Atlantic Make and Break Fabrication Project @MIT

[DoctorB]

Continuing to chip away at the Atlantic. I have one student working on it with support from lab staff Bill.

Boring out the connecting rod main (it gets babbited):



Couldn't resist putting components together.



Machined base. Still needs flash ground out and babbiting, but otherwise done:



Next up is a new set of patterns for the upper base. That's an interesting part and we're uncertain how it was originally done, but I have a plan. With the upper base, we can pour the bearings and hoping this can get done by February.

[DoctorB]

Peter Kinley came to town and brought the records book for the M&B engines - customers, serial numbers, configuration, etc. Really cool. It looks to cover all engines through 1969. Note the tabs for the various horsepower engines.



Also, our student, Tom, who is now a grad student here, continues to work on the piston when taking a break from more heady researchy things. Happy to have him!

[Breakaway]

Very cool!

Kevin

[DoctorB]

Back in the saddle (sort of) after an 18 month hiatus from campus. Tough to pour iron on zoom.

After shaking out cobwebs and going through the class storage area, we found our material and I have a student who will dive into the manifold. This class is now totally out of sequence and we've lost the continuity of our rotating bench of juniors and seniors, so this activity is now a volunteer side activity for our students - at least through January.
The manifold is actually a little bugger of a part with its cores, but we now have a strategy of which I'm reasonably optimistic. Basically, we ought to pour this part vertically. To facilitate this, we made a three part flask - a standard cope and drag, but with a bottom that allows us to add proper runners and gates. This arrangement should make filling the internal passages much easier. We'll see.

[Canoeyawl]

Good to see it progressing!
Thank's for keeping us in the loop

[nrs5000]

Excited to see this thread/work come back!

[nedL]

Yes! Great to see you up and going again.

[DoctorB]

setting up the piston for the wrist pin




There were two versions of the wrist pin. One was to bush the wrist pin holes and clamp the pin itself onto the bronze connecting rod. The other was to set screw the wrist pin to the piston's internal bosses (an approach to which I'm allergic). In the latter, the connecting rod rotates on the pin. We decided to go with clamping the pin to the rod, but no bush. That is, we have a CRS pin clamped by the connecting rod and rotating in the cast iron hole. We'll see how it goes. We can always bush later.

[nedL]

Beautiful!!!

[Canoeyawl]

Beautiful work! (I want to build something now...)
Your students are getting great experience, which they will remember for a lifetime.

Is that fixture for the piston Delrin and how does the wrist pin get lubrication?

(Engines that use a connecting rod pinch bolt (typically) use a groove or notch in the wrist pin for the pinch bolt itself to "key" the pin and prevent said pin from ever contacting the cylinder walls... If you have seen the damage from a wrist pin dragging on the cylinder you will know why, it is ugly. Pinch bolts are notoriously unreliable in that application and often safety wired)

[epoxyboy]

That's a serious big chunk of cast iron to be hurtling up and down a bore - I didn't realise how big until I saw the pics above. Looking at the big-end of the con rod, I can't help but think that the little hinge pin looks small, but I guess it never really sees much load if this is a two stroke.

Pete

[DoctorB]

Is that fixture for the piston Delrin and how does the wrist pin get lubrication?

Thanks! The fixture is delrin - we have buckets of it for our classes.
Good question about lubrication. There were no provisions in the drawings! I am tempted to drill a couple holes in the inner piston bosses. I'm open to suggestions.

Interesting comment about the groove and the pinch bolt. Kind of like a tailstock lock. I get it. Yeah, there are no details in the drawings about this, but the pinch bolt is moused with wire. I'll consider a way to lock the pin laterally.

[DoctorB]

That's a serious big chunk of cast iron to be hurtling up and down a bore - I didn't realise how big until I saw the pics above. Looking at the big-end of the con rod, I can't help but think that the little hinge pin looks small, but I guess it never really sees much load if this is a two stroke.

Pete

The pin diameter is to the print, though we put in a shorter pin just for machining the rod's faces. Hate to whack it with our precious shell mill.
It is a 4hp two stroke, so we're running with that.

