[clmenz]

Momentz, maybe totally off base but last second info/hints on CSA based concrete not Portland based cements, (dont know much about the availabilty/properties of the HPC concretes) . For us DYI hobbyists the CSA cretes have advantages including -very low native shrinkage rates (0.1-3% !!, High strengths early and late- 11k psi comp but sigh more $. For mixing/prep. If you are ?using higher stength rapid setting CSA based concrete (( in the U.S. - " CTS Rapid Set" mixes available @Home Depot or Whitecap)) , not Portland based cements, FOR CSA 1. mix water and cement/concrete by weights 2. Water/C ratio ideal about 0.36% to 0.38 %. 3. To slow setup times of the mix to 1 -3 hrs by. A. using ice in place of water 1hr plus B. use citric acid 0.1% to 0.8% (to gain an hour, also use citric acid dose - 0.4% at 50 °F-10°C), use cheap food grade from. 3. GFRC well worked out. Lots of relevant concrete info on YT from concrete countertops vids but I think best engineering info and great practical & detailed information about end user use of using pure CSA from Jeffrey Girard P.E. on YouTube with several videos & PowerPoint on "TheConcreteCountertopInstitute" his 1hr 20 min video fantastic detailed use of CSA Chuck

Sent from my SM-S908U using Tapatalk

[peteeng]

Hi All;- started researching 5 axis stuff. Heres some heavy metal removal. Peter

https://youtu.be/mTQonWqiU4w

https://youtu.be/tqN-pipUMac

[peteeng]

Hi All - The laminated part method is now mainstream. This article claims its novel but we've been doing it in the marine industry for decades to make moulds... or plugs. Peter

Thermwood announces large-format Cut Layer Additive process | CompositesWorld

https://youtu.be/GsofRj2KCMo

https://youtu.be/X10SvaUGhGU

[peteeng]

Evening All - I've started adding features like radii and cut outs to the parts. I go away for a week and when I get back the Fusion Post does not work with radius's (crap) I've had to use the Haas router post to get things moving. Will have to decompose the post to figure out what's happening there... At least I got some sawdust made tonight. Been banging my head on it all day.... Peter

[peteeng]

Hi All - Been working on my 5 axis machine design. The base is intended to be cast UHPC. I've sent some drawings to offshore and local aluminium casters for quotes. Waiting on feedback. I compared a hexgrid to a flat plate, to a diagrid base. 60mm thick for the grids. I changed the solid plate until it was close to the hexgrid base, so the plate is 41mm thick and weighs 45kg. The hexgrid deflects more then the diagrid so the diagrid id the go. The hex weighs 28.8kg the diagrid weighs 27.4 kg. The hex is attractive visually but stiffness wins. I also rushed the diagrid model so its a bit rough but it still deflects the least >>> Peter

[peteeng]

Morning All - For those using Fusion 360 simulation beware when using bonded connections. Bonded connections are stiffer then they should be. Use remote bonded and it will be closer. I made a solid T beam and a two connected beams of the same geometry. One using bonded the other using "remote bonded" The solid beam deflects 166um (LHS Beam in image), the bonded beam deflects 149um (middle beam) and the remote bonded deflects 169um (RHS beam). I'm about to do some trials on the fusion bolted connections to see what they are like. I'll compare to my other FEA system. Peter.

I upped the mesh density and the results are good. So do mesh checks and use a finer setting. These beams are 1500mm long and 20mm thick. The fine setting is 5mm mesh vs the default of around 15mm. Solid 183um bonded 181um and remote bonded 182um....

[peteeng]

Hi all - I ran the same model in Simsolid and the solid beam has the same deflection at 183um the bonded beam was slightly stiffer at 180um. Seems the bonding makes it slightly stiffer. Now to add the bolts.... Peter

[Buoyen]

Good find Peter.
The bolts joint will be interesting to see.

[gracesintec]

Hello ,thank you for your sharing ,Grace

[peteeng]

Morning All - I have been playing with bolts in Fusion. If your going to do this you need to use "separating" contacts between the bolted surfaces and set a friction co-efficient for that connection. This is not clearly explained in the help. Each bit is discussed in different areas so I had to put that puzzle together. Here is the result of run. I used 8 M8 bolts with 14kN preload and a u=0.35. The beams are steel. The bonded and solid beam deflects 183um and the bolted version deflects 857um. This is totally arbitrary model so don't pay attention to the numbers. The aim is to get a model that works, which it seems this does, The beam stress is 36MPa which is a bit high for a machine element. Setting up the contacts and the friction and the bolts is pretty straightforward. My one complaint is that you have to set up one bolt at a time... Some of my commercial models have 200+ bolts and this would be a mission... Next is to compare the F360 model to the same simsolid model.... Peter

[Momentz]

Forgive my ignorance, why be so pedantic about all these things, aren't you just building a hobby grade machine? 2+ years in drawing and tests seems a bit extreme.

