Dake Engine (drawings by Pat J)

[GreenTwin]

The first thing I need to do is open each part in Solidworks, and figure out the mass of each part.

I scaled my original drawings up by a factor of 1.5238 (don't ask me where I got that number, I am sure I had some logic reason for choosing it).
So my flywheel went from 4.3" to 6.55" diameter.

I will list the masses here, in gray iron.
I need this information to determine which size crucible to use, and how much iron to melt.

An A10 crucible will hold about 25 lbs of iron.

Flywheel mass: 3.9 lbs.
Base: 12.2 lbs.
Crankcase: 14.4 lbs.
Crankcase cover: 10.5 lbs.
Crankshaft housing: 4.2 lbs.
Valve body: 3.7 lbs.
Valve body cover: 0.80 lbs.

Additional iron will be needed for each mold for the sprue, runner(s), gates, and possible risers.

.

 

[GreenTwin]

The idea is to use these two auto body compounds, called skim coat and seal skin.

I have never used this material before, but I am hoping it will work well to fill in the lines in my 3D printed patterns.
I just received this today, so perhaps I will try it out tomorrow.

The skim coat is for vertical surfaces, and it is not suppose to sag.

The seal skin is for flat horizontal surfaces, and it is suppose to be self-leveling.

These have to be mixed with a hardener, but I think they will work well.
I have seen the autobody folks use a plastic skimmer to apply this material very thin, and then just do some light sanding.
We shall see how it goes tomorrow.

Edit:
I was hoping to stay with a water-based filler, but I think this material will adhere to the plastic better.
This material is toxic though, so I will have to wear a commercial chemical respirator unfortunatly.
.

 

[GreenTwin]

And in my haste to 3D print some patterns, I totally ignored things like core prints.
I think I can add those to the patterns without too much trouble.

The core prints will allow the cores to be supported by the sand mold.
.

 

[a41capt]

Okay, I followed you over here next! Still looking forward to seeing the pours and machine work.

Thanks,
John W

 

[GreenTwin]

The question was asked by joshsapps:

Please forgive my ignorance. What is the valving system on the left side of steam chest in the photo of post #26?

ie: Referring to the photo below:

Answer:
That is actually an oiler.
The horizontal thumbscrew adjusts the drip rate.

 

[GreenTwin]

You can see what looks like an oiler in this original Dake engraving, and so my dad copied this engraving as far as the oiler.

Other things that can be seen in this engraving are petcocks on the lower sides of the crankcase, to drain condensate, and adjustment screws on either side of the crankcase, to adjust the shims in the bottom of the crankcase.
The shims in the bottom of the crankcase allow a sliding fit of the shims against the bottom of the outer piston.

The engine below appears to have a belt-driven governor with flyballs, and an associated throttling valve.
A governor would seem to indicate that the engine below was run in one direction only, since the governor would not work in this position if the steam and exhaust were reversed.

 

[GreenTwin]

As far as machining the casting for the Dake, I would use the same aproach that I used for the green twin, which is as folows:

1. Machine the inside of the crankcase, using the lathe for accuracy.
2. While the crankcase is still mounted in the lathe, face off the mating flange surface.
3. Machine the face of the crankcase cover in the lathe.
4. Fit the crankshaft assembly to the crankcase, making sure the crank disk is flush with the interior surface of the crankcase.
5. Measure the interior depth of the crankcase, and machine the inner piston thickness to match to the exact dimension of the crankcase interior depth, plus perhaps 0.003".
6. Machine the outer piston to match the thickness of the inner piston.
7. The original Dake engines (as I understand it) were supplied with a range of metal crankcase gaskets/shims, in increments of 0.001".
The fit between the pistons and the crankcase cover was adjusted using these shims/gaskets.
This tells me that the original engine must have had pistions that were a bit taller than the crankcase, to allow for wear adjustment.
8. I think the original Dake engine internal parts were precision ground.
I can't precision grind my parts, and so the plan is to machine the pistons, dye them along with the interior of the crankcase, and then assemble the engine, and rotate it slowly by hand.
By observing the high spots (the spots where the dye is scraped off), on can scrap off the high spots, much like scraping the bed of a lathe to achieve a flat surface.
9. The pistons could be lapped to the crankcase using a bit of 600 grit lapping compound, being careful to avoid getting compound on the crank pin, or in the bearings.
The engine will wear-in to some extent, so I don't think lapping is really critical.
It should be noted that a Dake crankcase wears unevenly, and it wears more as you move away from the centerline of the crankshaft, since the velocity of the pistons is greater as you move out radially.
With the correct lubrication, the wear of the Dake internal parts does not seem to be a problem, and I have seen some very old Dake engines that still run perfectly.
10. The funnel-shaped crankcase housing should have the flange faced first, then it should be mounted in the lathe with that finished face hard against the lathe chuck. I plan to press in the bearing into the crank housing, and then with the housing in the lathe, final bore the bearing.

As has been mentioned in some of the Dake sales literature, there are only 3 moving parts to a Dake engine, which is somewhat correct.
The three moving "parts" are the inner piston, the outer piston, and the crankshaft/flywheel assembly.

From the sandpoint of machinging parts, this engine should be quick to machine, since there are really very few surfaces and parts to machine.
The key to success with this engine is accurately machining each part.

I tried to make a few parts for a Dake years ago (not for this build), just to practice machining techniques, and I discovered that my mill lacked the rigidity to make accurate engine parts.
For this Dake build, everything will be turned in the lathe, with no critical surfaced machined by the mill.