The journal end will be babbitted and its bolt moused like the clamp bolt at the wrist pin.

Correction: I just looked at the print. The bolt at the crankshaft end will be longer than what we (temporarily) used and will have a castle lock nut and cotter pin.

[Thad]

I put a 5hp Atlantic (NOS) in a boat (Danish Seine Boat). Talking to a man from the Lunenburg Foundry, he said the 5hp is more like 18hp in reality.

[DoctorB]

I put a 5hp Atlantic (NOS) in a boat (Danish Seine Boat). Talking to a man from the Lunenburg Foundry, he said the 5hp is more like 18hp in reality.

It took me a sec to realize you were saying "New Old Stock" rather than "Nitrous Oxide". I was briefly thinking "Holy cow!!" You were, right?

[Thad]

New Old Stock, yes.

[DoctorB]

New Old Stock, yes.

Nitrous would be amazing though.

[DoctorB]

Third attempt at the manifold. First vertical pour. We shake it out on Friday. For now its a wrapped present under the tree.


You're looking at the upside-down mould with its bottom removed. We have a two-part cope (left and right side), and a single piece drag. Runners were carved in the drag on either side of the two-part-cope's parting line. Had we done the cores differently, we could have easily have a runner in the parting plane with vertical gates. The core was made in March 2020, that crazy week we evacuated our students. It is pretty well cured.





First pour of this academic year. 30# and uneventful - just as we like it. It was fun to wake up the equipment and remind ourselves how this stuff worked. "Where'd I put those gloves?" "You have a thermocouple?" "Where's the refractory? "Where's the sodium silicate? "Is the chainfall working?" "Where's the sand?" We took our time and got our bearings. It was like being reunited with the dusty Millennium Falcon.

[robm]

Good question about lubrication. There were no provisions in the drawings! I am tempted to drill a couple holes in the inner piston bosses. I'm open to suggestions.

Are there oil drain holes from the ring grooves? If so, perhaps drill one to the wrist pin boss/bore. Or does that bottom ring act as oil control, so this wouldn't work?

Depending on splash lubrication that deep into the piston looks a bit iffy to me.

And to prevent the pin from moving laterally: teflon buttons at the ends of the wrist pin are sometimes used.

[epoxyboy]

Are there oil drain holes from the ring grooves? If so, perhaps drill one to the wrist pin boss/bore. Or does that bottom ring act as oil control, so this wouldn't work?

Depending on splash lubrication that deep into the piston looks a bit iffy to me.

And to prevent the pin from moving laterally: teflon buttons at the ends of the wrist pin are sometimes used.

It's a two stroke, so the lubricating oil is mixed with the fuel - there is no sump or splashing oil or oil control ring.
The closest thing to this that I've owned was a 1950's Daimler-Steyr-Puch scooter. That had circlips to retain the wrist pin, and a bronze bushing in the con rod.
There were about 5 1/16" holes spaced around maybe 90 of arc on the con rod, and a slit cut around the bushing that lined up with the holes to get some oil into the middle of the pin. If the wear on the pin and bushing were anything to go by, it wasn't a hugely effective system. 150cc, 5hp - one of these ugly beasts FWIW
Puch-1959-Scooter.jpg

Pete

[Canoeyawl]

With a fixed wrist pin the only thing I can come up with is a hole right through each boss on the piston to allow lubrication to enter top and bottom.
A healthy chamfer or slot won't hurt anything and would act as a funnel although the upper one may have to be cut by hand or with a sexy cutter with a removable head. I dream these things up while I'm sleeping...

(I do know that many/most of these old slow turning engines suffer from wrist pin to piston lubrication)
A chamfer on the inside of the piston wrist pin bore might encourage a bit of mix in there ?

[DoctorB]

Nice scooter!
For lubrication, we'll drill some holes in the piston bosses. Other than that, we'll stick to the prints - it worked for Lunenburg, it'll work for us. Plus, we're not exactly running rum downeast.

Here we are unwrapping our little box. Turned out pretty good. Some grinding and machining is next.








