MAN Diesel engine from 1907 "DM 2 * 100"

[MuellerNick]

Clamping those patterns:

You might ask yourself how I do clamp all those patterns, no matter how small or big.

Easy for core boxes, they can be clamped in a vise.

Well, I asked that when I started milling them. The most obvious option is using a vacuum chuck. I didn't like that for two reasons:
I don't have one (might cast one next year).
I don't like the idea of having a pump running for hours. Making noise, costing money, and what happens if you have to pause a job?

So I thought about a way to clamp that block material in a way that is solvent based. Well, I had no good idea exactly like that, but something along the line:
I use cut-offs of the block material, mill ribs and glue the stock onto the ribs with CA glue.
Why ribs and block material?
This has four advantages:
I can easily saw off the work with a saw blade (hack saw),
The CA is harder than the block material, so it works like a shield that prevents cutting into the work. Going too much down, you cut the solid material -> blade goes up. Cutting too high -> cut into the hard CA -> blade goes down.
Furthermore, the ribs do give room for the clue the be squeezed to the side, thus making a very small layer of glue (precise reference).
Also, I do have a precise reference for Z = 0, as I mill the surface with that milling bit.
And last but not least, you can mill into the chuck. That's a requirement if you use ball nose/bull nose mills.

After cutting off the work from the "chuck", I file off the remains of the ribs and dissolve the CA with acetone.

Medium viscosity CA works best.

While milling those tiny parts today, I thought I make some pictures.


This is the smallest part I milled for the patterns. Its footprint is 5.5 * 8 mm, 4.5 mm high. Mill is 1.5 diameter, and the cigarette lighter is scale 1:1.


Different view of that part. You can see the thin skin that remained from roughing and the fact that I cut into the ribs while finishing.

Usually, I make those ribs 1 mm wide. Width of the groove depends on the mill's size. Here it was 1.5 mm. Usually, it is 4 or 6 mm wide.


Nick

 

[idahoan]

Hi Nick

I wanted to tell you that I have really been enjoying this thread; your pattern making skills are amazing.


Thanks for taking the time to share your project with us.



Dave



PS. we run quite a bit of Renshape 460 at work making industrial design models. We pretty much tape every thing down to an aluminum fixture plate with 3M double sided tape.

When parts get very small as yours the surface area becomes an issue as the parts want to move around so the fixture with the CA glue would probably be better; I will have to remember that one.

 

[MuellerNick]

Yes, double sided tape is an other option. Doesn't work too well for parts that are high compared to their footprint.

I have used my method for more than 100 times and only once, a part came loose. That was, when I didn't give enough time for the clue to settle (was an old one that got quite slow). No matter what shape, all parts came off easily without breaking them.

For "big" parts, I don't put glue on every rib, but leave out two or three.


Nick

 

[MuellerNick]

The shape of things to come …

Won't be able to cast that tomorrow, as one half takes 8 hours to mill. And the first one is still making chips. 50 minutes per spoke to finish mill.

After roughing:

Will be much more elegant, the spokes are quite rough. No wonder, it is just roughed.

Diameter of the pattern is 430 mm!


Nick

 

[johanvanzanten]

Hello Nick,

I follow your thread for some time and I must say that I admire your work. I made a working diesel engine myself for a few years ago and I know the amount of work involved. I have experimented 1 1/2 year to get it running. Blast injection became a fiasco; I even was unable to get compressed air at 80 Bar safely. At last I switched to solide injection which worked rather well at the end. The engine involved is a 1:10 schale model of the 250/400 engine made by the Diesel team. Please browse my album on the Model-Engineer page to see some photo's.

http://www.model-engineer.co.uk/albums/member_photo.asp?a=6778&p=123122

I whish you lots of succes with this ambitious and interesting project.

Kind regards, Johan van Zanten (The Netherlands)

 

[AussieJimG]

This is a wonderful thread. Merry xmas Nick.

Jim

 

[MuellerNick]

Hi Johan!
Thanks for your wishes! I already knew your wonderful work. And I have read about your tries to get blast injection working.
I already learned that it is almost impossible to make a compressor to 80 bar working at that scale. Getting him tight is not the problem, but dead volume is.

