Two Stroke Expansion Chamber Design

Professor Blair's tests proved that the return waves timing can be calculated fairly closely using the straight distances and the pressure wave speed in the pipe for that distance. He also proved that the return waves are generated all along the length of the diffuser and baffle cones. So I put formulas based on that in an Excel workbook file to analyze an existing expansion chamber, or help design a new one, based on the timing of the return waves in relation to the timing of the ports. In 2016 I discovered Blairs research paper about his dyno and a karting engine where he gave full details, including exhaust gas temperatures and pipe return pressure waves graphs. I then fine tuned my formulas to match his results for the pressure waves and horsepower graph. It also agrees with different customers horsepower graphs. The spreadsheet includes my equations for:
1) calculating for the temperature loss in the exhaust gas all along the pipe based on the outer diameter which serves as a conduit for the heat to escape through the metal.
2) simulating return waves strength and form based on the psi changes every 10mm (which creates the return wave) from expansion or contraction of the cone diameter.
3) calculating how to use the return waves graph in relation to port open timing to derive an estimated pipe power and powerband.
4) calculating the exhaust pulse time duration based on RPM, engine size, and exhaust port dimensions. This is important because it determines the time duration of the primary return waves.


 

Out With the Old:
Yes you can use Blairs formula for determining the distance of the baffle from the beginning of the exhaust port (using a generic beginning exhaust gas temp) but he didnt provide a formula for determining the change in exhaust gas temperature along the pipe length. So the distance comes out right for some engines but not for others.
One other old way of calculating the dimensions of an expansion chamber relied on generic formulas that put the emphasis on the wrong things. For instance pipe calculators would determine the belly diameter based on Blair's maximum size (2.5xheader dia.) and the header length based on a generic idea of length for engine type, and then let the diffuser be a byproduct of those two. The result was a disaster as anyone who has used them can tell you. The pipe builder would make a pipe as the program dictated and then chunk the lousy thing in the trash and try different dimensions till he found something he liked. The program creators had read and used some of Gordon Jennings generality statements while ignoring the fine points that Blair had written that make pipe making a thing of precision and not guesswork. How program makers could of overlooked the utmost importance of the timing of the waves in relation to the port timings is beyond me. But they did. The main bike factories make decent pipes for their bikes and so they either make and try 20 different pipes for each bike or they know the same tricks of the trade that I am sharing with you here. Any way now you can make a pipe that does what you want it to do. But, of course, if you want to stick to the old way you can. Sheet 1 of my Excel file uses formulas for its calculations but they should only be used as a starting point since the most accurate determination is after designing the diffuser and baffle and then looking at the combined return waves at sheet 3 in relation to port timing, and using the power grapher and then fine tune it from there.


I got sick of the silly free programs and the expensive "for sale" programs (and that waste-of-time 2 Stroke Wizard) and so made my own expansion chamber calculator spreadsheet. (Do it right by doing it yourself)

This screenshot is sheet 1 of 4 sheets:



ECcalc versions: I have been working on this for 5 years off and on, and previously I made the early versions available free of charge because essentially I considered them to only be "beta" versions, awaiting the final product. My latest are versions 28 and 30, and they work as good or better than any of the expensive pipe design programs you can buy, but they are easier to use. Get them now and make yourself an excellent pipe! Both versions are included when you buy ECcalc. No-charge assistance is part of the package also. Just fill in the spreadsheet and then send it to me to check for errors.

ECcalc28: It is a simpler version of 30 because it doesn't require the measuring of the exhaust and transfer ports. It's a little less accurate but appropriate for those with no time to take the head off to measure the ports. It's not limited to a piston stroke of 28mm - 81mm like versions 30 is. It can use longer and shorter strokes. Click here for the website usage instructions.

ECcalc30: This will calculate a pipe with a straight header for a street/trail pipe, or a flared header for a race pipe. It lets you enter the measured widths of the exhaust port or let the program estimate the widths based on what style port it is. Click here for the website usage instructions.

The usage instructions videos and other instructional videos on YouTube are here : videos page

ECcalc-hiRPM: Since 28 and 30 aren't for engines reving higher than 15,000 RPM I've developed a version for high RPM engines calibrated with the specs from the world champion mini race car. Click here for more info.