Sent from my SM-G781B using Tapatalk

[peteeng]

Hi Mom - Seems your message hasn't come through yet? I have two machines to play with and I'm building a large 8x4'machine at the moment. So I have lots of hobby/maker level stuff at the moment. Milli was intended to be a mill to machine composites and to use cold casting or composite infusion as its main build method. It's a development project that has wondered around a few materials and processes that haven't worked out. By that I mean the materials have not hit target stiffness. Material development programs take much longer than machine builds.

As I go along I've looked at Fusion 360s FE capabilities and compared it to my other two FE systems. I publish results here for the interest and benefit of interested parties. So in the two years of this thread I've built two other machines and sold them, this has allowed me to prove/disprove a couple of things and that knowledge will go into Milli (or whatever) when the time comes. Since I do FE pretty much daily I share results here eg fusion with u=0.35 friction in the connection, 14000N preload M8 bolts predicts that the beam will deflect 857um. Simsolid predicts 194um for the same model and conditions (184um for the solid beam),. I tend to believe Simsolid as I've done lots of checks and balances with SS over the last couple of years. So now I have to dig into F360 a bit and figure out why it deflects more.

Being pedantic means I don't start building until I have understood what I can and can't do. Each machine build has an objective. Milli has not met the objective yet. The fallout of this thread is: 1) Use engineering grout that has a published modulus if you need to know the modulus for design work for UHPC or have it tested 2) The only material that I have made that has achieved the stiffness of aluminium is carbon fibre at 70-80GPa 3) laminated aluminium/plywood is a great material to build and work with, stiff and damp 4) good grade 100% plywood makes stiff and damp machines, accuracy for metal cutting is a question as it does move around proportional to humidity. Metal laminated parts are a solution to that issue 4) its really tough to better steel for cost and stiffness. As an aside that's why simulating bolting is important as for a hobbyist welding has so many hurdles to leap. So I want to understand Fusions FE bolts a lot better

5) Simulation has become very accurate and accessible over the last few years but for the novice its a steep learning curve to get good outcomes 6) Time flies when your having fun it seems... back to painting Lanky. Peter

[joeavaerage]

Hi,
'Don't underestimate pendantry!!!!!'

Couple of points:
1) How do you suppose a hobby machine gets tweaked and re-designed into a commercial product? It is the exact same attention to detail that turns a one-off hobby machine into
a commercially ready machine.

2) For all of peteeng's investigations it has revealed that epoxy granite is actually a rather poor material for CNC machines. The best peteeng was able to achieve was a Youngs mod. of 18GPa.....
when I eat el-dente pasta I expect better than 18GPa!!. This rather suggests that figures published by others, be they hobbyists or manufacturers are wildly optimistic. I trust peteeng's numbers,
so that means I perforce distrust all other numbers, especially those that report 40GPa and more.......I smell BS.

3)peteeng has come to the conclusion, by dint of experiment and analysis, that steel is about as good as it gets, 'bang for your buck'. I've always been of the opinion that iron and steel were overall
the best (stiffness per dollar spent)....but that is just my opinion. That peteeng has come to a similar conclusion after looking VERY critically at the alternatives is a very much more substantial recommendation than my opinion.

4) peteeng's professional career has included a lot of FEA using a variety of software platforms. I on the other hand am a rank beginner when it comes to FEA. Circuit simulation is what I know..... The important point
here is that peteeng is very good at sniffing out those items that do not result in realistic calculations. The whole point of FEA is to inform your decision process, and yet you can make a small error in modelling your part
which in turn leads to wildly optimistic FEA results, you wrongly conclude you have a good design when in fact you don't. I have learnt a great deal from peteeng's discussions on Fusion FEA, and have a much better,
albeit still way too simplistic, FEA BS smell detection! Intelligent use of FEA is what turns a flimsy hobby machine into a good, if not world class design. We all need to learn to use it with the fluency that peteeng demonstrates.

5) I have made comments in the past that some of the investigations have proven fruitless and that time and energy may have been better spent, ie not disimilar to yours "why be so pedantic about all these things".
What must be said is that this is a wide ranging discussion....and it costs next to nothing. Once you start building something then you start spending money. Its rather too late to appreciate that a particular
aluminum section that you want to use actually turns out to be a poor choice....when you're already halfway through the build. Many an otherwise good build has been compromised in this way. Another is when
you get well into a build only to find that you've blown your budget and now have to make poor compromises just to finish.

It can be a substitute for rolling up your sleeves and actually building something, but expenditure of mental energy on reading, participating in a discussion, and research is all free and energy expended has
zero carbon emissions! (Excluding BS that is!!!)

