealing the edges with a coat or two of shellac sealer helps to reduce the leakage a bit but the MDF normally allows for a good hold on the workpiece.I have never used Mach 3 and have no idea whether it has a surfacing function built in.I have skimmed spoilboards by either hand coding-a subroutine call with instructions to travel the length of the table and back however many times it takes to cover the whole area is easier than typing the whole thing-or alternatively use a CAM program to generate a facing toolpath.A tool similar to this will do the job https://www.ebay.com/itm/CNC-Spoilbo...I3mp:rk:2:pf:0 It only needs a very light cut maybe 0.6mm maximum.

If, as it appears,your machine has a single lead screw running beneath the table you shouldn't get too aggressive with any cut as there will always be a possibility of skewing the gantry. The need for a beam travelling beneath the table also means you have a risk that the table might sag if you need to work with anything as heavy as say 50mm MDF.

[Gary-Wiant]

If you put a spoil board on the top you should look into LDF (low density fiberboard) it will be more porous.

I never had the same machine you have but I did have a machine similar that only allowed the table to be supported on the ends & I had MAJOR issues with rigidity.

Good luck & don't push it

[he1957]

I use either a 25 mm or 32 mm 2 blade router bit for surfacing. If I could fit my MT3 50+mm carbide insert mill facing cutters into the ER25 collet chuck I'd use those but my spindle won't go at less than 6k RPM (yet). The wider the bit the faster the process but with the 25mm bit and DoC of 0.015 mm over 1200x2400 mm it takes about 30-40 minutes with 65% stepover at 3000 mm/minute. As mentioned, take as little off as needed and repeat (gets the higher travel feeds too). The better the spindle tramming the better the finish and minimal tram/witness marks.

For standard MDF, you could skim both sides to increase porosity. The outermost factory layer finish is denser than the core. The rubber fill strip goes into a slot closest to the outer edge but still under the board as needed to keep vacuum under the material being held. You want as much vacuum under the part as you can get.

Mach-3 has some wizards which include both pocketing and surfacing but I don't use them either. I use a simple minimal DoC pocket operation in my SimplyCam software over the area to surface (often a secondary layer spoil-board clamped to the main for which I use screws with to protect my main spoil-board). This is also what I'll use to pre-prepare a top surface to get it flat before carving or engraving (if I can).

If you have a touch-probe, there is software that can be used to generate DoC compensation (auto-leveling) for table or material thickness variations, see: G-Code to DXF with G-Code Ripper | Scorch Works Blog

As mentioned, consider surfacing the rubber surface after shimming/tramming but as also mentioned by others be mindful of table sag with heavy work pieces.

[mactec54]

I should be able to shim it reasonably easily unless its glued down, as you can see in the cross struts under the machine, so I will give that a go...the bed does appear to be a very hard rubber, its all 1 piece and been routed out on the machine, but I dont think it was done with the machine.

The vacuum bed runs with 2 x 2.2kw vacuum pumps, vacuum ports are routed into the lower part of the bed as you can see in the pics also, the bed is broken into 4 vacuum sections which I can turn any of on and off, they also supplied rubbers to go into the grooves of the bed for isolating sections.

I wondered how the vacuum bed worked with a spoil board, what is the recommended thickness mdf sheet for that?
Do i use the rubbers in the grooves with the spoil board?

Does mach3 have a surfacing function or do I need to setup a job in my cad/cam software?

I dont have a surfacing tool, what would be a good one to get without spending hundreds of dollars?

The cuts I have done so far are not too bad as I work out feed rates etc..

The machine has delta servos and drivers and runs of an ethernet controller and uses a 3kw spindle.

Really appreciate everyones help and feedback its been great thank you.

You should post who the manufacture is of this machine, what I see under this machine is alarming for this size machine a single Ballscrew for this size machine is a bad idea

From lack of table supports is why the table is not flat, when you machine plastic like this it is going to move all over the place, they machined this top after they assembled the machine

Adding a spoil board is your best solution, and yes you will need rubber seal / gasket around the outside edge of the spoil board or how ever it works out with the slots to your sheet size

I don't know how the spoil board is going to work out though, because once you turn the vacuum off the spoil board could move, you will want to mount it some way so it can't move after it has been surfaced and would have to use the vacuum every time you use the machine, not a great idea having to use the vacuum every time you want to cut something

You may be better off to take this table off and attach a 2 layer MDF table top

[routalot]

I would just screw the spoilboard in place with enough brass screws to prevent the MDF sliding around.Brass because when-not if-you make a mistake it will do the tool less harm.For cutting sheet material,I wouldn't bother with a rubber gasket as the sheer area allows the sheet to remain in place by virtue of it's surface area needing relatively little vacuum to generate a large clamping force.

