Moar!!!! Got all the patterns finally finished.(at last) The sequence I took the pictures in is a bit screwed up as I went filling some boxes while paint was drying on others, but Ill re-order now to make sense, Got the box below done and the loose insert made and fixed/mounted, This insert is 'loose' and gets unbolted before mould is taken out of box - as described somewhere above, The insert has to contain the fillet where it meets runners also, as the fillet cannot be applied to the runners or it would lock the sand in, The end of the 'composite runner pattern' gets trimmed flat for the insert to sit flat against, Two M8 bolts both position, and hold the insert in place. Brazilian Mahogany can be tapped as you would steel with a normal tap and will hold threads for a good while. ''Not a lot of people know that'' lol but you can > I got the block in for the filter recess too, the sprue pushes down into this - The above are now totally done and sitting in blue waiting to dry, Onto the sand at last, Started with the insert box, Filled and cured with Co2, The short sprue is first removed - the sand cures instantly once the gas infuses through it, The outer main box is then slid off, - spot the tiny bit of breakout on the core prints - on inspecting the mould there was a tiny undercut here which Ill fill for the next round. The undercut is approx .2mm high so it gives you an idea on how accurate the mould needs to be. Im not that bothered with mould finish in these areas though as its only the core that sits there, and it doesn't see metal, worth a mention though... Its then turned on its side and the base form removed, The sand is very robust so takes handling pretty well, a view down into the insert showing the blind riser and the runner/gate, A few other shots showing different parts of the insert incl the gate, Its cross-sectional profile, which you can now hopefully compare to the shape in the original side view drawing, From the bottom, Onto the next insert that sits at the top at edge, Filled and gassed as before, Some may have noticed that the boxes have square sides with no draft - they do, but if you concentrated the paint to go heavy at the bottom you can actually create acceptable draft so that the box slides off without too much scraping. Im guessing .25mm thicker paint at the bottom is fine, How we are looking so far, I didnt bother with location dowels/pegs on the first runner mould set as I want to see how well it builds by just going off the side walls for ref - if it lines ok internally Ill just go off those - once the glue is set they wont stir. I have to make the core box for the sand parts to form the void/runner internal 'tube' but these are two very simple boxes in comparison to the main moulds above. The next time you see a head, block, or intake you will hopefully think a bit more about how they came about now lol Brian,
As usual, great progress and detail ! Can you explain a little more about the sand addition and the CO2 ? I could google it, but it`d be nice to have it in this thread for completeness.
Sure Nige, I probably should have mentioned it before this. I have to start from the start so here we go! The first sand I used to cast the twin port pieces with the butterflies is oil bond sand. Its very fine - approx 220grit silica sand(same measurement as sandpaper grades/grit size). It is held together with ordinary engine oil - a small bit is mixed in and it binds it so that it holds shape. Due to the fineness of this sand the parts come out of the moulds shining as you saw. The oil bond sand I used at the start is about 6yrs old - it lasts forever. The oil in the sand near the hot metal gets burnt away so you need to re oil the sand every once in a while. Ya can do this by hand mixing it in, or with a machine called a muller which is basically a drum with rollers within to squash the sand altogether as it rotates - coating every particle with oil. The main down side of oil bonded sand is that it is very easy damage - a stray finger nail will ruin a mould, or a not so perfect pattern will also destroy it on withdrawal. For that reason its only really suitable for simple shapes containing a top mould, and a bottom mould. You cannot make a multi tool with a collection of sand parts with oil bond as it would all fall to bits under its own weight. Another downside is the smoke - when you open even a small mould the oil burns even more once air gets at it and will fill a decent sized workshop to 6ft visibility in about 4min. The open hot sand can also spontaneously combust at any time too if the surface area of the part is large and theres a large volume of hot sand 'on view'. The flames are pretty cold and harmless but if there is someone watching that never saw it happen it provides some spectacle. On the upside its dirt cheap(get it!) approx 35pounds for 40kg and lasts forever. Its a handy sand for doing something simple very fast and for no cost. Next is the sand above - thats is the chemically bonded sand that you asked about. The base ingredient for this is again silica sand, but this time the grit size is larger - mine is the finest grit I can get in 20kg bags and costs about 8 pounds a bag. Its just plain silica sand with no moisture(big explosions) plant life, or sea salt(rendering beach sand