An actual Dake engine inner piston has an external flange, and an internal flange.
I omitted the internal flange in my drawings, but for accuracy, I think I will add it back in.
The engine runs fine without the internal piston flange, but the flange gives a larger bearing surface to ride over the crank disk.

Below are a few of my practice Dake part machined pieces.
I recall getting one Dake piston at about 99% complete, and then the part slipped out of the jaws, creating huge gashes in the piece.

It took me a lot of ruined parts to figure out how to machine metal accurately.

.

 

[GreenTwin]

This was an attempt to cut out a section of the crancase.
Not too bad of an attempt, considering this was some of the first machining I had ever tried.

One person commented "You got the bottom of the crankcase machined wrong".
Nope, the bottom of the crankase is suppose to be sloped.

 

[GreenTwin]

This is the inner piston that I machined out of gray iron.
I ruined this piece at the last second.

I made one of these using the milling machine, but its thickness was not consistent.
I made a second one in the lathe, and it was consistently flat.

And this piston has the back flange on it, as was the original Dake design.

The scribed lines on the face of the piston indicate the hidden passages inside of the piston.

I did not consider this to be a particularly difficult piece to machine, even though I did ruin this piece.
A machinest with any skill at all could make this piece accurately.

 

[GreenTwin]

This is Charles Docksteader's open-source program that he used to model a Dake engine.

It should be noted that the ports in the inner piston often varied in shape from one Dake engine to another, and could be rounded symmetrical, squared-off shape such as shown in the simulation, or assymetrical teardrop shaped, and this is not reflected in the Docksteader program.

I think the Dake port shapes depended on the intended use of the engine, ie: whether it would primarily be used running forward all the time, or be used alternately forwards or backwards.

The exact port shape also determined such critical factors as admission, cutoff, compression and release.
If the Dake needed maximum torque, then a very late cutoff could be used.
For maximum efficiency, and early cutoff could be used, while sacrificing some torque at low speeds or no speed.

If I remember correctly, the assymetrical ports were used to create a fairly typical set of valve events.

Edit:
It should also be noted that the program below is only showing what is happening on the left side of the outer cylinder, for simplicity sake.
But the steam admission and exhaust also happens on the top and bottom of the inner cylinder, as well as the left and right side of the outer cylinder.

 

[GreenTwin]

These are my illustrations of the asymmetrical port configuration, and the associated events.

 

[GreenTwin]

Here are some early attempts I made to make Dake patterns using wood.
This was before I learned 3D modeling, and before I purchased a 3D printer.

.

 

[GreenTwin]

A few more wood pattern photos.

One lesson I learned along the road to making wood patterns was to use a router mounted on the lathe toolholder, and turn the lathe chuck by hand, to avoid the disaster that you see below where the wood piece hung up on the tool bit.

I don't care for wood lathes, and I don't trust anything spinning that fast.

These were some of my rather crude early attempts at making patterns in wood.
.

 

[minh-thanh]

Hi GreenTwin !
I have seen many of your topics about casting., there is always something new to learn
That will be the document I have to refer to when I start casting
Thanks for sharing !

 

[GreenTwin]

Thanks for the kind words minh-thanh,

I really enjoy the metal casting hobby, and I am glad others like to read about it.

When I was learning how to cast engine parts, there were very few examples to be found anywhere, so helpfully this information will help others see how it is done.

Pat J

Edit:
I have more backyard casting info at this post:
https://www.homemodelenginemachinist.com/threads/home-foundry.33291/.

 

[kvom]

Here's a build thread from 10 years back of a Dake model. Potty Dake Engine

 

[GreenTwin]

I posted my Dake drawings on an old website in about 2011, and a fellow named Stew Hart saw them, and asked if he could build a Dake model from my drawings. I told him that one day I would build and perhaps publish a Dake myself, and so I told him he could use my drawings, but that he should morph his build into something a bit different, and not just outright copy my design.

My "P.Jorgensen copyright 2010" Dake drawings above (posts #5 and #6) are the drawings that Stew used as a go-by to build his Potty Dake engine.

Stew states in his first post on his build:

I've had an ankering to build a Dake engine for quite some time, when I came across information on Pat's fathers Dake engine, since then I've researched the engine quite a few times each time I've advanced my understanding and spent a long time thinking about the best way to machine one from bar stock.

The "information on Pat's father's Dake engine" that Stew came across was my drawings above in posts #5 and #6.

Great Dake build by Stew.
He is a talented builder.

.

 

[GreenTwin]

And a bit more work on the Dake base pattern.

Here is the epoxied base pattern pieces, and you can see there is some misalignment.

I epoxied the pieces together with them resting on a flat surface, upright, which misaligned the top surface.

I can deal with leveling the top surface with filler, and this will keep the bottom of the base level and flat.

 

[GreenTwin]

And here is the filled piece.

The filler looks to be big gloppy mess, but I would say 98% of the filler will get sanded off.


One method of sanding wall patch I discovered a few years ago is using sanding screens.
Sanding screens can be used with a stick-mounted sanding block, or a hand block.


The screens have an open mesh, and so they allow the dust to pass through, and allow the screen to continue cutting.

Sandpaper tends to clog with wall patch compound to some extent, but I do use a lot of sandpaper for pattern work.

Luckily this wall patch filler is not too dense.
If this were bondo, I would be sanding the rest of my life.

 

[krypto]

A very interesting engine and I appreciate the time you have already spent on making and posting the plans and documenting your build.

Just add your .STP files into a ZIP archive and then you can upload that file into this forum. It's also easier to download one container file over a multitude of separate files.