[dragon_sa]

Hi All

Thank you for valuable information...Im a complete newy to these machines..my main products will be mdf and acrylics....template and cabinet based....maybe some signage..I have cut a few things and it seems to be ok apart from me working out feed rates and depth of cuts...correct cutting tools....actually lots of things

The company is printed on the front...Acctek...the machine is completely hand made and you can see...there are lots of little...hopefully small issues to deal with...

I should have room for some counter sunk screws in the mdf on the sides as the bed is wider and longer than the cutting area.

You have brought to light the fact that the machine bed is only suspended from end to end...which is something I never considered/thought of...before making the purchase.
I think if it is unable to handle the materials I want to cut ( mdf, laminated boards, acrylic) I may look at reinforcing the beams with an extra steal c section welded to each side to increase the span weight capacity...depending on cost and current performance aswell....I may be able to add an extra servo and ball screw to control the y axis..spread apart to the sides..possibly allowing the cutting of the sliding beam.

Based on your feed back I am going to shim the bed on the rails...surface the rubber and add a 12mm mdf spoil board also surfaced and fixed down on edges beyond cutting capability.

cheers all still open to any more feedback.

[mactec54]

Hi All

Thank you for valuable information...Im a complete newy to these machines..my main products will be mdf and acrylics....template and cabinet based....maybe some signage..I have cut a few things and it seems to be ok apart from me working out feed rates and depth of cuts...correct cutting tools....actually lots of things

The company is printed on the front...Acctek...the machine is completely hand made and you can see...there are lots of little...hopefully small issues to deal with...

I should have room for some counter sunk screws in the mdf on the sides as the bed is wider and longer than the cutting area.

You have brought to light the fact that the machine bed is only suspended from end to end...which is something I never considered/thought of...before making the purchase.
I think if it is unable to handle the materials I want to cut ( mdf, laminated boards, acrylic) I may look at reinforcing the beams with an extra steal c section welded to each side to increase the span weight capacity...depending on cost and current performance aswell....I may be able to add an extra servo and ball screw to control the y axis..spread apart to the sides..possibly allowing the cutting of the sliding beam.

Based on your feed back I am going to shim the bed on the rails...surface the rubber and add a 12mm mdf spoil board also surfaced and fixed down on edges beyond cutting capability.

cheers all still open to any more feedback.

You would want screws in the middle as well or where ever there are supports then you won't have to run the Vacuum each time you want to cut something, 12mm would be too thin 19mm would be more ideal, this would give you more resurfacing options just remember you have to take probably more than the 1.5mm off to start with and counter sink the holes there would not be much for your screws to hold it down if you use 12mm

[routalot]

The reasons I advocate 12mm MDF are that it is thin enough for the vacuum pump to draw air through easily and that in my climate a change in atmospheric humidity won't give too much of a change in thickness.As it is I currently have 95% humidity,in a couple of days it might be 65%.These changes cause MDF to move a bit and the thicker the material,the greater the amount of movement.

For several years I have held 12mm MDF on a router using 1"X6 brass screws it was never a problem as there were just enough to prevent the spoilboard sliding and we didn't have any other hold down arrangements than a vacuum.If a smaller job needed to be held I used a carrier sheet of 9mm ply screwed to the workpiece with brass screws or several pieces of good quality double sided tape.For the machine we are discussing here I would be seeking to minimise the weight I added to the table to keep any sagging to a minimum.

[mactec54]

Based on your feed back I am going to shim the bed on the rails...surface the rubber and add a 12mm mdf spoil board also surfaced and fixed down on edges beyond cutting capability.

cheers all still open to any more feedback.

Here is a video on how to make a duel purpose table and you will get the best of both world vacuum and a T track top, you can make this any size you want to suit your machine

https://www.youtube.com/watch?v=5hKCy8TkUJo&t=622s

[dragon_sa]

Thanks everyone....I have shimmed the bed now..wasnt that hard in the end..I am going to go with a 16mm spoil board, I will run the surface planer over my rubber bed first, that will act like the first layer in that video, then take a skim off both sides of the spoil board then laminate the edges, I will then fix it down to the bed with counter sunk brass screws and hopefully good to go

[routalot]

Its good to know that the shimming went well.You appear to be well on the way to success and your plan looks very promising.I do believe that you would be able to just skim one side of your spoilboard, rather than both as the objective is to ensure that the working surface is parallel to the Z plane of the machine.