useless). Its the same sand used for blasting and its actually marked on the bags. The binder to hold it together is sodium silicate - also called waterglass and is the same active chemical used in rad weld, and also what they use to destroy the engines in the USA under the ''cash for clunkers scheme''. The sodium silicate is mixed with the sand to form a 'fluffy' dry mix. The idea is to totally cover each grain with it so that it will bond to the grain next to it similar to open cell foam - you cannot use too much binder as the sand will then become closed cell and lose its porosity required for venting hot gas. Once the Co2 is introduced into the sand it causes the sodium silicate to solidify locking the particles together. The reaction is exothermic and the moulds do get quite warm. The beauty of this binder is that water un-does the bond and you can simply wash out the internal cores once the part is taken from the mould. The sand can then be re-used again if you wish once you sieve it. Bonded mould parts with this method are very robust - think big blocks of digestive biscuits and can be built up to any size by gluing them together with either epoxy, or more sodium silicate. You cannot use water base glues as they will break down the moulds and also could cause explosions due to the dampness. One downside is the rougher sand - but, the moulds can be painted with an alcohol/graphite/zircon talc in suspension to make finished surface smoother. Another binder method which the industry use mainly is a form of epoxy resin binder. Same principal as the sodium silicate just that no Co2 is needed to cure and its hardening time can be tailored via hardener ratios. I dont ever intend using this binder though as the Co2 is fine for small stuff, as in less than 1000 parts a day!! Both of these methods, and the way I am making the curved runner moulds - that is many sections built together with glue are common in industry and are known as the ''Core Package System'' or ''CPS'' way of mould building. 10% of the worlds engines and heads are done this way and are mainly used in the smaller and Exotic plants or prototyping halls where numbers are relatively small or the time to development short. Blocks are made mainly with high pressure die casting, where the outer parts of the head are formed with steel tools. Very high production rates are possible with this method since you have no sand moulds to build. The tool closes, metal gets injected in, it opens, you pull part. Its insanely expensive though with one tool costing over 4 million euro. Generally you would not be placing sand cores within the tools to form voids as it would just take too long. This is the reason a lot of blocks are open deck - steel fingers extend down into the tool to form the coolant paths around the bores. They are pulled up before/when the tool opens and the galleries are formed that way. For that reason its mainly used for trusted designs and long runs. The other last one worth a mention is gravity die casting - here it is similar to above, but instead of injecting the metal in it just gets poured in from a ladle as with the sand moulds. This is a fair bit cheaper since you dont need a massive injection machine bolted to the side of it - and it all running at 800 degrees. Its common practice to install cores into the tools in this case to form voids/ways in the various parts made this way. Actually, now that I think - the 3 cylinder vw engine uses gravity die casting for manufacture. Ferrari use gravity die casting and CPS systems(multi part built sand moulds as you see in thread) as their production numbers are pretty low for heads and blocks. There are a few other die casting methods too involving low pressure to fill the mould, or vacuum and pressure but the basics are 'similar' I could go on for ages and ages its a very vast subject - it started back in 700 BC and kinda went from there! Ive been studying it for yonks - its an amazing way to make something but not for everyone as the moulds can get 'interesting'. Brian,
wow wish I could do that, amazing read, looks so epic cant wait to see the finished product! you really are a man with many talents haha
Anyone can do it - you just have to really want to lol Got the lower mould filled this eve - it all went a bit arseways as I had to leave it mid way for a phone call so I just packed up 3/4 of the box flattened it off, and filled another piece up against the flat bit when I got back to it. It starts to skin up pretty fast with the Co2 in the air so Id have to chuck the sand otherwise. Anyway, its still the same just that the lower mould is in two pieces if you look close. Ill just glue these together and do the other three in one piece(with no phonecalls!) Packing up the box pre-call! The heavy bit on the pattern got a 3/4 round chill as its a little heavier in thickness than the walls, Chill seated beside section, Sand cured and box slid off, The top insert sits ontop, Side insert sits on the front, Like so, Its very hard shoot the curves as they look in the flesh, one with flash, Another, Didnt bother with the top section as the heater ran out of oil so I called it an 'evening'! Hope to get the core box plus cores done tomorrow eve and get the top box filled - ready for final building then and some mould coat + metal Bloody freezing out there this eve Brian,
