Map Data

The spreadsheet below was all done manually using K-Scan and an arbitrary function generator to supply the crank trigger.

Of course, it would be infinitely better to just extract the data directly from the map, but I'm a long way from doing that.

A few caveats: I would not bet my life on the absolute accuracy of the numbers in the spreadsheet. The timing numbers seem quite stable. The fueling numbers are more of a moving target -- especially at the extremes of RPM. All of the Evo6 maps use similar values during kickstarting. Although these are “steady state” values, the transient fueling can be somewhat different. I began taking map data at an engine temperature of 63° C, but at some point the dial got bumped and other data was taken at 67° C. Probably not a big difference, really.

Ignition Timing @ 24-degrees Butterfly Rotation

RPM versus ignition advance in degrees BTDC:

731, 8

1055, 9

1097, 10

1109, 12

1400, 16

1500, 18

1600, 24

1750, 25

1926+, 26

A trials bike gets to max ignition advance very quickly!

OSSA Maps Data.ods

All data at a barometric pressure of 98.5 kPa, air temp 17° C, engine temp 63° C.

Data taken at zero throttle (5 butterfly degrees), part throttle (60 degrees, default value for failed TPS), and WOT (85 degrees).

Ignition map is 3D, but not a lot of difference.

Note transition from firing ATDC to BTDC occurs at 732 RPM.

Rev limiting occurs ~7000 RPM, depending on throttle position and map.

Throttle Angle vs. Area Ratio

Even though K-Scan will report map data for throttle angles all the way to 105 degrees, the butterfly won't open more than 82 - 85 degrees. And, when I think about it, anything past 90 degrees would be diminishing the throat area again anyway.

The following spreadsheet was constructed using part of equation 3.7.7 in Blair's 4-stroke book.

Note that that the spreadsheet needs angles expressed in radians and I added a column of measured TPS voltages.

Blair Butterfly equivalance eq. 3.7.7.ods

Credit: Gordon Blair, Design and Simulation of Four-Stroke Engines

Degree Wheel to aid visualization

Ignition Timing for Starting

Viewed from the flywheel side, the crankshaft rotates counterclockwise. The normal range of ignition timings while running is approximately 20 to 30 degrees BTDC.

K-Scan reports some very retarded ignition timing at kicking speeds. For example, -44 degrees (or 44 degrees ATDC). Combustion progresses very slowly at kickstarting speeds due to low turbulence. Firing well after TDC makes the motor easier to kick and ensures it won't “kick back”.

With a 60mm stroke and a 118mm rod, at -44 ATDC the piston is 10mm down from TDC.

But I was shocked to see that my favorite TR250i map (Evo 2) has a timing of -125 degrees at the very slowest kicking speed I investigated (480 RPM).

At -125 degrees, the piston is 50mm down from TDC, and I can't see any point in firing the spark plug there. It appears that none of the final maps (Evo6) have the -125 degree data. I wonder if this was an attempt to stop the crankshaft rotation at a spot beneficial for restarting?

Factory Mapping Process

I gained a fascinating insight into OSSA's map-making process from a factory tour documented on Retrotrials.com. Unfortunately, I was refused permission to use a photo of the dyno room on my website, but you can follow the adjacent link to see it for yourself. Below is my interpretation of the salient points.

Page 2 Ossa Factory visit 2012After I had caught a few snaps we were then approached by the friendly receptionist who did a grand job of walking us thorugh the factory, introducing us to many people in the different fields of...

OSSA's Dyno Room, Salient Points

The photos were taken October 25, 2012. They show a Dynojet dynamometer. The bike has white “Six Days” graphics and is fitted with a small 18" street tire.

Dynojet software can be seen running on the main computer. A Dynojet RPM pickup is being used both on the coil primary and the spark plug lead. Cooling water is supplied via the shop, not the bike radiator.

The right-side photo shows the Kokusan Denki interface box. The left-side photo shows the ECU dangling alongside the bike.

But, by far, the most interesting observation is the use of a wide-band oxygen sensor. It can be seen fitted about midway in the exhaust header pipe.

I am not a fan of using WBO2 to tune to some arbitrary air-fuel ratio -- especially on 2-stroke engines! I intend to eventually write a section explaining my reasoning, but it probably won't be soon or brief.

Obviously, we don't know to what extent OSSA relied on WBO2 data in creating the maps. I will say that it can make quick work of creating maps, especially if the feedback process is somewhat automated. It would explain why the fuel maps seem somewhat “lumpy”.

I expect the maps could be improved.