[peteeng]

Hi All&Sundry - Here's where I'm at. Doesn't look much but it's a nice clean start to the P3 build. I have set up the assembly directory and will start making dummy drawings of bits next sitting. Need to go and buy a folder and a ream of paper, there's going to be a few dwgs!! The first small assm was the return assm and I rebuilt the rails so they are scalable. Building P3 won't take long as all the parts exist. Its the drawing stack thats the slog. Tomorrow I work on getting the laser on Brevis working that should be fun... Peter

Holes in sides of walls? I was going to place the electronics in the walls. But I think I'll leave these intact to maximise stiffness. I'll make a drawer under the bench somewhere for all the electrons.

[peteeng]

Hi All & Sundry - Question, can you get metric fasteners in USA easily? Maximus is designed for iso metric screws and I intend to provide a fastener list vs selling screws. No point in me selling screws & nuts. Is this a good plan? Peter

[peteeng]

Hi Sundry - Maximus is becoming real. The dwg stack is happening!! Peter

[ericks]

Why don't i have such good cad skills

[peteeng]

Hi Ericks - Here's some more CAD bling. Peter Practice, practice, practice.... 10,000 hrs

The process for manufacture is to 1) Create sub-assemblies that make sense. Then check these for clearance, missing parts and any thing that becomes apparent in the reviews. For instance in the motor assm dwg I can see that I'd prefer more clearance on one side of the pulley so I'll fix that. It's not that easy to see these things in a full CAD assembly. Its quite easy to design something that you can;t put a bolt into but there it is on the drawing, oops got to get creative in the workshop... 2) Creating a BOM bill of materials or BOQ bill of quantities. This then is broken down into orders or requests for quotes. 3) Once assemblies are dealt with then individual parts are drawn with manufacturing information. Then more RFQ's go out.

I have been sending off various models for comment along the way so there are no surprises about bending them. Sometimes they say no can't do, yet I have a similar part in my hand from them from history. So I remind them of this and they say OK. Sometimes the Std set of bending rules do get broken.

For manufacture you need to supply a STEP solid model of the part so the CNC guy can program the geometry and the correct issue drawing that marries to the model. I hear stories about files and drawings not matching and people getting very upset when things come back that aren't expected!! The companies I deal with have automated lines so sometimes things don't get the human eyes and hands on them and the storeman who packs it surely does not know!! Cheers Peter

This sort of "risk" is even higher on complex CNC machined parts. Everything must be checked very very well before you push the big green button on the order and the machine. Its very easy to have 20 or 100 things all wrong and difficult to correct! When things were done by a machinist he'd stop and say that's not right and give you a call. When most things are automated and programmed from the geometry and the contract says its your geometry so its your fault, the machine just ploughs on making that same beautiful thing or that horrible error over and over.

I've also attached where P3 is up to. Gantry assm is next...

[ericks]

For sure you are correct, i would have saved a lot of time and $ if i first designed my router fully in cad. I really need to get a good cad program and start working/practicing

[peteeng]

Hi Ericks - Have a look at Atom3D its the hobby version of Alibre that I use. Not sure of cost but its economy grade. Very good feature list. Plus Fusion360 is free for personal use. Or look at FreeCAD it also has a CAM package and its all free. Its getting more and more powerful every time I look at it. Peter

https://www.alibre.com/atom3d/

[ericks]

Thanks Peter, i will do that

[peteeng]

Hi Boys and Girls - I've built P3 from scratch up to the gantry and removed all the model niggles as I've gone along. I've aligned everything to the Nth degree and the model slides around nicely now. I have an assembly dwg for all of the 5 sub assemblies so far, there will be 11 subs. I rechecked the belt figures and geometry so that's all clear. The ATL10-25mm belt has a breaking strain of 1646kgf not that means much. Its recommended max service load in this configuration is 425kgf, I think it's half that for a U drive. So it will be very reliable. The recommended idler diameter for the U drive with this belt is 120mm! for the straight drive its 50mm so we are 55mm so all good. I have to estimate the pretension now these things are finalised, maybe 25-50kgf. I have found that a bend on the end of the gantry buttress comes and nudges the bearing bridge buttress so I have connected them together with a screw. I'll try to make this two screws soon. Once the assm dwgs are done its onto part dwgs, then nuts and bolts. Hmmm some way to go. It's the working week tomorrow and I have a 600Tonne trailer to analyse, slightly different to Max. The recommended belt tooth load at 100rpm is over 150kgf so I think all bases are covered with the belt. More to come... Peter

[peteeng]

Hi Lads & Lassies - Today I made the first part dwg and tidied up the column bolts. So many holes to make and align and fill with a screw... On Scoot I have a tool board for clamps and spanners. Very handy. It also is the high spot for catenary cables to the gantry. I haven't used cable chains on my machines to date. Maybe this is the one. I use a 12 or 20 core flex cable. If I use the tool board I may be able to still go with the catenary....Peter

[peteeng]

More work - I have started adding the curves to the parts as well. Not sure what the curve on an end of something like this is called. More holes... Peter

[peteeng]

Hi Y'all - Spent the evening watching a movie for a change but have managed to put in some backers. Try to get some parts going everyday. I once managed a Design Dept for an Automotive Accessories company and we designed and tested towbars to destruction including the car sometimes. Volvo has never forgiven me...We found that large backing plates did much better in transferring load then washers & nuts. Plus its easier to get the bolt started and snugged vs mucking around with nuts, washers and a spanner. Costs more but will provide a stiffer connection. Gets the metal out to the edges of the bends. Cheers Peter

[ger21]

Is this bracket welded together? If so, how do you prevent distortion?

