Rotax 912 1:2 scale model engine build .

[ThomasM]

Hello everyone,

It been over a year now without building anything like a modelengine and i decided to start a new project.

I hope my build will be of interest and you guys are not angry because i did not post anything since the finishing of my model A engine  ....

I will not tell what the original engine is now. Maybe it will be fun to wait until one of you figure it out when the building progresses.....

One original engine i have here at my home and it is teared down because it needs an overhaul. So it is easy to take measurements for the 1:2 scale model engine.

The engine is a four stroke four cylinder engine and still in production......

My modelengine calls for a build crankshaft like the original. I really don´t like build crankshafts and connecting rods that are not removable but the original has a build one so i have do deal with this.....

So here are a few pictures of the build of the crankshaft. I have also startet the mould for the crankcase and maybe i will cast the first part tomorrow or on sunday but the mould is not dry enough now.

The pictures are :

1 : Drilling and reaming of the crankshaft webs interference fits. The stroke of the crankshaft will be 30 mm.

2: Milling the lateral surfaces of the crankshaft webs to exactly the same size. This surfaces will be the reference surfaces when pressing the whole thing together ....

3: Milling the outside shape of the webs

4: Drilling the oil passages. The engine will have pressure dry sump lubrication like the original.

5: Hardening of the crankshaft webs. I think i will only have a chance to get strong enough interference fits when all the parts are hardened. All of the parts are made of C45 tool steel. This is a beautiful and also cheap material ( 0,45 % carbon ). It can be hardened at 750 ° and tempered at 400 °.

6: Crankshaft web after hardening and sand blasting. After the sand blasting i tempered the webs at 400° and cooled it down in oil to get the black finish you see on the other pictures.

7: Most of the crankshaft parts finished. The diameter of the interference fits of the pins is 12 mm . I am a little afraid of them being so small. But the crankshaft of the original engine is also without much safety reserve so we will see.....

8: Setup for cnc milling the four connecting rods. The rods are also made of C45 tool steel. The are not hardened because they will get a 16 mm bronze bearing at the big end and al 10 mm bronze bearing at the piston pin.

More to follow.... Only 8 attachments per post alowed.

[ThomasM]

9: Milling one side of the connecting rod.

10: Finished connecting rod without bronze bearings.

11: First crankshaft segment with pressed in journal.

12: Machining of the bronze bearings for the big ends. The bearings will be 16 mm diameter.

13: Setup for tuning the outside diameter of the bronze bearings to size. I like to use a mandrel for thin walled pieces like that. Outside diameter of the bearings is 18 mm, 7 mm wide. They are pressed in the connecting rods with interference fits ( + 0,05 )

14: After pressing the first segment together i checked for straightness and got 0,15 mm runout. This is not to bad for a build segment but to get it closer i tweaked it a bit with the shown setup. Now i have 0,04 mm runout from the first main bearing to the second one.

15: Milling of the key nut of the front crankshaft part. The key will be 5 x 5 mm and drives the small gear of the gearbox of the engine. The engine will have a reduction gearbox.

16: Hard turning of the hardened and anealed front part of the crankshaft. Hard turning i something i have never done before. In the past i grinded all hardened parts. But turning give also a beautiful surface and the turning insert likes it.

[Roger B]

So I can get as far as a 4 cylinder 3 bearing crank with ball/roller bearing big ends. Every build is good 

[ThomasM]

18: Finished crankshaft in the lathe . The end segment is not pressed in yet but the runout is 0,1 at the last drilling witch is for the end segment. This is better than i expected and more than enough.

Thats all for today, i will post some pictures when i did the first casting of the chrankcase soon.

Best regards,

Thomas Müller

[ThomasM]

Hello Roger,

There are 3 main bearings on this engine... The first one on front of cylinder 1,2 a middle bearing between cylinder 1,2 and 3,4 and one at the end behind 3,4. This bearings will be bronze bearings, pressure lubricated.

The additional bearings at the front ( needle bearings ) and the very end of the crankshaft are to support the gearbox drive ( front ) and the camshaft drive and geared drive for the electric starter at the end ....   

Best regards

Thomas

[Bearcar1]

Thomas, there is nothing here to worry about in terms of no posts for a bit. Life in general gets in the way and at times we all get a bit worn out by the continual strains of constant modelling. I am delighted to see you back at it, thank you for sharing.  


Bearcar1
Jim

[Johnmcc69]

Nice parts!
 Looks like an inline motorcycle engine.(?)

 John

[ThomasM]

Hello John,

I am sorry but it is not a motorcycle engine. It is also not an inline engine. 

Tomorrow i will make my first attempt to cast the right ( ! ) crankcase half. I hope it will work because i have never used vacuum casting with a lost PLA pattern on such a relatively huge part. I have used vacuum casting on my " Model A " engine for the small exhaust and intake manifolds . Now i will try to cast a part 20 times larger with the same system. We will see if it will work tomorrow......

There are some more pictures to show the casting setup. Please keep your fingers crossed everything will work !

20: 3D printed PLA pattern for the right crankcase half. The different colors are because i ran out of white filament while printing. Also my printer is not able to handle the whole pattern at once because it is slightly over the maximum capacity of 160 mm in x-y- direction. So i glued two parts together. Total printing time was 28 hours 

21: The inner cavities where filled with investment plaster....