Got the top box filled, you might notice the two tabs I added onto the sides of the runners too. These are for datum jig points when It comes to machining them - and doing them all the same. These are the only flat surface on the entire part so at least Ill have those to work with. They will get cut off when the parts are machined. These tabs are evident on a lot/all odd shape cast/diecast parts and normally get left on. These can also be seen on the oil pump below it, they are normally seen in 3's as a three legged stool never rocks(in the jig) The box was gassed and the two m8 bolts removed that hold in the insert, the insert then stays behind in the main mould as planned and gets removed with two screws for 'pullage' The recess for the filter too - its not the best location for the filter and I would have liked it at the entrance to the runner but Id then have to make another insert for that then so to hell with it - Itll be fine, The side of the filter is sanded at an angle and it taps in - interference fit, It sits over the riser on the lower mould as planned when the mould is built,(this is the lower mould) A shot of the rear on the 'bad light' side, In other news I bored out the pulleys as they were pilot bores, these were drilled .2mm under and reamed to size, They fit pretty good - they have to get grubs yet. A timing pulley supplied pre-tapped for a grub is best avoided as you have no way then of clocking the pulleys in ref to the flats on your shafts, and where you want the teeth to end up in respect to that. I have no flats on the spindles yet but if you had you would then have no indexing/timing control with a pre-tapped item. The top tapered bore timing pulley with no keyway on an engine is a good example of this in practice. It also saves a machining process too in terms of the keyways having to end up in the right spot on the camshafts. The main advantage though is that the designer can relax a bit in terms of tolerance between the crank pulley and the cam pulley and whatever gets driven in between + the belt itself. It also has its advantages if you skim the head - you just fit the pulleys loose - lock shafts - and turn back pulleys to tension belt on the pull side. The slack gets taken up on the opposite side with the tensioner and then the pulleys get locked on. If you had keys on all the pulleys then this is not possible without vernier items. I would think the main advantage is cost on the line though, in terms of cutting keyways...damn beancounters! Back on topic.... The belts I'm using, These are the width of the pulley - I couldnt get ones half width so Im going to slice them down the middle on the lathe. Its how they come to width in the first place anyway as they are made maybe 200mm wide and sliced up after. Brian,
Not much to report this evening. Got a thick block of MDF squares glued up for the core box. Also waiting on Isopropyl alcohol needed to thin the mould wash. The mould wash Im using is a new one that I haven't used before as is the core glue so Ill have to test those first to see how they handle before going all on with them here and something unexpected happening. I learned in an almost disastrous way that you have to be very sure of your materials when casting anything. Its not from the aspect of getting a good pour or part, but its in the interests of your own, or my own safety. I dont ever harp on about safety regards pouring metal as I assume most people have enough cop but this is worth mentioning I suppose. I had bought core glue with no data sheet of its composition a good while back. I glued up all the moulds and let it sit for 12hrs before pouring 3 litres of aluminum. As it turned out the glue had small traces of water in it and the result was a large explosion approx 2min after pouring. I suspect the delay was due to nucleate boiling or some other effects once the metal contacted the glue. The end effect was the all 3 litres of the alloy exploded out of the top of the mould and showered everywhere. 2/3s of it stuck to the roof above. I got covered in the shower but amazingly none of it touched my head, face or hands. I did ruin all the protective gear though and I had to strip off most of it there and then as it started to burn through. Ever since that I always test new materials before hand in a small mould and go over the data sheets with a fine comb. The one thing that will kill you or leave you with no face when casting is dampness, either damp ingots dropped into the crucible, damp moulds, or leaving the crucible on damp concrete. Letting the crucible fall or it spilling out of a mould at the parting line is trivial in comparison. Here is said roof - it was maybe 13ft over the mould at the time of the explosion. A large piece about an inch thick and 2ft in diameter had to be hacked off it when it cooled. Here are two videos of what it was like - Id rate it midway between the two so PLEASE TAKE CARE ALL and be sure of what you are working with. [video=youtube;hdtZVgOIPC0]http://www.youtube.com/watch?v=hdtZVgOIPC0[/video] [video=youtube;o44IFQvTiC8]http://www.youtube.com/watch?v=o44IFQvTiC8[/video] Did a few dry builds of the various mould components to see how they fitted together and how it all lined up. The internal features