Author Topic: Crank windage  (Read 37423 times)

0 Members and 1 Guest are viewing this topic.

Offline Jack Gifford

  • Hero Member
  • *****
  • Posts: 1568
Crank windage
« on: January 07, 2014, 02:54:23 AM »
Needing to have a crank machined from billet for my lakester's four-cylinder engine may be the excuse I need to experiment with a novel (eccentric?) scheme for minimizing crank windage. The common knife-edging of counterweights, in my view, doesn't eliminate the majority of the turbulence created by weights (and similarly, any un-weighted crank webs). I intend to convert all of my crank's webs (all eight will be counterweights) to full-circles by attaching light weight "filler" segments (aluminum? magnesium? carbon fiber?) to the weights. I'm looking at various attachment schemes, from the standpoints of margin-of-safety retention (to reliably withstand about 8,500 G's!) and that the special machining can be done mostly on my mill.

Of course the primary turbulence-generators- rod journals, rods, and pistons- can't be eliminated, but the crank webs/weights are not an insignificant contribution.

My main reason for posting this: to learn about any prior efforts like this. Anybody know any examples? (other than full-flywheel style built-up cranks)
M/T Pontiac hemi guru
F/BFL 1-mile Loring record 2020

Offline fordboy628

  • Hero Member
  • *****
  • Posts: 2342
  • GONE FISHIN' . . .
Re: Crank windage
« Reply #1 on: January 07, 2014, 09:11:39 AM »
Jack,

I've never heard of such a setup being utilized or tested, although I do not know everybody/everything.   There are DRAMATIC gains to be had by reducing and/or eliminating "windage".   The most productive avenue for windage reduction I have seen, is to use an oversized dry sump pump system to create negative crankcase pressure.    Negative crankcase pressure creates other "opportunities" for power gains, in addition to creating a myriad of seal and gasket issues that need to be addressed.    You will need to create about -5/-15 inches of water pressure for a useful gain.   At these negative pressures, air influx to the crankcase becomes a serious sealing issue . . . . . .

Some issues to consider for your setup:

1/   A fully "rounded" counterweight might eliminate the "churning", but some windage will still result from "surface attachment".   The best results I've seen use a combination of polished surfaces, crank "scraper(s)", windage tray(s), and the aforementioned oversized scavenge pump(s) to remove the frothy oil from the crankcase as rapidly as practical.

2/   Even with the lightest mass of "filler segments" engine balance will, of course, be affected.   There may be some clever way to negate any harmful balance factor issues for your engine configuration.    If not, you may want to analyse whether the possible benefit will be. worth the effort.    Thinking of the problem in linear terms, what percentage of the "exposed" crankshaft length is purely rotational Vs. the percentage that is rotational/reciprocal?

3/   How can the "filler segments" be "effectively retained"?   The figure you mentioned of 8500 G's, is that a calculated number or a speculative number?   Of the racing engines I have been involved with, the engineering/design limits are typically 5000/5200 G's maximum.    And there are various material failures at those values.   Perhaps you can calculate the load in pounds/feet.    Very high loads will have a very limited useful life in terms of cycles to failure.    Think drag racing connecting rods.    Failures of this type will be both spectacular and expensive.    I like tension loads as low as possible, and under a max of 12,500 pounds/feet.    At loads of 15,000 pounds/feet and upwards, material failure is annoyingly common.    Between 12,500 and 15,000 pounds/feet is drag race/crap shoot territory, using the best (the most expensive) material(s) available.

4/   Will the engine configuration be inherently balanced?   (ie: flat opposed four cyl, etc.)    Or, inherently unbalanced?   (ie: inline four cyl, V-twin, flat plane crank V-8, etc.)   Vibrations and torsionals can really play hell with parts attached with fasteners.

Just my two cents.   Hope this helps you out.
 :cheers:
Fordboy
Science, NOT Magic . . . .

I used to be a people person.  But people changed that relationship.

"There is nothing permanent except change."    Heraclitus

"Only two things are infinite, the universe and human stupidity, and I'm not sure about the former."     Albert Einstein

Offline rouse

  • Sr. Member
  • ****
  • Posts: 563
  • Impound is the place to be
Re: Crank windage
« Reply #2 on: January 07, 2014, 09:27:41 AM »
"Comet" GoKart engines all have a polyethylene stuffer riveted onto the cranks. They are machined smooth and round to make a disc. Those engines work just fine up 13500 RPMs, although I tried 18000 and the engine bought the farm when the plastic came apart.
Johnnie Rouse
Bike 4680 P-PP2000 SCTA record 153.325    A-PF3000 182.920
                              Texas Mile 152.518 PP class  186 A-PF Class
If you love your freedom thank a vet.