[mactec54]

Hi All & Sundry - Question, can you get metric fasteners in USA easily? Maximus is designed for iso metric screws and I intend to provide a fastener list vs selling screws. No point in me selling screws & nuts. Is this a good plan? Peter

Yes no problem with metric anything in the USA they have both inch and metric

[peteeng]

Thanks Mactec and sharp eyes Gerry.

The BRACKET ISSUE - Initially I tried to design some of these parts without welding being used. But some geometry and stiffness requirements just needs some sort of rigid connection. This part looks simple but is actually a complex item. Maybe a little long winded here but worth putting this out there I feel as it crosses many other issues.

When I was close to the sheetmetal guys I personally brazed this sort of item or silver soldered them to minimise distortion. Which sometimes meant a double run to their place but being close didn't matter much. 2 years ago I moved 1.5 hours away from them so it's not a short run anymore. So two things to consider. 1) Finishing the parts either powdercoat or electropolish the parts need to stay at the metal place. Not economic for me to go there come home do the brazing take it back for them to finish 2) I tried a local powder coater and they applied 750um plus of paint, filled up 6mm holes with paint, tried to fix it blew it then tried to charge me for the repaint. So I have sworn off powdercoat. The powdercoated machine was intended for a trade exhibit here and the screwup meant we missed it which mucked up our intended product launch... so 3) The intended out come is to design the brackets with a slot and tab arrangement and slot TIG weld the back lightly with special instructions on flatness and squareness for the part. If we weld it then it can be electropolished. The two epolishes don't want brazed parts in their tanks. I have looked up the tank chemistry and it would be fine but that's their call.

I have started testing my own DIY epolish system as the SS looks great when its polished. Also recently I had the company make a stainless steel mould for a carbon fibre part and the tab/slot TIG tack and small weld approach worked very well and that was a 3m long by 600mm wide mould. So I'm confident that it can be welded OK. Once I get my workshop up next year I'll go back to brazing or buy a tig but this maybe a good path. Got to do these things to find out.

Peter

[ger21]

Question, can you get metric fasteners in USA easily?

While we may be behind in some areas, we can get just about anything, and rather easily. Take a look at mcmaster.com. Anything you could want, usually delivered the next day for about $6 for small packages.

[peteeng]

Hi Gerry - Thanks for the info re bolts and others who have responded on that. In Oz companies use courier bags and it costs $20AUD to get most things from A to B. If they took the same package to the post office probably cost <$10AUD but they don't want to go to the post office. We do have mail order companies but not quite like Mcmasters etc. Suppliers in asia can supply small items FOC mail and that seems to be the trend here. Bearings, nuts bolts, metal bits etc probably put some of the Oz companies out of biz the way it's going. Times are a changing, back to mail order... Peter

Gerry here's an image of the back of the bracket so you can see the tabs. Assembly is a bit tricky as you can't have two tabs at 90degs and have them engage. But it can be done just have to work out some entry angles. The surface of the tab is about 3mm under the surface of the bracket and they do about a 2mm weld. In this way the shrink direction pulls the tab into the slot. If you welded it on the outside it would cup the bracket. Cheers peter s

[JFL4066]

Hi Gerry - Thanks for the info re bolts and others who have responded on that. In Oz companies use courier bags and it costs $20AUD to get most things from A to B. If they took the same package to the post office probably cost <$10AUD but they don't want to go to the post office. We do have mail order companies but not quite like Mcmasters etc. Suppliers in asia can supply small items FOC mail and that seems to be the trend here. Bearings, nuts bolts, metal bits etc probably put some of the Oz companies out of biz the way it's going. Times are a changing, back to mail order... Peter

Gerry here's an image of the back of the bracket so you can see the tabs. Assembly is a bit tricky as you can't have two tabs at 90degs and have them engage. But it can be done just have to work out some entry angles. The surface of the tab is about 3mm under the surface of the bracket and they do about a 2mm weld. In this way the shrink direction pulls the tab into the slot. If you welded it on the outside it would cup the bracket. Cheers peter s

I have Tig welded similar assemblies and the issue is not to have any gaps or poor fits. If the bend is accurate enough and will conform to the "brace" with mild clamping to achieve full contact then movement is really minimized. Plus allow to air cool and not be in a hurry and quench cool. Laser cut tabs and slots are awesome IMO for the welder.

[peteeng]

Hi All - This mornings stuff is about details. More holes and checking things are exactly where they should be. A development model uses geometric references to align parts such as their mid-planes and axes. In this way they always line up where they "should" be even if the features aren't quite right. Development models need to change frequently so this is acceptable. But when it comes time for a manufacturing model you use features to line things up because it's the features that matter. If you use a plane you run the risk of unassociated holes or edges being misaligned. Another approach is to use related holes so when a master hole is moved or resized it moves all the slave holes but this takes more model management. The cam follower I wanted to use for the return pulleys is hard to get with a flat race so I have used a 5201 bearing. Same cost, easy to get. Same cost at VXB USA as from my local so that's now done. Back to more holes and screws....
Peter

[Ntl]

I was just looking at your drawing and dang that machine is going to be expensive to build, is the whole thing made from formed plate?

Dan