Firstly, you must do a base run.  (I like to do several runs to blow the "cobwebs" out.)  It would not be fair to use the first run as a base because this is often down on power.  The bike in this example only put out 94hp (and a cloud of smoke from the carbon build up) on the first run.  The second run made 103, the third 104. 

The owner of this bike complained that it had a flat spot about 3500RPM.  The bike is a 1999 955i with a Black Widow pipe and perhaps a K&N air filter (not 100% sure about that.)  The data here is arranged in table format to make it easy to follow.  Each run is at least two runs because the first run is needed to bring everything up to a good running temperature  without overheating the airbox.

The graphs show HP and Air/Fuel ratio.  You need to remember that the Air/Fuel line on the graph will not be at the correct RPM because it takes time for the gas to reach the probe.  I have found that a difference of 500 - 1000 RPM is close to the mark (more on this below.)

Click on images to enlarge.

	This first graph shows horsepower with the standard Triumph sports map.  The dip in power is very obvious.  From the graph it appears to be very lean in this area.  (The reason this graph looks like it has been touched up is because I didn't have a graph with just run #3 on it so I have erased the other run on this graph.)
	Because of the severity of the power drop I decided to have a look at the map values rather than just use the percentage trim maps.  This image shows how the numbers in the map drop by 33% at 2750 and 3000 RPM. this can also be seen in the 3DGraph below where the red line runs through this dip.

This dip can also be seen in this 3DGraph where the red line runs through this dip.

	This shows the changes I made to the map.  All I have done is take out the dip by bringing the numbers up to a value that gives a smooth transition between the numbers either side.
	At the same time I added these trims to the fuel % trim table to try and lean out the rich condition up top.  You can see that I don't just trim the top row.  This is because the trims that apply at full throttle will also apply at the throttle points just below full throttle, likewise when I trim the 50% area I also trim above and below that point.
	The result (run #5) shows that the dip in power has been removed and power at the top end has also increased.
	From this point on I use the fuel % trim table for all changes.  This screen shot shows the changes made after looking at dyno run #5.  If you look at the previous graph you will see a lean spike that appears at 3000 RPM on the graph.  Because of the lag in the exhaust gas sensor, you need to look at the power graph to find a corresponding point where there is a dip in the power.  (This is where the lean condition actually exists.)  In this case you can see that power fell a little at the 2500RPM mark and for this reason I have added a 6% trim in the 2500RPM area in the trim map.  I also increased the trim between 4500RPM to 6500RPM and added trims at 3500RPM and 4000RPM to try and bring the exhaust gas readings down toward the 13/1 area.
	Run #7 shows that the power has increased at 2500RPM and in the 3500-4000RPM area.  The air/fuel graph is now starting to look much better, as is the power graph.
	The top end of the graph is a little lean at 8000-8500RPM and a little rich at 9000-10750RPM.  This screen shot shows the changes I made to try and fix this.  I also increased the trim at 2500 and took a little fuel out at the very bottom of the map.
	Run #10 shows that the resulting power was increased again at 2250-2500RPM and in the 8000 to redline area.
	This graph shows the final result (run #24) along with the starting point with the standard Triumph high pipe map (see Run #3.)  As you can see, at this point I am up to dyno run #24 - most of the dyno runs from 10 onward have been used to clean up the part throttle portions of the map.  The procedure for this is the same as the full throttle tuning except you only use part throttle during the run.  I do these at 25% and 50% throttle.
	This shows the same data as above but displays HP and Torque instead of HP and Air/Fuel ratio.

As you can see from the final graph this bike is now running much better and the annoying dip in the power curve is gone forever. 

If you have any questions about the methods shown here please feel free to email me.