[peteeng]

Hi Ralph - 1.5 gal = 5.67 liters and most of those resins are very close to 1000kg/m3 so $90USD /5.67 = $15.87 US$/kg x1.32 = $20.95 AUD/kg. I pay $16-$17 AUD/kg for similar stuff here in 20kg kits from the manufacturer so price is good. Peter

[catahoula]

(sorry Pete, hate to turn this into yet another EG thread...)

I just bought a 1.5g kit of this stuff for $90:

spumco- nice find! I just found this, https://entropyresins.com/product/cc...casting-epoxy/
Which is 270 cps for the slow hardener, pretty good. And from Gougeon brothers, like West system, so reputable. But hard to argue with 170 cps and nearly half price. Might give that a go. For some reason I was worried about the cost of the EG fill, but I re-ran the volumes and to do a complete fill on my machine frame is only 10 gallons of mix, so 1.5-2 gallons of epoxy which is very manageable.

........
........
but, on an actual CSA related note, I called Rapid Set today re: UltraFlow 4000 high precision cementitious grout, and asked for some additional data on long term shrinkage. What they could provide included some test results from ASTM C 827 that indicated a total height change of +0.1%, BUT this was only for the first 170 minutes. So not indicating long term anything. Interesting though to see a positive figure. Must be the compensatory additives doing their thing. I wonder if it settles to zero after some time. As the potential buyer of a total of like 2 bags of this stuff, I didn't feel like running it all the way up to the engineering team, so I'm afraid taken alone these figures aren't too useful.

[peteeng]

Happy Christmas Eve All - I have been looking at the top and bottom height of the head while listening to Christmas Carols. On a side note I noticed one of the present bags has a spelling issue. Spelt and made in China

So it seems I will have a bottom nose height of +100mm from the table and a nose top ht of 450mm giving a travel of 350mm.... Any comments from people that use BT30 collets would be appreciated... I have started trimming things down and adjusting dims to suit this. I have been looking at the main parts and deflections to try and identify the culprits. The X load is still the hurdle so it must be the combined torsion/bending. The base and column can be E=40GPa and still get the 20N/um. The column can be aluminium or steel and its OK. I'd prefer Al for the saddle... The arm although short dips under the 20 if E40 material is used. So that maybe a candidate for machining as well as the saddle if I can't get a casting up to E70.... I look at the column and saddle and think a gantry would be better use of all that material.... So once I tune up Milli-M to the right dims I'll build a gantry to check this out. Milli-M still has to resolve bearing mounts and drive gutters so now's the time to do the comparison....

I may need a little land at the front of the "table" as well so it has somewhere to clamp to front to back...

Now the CD player is churning out "Time to say goodbye" by some tenor so maybe that's a hint... Ho Ho Ho to everyone out there.. Peter

[peteeng]

Hi Cat - No all good. All info will help pave my path eventually. The CSA data I have here shows a slight increase in early volume then a flattening out close to zero. especially if post cured at 60-70dges C for a 8 hrs. Can't find that chart at the moment... Peter

I look at the skyfire frame and think Milli-M is soo much more material would be good to model it to use as a benchmark. Would need all its dims.... Peter

The epoxy I usually use has a viscosity of 220-230cps so the above ones will be fine. Its specs are good, I've used similar specs on multi million dollar yacht hulls... will be good if post cured....Peter

[peteeng]

Hi All - Here's one of the CSA specs I was looking at. E=30GPa out of the bag, low exotherm, 60mins gel 590mins set, add blue metal and it gets to E=40GPa....Peter

[spumco]

Peter,

You want .stp files for the skyfire castings? Or some other format?

Column & head are done, but I haven't modeled the base/saddle/table yet.

-Ralph

[peteeng]

Hi Ralph - I need it all to do the FE so I'll wait until you have it all if that's your plan? Thanks for the offer.... step is fine when you get there. Does it use 20 or 25mm rails? Ho Ho Ho its Christmas and there's not a mouse in the house....We do get them in the roof occasionally Peter

[spumco]

.stp file of the complete assembly is ready. Stand, base, column, head. I left the spindle out, but it's a 90mm BT30 standard cartridge thing. My original spindle was an 80mm R8 cartridge (now gone), but Skyfire are using the 90mm one.

All components are cast iron, but it's pretty decent stuff - very ductile when drilled/cut and moderately hard. Lots of minor surface defects under the paint, but haven't found any inclusions or major issues yet.

Dimensions are very close, but due to the nature of castings (and typical QC) things can be off a hair in places. Also did not model the draft angles or the various as-cast surfaces; I tried, but gave up. It's about a 1.5 degree draft angle on most non-machined surfaces.


