But why they say "It is NOT intended to be used between a CPU and the CPU heatsink" (they cap the 'NOT') ?? They further go on to recommend using their compound/grease instead. Is it just so one can easely replace CPU and/or heatsink ?

 

To quote their web site :

"Arctic Silver Thermal Adhesive is a two part permanent adhesive for thermal joints in minimum bond line applications. It is NOT intended to be used between a CPU and the CPU heatsink. On a CPU please use a thermal compound such as Arctic Silver 5, Céramique 2, Arctic Alumina orMatrix. For proper use refer to the instructions here."

Is it just because "Components you attach with Arctic Silver Thermal Adhesive will stay attached forever." ?

 

I was thinking about that. But wondered about the drivers and steppers running hot 100% of the time being always powered (?) The power supply would need to feed full power (143% or so of rated stepper amps) to all steppers all the time, about 15 amps total in my case (if not more) ?

Does the .9 version of GRBL allow to control this per stepper instead of a all or nothing ? I could then do upgrade and just keep the Z powered as the X and Y seem happy with the current setting of ... 25 ms for idle delay.

So I need to make certain the heatsinks stay on the drivers (adhesive it must be rather than grease/compound) AND I need to get the fans mounted to keep temps as low as possible, especially with the jobs which could run for hours non-stop. Might need to mount a prop on the rear shaft of the NEMA 23s to get max air moving on them as they jet around. Maybe I can just wire in those tiny fans instead, I've got them.

 

True, CPU heatsinks are usually clamped on tight, no need for glue in those cases. Just odd how they phrase it ...

 

Shields up ! Here is first public draft of the plexiglass pattern to make a shield which will wrap about the Y gantry plates (of at least myOX). You could always work from the base and do one for your own OX, if using Chris' plates.



So I've being "playing" with SketchUp (Make 2015) for a bit now ... Putting up with some of its odd behaviour - maybe from me just experimenting rather then reading the documentation (I have looked at some videos), but I'm managing my way through it. Nice program, especially for the price (free !)

Next step is to convert to g-code. I could use SketchUcam with just inside and outside cuts, along with fold cuts, but I really want to see what it does with generating "3D" g-code.

I have not yet cut the 'new and improved' shield (see attached SketchUp file), but thought I could share and, especially, see if any one spots any problem before I get myOX working at it (first on a piece of scrap wood, then on the target plexiglass). The ones on myOX (see past pictures), where handmade on a table saw a while back (third attempt was the good one). CNC version should be much cleaner and shareable.

Note the target material will be 1/4" plexiglass/lexan. I'll work just one side (for now). So things like the myOX "logo" benefits from the transparency of the material. It will be 'frosted' in contrast with rest. So it is also 'mirrored'. Once I get to a two sided job, it could be engraved on the outside face ("bottom" view of current file, sorry) and filled with paint (ie. coloured plasti-dip ?) The three round holes are for the 'abort', 'pause/stop' and 'resume' switches. The fourth larger round hole is for the "Emergency Stop" button, although it might be wiser to have this one mounted to the table ... I am thinking of having a 2nd one at other end of myOX. The rectangular hole is for a two line display (volts/amps; not certain what I'll plug it too to be of any use). While the opening for the fan is generous, with a tiny bit of 'style' (I'm no artist and I didn't want to copy/paste from others), I do plan on having a filter mat sandwiched between the fan and the opening so I don't suck in dust/chips (nor the mat hopefully). The 'slots' are to hold the controller board support. Yes, I need to 'drill' a few holes for few screws (no glue so I can take things apart). And obviously, there are the fold lines - as wide as the material is thick and 1/2 as deep to get a nice clean bend under the heat gun.

I'll film the cutting, probably do a 'build' for it (or would it be a ressource) and also do blurb on the bending as requested a while back by some (ok, just one or two myOX 'followers'). Since I'll be back at work, it may be next weekend before things show up ... However I do want to start regular use of myOX, so I could turn this around sooner than later.

 

The shield is already a (beta) ressource here.

 

Definitely. I just have to CNC the part ... I'm hoping tomorrow after work, if I can get off earlier than usual. So the video will definitely follow shortly thereafter (just don't expect a Spielberg production).

With the 'folds' and a good heat gun it is relatively easy (especially the 90 degree bends). Keep an eye on the ressource (I just created it). The bending will probably be 'repeated' (or linked) as a tutorial as well.

 

Well, it hit me earlier in the week ... a revelation of sorts.

In CAM there seems to be too options :

1) The first is along the lines taken with SketchUcam - one pretty well optimizes the tool path and thus the job time, tool wear, etc. by pretty well hand picking the areas being machined. You tell it to do an inside or outside cut for this 'curve' or that 'pocket' and drill holes here and there.

or

2) the CAM software, like MeshCAM, "simply" goes back and forth, offsetting the parallel paths of the tool which covers the entire surface (!), occasionally lifting it above the piece being worked on and doing a rapid to an other part of the surface.

More on this revelation in the resource : myOX Y gantry plate bent plexiglass dust shield (trying to keep things close to most appropriate build/resource).

