Mercedes or BMW Inline-6 Info

Thanks so much for this Mike. A water flow path: pump outlet > lower jacket tubes > flooding upward within each cylinder jacket in parallel > head cavity > along inter-head horizontal tubes > radiator > back to pump inlet is how I envisioned it. That seems to be the Mercedes mode, at least in these particular pics. You can see the feed tube from pump starting on LHS side of engine, but immediately ducking behind rear cylinder & then into the main horizontal lower line array on RHS. If so, this means each cylinder's jacket cavity is connected to head cavity.

The BMW picture shows what looks like pump outlet directly connected to upper head tube of rear cylinder. Its not immediately clear to me how the lower jacket tube is connected. Potentially missing plumbing in the display engine does not help matters. Or maybe a different cooling arrangement? The Airplane Engine Encyclopedia excerpt mentions dual flow which I highlighted pump has two outlets, one leading to the bottom and the other to the top of the end cylinders. Initially I thought maybe the reverse of Mercedes: feed flowing horizontally along upper tube & draining down each jacket into lower tube? But that does not sound efficient. Hot head water goes down into jacket & potentially getting progressively hotter as it moves forward head to head. Or maybe I'm overthinking how cooling actually works. A normal auto engine has an inlet to & outlet from the whole bank of cylinders annular passage.

Back to the model version I totally agree - complicated! I am aware of Stephens amazing copper work. Somewhere on my PC I have some documents describing his methods. Maybe from an article or maybe from YouTube excerpts I forgot which at the moment. What an achievement, it made my head spin.

What is in the back of my mind is a modelling cheat of sorts. Not easy but hopefully 'easier'. The cylinders would be one circuit from the pump meaning the annular jacket & lower tube segments, flowing from cylinder to cylinder in parallel. The head would be an independent circuit from the the pump. Probably naively visualizing a horizontal hole drilled in the center of an otherwise normal solid head located between the valves. Water flows from head to head via the upper tube-etts connecting this hole. Some of this brainstorming might be clearer with some section sketches which I will embark on. But just wondering out loud if its a train wreck idea for some reason beyond my limited understanding. For one thing I'm having trouble visualizing flow in the lower line. Its not a conducive flow path to circulating around the jacket of cylinder A & then into cylinder B because the pipe is 'low'. Seems to me it would want to flow predominantly along the pipe & water in the jacket space being relatively stagnant?

Quote from: steamerAs far as a crankshaft for a running 6 cylinder engine, you might want to look and George's 300 6 cylinder engine, or any of the Shillings drawings.  There are many running Shillings out there. 

Dave

The Shillings concept was the complete opposite of mine. He went for simple mechanics and a good external appearance to match the prototype. I am concentrating on the internal technical details and making a box to keep them in.

His book is currently available, but obviously in German, the pictures are good:

https://shop.vth.de/boxer__reihen-_und_v-motoren_als_modell__pod_-3120033/

I do agree that the BMW/MB engine cooling concept seems to be a feed manifold at the bottom of the cylinders and a collection manifold at the cylinder heads giving a  parallel feed to all 6 cylinders and reducing the temperature unbalance.

The aircraft engines also seem to have a significant hot water feed to the carburettors to stop icing.

Quote from: steamerAs far as a crankshaft for a running 6 cylinder engine, you might want to look and George's 300 6 cylinder engine, or any of the Shillings drawings.  There are many running Shillings out there. 

Good leads Dave. I thought the GB-300 was a gear bank driving camshaft but maybe I'm mistaken. Adding to the list, Ron is has a similar bevel gear shaft shaft drive on his Hispano 8-cylinder work in progress.
Schillings had a couple generic section views of bevel gear shaft drive, not sure if it was applied to an inline, but same generic principle. Many 'details' to conquer!

Quote from: petertha

Quote from: steamerAs far as a crankshaft for a running 6 cylinder engine, you might want to look and George's 300 6 cylinder engine, or any of the Shillings drawings.  There are many running Shillings out there. 

Good leads Dave. I thought the GB-300 was a gear bank driving camshaft but maybe I'm mistaken. Adding to the list, Ron is has a similar bevel gear shaft shaft drive on his Hispano 8-cylinder work in progress.
Schillings had a couple generic section views of bevel gear shaft drive, not sure if it was applied to an inline, but same generic principle. Many 'details' to conquer!

Apparently, I misread your post.  I thought you asked for a crankshaft from a running engine.  My bad  Yes Shillings does have a few bevel drive camshafts in his plan book as well.  The BMW model looks fairly straightforward.

Dave

You might want to look at his DB603 model as well, which is also bevel gear driven, and fuel injected.....sort of.    He was very prolific!
He has build photo's....


Dave

Quote from: steamerApparently, I misread your post.  I thought you asked for a crankshaft from a running engine.  My bad  Yes Shillings does have a few bevel drive camshafts in his plan book as well.  The BMW model looks fairly straightforward.
Dave

I am apparently asleep at the wheel this morning. I was aware of Schillings sketch you show, caption even references a BMW IIIa. Somehow I confused myself & posted a bevel shaft drive from another variant, maybe it was a boxer. Anyway, he has a nice layout & even depicts a water jacket on the cylinders (only) which is encouraging.

I was giving the construction a bit of a thought while the CNC was pattern making.