line perfectly when the parts are constructed with the outer sides flush so Im not going to bother with location pegs on the other 3 build sets. While all this may seem slow the next three runner sets are going to be a breeze and only take 2hrs to complete. Once the first one is checked and fits ok, as long as you do all the rest the same then they too will be fine. You would always build one first and not all 4 moulds incase there was a problem somewhere along the line. That way if anything crops up you are only one mould down, and not four. Like everything consistency will ensure good results with all parts. Consistency covering pour time, temperatures, sand tightness, glue amounts, and so on. Here are a few shots of various parts in place - here the entire mould is upside down and will get built the right way up but since its just for checking it doesn't matter, You will notice at the parting line below where both halves meet and are radiused at the edges. This radius is a result of the runner fillet in the mould box and deliberate. It has the benefit of forming a wider flash/parting line on the runner and tends to camouflage any miss match between the two mould halves when its ground off. Brian,
Only a bit done this eve, said block that got glued up ready for milling, Machined, Two pieces displaying the parting line get screwed and glued on before un-screwing the jig board on the other side, Turned over - the jig board which the thick block was screwed to prior to machining - mind the screws when milling! - Jig board now off - doing it this way means everything gets fixed together accurate, I have yet to fix on the other side to match - all out of scrap MDF so need to pick up some more. The veneer curve guide is out again and will feature in the next update, Brian,
More work on the core box, Instead of cnc'ing the other piece for opposite side I cut out with jigsaw and trimmed with a 1/2'' flush trimmer. The MDF scrap was small, and jigging awkward so this is the other option, Oversized rough cut piece fixed to side for trimming, Router, and flush trimmer bit, Off, and fixed now to other side, Mr Veneer features again... Its still the correct size...just checking.... Two pieces of Mahogany get inserted into ends, The Mahogany gets coped out to match the half section of core shape either end and more goop is applied between, The Veneer template is tacked to side, The goop is raked out in a shape closely matching the runner core profile - this is to provide backup for an accurate skim of filler to give it its final shape. The goop is dirt cheap, filler is not, Once that sets up it will get an accurate final skim of filler to bring this core box half to its final shape. The veneer is used to guide the scraper as with the main mould boxes. The other half of the core box can be done in a far quicker way once I have the shape of this one pegged down accurately, Looks like the sand needed to cast all 4 runners wont be here until Friday so there is no major panic with epoxy/filler drying times on the core box - plenty for doing in the mean time in the line of core glue and coating testing which I may as well show too. Got the carbon tubes ordered for the runners over rocker - was going to make on a mandrel with pre-preg and heatshrink tape but I think it would be an un-necessary waste of time considering I can buy what I want off the shelf ready made(for once!) Brian, Edit, forgot these from the last post! Chopping the belts to width, belt loaded onto a tight fitting mandrel - lathe on, slice with stanley knife, simple as that! Ya can cut belts this way down to .2mm wide if you want to, not that ya'd want one that narrow...
When this is finished brew up some 4 valve heads for flathead fords and the guys on H.A.M.B will wet themselves
The blocks on those are interesting with the side valves...you would have to deck the hell outa them to get rid of the path between chamber and valves to make a gasket face! Brian,
http://www.museumofamericanspeed.com/Ford-Flathead-V-8-Ardun-OHV-Conversion,9123.html lots of companies tried it, did you see Mark's investigations the v12 308 guy? can't wait to see more of this manifold
I read bits of Marks stuff alright - hes going putting one into a boat isn't he? As you know anything is possible but I think you would need guaranteed sales to start into that jigsaw! Not much to report on progress - waiting on stuff, Im guessing the flooding and general havoc in the Uk is to blame Got the top section of corebox finished and all smooth, Thats sitting in blue now waiting for the next step(once a delivery arrives!) Brian,
Next project is the cylinder head, Just checking tracking numbers there and my stuff is finally on Irish soil so hopefully they will show tomorrow and I can pour these babies Sunday, Brian,
While off F1 yet Rob lol Not much to report sadly in the line of deliveries - hoped to get the main one for the weekend but it didnt show. I did get some bits though so here they are, The bearings, yes they are needles, overkill - probably, but why not. Got some circlips too - about 6 times too many lol, and some viton O rings, And onto something totally different but vital > Rolled to a tapered tube - Fitted onto the core mould half, Thats all to report for the minute until my truck turns up! Brian,