I'll see how injection works. I have collected a lot of ideas over the time. They "just" have to work.


Nick

 

[MuellerNick]

It is Sunday, and you deserve new pictures!

The not-that-much-shape of things to come:

That's how it looked when I came back from dinner. I started a roughing job right before leaving.

Casting #38. Fence post, the short one.
After having too much trouble with the long version, I thought what I could improve. Added more ingates. I tried that with the long version, but also thought that I should try it with gravity feeding. So I did that with the long version and it worked. Being confident, I did the same with the short version. And it worked too!


As cast.


Detail of the crown, no deburring etc.!


Core box and core.

Next two parts will need more pictures, so a new posting. I'll be right back …

 

[MuellerNick]

Casting #39 & #40. Elbow for the exhaust pipe.
I have to admit, it is not finished. But casting-wise it is. The pattern needs some re-work and I need to make a mandrel and a roll for my mandrel bender. More about that later.


This is the left and right one.


Left one held at the cylinder head (a reject).


This is the core box for the outer contour of the pipe. On my left finger sits a tiny loose insert. It is color coded, as the left and right one are different.


After rattling out the core, the loose insert still sits in the core. Pulled out inwards. It just slips out of its pocked in the core box, so that's why they call it loose insert.


The core box for the inner contour. In the lower left cavity, the core still sits in. The two red circles are reminders for me. The two red lines is the "ingate" for the sand. This is the only access, because part of that core will be a stainless steel tube ...


That SS tube (to come) will replace the cylindrical part (red rectangle of the core) and keep the core in place. That core is a shell around the tube and builds a water jacket. Yes, the exhaust pipe was water cooled.
On the left core that I'm holding, you see pockets that allow enough wall thickness for the pockets you can see on the outside of the casting.


Cores for left and right (cast both in one go) and with the cores inserted. Just the cores alone doesn't work that well, the construction of the mold relies on the tube. That tube will close all accesses to the outside, just leaving one bore for the metal do go in (middle bore at the very lowest end of the picture).

Well, that's it for today!

I have two more patterns finished. Maybe I'll cast them tomorrow. They are too big (one of them) to be cast inside. Depends on the weather. Forecast is 12 °C.


Nick
... and a merry X-mas to all!

 

[MuellerNick]

I thought, I'll make myself a Christmas present … but it went wrong.


This should be the exhaust pipe. Here, with all the runners and ingates etc.
It wouldn't be me, if I wouldn't cast it double walled like the original. There is a waterjacket around the inner tube, and I wanted that too! Wall thicknesses would have gotten too small, so I cast that around a stainless tube with 10 mm OD.
Does look like a near miss.


But from the rear, it is a clear fail.


Looking at the face, the SS tube shifted. It was bent by the heat of the melt. The reason was (I guess), that the runners weren't fed with the same timing. One got the melt a bit earlier.


Core around SS tube in the cavity. Yes, it is almost 400 mm long.
You see a gap in the core that allows access of the melt to the tube. That was intentionally. It seves as a runner and to make all a bit stiffer. Maybe that was a bad idea and I should isolate the inner tube completely. I'll partially patch up that gap with oil bound sand the next time.


Detail.


Pattern and core & tube. The nibble on the core points upwards when casting.


Nick

 

[AussieJimG]

Still absolutely amazing. Merry Xmas.

Jim

 

[Banjoe]

What an incredible journey. Many thanks for bringing us along on this adventure!

 

[starnovice]

Do you need to pour at the middle of the pattern instead of at the end?

Pat

 

[MuellerNick]

Do you need to pour at the middle of the pattern instead of at the end?

I think the runners on both sides distribute to the ingates evenly enough. The runner going to one end at just one side was more like a test wether it makes a difference. It didn't.
As far as I see for now, the main problem was the core. I rushed that a bit too much and didn't pay enough attention to equal thickness. Also, the way I glued the core to the tube was a bit counter-productive. I used CA. The problem with CA is, that I made that at the place where I cast the cores. The amin (ammonia) activates the CA in a funny way, so it expands a bit. I'll try a different clue and pay more attention that the OD of the core is to specs. Also a better runner (pay attention to equal timing) and the runner in the core different (haven't decided yet how exactly). And higher temperature (700°C).

