Thursday, July 31, 2008

68 Inch Racer, part 3

In part 1 and part 2 of 68 Inch Racer, I outlined the physical dimensions and bottom end parts for this "dream" motor. In this installment I would like to wrap it up with the top end and valve train.

As to the heads, if I had a set of dual carb Knuck heads laying around (which I do) I would probably use them. After all, if I am going to label this a "racer" then why would I be worried about the leg clearance issue raised by two carbs sticking out the left side of the bike. Intake valves; 2.060". I had a set custom made by Ferrea Racing for a motor I did last winter. Beautiful valves, and no need for a lash cap, but I was disappointed with their limited selection of tulip choices for a 3/8" stem diameter. The only one they offer had more tulip than I wanted, which necessitated shorting the bottom of the guides slightly. It was also quite a bit heavier than I would have liked. Perhaps a 5/16" stem next time with a custom made valve guide.


It goes without saying that the piston domes, exhaust ports and combustion chambers would get the ceramic treatment like this.

Working off a couple of Knuck heads that I have oiled out in the past, it looks as though a head volume of 128 cc is a realistic starting point. Now, if we bore the Knuck heads out for the Pan cylinder's fire ring, we actually increase the head volume by about 5.71 cc's. But of course the fire ring itself will take up most of that increase. The total head volume when bored for the fire ring, but taking into account the amount displaced by the fire ring, calculates to 128.38 cc.

But here is where it gets interesting. The 74" Pan cylinder is shorter than the 74" Knuck by .200" when measured gasket surface to gasket surface (5.530" vs. 5.330). But if you add the height of the Pan fire ring they are the same length. That means that the piston will come to the same point in relation to the gasket surface on the Knuck cylinder as it does to the top of the fire ring on the Pan. But in effect you are lowering the head .200" down by using the Pan cylinder (think of it like shaving the head .200") Now we are talking some compression! When I plug all this into my Engine Analyzer program with a .020 thick copper head gasket, it comes out to 12.26:1 compression ratio. And after double checking all my figures a couple times, it still comes out to that!

Cool. Now we are into race gas territory (68 inch Racer - remember). That kind of compression will allow us to run a fairly radical cam as well. Any time you talk about high performance on a Knuckle engine, in my book there is really only one brand of cam to consider, and that is Leineweber. I am thinking his #3 Knuckle cam. It has .420 lift and 312 degrees of duration, but keep in mind this duration is measured at .020" rather than the customary (for Harleys) .053". That is still probably near 275 degrees duration if converted to .053".

Keeping with the "racer" theme, everything seems to point toward an engine that will not be easy to start. So we might as well add a magneto. I'm envisioning a bike that does not even have a kick starter.

Since this "paper engine" is getting a set of real life vintage dual carb Knuck heads, I would also use a set of real life 40mm Dell 'Orto pumper carbs that I have left over from my first drag bike. They were a little small for the 114" Knuckle, but I suspect they would work well in this application. Since the heads have a 4 bolt Linkert pattern, I would weld flanges onto the Dell 'Ortos to match. Jetting them to run on E85 (85% ethanol) would be really tempting. It has enough octane for this kind of compression ratio, plus the government helps pay for it (with my money). Good luck on getting them to kick in for your $15 a gallon racing gas.


One last thing. If the motor was to be used on the drag strip, then I would leave the valve train open, with no oil going to the top end. This is not a huge advantage, it just simplifies the oiling system and eliminates any clearance issues between spring and cover. But that's the way we used to do it, and it looks pretty cool. For any other application I'd enclose the valve train. Either way I'd use a pre '40 oil pump for the thinner gears and less oil volume in the motor.
That pretty well wraps it up. Did I miss anything? And.... an even better question: will it ever get built?

2 comments:

Anonymous said...

I'm following your 'paper engine' a bit. Will go back and re-read them again, so that I can digest it all. If you have the time, I'd like for you to run your numbers on a piece of software that I am working on as time permits. It will probably stay in beta format for life, as I will always be changing it. Here is the link:

http://www.terra-glide.net/harley_tech.html

Scroll down and look for the program called EngUtils.exe. Right-click on it and save it anywhere. Or, you can run it straight from my page, as it requires no installation. I write software helps for myself, so that I don't have to scratch my head over calculations. The graphics are Romper Room at best. Drop me a note with feedback when you can. If you see any fixes that I need to make, I'll catch them the next time I have the notebook computer out. That's where I do my programming. Have a good one.

Jack

St. Lee said...

Hi Jack, I have been playing with your program, and have some questions/comments.

On the Engine Profile Utlities screen I found the Crank Angle/ABDC calculator a little confusing at first, but figured it out. For the Intake Close Angle, what is your program based on for the timing event? Measuring at .053" is pretty much the industry standard for Harleys. However the standard for automotive is .050", and Leineweber only gives his Knuck cam specs at .020". I am sure this difference will have a big effect on Effective Ratio results.

On the Compression Ratio with Calculated Chamber Volume screen, I came up with 12.15:1 (nominal) verses 12.26:1 on my Engine Analyser program. Pretty close, the difference being the EA program does not take into account the volume between the top ring and the piston deck, whereas yours does.

However, I had to add in the base gasket thickness to the cylinder height. Also, if I enter the cylinder height without the firering, it makes the piston to deck height figure confusing. On the other hand if I do add the firering height to the cylinder height it throws the comp ratio way off.

I really like the dome volume calculations menu.

It would be cool if you could incorporate something for the fire ring specs.

Overall, I think it's a pretty cool program, and I put it right on my desktop. Thanks for sharing it.