 

I'm using GT3 and, although I'm not looking for super precision at this point (think more of 'art' than 'jet plane parts'). I don't seem to have any stretching over a large work area (ex: family tree first run is about. Of course, the stretching might be more obvious after some time as things wear and fatigue ?

 

Just a tangent... not the end of the world as we know it.

I do have some extra GT3. Maybe there is enough to give it a try ... I am just wondering how dust and chips will play in the mix ?

I have to get the vacuum installed and/or maybe some air blowing unto the belt (compressor sits nearby ...

Someone had mentioned/tried placing the Y gantry plates on the inside, thus having the belt and wheels facing outward. I do not recal seeing anything more on that idea... It would make my shield way different, but, with its fan(s), it could keep the wheels and maybe the belt clean(er) ?

 

I don't know, the dust from working MDF just gets everywhere. I mean the entire garage now has a thin layer of this dust. Takes forever to clean up. The barrier you suggest would be great when working other material, I suspect.

 

Well, as you would see from post in the ressource myOX Y gantry plate bent plexiglass dust shield I flattened the layout from the orignal 3D and used SketchUcam to do pockets, traces and whatnots, tweaking generated "paths" to get rid of a few ... unruly areas by editing segments.

I just could not go to bed without seeing what SketchUcam would do, especially since I tweaked and played a bit of shell game with it... Taking a peek at the resulting G-code with Universal G-code Sender's viewer I get this :

View attachment 6422

Get the full post over at the resource linked at beginning of this post. Yes, it is missing the outer cutout to separate the shield from rest of raw material. But, this is looking AWESOME and cutting it should only take minutes.

If it wasn't nearly 3 am, I would run to myOX and give this G-code a try right now ... on a piece of scrap wood ... but I have to catch some before work - only a few hours away now.

 

Nothing special done to my X axis as of yet ... But I'm working on something (can't go into details right now).

 

It's been a while, but I'm back with a few more tidbits on myOX.

First, I had to use the snowblower three times in 24 hours as we got buried under over a foot of snow last Sunday (not counting the drifts well over 2' all over town). I know, a foot is not that much for many, but, here, in the Canadian Tropics, it was a record which nearly closed the entire region, not just the city To make things worse, my trusted and rusting big John Deere blower could not take the load ... it had turned into a shed potato I guess ... but it was a quick repair (5$ worth of parts !) It will need a sand blasting, some TLC and a fresh coat of paint. Maybe a job for the coming summer ?

In further testing myOX, as I cut my dust shields, I came to realize a few more facts of DIY CNC and myOX. In no particular order :

• I need to fix my sliding Z : vibration, NEMA17 and heavy router don't go together with precision. I did find a simple solution which I will test later today. It involves use of a bungy cord ! The guy apparently swears by it ... I'll have to get NEMA23 upgrade though (but not quite certain it will fixe the problem for some reason). Here's a stupid question: what if I flip ACME around ? I mean I can't force the Z upward but it only takes a (blood filled) mosquito to force it downward ...
• Otherwise, the OX design is nice and strong as myOX plows through wood and plexiglass nearly like butter on a hot summer day. The NEMA23 have no problem moving the big router around the full working areas. In fact, it pushed a clamp out of its way (ok, just a few mm, but still) !
• Walking into Rockler's (there's one less than an hour away, customs time included) is like walking into a candy store ! They also sale some 'fancy' CNC machines with the one closest in dimensions to myOX at over 7k$ without the table to sit it on - add 2k$ !! Sure makes the learning curve worth the investment in myOX or any other OX build. Best part of it, this Rockler's is few minutes away from one of the wife's shopping area ...
• Limit switches are for the weak of heart But there is a 'simple' solution, if the GRBL authors don't mind doing a little tweak (more on that in a separate post). It surprises me that no one actually thought of it before newbie me. Maybe it's my 'out of the box' thinking at work again ... For now, as I always admire myOX at work, for hours on, I disabled the limit switches, hand near power cord for a fast disconnect in case myOX goes wild.
• Always have your video camera fully charged before filming a job in progress ! So I unfortunately don't have much to show... for now, as my video camera is finally plugged to a charger. I'll have to use a webcam powered through the computer, eh ? Wait a second, I'll try one of those remote cameras for R/C planes and such, powered through myOX and wireless to computer ...
• DON'T plug everything into the same circuit, especially if only a 15 amp breaker ! , and I made the mistake of adding small room heater to bring the garage temperature to a more confortable level and, as I realized I didn't turn the vacuum on, the power went out at the flip of the vacuum power switch. Stupid electrician putting only a 15 amp breaker in a garage, eh ? At least I had him separate all garage outlets into two circuits (both turn out to be only 15 amps !) And me that had myOX's power plugs on a 20 amp powerbar just to be be safe.
• ...

More to follow.

 

A short (13") bungy cord looks good. I have to force to freely move the Z down. Turning the spindle up or down seems just fine, at least for the short travel needed for 2.5D type work. Before I get into more than the 1 or even 2" Z travel used to date I will need to upgrade to a NEMA23.

At 150 ipm, X and Y, I was also a bit agressive in the plexiglass, but it was cutting and making nice looking chips. In addition to slow down the routing, I'll tweak the design of the dust shield (go see the resource for details).