I'll have a mid-year's party tomorrow with some of my friends. So I desperately need to clean my shop (doesn't hurt at all). I'll have time for casting this Saturday.


Nick

 

[MuellerNick]

25 minutes ago, I dropped the last pattern/core box in the vise.
Just now, I made the last toolchange!

Supposed all works well, I should add.

I'll have a beer now. When my Maho is finished at 2 o'clock in the morning (30 minutes from now), she'll get a real strong hug and a big kiss on here oily spindle.


Nick

 

[metalmad]

Hi Nick
Does this mean your nearing the end of the project?
Please say its not so!
I love this Thread
Pete

 

[MuellerNick]

Does this mean your nearing the end of the project?

Do you think I'm finished when I have a pile of un-machined castings?

I've got 4 more small casting, fresh from the foundry. All were cast with vacuum-assistance.

Casting #41. Coolant passage between cylinder head and A-frame. That one is similar to the passage on the compressor. But it is a bit bigger.


Two with one shot. Will be cut off at the red line.


Different view.


Core and core box in the background. I need one passage per cylinder.

Casting #42. Lid for the flywheel bearing. It should look familiar to you, but it is a bit bigger than the other one.


Two of them.


View from the back. The right one is the smaller (old) one.


Core box. No, that's not a crack in the lower left. I patched up some cutoff. The gap doesn't hurt in that place.


And core.

More in a moment …

 

[MuellerNick]

Casting #43 and #44. This are two different parts, so they do have different numbers. But they are cast in one mold. Both belong to the service platform.


This is a clever (by my standards) mould. One part has a split line, and I don't need that much of them. The other one is a simple pattern and I need more of them. So the split one went to the side of the runner, the simple one to the top. As two of the core molds are needed (same, just flip one over), I get 6 of the simple one.

Casting #43. Mount for the ladder to the platform.


It needs some milling and fitting at the red line.

Casting #44. Mount for the platform's fence posts.


Will be cut off at the red line. The eyelet will be bored, and the post will go into that bore.


Core box and core.

Well, that's it for today. I'll prepare a bit for casting tomorrow. I'll see how far I get.


Nick

 

[MuellerNick]

Things not always work as planned …

On this Sunday, I thought I'll cast my last parts (at least for the first time, The A-frame, the flywheel and the cylinder head need one more casting). So a nice coffee, some relaxing, smoothly prepare the moulds…
Smoothly?!
The cope for the flywheel broke when I assembled the mold. But it looked like it still works. So I fired up the furnace outside and waited for the melt.
While pouring, I knew that this won't work. I throtteled the gate, as there is a lot of mass to be cast. But not that much! It took to long and I had my doubts about the volume.

Breaking the mould, it didn't look too bad …


Until I turned it over.


So I decided to make an other mould for it, while cranking up the furnace to try that one in CI.
I made the mould, assembled it, and … it broke. So I went out to have a look at the furnace. It burned. It burned inside and at the outside. Oil dripped out of it and started to burn on the floor. Disconnected the fuel and let that crap burn.

Saturday, I had prepared a new core for the exhaust pipe. This one, I can cast in my shop. All went good. Mould did not break. Casted it and … well. At least it got way better:

There is an other void on the back, even smaller than the small one. Judging from the surface, I have left room for an even hotter pour. This one was 700 °C.

Hah!
I have an other casting to do. This is the dumbest of the whole project! Maybe this one works. Lets see…

Casting #45. Coupling between flywheel and generator. I need two of them.


I think I don't need to show the pattern for it. The pattern was made in the kindergarden (right here in my shop, I guess).

Tomorrow is another day. :fan:
Nick

 

[MuellerNick]

Thought I like this. Maybe you like it too.

Note: There are a lot of rejects in this mockup. And hot-melt glue doesn't allow precise alignment.















And a happy 2013 with no rejects to all!

Nick