X/Y uses 20mm linear rails, Z uses 25mm (actually 23mm). Both are PMI brand, MSA20/MSA25 with E-trucks (flanged). I believe the Skyfire and later Mikinis use HiWin rails of the same size/type. Not all the trucks are present, but the rails & trucks in the model are taken right from the PMI site, so bolting size/pitch and other features relevant to your FEA should be accurate.

I didn't put them in, but the ballscrews are 1605 size, claimed to be C3. They're definately ground but I can't measure C3 vs C5. Not sure of the manufacturer - part number is there but no mfgr name.

File is just under 5meg. Email, dropbox?

Let me know how you'd like the file.

-Ralph

[peteeng]

Hi Ralph - 5mb can emailed peter at cncrouterkits.com.au. Peter

[mactec54]

Hi Cat - No all good. All info will help pave my path eventually. The CSA data I have here shows a slight increase in early volume then a flattening out close to zero. especially if post cured at 60-70dges C for a 8 hrs. Can't find that chart at the moment... Peter

I look at the skyfire frame and think Milli-M is soo much more material would be good to model it to use as a benchmark. Would need all its dims.... Peter

The epoxy I usually use has a viscosity of 220-230cps so the above ones will be fine. Its specs are good, I've used similar specs on multi million dollar yacht hulls... will be good if post cured....Peter

Even though the Skyfire is a good little mill it lacks rigidity there small machines where not bad, once they went to the larger build they are no better than any other mill that size which all seem to have the same problem

[peteeng]

Hi Mactec - Yep rigidity is the prime necessity for a machine and that's what I'm trying to understand and build into Milli. The skyfire model will be a good benchmark, something from reality... I got thinking about a gantry again so built a machine with the same footprint as Milli-M. The gantry Y stack takes up a lot of room. Would need to make this machine deeper to recover some envelope. Its heavier then Milli-M and less stiff. This is because the Z axis when in the down position (eg probably its most used position) is longest and we are back to the long cantilever issue. This gantry is compact because its block geometry with no concern about how to bolt cars on and then there's the saddle packaging puzzle on top of that. So seems the gantry again is not the way fwd. The Milli- M concept is looking strong so shall detail that more.... Nearly New Year... Hope Santa squeezed that new VMC down the chimney for you... Peter

[spumco]

Even though the Skyfire is a good little mill it lacks rigidity there small machines where not bad, once they went to the larger build they are no better than any other mill that size which all seem to have the same problem

Mactec,

No disagreement from me on the lack of rigidity. It's significantly better/stiffer than a G0704-class, or even a 770, but it's not an undersized commercial-class mill.

It's usable now, but I'm hoping that the upgrades I'm in the process of completing will help. The biggest problem was the R8 + TTS setup. After that, I think it's column stiffness - but Peter's FES will hopefully shed light on the weak points. It may be that the head itself is squirming around.

[peteeng]

Hi Ralph >> Did you send the file? Not received yet... Peter

[spumco]

Hi Ralph >> Did you send the file? Not received yet... Peter

File sent, no bounce-back received. Maybe check junk mail folder?

[peteeng]

Hi Ralph - Interesting, sent myself a test email no ping and no email. Will have to check that out. Send to peter at carbon-works.com.au that works just checked.... Peter

[spumco]

Sent. We'll see how slow the electrons are this time of year.

[peteeng]

Hi Ralph - All received and it imported "no errors" so will pre-process and get into FE asap. Best regards and Seasons Greetings. Peter

Hi ralph - Model is good, missing a couple of cars but I'll copy those in. What is the distance to the spindle nose or a typical tool from the arm? Peter

[spumco]

BT30 model sent.

By the way... those extra pads in the Z-axis ballscrew trough are not OEM. They're plates I made to mount the ballscrew bearings.

The Skyfire/Mikini come from the factory (maybe still, instagram photos show many design changes over the years) with the Z-axis motor at the bottom. It resides in that pocket in the base.

Early Mikinis had the trough closed in with a little drain hole at the back of the base. Total design fail - it collected chips and coolant. My stepper was submerged 3/4" deep when I pulled everything apart. Means you had to pull the column to get the motor out. Utterly retarded.

Later models had the pocket open at the bottom so the motor could be pulled out (barely) from below.

I've decided to swap it around and put the motor at the top like every other mill designed by grown-ups. Hence the bearing mount plates.

[peteeng]

Hi All Christmas survivors - I've started looking at Ralphs model, checking for gaps and details that may be an issue etc. One thing that has crossed my mind is the class of Cast Iron it maybe. CI has a range of modulus from 50-110Gpa I think. Any clues to the type of CI in these things would be appreciated... I usually use 100GPa as its a round number... Peter

[spumco]

Peter,

Use ASTM A159 G3000 for the CI material properties. The CNCZone thread documenting Skyfire's original build from 2011 indicates this was the grade he selected at the foundry.

-Ralph