Internal passages look good:




There's some swell and a couple little defects from sand collapsing. But nothing a little welding can't solve. I may weld up a steel flask similar to the original cylinder flask from Lunenburg. It will be simpler to pack and still be a vertical pour - which I think is so important for this type of part. Our apprentice, Chiaki, is doing this on a volunteer basis. I'll post as the finish work progresses.

[DoctorB]

Here we are, over a year later. I've had a number of projects going on, but thought I'd share the latest and greatest from the Atlantic project. We re-poured the upper half of the crankcase. Came out near perfect....really nice. We changed our strategy for pouring from previous attempts and had a standard cope and drag with the parting plane at the obvious bottom surface of the part; however, with all the undercuts of that upper section, we made a NINE part upper pattern that jigsawed together and we delicately removed from the mold.


Main pattern part removed



Removing the pattern pieces for what is to become the upper-most section of the crankcase. In this photo, I removed the center three keystoned pieces, and just slid out the upper-right corner. All side-pull stuff for the undercuts.




The part emerges






The student is with me through May, and we only really have the cylinder left. We are going to re-do the manifold first.

It's been a while. I hope you are all doing well.

[hawkeye54]

Thanks for the update, DoctorB - Very informative and great photos









Rick

[nedL]

Thank you for the update. That looks like a great success!
Always following.

And what would you think about the idea of completing multiple engines a day? Lol

[JimConlin]

Wow!
How about some pics of the pattern?

Jim

[Woxbox]

A nine-part pattern? Was that ever standard practice?

[DoctorB]

A nine-part pattern? Was that ever standard practice?

Probably not, but I wanted to make sure the side pulls wouldn't get hung up on anything, so I erred on the side of caution. Not the most efficient method, but we got it done.

A four-level flask would have done it, with the upper-most parting plane being that "top-hat" section. Or more complicated cores to create those upper bosses.

Our first attempt a couple of years ago had a simple vertical parting plane ( the pull direction was along the crankshaft axis). This meant the pattern had a long pull to remove, and the cores had to be dropped deep into the mold. We invariably knocked sand around during assembly. We could have done it again this way, more carefully, but I was more confident in the side pulls of a multi-part pattern than a deep draw. The method was very similar to the way we did the Herreshoff steam engine bed. I'll post pattern pics when I'm back in the shop.

P.S. The top hat section was only seven parts. My mistake.

[DoctorB]

Pics of the pattern.
1. Right-side up showing the seven piece top.
2. upside-down showing the orientation when pulling it out of the flask.
3. Top section as seen once the main body is removed from the mold. (see pic1 from post #163)

You can see from the first two pics in post #163 how we fished out that multipart business - first the center three parts (note the generous drafts), then two parts on each side with side pulls. We made the sides two pieces each because I wasn't sure we could sneak them out of the hole as a single part. Looking at it now, we probably could have done it.

[hawkeye54]

I am always learning something new from this thread - thanks for the updates, DoctorB ! !



Rick

[Ben Fuller]

Be interesting to compare these patterns with the wooden ones for this or a similar engine. I wonder if any exist. Can allowing for shrinkage be calculated or is that black art?

[DoctorB]

Glad to hear it Rick!

Ben,
For grey iron, we allow for 1% pattern shrinkage. The sections are fairly uniform, so we also weren't that concerned about solidification shrinkage, except maybe near the upper cylinder mounting bosses. Even the 1% pattern shrinkage isn't much to worry about, actually. The length is oversized and machined to spec, the bearing shell sections have a nominal 3/16" annular clearance for babbitt, and we face the top and bottom to mate to the bottom base and cylinder, respectively. Then there's tiny, tiny mold swell. But, all this being said, we did account for 1% and the non-critical width was within a few 0.01".

The student who made this returns in early February and will start machining, which I'm looking forward to, because then we can babbitt the assembled base with the crankshaft.

This coming semester, I'm spending the first half of term with iron - students will make patterns, cast, machine, grind, and hand scrape! their own camel-back straightedges. It's a bit of an experiment, but we'll see how it goes. After spring break, we will focus on copper alloys, the objective to make headway with the Herreshoff windlass. We have to cast and machine the capstans, wildcats, and central gear - should be fun.