Offline manta22

  • Hero Member
  • *****
  • Posts: 4137
  • What, me worry?
Re: Crank windage
« Reply #3 on: January 07, 2014, 11:07:32 AM »
Fordboy;

Re: Your point #4:  Horizontally opposed 4-cylinder engines are not inherently balanced; there is a couple in the horizontal plane. A 6-cylinder in-line engine is inherently balanced; so is a V-12. Those are sweet. Why Dodge developed a V-10 is beyond me-- it could have been a V-12.

Regards, Neil  Tucson, Az
Regards, Neil  Tucson, AZ

Offline panic

  • Sr. Member
  • ****
  • Posts: 845
    • My tech papers
Re: Crank windage
« Reply #4 on: January 07, 2014, 01:33:21 PM »
I agree that regardless of how effective the nose & tail shape are, an un-interrupted cylinder generates less drag.
A low density stuffer sounds good, don't know enough about the attachment method to comment.

JM2¢?
How about a hollow box of 16 gauge sheet, welded in place? As long as the box is liquid-tight it doesn't need to have continuous attachment.

Offline tauruck

  • Hero Member
  • *****
  • Posts: 5126
Re: Crank windage
« Reply #5 on: January 07, 2014, 01:48:34 PM »
The effort will outweigh the reward.

My 2c. 8-)

Offline panic

  • Sr. Member
  • ****
  • Posts: 845
    • My tech papers
Re: Crank windage
« Reply #6 on: January 07, 2014, 05:43:55 PM »
That places staying home above building cars, doesn't it?

Offline Interested Observer

  • Sr. Member
  • ****
  • Posts: 433
Re: Crank windage
« Reply #7 on: January 07, 2014, 06:48:44 PM »
Fordboy,
Your use of the units “pounds/feet” is puzzling.  It seems to be intended to indicate stress levels, and the numerical amounts discussed seem to correlate with stresses expressed as “pounds/square inch”, but “pounds/feet” has me stumped.

Back to windage--
I would agree with Fordboy that it is probably much easier and more effective to eliminate the wind rather than try to create an ineffective propeller.  If there is little fluid to churn, one gains from not only the counterweights, but also the throws, connecting rods, piston pumping etc.
A clean disc vs. conventional counterweights comparison might be a nice CFD exercise--which could likely give a quantitative indication of any gains that might be obtained.

And I agree with Taurek that, in actuality, the effort will outweigh the possible rewards.  Seriously messing with a high speed crankshaft is playing with fire.

Offline Jack Gifford

  • Hero Member
  • *****
  • Posts: 1568
Re: Crank windage
« Reply #8 on: January 08, 2014, 02:47:14 AM »
... That places staying home above building cars, doesn't it?...
That would be my response to "is it worth the effort?". After many years of blown-alky competition (up to 9,000 RPM), using all the common remedies for windage effects (aerated oil, etc.), I'll gladly put lots of effort into this experiment.

Thanks for the responses. I'll try to answer some of the concerns:

> Dry sump- that's a given for this project (along with windage tray and scraper). But like my other hemi (V8), I won't be pulling more than about 3 in.Hg. depression in the crankcase. Sizing the scavenge pump larger creates more oil aeration (by nature of any drysump system). As for reducing the density of the crankcase "fluid" (mostly air) via more scavenging, I see that as a pretty insignificant contribution to windage reduction.

> Surface attachment- order of magnitude smaller effect than going to full-round.

> My additions will maintain the center-of-mass of each counterweight at its original location, thus will not alter the balancing considerations.

> I acknowledged in my original post that nothing can be done about windage created by the reciprocating stuff. But this still leaves "interrupted" counterweights as a very significant factor.

> Radial acceleration of 8,500 G is calculated- 10,000 RPM, but +/- about 5% on radial distances of filler-segment centers-of-mass from crank centerline. Acceleration vectors other than radial are not known (would need comprehensive torsional displacement analysis throughout the crank),  but I can almost guarantee they would be at least one order of magnitude less than radial.

> Cycles to failure- just for a moment, ignore any crank twisting, and consider this example: say I make a clean run, accelerating the engine to some speed, then decelerate to a stop. That whole run would cause one cycle of stress in my attachments, while the reciprocating pieces underwent many thousands of cycles. But, yes, cranks do dynamically twist- but do not create tangential accelerations on the order of 8,500 Gs.

> Hollow box construction- rigidity at 8,500 Gs could be problematic. Unless I got into ultra-expensive carbon-fiber science, which doesn't correspond with my shade-tree approach.

> I'm purposely not showing any attachment schemes, since I'm still studying various methods.
M/T Pontiac hemi guru
F/BFL 1-mile Loring record 2020

Offline DND

  • Sr. Member
  • ****
  • Posts: 395
Re: Crank windage
« Reply #9 on: January 08, 2014, 05:55:27 AM »
Hi Jack

Spend your time building a flow bench and play with the intake & exhaust tracks, that could gain you a lot more HP that some air bumping up against the crankshaft bumps

Plus crankshaft harmonics are bad news!!!

