Like this Frits, just add a lever arm and we are good to go.
Like this Frits, just add a lever arm and we are good to go.
like this, nice and tight.
It lives, runs pretty rough but it does go. Sliding gibs work perfectly, as a throttle so far. Spark plug is impossible to extract with the fuel tank in place, Ill need to make a fancy plug spanner. Default car fuel map is clearly way wrong. Typical raspy sounding rotary valve engine though.
Fast and as ever brilliant work ... but i think you need to give the lawn the Burt Munro treatment ...
There you go, the bugger runs, and idles even.
https://youtu.be/mS0cyt7F9js
This is a cylinder I designed and made moulds for sand cores with 3D printing two years ago. Moulds are painted with primer and filler and hand sanded to nice finish.
I used CO2 method for cores and outer mould. AlSi12 alu. 700°C (1290°F).
First try was a failure. A lot of bubbling (like boiling) when casted. Large bubbles in finished product. Cores did move a little. When casted, I let mould in wooden box. After removing box, I noticed a lot of moisture on walls of a box.
Second casting was also a failure. I removed wooden box prior to casting. Alu temperature was increased to 720°C (1330°F). CO2 mould was buried in reusable "black sand". Result was better but still a lot of bubbles and unusable product.
Third casting was almost a success (usable product). I put CO2 mould in electric owen over night. Temperature was increased from ambient to 80°C over 4hrs and was kept at 80°C (176°F) until morning. We casted next day. Alu temp was 720°C (1330°F). There was almost no bubbles, only one large above exhaust duct which was later welded. Cylinder was CT scanned which revealed also two small air pocket channels between A transfer and water pocket. That was also welded.
T6 heat treatment was done. I did all machining on manual lathe and mill. After that I had no more energy to repeat casting as I needed like 8hrs to make moulds and than whole owen procedure and casting. We were casting in friends foundry that is not close to where I live.
Cylinder design is similar to Aprilia RSA, I did rear wall of B transfers straight. If that will not work I will fill it in hook type. It has exhaust valve as small as possible to ensure good water cooling around exhaust duct. From the outside, I kept cylinder as original Tomos GP cylinder from 1977 (it will go on classic Tomos GP77 replica).
After that, I sent it out for nicasil treatment and now I just need to make final touches and finish rest of the engine for testing. Cylinder head moulds were sent to my friend, so I am waiting to get them casted.
I am wondering, if someone knows, what did cause moisture in the sand? Is it from CO2 getting cold exiting gas tank and than condensing air moisture in sand? I believe that made bubbles in casting.
Some more pics..
At third try I made attachments for transfer and exhaust core to central core for better stability. Over pour channel and risers additional tube was added for higher pressure. Also sand was better mixed with "liquid glass".
More pics...
Cylinder was dimensionally pretty exact. about 0,3mm misalignment. Exhaust duct was lower about 0,5mm but I made it smaller as I must finish it afterwards anyway.
You can see First try on first and last pic.
Two cones from side are removed after box is filled with sand. Than two steel cones are inserted that are holding water core in place.
Right, several things to note. I use a product called core fix, a sand glue, once fixed in place the cores do not move.
Ingate shape is important, you need a good consistent flow but not a turbulent or too fast flow into the mold, catching gas within the molten aluminium. A wide flat ingate seems to work best. Ingate should be the smallest area so the runner stays full throughout the whole pour.
Sand cores are porous and contain air. When hot metal arrives in the mold this gas expands and needs a direct path out of the mold, not through the molten metal. Gas needs to be able to escape through the core prints.
What sand are you using?
If the sand cores sit around too long they will absorb some moisture, gently pre heating the mold can help. I've got a steel plate with a gas burner underneath, just enough to drive out moisture, too hot and you will wreak the sand mold.
Nice job, most impressed.
Wow, that is a nice job. I used the same method to make core boxes by 3D printing. And the cylinder looks like it could have a lot of power too.