22: I inserted two pieces of threaded rods to ensure save connection between the inner investment cores and the outer shape which was also made with investment plaster.

23: Finished mold before the PLA burnout in the melting furnace.....

24: Vacuum casting setup for tomorrow. There are holes in the surface under the investment plaster mold and a vacuum cleaner will apply vacuum on the casting progress. Also there will be a flask on the lateral sides of the mold to apply vacuum on the lateral sides of the mold also.......

25: Burning out the PLA. Hopefully the mold will not break....

Best regards,
Thomas

[Admiral_dk]

A Flat Four - also called a Boxer -> VW Beetle is the first one that comes to mind (the original Porche  ) - followed by a few Aero Engines like Lycomings, Rolls-Royce etc. and a Subaru as the last one .... I'm sure there are more to choose from.

[ThomasM]

Hello Admiral-dk,

You got it right. It well become an Boxer ( Flat Four ) aviation engine.

The engines you named in your post are more or less very nice looking engines except for the subaru. I dont want to overdo the guesswork so i can tell it will be a Rotax 912 aviation engine. Maybe not the most beautiful aero engine in the world ( to put it mildly ) but very common.

Unfortunately the casting today was a fail but i thing i have sorted out the mistakes already.

You can see in the pictures i posted yesterday ( #22 ) that i have used modelling clay for children ( child kneat ? ) for filling the 3D printing lines. This was a mistake because while the PLA burns to nothing in the melting furnace the modelling clay does not. Everywhere i used this modelling clay on the casting model there where huge casting defects and i found residuals of it at the very bottom of the mold.

Also it is better to fill the mold from the bottom to the top with a better design of the sprue and gates of the mold.

Hopefully next time it will work.....

[Admiral_dk]

I knew that Rotax makes a nice small Flat Twin, that is very popular among the micro-lite crowd - and I can't say that I'm surprised that they kind of doubled it up.

I will most certainly follow this thread   

Best wishes

Per

[petertha]

I dont want to overdo the guesswork so i can tell it will be a Rotax 912 aviation engine.

Nice. Looking forward to your build. Especially at that scale.

It was pure coincidence I was looking at that same engine crankshaft layout for something completely unrelated. It looks like the inner throws have counterweights. Are you adding that on to your crankshaft separately or different ideas?
https://www.leadingedgeairfoils.com/crankshaft-set.html
http://www.ultralightnews.ca/rotax912/912-914crankshaft.htm

[ThomasM]

@Admiral_dk: The old two cylinder aero engines from Rotax were all inline two stroke. I did a model engine build of the most common Rotax 582 a few years ago shown in the picture. The 4 - stroke 912 was build as a four cylinder from the beginning in the 80´s.

@petertha: Please correct me if you disagree. When you look at the forces of a four cylinder boxer engine you can see that the counterweights are not necessary. Actually there are necessary of course but for a different reason someone could think.

It is a little different to explain in English .... The forces ( first order ) sort themselves out . Look at the picture and imagine that the crankshaft rotates no matter which direction. You can see that the forces of the rotating and oscillating masses of the front two cylinders cancel each other out. The same happens on the rear two cylinders. The only " problem " is that this canceling out of forces happens in a slightly different level because of the cylinder offset of the opposed cylinders. This creates a yaw moment ( the circle arrows ) on cylinder 1-2 and 3-4. The two yaw moments cancel them self out also but this leads to a pulsating bending moment on the crankshaft. On the full scale engine they want to eliminate this bending moment as good as possible to prevent the crankshaft and housing from fatigue breakage.

On the model engine i think this can be neglected.

[Vixen]

ThomasM

Good explanation of Boxer balancing and why counterwieghts are not necessary for the primary balance. With your smaller engine size, the secondary balance moments should be small enough to ignore.

Mike

[ThomasM]

Hello everyone,

After a few mishaps while trying to cast the two halfes of the crankcase i finally had success.

I tried to cast the crankcase with regular investment plaster the first four times. The problem is that investment plaster becomes very fragile when burning out the PLA in my hardening furnace. I also wanted to apply vacuum when casting and the molds broke a few times.

So i changed my procedure. I used sodium carbide bonded silica sand on the mold and the cover with the sprue and risers.
I also added a sprue and gates to fill the mold from the bottom to the top while casting to avoid casting defects.

This led to very solid molds and a successful casting of the two crankcase halfes. Unfortunately the surface finish would have been much better with the investment plaster but with a little filing and grinding the castings are usable.

The pictures :

30: The green 3d printed part is the sprue and the gates.
31: The inner cavities of the crankcase were filled with silica bonded sand first. The steel wires are for connecting this cores to the rest of the mold.
32: The rammed the sand to the outside shape first without a molding box and later i added the gates and the molding box.
33: finished mold before the burnout
34: The cover of the mold is also made from sodium carbide sand. The bigger hole is the sprue and the smaller ones are the risers.
35: It is so much fun to machine a casting for me ! After i established a reference surface on the bottom of the casting i started with the finishing of the cylinder sealing surfaces.
36: The finishing of the bores for the cylinder sleeves with a little cnc program. This bores are 45 mm diameter.

Now i will continue with the machining....

Best regards !
Thomas