G Don

Offline fordboy628

  • Hero Member
  • *****
  • Posts: 2342
  • GONE FISHIN' . . .
Re: Crank windage
« Reply #10 on: January 08, 2014, 12:24:29 PM »
Fordboy;

Re: Your point #4:  Horizontally opposed 4-cylinder engines are not inherently balanced; there is a couple in the horizontal plane. A 6-cylinder in-line engine is inherently balanced; so is a V-12. Those are sweet. Why Dodge developed a V-10 is beyond me-- it could have been a V-12.

Regards, Neil  Tucson, Az

Neil,

After reviewing my Charles Taylor, you are correct.    There also seem to be some other types inherently balanced.   Inline 8, 90 degree crank, with a certain firing order.   Along with some others, ie: flat 8, 180 degree crank, firing 2 cyl simultaneously;  flat 12 with 60 degree crank; etc.   Anyone wishing to inquire further should refer to: Volume 2, 'The Internal-Combustion Engine in Theory and Practice', pages 298/305 for self-gratification.


According to a reliable source I knew who was working for Dodge (Chrysler) at the time:   "They wanted an advertising tie-in to F/1 technology."

V-10:  equals 2 inline 5 cylinder engines coupled at the crank; OR, 5 V-twins coupled end to end to end, etc.     Just goes to prove:   You can come up with an "acceptable solution" to any problem.    You just have to throw enough money at it . . . . . . . .

Oddly enough, 4 cycle inline 5's and V-10' engines are not listed in Taylor's book.   They are however listed under 2 cycle engine possibilities . . . . .
:cheers:
F/B
Science, NOT Magic . . . .

I used to be a people person.  But people changed that relationship.

"There is nothing permanent except change."    Heraclitus

"Only two things are infinite, the universe and human stupidity, and I'm not sure about the former."     Albert Einstein

Offline Seldom Seen Slim

  • Nancy and me and the pit bike
  • Administrator
  • Hero Member
  • ***
  • Posts: 13158
  • Nancy -- 201.913 mph record on a production ZX15!
    • Nancy and Jon's personal website.
Re: Crank windage
« Reply #11 on: January 08, 2014, 12:31:23 PM »
"should refer to: Volume 2, 'The Internal-Combustion Engine in Theory and Practice', pages 298/305 for self-gratification."


If I do -- will hair grow on my hands and I'll go blind? :evil: :evil: :cheers:
Jon E. Wennerberg
 a/k/a Seldom Seen Slim
 Skandia, Michigan
 (that's way up north)
2 Club member x2
Owner of landracing.com

Offline fredvance

  • FVANCE
  • Hero Member
  • *****
  • Posts: 2297
    • Vance and Forstall Racing
Re: Crank windage
« Reply #12 on: January 08, 2014, 01:42:10 PM »
That should have happened when you were 13!! :-P
WORLDS FASTEST PRODUCTION MOTORCYCLE 213.470
Vance&Forstall Racing
WOS 2011 235+MPH
Engine by Knecum, Tuned by Johnny Cheese.
Sponsers Catalyst Composites, Johnny Cheese Perf, Knecum Racing Engines, Murray Headers, Carpenter Racing

Offline Crackerman

  • Full Member
  • ***
  • Posts: 206
Re: Crank windage
« Reply #13 on: January 08, 2014, 03:21:23 PM »
"should refer to: Volume 2, 'The Internal-Combustion Engine in Theory and Practice', pages 298/305 for self-gratification."


If I do -- will hair grow on my hands and I'll go blind? :evil: :evil: :cheers:
well.. not if you keep the hair worn off!

nothing for eyesight though... (carrots?)

Offline JimL

  • Sr. Member
  • ****
  • Posts: 799
Re: Crank windage
« Reply #14 on: January 08, 2014, 04:06:56 PM »
Regarding the V10 engines that came along in the 90s.  I was at Cobo Hall for the SAE conference when EPA sat on the stage and we (manufacturers) sat in the chairs.  It was the decision point for OBD2, and the most heated discussions centered around all-load, all-speed MISFIRE detection standards.  Only AUDI had a pretty good working solution at that date, but it used a complicated sensor arrangement looking at both ends of the crankshaft, simultaneously, to take torsional twist calculations into the equation.  That method would have been the death knell for manual transmissions.

An agreement was made to defer some measure of the requirement (misfire detection) for "more than 8 cylinder" engines.  That moment was a head-start for getting some big displacement engines into the market, while buying some time for technology to catch up with aspects of OBD2 goals.  For two manufacturers, large V10s were the cost-effective way to proceed, while another chose to back away from the extra-large engine for OBD2 level engine management......but....

.....Keep in mind, that there was allowance for delay of OBD2 application until '96 model year for any engine that was not fitted with a "newly designed engine management system" in '94-95 model years.

The mid-to-late 90s was an intense time for everyone in the industry; in any event, they all did a great job.  Thats what I know of the whole V10 story, just from the emissions side of the puzzle.

JimL