![]() 07/30/2018 at 15:05 • Filed to: neat | ![]() | ![]() |
Apparently,
!!!error: Indecipherable SUB-paragraph formatting!!!
(paired like a Twingle) two-stroke water-cooled radial.
No pics in post
because there isn’t a source other than the Smithsonian and they have policies
, so you’ll have to go there.
![]() 07/30/2018 at 15:09 |
|
Can’t you post the pic and then site them as the source?
![]() 07/30/2018 at 15:11 |
|
No reproduction without written permission, etc. etc. Right there on the page. I could offer a fair-use-friendly descaled image or something like that contrary to their request (but probably legal)
, but I figured I’d just give a link.
![]() 07/30/2018 at 15:11 |
|
This:
https://oldmachinepress.com/2014/08/17/general-motors-electro-motive-16-184-diesel-engine/
Is also quite goony.
![]() 07/30/2018 at 15:13 |
|
Fair enough.
![]() 07/30/2018 at 15:17 |
|
Wow that’s got a lot of weirdness going on - starting with the fact that it’s a radial with an even number of cylinders (because two stroke).
A 1942 engineering report written by GM’s Research Laboratories, led at the time by its highly regarded director Charles F. Kettering, stated that the engine had excellent power/displacement and power/weight ratios, low fuel consumption, and very low vibration characteristics.
Hmm, and yet they killed it...
![]() 07/30/2018 at 15:25 |
|
I was on a white whale search for an eight-cylinder radial I know I’ve seen before - came up with that, and was like “huh”. Still didn’t find what I was looking for, but apparently Wright made a radial eight and Anzani made a radial ten, so those are both something.
There’s no *real* need for a four-stroke radial to have an odd number of cylinders if it’s not common journal... but there’s also almost no reason not to build common journal. It’s been done in production with gears or camming, but it’s not the usual.
Even with a common journal, you can have even cylinder count, it just means that your power pulses have to come in
pairs which have
irregular “back” to
”front” spacing
, like a Harley t
win
. Which favors higher cylinder counts because they’re *closer* to evenly timed
and less vibratey.
![]() 07/30/2018 at 15:35 |
|
I think, though, that with a 4 stroke radial, anything you’re doing to build an engine with an even number of cylinders *per bank* is just adding complexity for no good reason, though.
![]() 07/30/2018 at 15:43 |
|
It depends. There is something of an issue with a standard sequential-offset-fire sort of odd count radial that the power you get is really pulsey, and that there’s some gyroscopic action from the *big* common journal drive plate. The radial-engined LeMons MR2 kept breaking drive components from the vibration because it didn’t have something big and flywheel-like like a propeller to smooth it out (and it was a five, so...).
There’s also the issue of exhaust bundling and intake bundling, which potentially makes pairing advantageous. Not to mention the ability to potentially break out ignition to full independence per cylinder if you aren’t using a common journal - allows use of simple magnetos and total redundancy.
I wouldn’t say no good reason, but maybe
reasons that are secondary for most applications.
![]() 07/30/2018 at 15:50 |
|
Ha - yeah, I’m coming from a mindset that radials are for airplanes, basically, and for airplanes, the simpler, the better. You’re right, I wasn’t really thinking past that--
![]() 07/30/2018 at 16:12 |
|
Airplanes endorse complexity on a regular basis... as long as it’s related to being redundant or lasting a million billion hours. Some mutations of a normal radial are along those lines, but would be useful enough to outweigh “simple” only some of the time.
If the radial is to be in a boat or stationary application (look in the other replies to this post for an example), then sometimes service improving things enter in.