My pipe building resources page

The mathematical basis for the calculations of the return waves in ECcalc.

An example of pipe analysis and redesign.

Pipes built using this method.

My video showing how to make a pipe by hand.

How ECcalc agrees with the real life exhaust pressure graphs from Blair.

Four examples showing how accurate ECcalc is compared to dyno graphs.

Pipe Design Questions & Answers

Competitors pipe calculators

Buy ECcalc via PayPal by clicking here

ECcalc Features

1. Limitations: Acceptable piston stroke is 28mm to 81mm in version 30. In version 28 there is no such limitation. Minimum to maximum acceptable exhaust pulse widths are .45 to 2.1 milliseconds in version 30. Click here to download a spreadsheet that will let you figure out your engines exhaust pulse time duration. For engines with ignition timing that doesn't vary with RPM I have made a new option for them but it requires the user to measure the exhaust gas temperatures.
2. This Excel spreadsheet uses sophisticated mathematical means to determine the return waves shape, length, and peak which are used in additional formulas to determine the extra engine power the expansion chamber is contributing to the engine throughout its RPM range.
3. With input of manually measured exhaust area it calculates the exhaust pulse time which determines the time duration of the return waves.
4. The exhaust pulse time changes with RPM. Past ECcalc versions didn't vary it with RPM.
5. Sheet 1 allows the user to find the basic pipe dimensions that harmonize with the engine port durations and RPM and then allow the user to fine tune them on sheet 3 in a way that no formula based calculator can do.
6. It uses formulas for determining the different wave speeds inside the pipe based on different exhaust gas temperatures determined by heat loss from the pipe walls according to the diameters and lengths. The method was calibrated by the data of end pipe temperatures listed in one of Blairs research papers.
7. It has an accurate method of calculating the return wave strength and form by using Boyles Law which teaches that as the diameter increases, the psi change decreases. Since the return wave is generated by the psi changes then this non-linear way of calculating is the only valid way.
8. It shows a simple way to create the "flat sheet" patterns needed to cut out the right pattern of sheet metal needed to make each cone. No computer program is needed, just a long metric ruler, poster paper, an ink pen, and a Marks-A-Lot.

Getting a Pipe Designed:
Email me at 19jaguar75@gmail.com if you want me to do it for you. Cost is $75 via Paypal.
Go here for details needed.

Here's a testimony from a happy user of ECcalc:

"I didn't like Bimotions pipe design software so I bought and used yours. I like the fact that I can see what pressure/vacuum is present in the port at the different piston/port positions. This 1973 Yamaha MX 250 has got 225 psi cranking compression. A little high, but it pulls hard and starts easily. The owner can't believe it starts on the first kick, and idles well while people around him are blipping the throttle in the staging lanes to keep their bikes running. On top of that he always gets the hole shot to the first turn, even against 500cc bikes in other classes. At the pacific northwest AHRMA award ceremonies the picture they put on all of the plaques is the Yamaha leading a couple of other bikes into the first turn. Pretty cool! The bike does epitomize the intent of the class with its old fashioned expansion chamber instead of a modern snail pipe, and looks pretty much stock on the outside. At one point I attempted to raise the RPM's a little by shortening the header pipe 25mm. I don't know what you use as a temp probe. I was using some assumptions when I designed the pipe as the motor was not running yet. So I put the top of a Fluke immersion probe into the pipe as I went for a ride. It seemed really low like 375*C so I did not trust the readings. I went ahead and shortened the pipe anyway. It actually seemed worse on the top end. I decided to run ECcalc with the lower temp #'s and everything matched up with what I was feeling. So, I retarded the fixed ignition timing like .010" to put some more heat into the exhaust. OMG the shortened pipe with less timing really came alive, more mid-range and more RPM's. In fact at the next race he started having clutch slippage in higher gears. It has new Barnett frictions and steels. So, I just installed heavier springs, and a new clutch cover as there was some hairline cracks in it. I am thinking I should lower the compression down a little so it won't retain as much heat in the motor and maybe make more power at the end of the race, I don't know. It works pretty damn good, I hate to mess it up."

All About Expansion Chambers