A couple of suggestions. Your pouring channel looks way too large. The aluminium will slosh like crazy inside that and pull air in and create turbulence, mixing oxides to the melt. The mixed oxide films are detrimental to casting quality. The pouring channel should be so small that you can easily fill it right to the top at beginning of the pour and keep it full, and the metal should flow in gently over period of about 10 seconds. To my eye that sprue looks so large that you can pour in the full crucible in just a few seconds. The good thing is that you have a filter there, that will help to slow down the flow and filter out oxides.
I recommend to check olfoundrymans video about the sprue design: https://youtu.be/m6W71s-cd_0
When I poured my iron cylinder, I used a sprue that was around 12mm at the bottom, yes 12mm and 14mm at the top and still that flowed so fast that I had trouble keeping up. I have also poured some smaller aluminium castings using a sprue with 9mm bottom diameter and that worked perfectly well. It would also be very good to reduce the height that the metal has to fall, but as the cylinder is oriented that way in the mold, it is hard to change it. I make a cylinder mold such that the cylinder is lying on it's side so the mold can be less tall.
The tall risers are completely unnecessary and a pouring channel riser makes the turbulence problem even worse.
About the moisture problem. I think the problem might not be necessarily moisture getting into the sand, but that the moisture cannot escape through the sand. How much % of sodium silicate you are using? Usually normal greensand can let the steam through very well when it's not rammed too tightly. It might also be that you have rammed the outer mold too tightly. You could also poke some holes into the mold that go near the cavity but not into it, letting the vapor out more easily.
I think the fill rate of the mold may also contribute to this. If you slow down the fill rate, the water has more time to vaporize and escape upwards through the mold before the mold is completely filled.
Edit: with aluminium, it's good to have a large mass riser to feed the shrinkage, but it should not be very tall just in hopes to increase pressure.
Re Flettner.
I used also this glue, but after first atempt, I made modifications to cores to have "hooks" to attach to central core. Misaligments on second and third try were only here and there on unimportant places. Base gasket fits on exactly, missalignment is around 0,1mm on two places. Definitely I would make it way different next time that I did it now. I wouldn't use this method again for only few cylinders as it was huge time consumer. Lucky for me, my brother did printing and it was free as total print time was like 150hrs and I had to repeat some moulds with modifications. It was my first alu casting project and I did quite complicate moulds for first time. Should pick something easier.
Cores were made one day and pour next day. I agree, moisture could not escape as when I did first pour with wooden box, after removing box, it was wet.
I was using silica sand. I don't have exact info as it was provided by foundry. They never had problems like this before and usually they make outer moulds from black wet reusable sand. I had to do it with silica and CO2 as I had to use glue on few places and glue does not stick to black sand. Edit: I was using 7% sodium silicate. (bonding liquid glass).
Re palezu.
It should have some power yes. I am expecting over 20hp on wheel with first tryouts, but with some time and experiments I aim for 24hp+. Will write more on ese thread if it is ok for you guys, as I am using also variable rotodisc and maybe I will try Frits idea of variable exhaust nozzle, so if that is interesting to you.
Too bad I didn't know that before about smaller (lower height) of inlet and riser channel. It looks like logic to do. Nice video.
Thanks for risers tip, I tought that was my first mistake that I made them too low.
I will attach video of pour somewhere (yt or fb) for you to see.
Here are some machining pictures and finished product..
1st and 2nd pic is machining of exhaust valve hole. 1.5mm milling tool.
3th pic is casted surface on transfers.
4th pic you can see that cyl was not welded enough around bolt hole (I must give it to laser welding shop as it allready has nicasil). Does anyone have suggestion if some JB weld or greenstuff will hold the side of M6 hole? Any other ideas to repair this?
Modifications to the clutch case pattern. Make it more like my 350 Kawasaki but might make the outside cover from carbon fiber? How difficult could it be? Machine out a male and female die set?
There are currently 1 users browsing this thread. (0 members and 1 guests)
Bookmarks