Craig

[peteeng]

Hi Mom and others - My interest is in building very good/best hobby / maker type router/mill machines. It's taken me 6 years and several machines and lots of material tests to sift through various manufacturing factors and processes to come to various understandings of what's possible in making ultra stiff small machines. It all started by saying I can do better than a ShapeOKO machine. Many consider a ShapeOKO a commercial machine as they make $$$ from them. This whole process and this forum has given me information that you can't get from reading a book or even talking to someone about it. You have to walk the walk and wear the rubber out before you comprehend... Rabbit holes, red lights and train wrecks abound. It's been valuable to me in other areas as well. I design machinery and analyse machinery in my day job. Its just as important to know what does not work as well as what could work in my line of biz and how to simulate it. Clients these days expect everything to be able to be simulated in say 1hr, when I say it will take a week to do something they glaze over... Then they want it yesterday... Some years ago it used to take weeks just to set up a big model and days to run it!!! Now its done in mins.

Anywho its nearly another Christmas down and I move on. Maybe next year is the year of the robot... Plus the mill and the land yacht and big surface plate and teardrop trailer... Peter

[Momentz]

That all makes sense, thanks for taking the time to share.

Sent from my SM-G781B using Tapatalk

[joeavaerage]

Hi,

and the land yacht

YEAH!!!! They are about the most fun you can have with your pants on! A friend of mine had some 'Blowcarts' a commercial design, and they are just awesome fun and remarkably forgiving.
I've already got a kick arse carbon/glass mast. The Blowcarts have a quite flexible mast so the mast would bend and depower in a real gust, saving an inexpert land yacht sailor from a crash.
I want a stiffer mast,I have been sailing for twenty plus years, and so should be able to depower in time. Still contemplating either a windsurfer style sail, which allows real nice sail shape control to the base
of the mast, or a more traditional mast and boom type sail but with a hydraulic boom-vang, to really get this sucker moving!!!!! We had the Blowcarts doing 50kph, and when your arse is two inches off the deck
it feels like warp speed......and I like to get to 80kph......and wear a helmet!!!

Craig

[peteeng]

Hi Mom et al - If someone came to me and said they want a one off specialist mill or even a router I'd be making the parts out of steel billet on large CNCs. Those parts can be made to a precision that they self-assemble to exactly the accuracy specified. Being able to do modal (vibration ) analysis means pesky vib modes can be eliminated immediately. If the machine was massive then a cast UHPC base makes sense, but the limit is on finish machining and how big a machine you have access to. Its always the case to make a really good robot or machine you have to use a technology level up from your intended machine to realise that machine. Cast iron or cast steel has its place if you want to make many many of the same thing, With the advent of linear rails the lubricity features of CI have been superseded so the mass production opportunity is the only thing keeping that going. 3D metal printing will surpass CI casting very soon... I keep getting badgered by Markforge about bargains in metal printers $40kUSD off a $500k machine is a bargain? I have a RFQ in at the moment with an asian aluminium caster to determine if its economical vs a UHPC version. Elon Musk has gone gigacasting in Al so should I... I've decided to update F360 with the simulation ext so I can do more generative design... There's always a Grail out there to chase.... Have fun... But I'm back to painting Lanky parts with a brush, that technology hasn't changed since kindergarden... except they don't use pig hair now... Peter

[joeavaerage]

Hi,
I still like cast iron, it is so damped, it is a pleasure to use.

This morning I tentatively booked a casting job to be done early in the new year. It is a new headstock for my mill. The one that I made out of steel a couple of years ago was as a result
of budget constraints and is too wobbly, but I have finally gotten around to saving the budget necessary to have one cast. I've been working of the pattern the last week or so.
I'll post a pic when I get home later tonight.

I find the trick is to take manageable bites at this thing....if you try to do it all at once the budget hurdle just stops you dead. I've finished my trunnion fifth axis, at least the money spending part.
I've bought another second hand 1.8kW Allen Bradley servo to replace the one I broke earlier in the year. Its pretty rough...so I hope it works. Its coming from the US, and is at my shipping agents address in Oregon
as we speak. Within a few days I will have paid the required NZ tax and shipping and it will be on the way to me here...by Christmas hopefully. Early in the new year, maybe mid January I'll have my headstock cast.
Just manageable steps.

Craig

[peteeng]

Hi Craig - The Blokart is a great muse. I want it to be a 2 person "boat" so its a bit more fun with your pants on. I'm stuck getting small 4" x16 wheels used on small ebikes. Its a backburner project but we have 20km long beachs with good on shore breeze so it will be an 80kph drag race. When I sailed medium sized catamarans I started wearing body armour. Once I was sling shotted out of the trapeze so high the flight time was remarkable. But the landing on the mast and rigging was not good....Peter

[joeavaerage]

Hi peteeng,
yes' I've come off the trapeze at speed and got strained through the rigging....hurt like hell...but the ride prior was blindlng.
25knots of breeze with a huge gennaker on a 12ft skiff with a 10ft prod, two trapezed TO THE MAX!!!
I've often had people ask 'Why sailing, it looks like an expensive way to get cold and wet', but to be planing at 20 plus knots of boat speed
with your butt skimming the wave tops out 'on the wire' has to be the most fun wit or without pants.

Craig