I think a cylinder could be done in three parts, looking at the Vintage aviation photos the main cylinder would be much as they have it threaded at the top and flange at the bottom coul dbe iron on steel. The "head" complete with the side entry spark plug boss etc again much like they have would screw onto the cylinder. However rather than try and enclose the head in a water jacket make from solid steel with the correct outer profile and solder on the 2 water, inlet and exhaust bosses. The third part would be a "tube" which would be sandwiched between the head and a rib part way down the cylinder and that would form the water jacket. If steel or brass then the connecting water bosses could again be easily soldered on. Seal the three with JB Weld and once painted the joint won't show. Might just need to reduce the bore slightly to give enough thickness for threading but something like a 0.5mm pitch thread won't take up much depth

As for cooling the "solid head" a simple straight horizontal hole passing through the middle of the Vee of the valves would provide a reasonable amount of cooling to the head. You could even sleeve all six heads with a length of thin KS type brass tube then only need worry about a water tight joint at either end.

Maybe these markups will help illustrate my idea (and potential limitations). The Schillings front view shows the cylinder water jacket being fed from one side & out the other via a single nipple. The inlet & outlet nipples are both 'low' n the jacket. They are tied together in series with a similar horizontal feed line. The left mirrors the right. To me, this is a more effective flow path, water has go around the cylinder to get to the other side. What I was trying to say before is: trying to stay with the FS engine SINGLE horizontal lower tube on one side alone would not achieve this. IMO, the water flow would just mostly short circuit mostly predominantly down the tube & not around the liners.

I also shaded the (separate water circuit) head cooling hole idea, roughly where the fore/aft hole would be drilled in solid head. It looks very crowded at this scale & presence/orientation of (glow) plug amongst other challenges. But a larger scale engine & smaller shop made spark plugs could help alleviate? I mean even if head water cooling was only 20% effective, that's 20% more than jacket water only. The jacket chamber top is vertically limited, the hottest area is proximity to combustion chamber.

Yeah not impossible, but a real PITA I would think

See sketch

Not sure you will need that honestly on a model, but I suppose if you're going to go exact, you need to consider creating the passages in the head that give you that

Quote from: steamerYeah not impossible, but a real PITA I would think
Not sure you will need that honestly on a model, but I suppose if you're going to go exact, you need to consider creating the passages in the head that give you that

yes, the water interconnection from head to jacket is specifically what I was trying to avoid by keeping the 2 circuits separate from one another. Water jackets of cylinders connected together on one line, heads connected together on another separate line.  So head water would basically be going in a straight line through the head center hole > interconnect pipe > next head hole etc. For sure a solid head has more mass than a thin walled shelled head like the FS engine, so less wetted area & turbulence, but I figure much of the valve features wont scale perfect anyway. The FS valves are quite large relative to bore as are their respective ports. Just a guess I think I could get away with say 3/4 valve size & that would make the cage diameter about right to traverse a decent hole between the step in the cage where it reduces for valve stem. What I don't have a feel for is relative pressure between the 2 flow paths. If one is more restrictive than the other (likely but to what degree) but they presumably meet at the radiator common pressure, how does that work itself out. Is some kind of check valve needed to prevent back flow?

Quote from: petertha

Quote from: steamerYeah not impossible, but a real PITA I would think
Not sure you will need that honestly on a model, but I suppose if you're going to go exact, you need to consider creating the passages in the head that give you that

yes, the water interconnection from head to jacket is specifically what I was trying to avoid by keeping the 2 circuits separate from one another. Water jackets of cylinders connected together on one line, heads connected together on another separate line.  So head water would basically be going in a straight line through the head center hole > interconnect pipe > next head hole etc. For sure a solid head has more mass than a thin walled shelled head like the FS engine, so less wetted area & turbulence, but I figure much of the valve features wont scale perfect anyway. The FS valves are quite large relative to bore as are their respective ports. Just a guess I think I could get away with say 3/4 valve size & that would make the cage diameter about right to traverse a decent hole between the step in the cage where it reduces for valve stem. What I don't have a feel for is relative pressure between the 2 flow paths. If one is more restrictive than the other (likely but to what degree) but they presumably meet at the radiator common pressure, how does that work itself out. Is some kind of check valve needed to prevent back flow?

Keep in mind    The heat generated will go as the 3rd power of the scale.    1/4 scale  or about 1/64th the amount of heat.  Lots of these models run without coolant of any kind as the run duration is short, and they're not trying to outfly someone to shoot them down.....or do 248 mph down the Mulsanne.....knowwhatImean?

Using my radial 'air head' with similar valve angle as crude example - fill in the fins as solid & ignore the big glow plug which would need to be smaller & parked perpendicular to cylinder like FS & Schilling, possibly a cooling hole about yay big could be accommodated?

Quote from: peterthaUsing my radial 'air head' with similar valve angle as crude example - fill in the fins as solid & ignore the big glow plug which would need to be smaller & parked perpendicular to cylinder like FS & Schilling, possibly a cooling hole about yay big could be accommodated?

Keep in mind that the heads are individual.    It may be really hard to get that pipe in there and have enough contact to have effective heat transfer with the head across all 6 cylinders.

Something to think about anyway

Dave

I suspect you are right. I have not been involved in RC nitro boats but as a guess looks the head has an inlet nipple feeding an annular cavity milled in the head, water is forced to go around the head perimeter to some degree. My straight through tube lacks this perimeter flow path. Pondering the cross section some more, maybe the upper edge of cylinder water water jacket cavity is reasonably near combustion area to do some good. A better method would have the tubes vertically displaced from one another. Detracting from the FS design some more. Compromises, compromises...

I would tend to agree.  No trapped air pocket at the head.  Do you think you will use this in a model airplane?  Or just demonstration runs?.  I suspect a 4 stroke will tend to run much cooler than a nitro 2 stroke at 40000 rpm.  Just thinking outload

Dave