Building an Expansion Chamber

LRSimons

Member
Local time
12:44 AM
Joined
Jul 26, 2016
Messages
37
Location
SE Wisconsin
Hi all, I recently (sort of) finished my tuned pipe. It's taken a couple months of intermittent work, but after I got it fitted and somewhat tuned, I made my long standing goal of breaking the 35 mph speed limit on my testing road (and chased down a harley to boot!).

To begin, I cut out trapezoidal shapes out of 18 gauge matching up to Jaguar's dimensions the best I could. By just creating a trapezoid, however, the cone does not come out with square ends, as I found out after the fact. The proper way to do it is to cut arcs instead of square edges on the openings of the cone, which requires more math and fun things like ruining bandsaw blades to cut curves through steel. So, I just squared up the edges of the cones on the belt sander. Actually forming the cones was somewhat difficult, as they never seemed to get completely straight. What seemed to work the best was getting one end somewhat lined up, tacking it, then hammering the next inch into shape and repeating. After the cones were finished, they didn't quite line up, so I just welded in some more metal where there needed to be a bridge. At this point, I was just trying to get a pipe made (it would still work to some degree) rather than stick exactly to dimensions.

Yes, the welds look like garbage. Using whatever machines were available at that moment, I started out with a full size stick welder (terrible), then used the TIG machine (slow), and finally a large MIG machine (which still burned through on the lowest settings but I could do a series of tacks much more quickly and cleaner). The frankenstein part in the middle was a lack of math and planning, as I mentioned above. Feel free to make fun of me now.

Because I wanted to retain my other exhaust (stock with extension and flattened muffler), I made the header and flange as well. At the shop I was working at for this bit, the tubing roller didn't quite have the right dies and therefore was mostly a tubing kinker, not a roller. To get around this, I stuffed the pipe full of sand with both ends welded shut. The tube bent just fine with minimal kinking. I made the flange at home out of some 3/16 flat stock, after I got some better equipment. I just traced the outline of the original flange for a pattern, drilled the holes and cut to shape. To offset the pipe from the engine a bit, I cut two bits of curved pipe and welded them together to end up with a bit of an offset. It was then a matter of attaching the pipe and tuning the header and stinger. As you can see, after I got the pipe attached there was way too much header. I somewhat stupidly decided to remove an inch or two at a time, test, and repeat. This took all day, and I eventually decided to just cut the pipe to about 8", what was recommended from just about everything I read. That was the ticket, and the top speed rocketed up to 35. However, now that the pipe was actually doing something, the AFR needed to be richened up. I didn't really care on my test ride, knowing that something would break from the heat, but I was too exited to finally have something to show for my couple days of tuning and testing. After I hit the top speed and speed matched that poor guy on the harley, pop, magneto failed.

Oh well, glad we still have pedals.

Next step is waiting for a new magneto to arrive, drilling jet (done), and fixing all the stupid stuff I've been putting off in the quest for speed. Now that I can actually reach the speed limit, I'll be able to commute without being run over.

Input welcome.


Logan
 
The cylinder was ported a while back, but not measured. I was pretty conservative with the file. Today I made a degree wheel and measured the timing for real. It was just a printout mounted to some cardboard, and this was the first time I've measured so the reading may not be completely accurate. Exhaust opened at 118 degrees, give or take a couple. The caliper reading was in the range of 1.14-1.16", from the deck to the top of the exhaust port. For metric friends, about 29 mm.

As for RPM, I have the standard single speed setup with a 36T rear sprocket and 29" wheels. Not sure if the 29" is with the tires on or off. Anyways, I'm not especially eager to do the calculations myself at this hour, and the mac can't seem to see the rpm calculators on the forum. I will try in the morning, or maybe someone will know.
 
My notes show that the 66cc engine has around 142 degrees open port duration so if yours is ported it should be even more than that.
The hard part of determining port duration is knowing when the piston top edge is equal in height to the top edge of the port. Looking down into the cylinder distorts reality.
You could retry measuring it by using a thin feeler gauge. Put it on top of the piston with a few millimeters of it inside the transfer port and then turn the crank till the feeler is pressing against the top of the transfer port. Then record the corresponding degree on the wheel.
Never depend on where you think TDC or BDC is.
After the first degree mark then bring the piston down and back up to find where on the degree wheel it will be when the feeler again hits the top of the transfer port.
Then from that range of degrees you'll know the port duration.
 
Thanks for the better technique. I measured the exhaust port duration today, with a razor blade as far down on the piston's bevel as I could get it. Duration is right around 145 degrees. Is that fairly conservative? The engine is not a screamer at all, it topped out at 31 mph with the stock pipe (6" extension welded in).

I could probably raise the port a bit, but am I wrong in thinking that tuning the pipe for higher revs would be better in terms of the low end power left? I know either way the low end will suffer, but having some torque left would be nice. The comparison I'm making between increasing rpm with porting vs pipe is like making power with a car engine- porting would be the equivalent to having high flowing heads and a lumpy cam (making the engine hard to deal with on the street), while tuning the pipe to raise rpm is like adding forced induction- more power without sacrificing much drivability.

Am I on the right track with this or totally wrong?

Either way I have to finish balancing the crank first.


Logan
 
my 60cc engine had a duration of 157 and with the torque pipe still had enough low end power to start off without pedaling
 
3000 rpm is the broadest range available for the best boost from a pipe, all at the top. but my torque pipe doesn't noticeably diminish the power at rpm below that band like other pipes do.
it rev'd to 9200 rpm.
 
no maths involved for the laying out of pipe templates.

google and download a lil program called "cone". gives you templates required to print out on paper, glue em to the steel, cut along the dotted line,(rough it out, save the bandsaw blades, use a linisher/belt sander to get it to the line) if you remember to select "include fold lines" it makes it a lot easier as well (something i always forget to do then struggle with rolling cones with a ring roller that never does a "perfect" job).

this one...lets you specify the angle of top and bottom... very useful.

https://www.conelayout.com/

ends come out square (or angled as intended for producing bends...)

just beware that, as freeware. it has some trick of adding 1 second of delay when opened for every day installed... mines reached about 18 minutes now... the developer expects people to pay for it? hmmph!

then you get something that sort of looks like this...

pict0267-large-jpg.51348

http://motoredbikes.com/attachments/pict0252-medium-jpg.48437/

label each piece as you print it or its easy to get confused! the more pieces the better. nice gradual smooth curves.
nasty sharp changes in angles do NOT work well! (ie, this particular pipe was about 30 pieces all up, and was possibly my best one... i made one in about 8 pieces...while it had a nice powerband, it really did not like getting up there, mainly due to turbulence in the gas flow...)

here we go...worst pipe i ever made for performance :)

http://motoredbikes.com/attachments/pict0377-large-jpg.48894/

months? takes about a day, even less... when you actually put your mind to it. spend longer figuring out the bends etc than the actual making of said pipe...

PS... use software that works for designing the pipe or you really are just wasting time and resources ;)
 
Last edited:
Back
Top