Fuel Injected Motorcycles Tuning Information

Fuel Injected Motorcycles

Created 28 October 1998 v01

How we can tune your modified engine by remote control.

Our UltiMap eproms are specifically mapped for each cylinder, for each usable point on the map. This is done using a Data Recorder and a 2 channel Lambda meter, which give specific information about every operating point that is used as the motorcycle is ridden. This produces real-world results from road testing, which gives far better results than dyno testing.

So What's Wrong with Dyno Tuning?

Nothing, as long as you realise that dynos cannot reproduce road conditions, especially with regard to air speed. We have found time after time that dyno tuning works in around 85% of the operating points that we test, but there are anomalies which appear when you datalog the bike on the road. These are impossible to predict, and usually show up as a really lean spot at lower RPM, and maybe another really rich spot at high RPM, even though we set the mixture to ideal values on the brake dyno.

So although we can get a bike very close to ideal on a brake dyno, we can't make it perfect, and we usually finish the job now with a data-recorded road test, to find and optimise these peculiar anomalies. In fact we often do all the road testing first, and then finish up on the dyno when we have optimal road results.

This is why we don't sell any UltiMap chips which are just "guesswork" as even with all our experience we still cannot accurately predict what a modified engine will do. Of course we know the trends, but the point is to produce a product which is as close to ideal as possible, for a given engine configuration.

How can I get a chip for my modified engine ?

For us to get even remotely close to mapping your engine, we need quality test information. So we regularly ask customers to make exhaust gas measurements on a brake dyno, and then furnish us with the data. From this data we can see what the engine is doing, and then adjust the mapping to get closer to the ideal. Frequently this mapping must again be tested the same way, and with progressive alterations we can get a very good result.

However, unless you can measure both cylinders separately, we cannot offer individual cylinder maps such as those we supply on UltiMap eproms.

So What's the Process?

Firstly

you need access to a brake dyno and a gas analyser. We use high-speed lambda probes which are mounted in the header pipes, but you can use a regular 4-gas analyser with a sniffer probe in the muffler. These are very accurate, but take a lot longer to make a stable reading, so consequently there's a lot more dyno time and engine wear involved.

With a Lambda probe in each cylinder we can read exhaust gas at each operating point in about 15 seconds, so the entire measuring job only takes about 30 minutes. With a 4-gas analyser, each reading may take 30 or more seconds to stabilise, so the entire job takes about 1 hour or more, assuming you can keep the engine temperature under control.

Secondly,

you need to make measurements at each operating point on the map. In the case of Weber- injected bikes, there are 16 RPM points and 16 throttle points for each cylinder, so there's a total of 512 operating points! We normally test around 150 out of the possible 256 for each cylinder, because some operating points are clearly unused. For example, you don't need optimised fuel delivery at full throttle at 1000 RPM, as no engine is capable of operating there anyway (unless it's a diesel..)

To make these measurement, you need a dyno which controls the engine speed and is capable of maintaining constant engine RPM. This is commonly referred to as a brake dyno, because the roller has a computerised brake which can offer resistance to the engine's acceleration.

Dynojet Dynos do not have this facility.

The new Dynojet brake dyno would be perfectly suited to the task, but the normal Dynojet 150 model is not appropriate. Typical dynos which do have the capability are Superflow, Borghi & Saveri, Heenan & Froude, and numerous others. To clearly establish whether the dyno you want to use can do this, simply ask the operator:"I want to hold the engine at 7000 RPM for ten minutes. Is this possible?" If they tell you that it is not necessary or important, or it's not possible on their dyno, then the dyno is no good. You need a dyno which can do this.

You also need to make your measurements at the specific RPM and throttle points used on your bike. We call these break-points. Every Weber-injected model has an EPROM which stores not only the maps, but the RPM and throttle points as well. These are usually customised to suit a particular engine configuration. For example, a Guzzi Centauro doesn't need an RPM point at 12000 RPM, but a Ducati 995 Corsa does.

So you must use the break-points for your model. If you don't know them, we can supply them. The throttle points must also be tested, as most usefull engine operation is not at full throttle. We test every throttle point from idle to full throttle, at a given RPM. Again, if the dyno operator tells you that you don;t need to do this, then find another dyno!

The easiest way to measure throttle position is with our Hand Terminal, which reads out the throttle in degrees of opening. If you don't have one, we can supply voltage readings which you must take from the throttle sensor to achieve the specific positions used on your bike.

Thirdly,

you must control the engine temperature so that the readings are all taken at the engine's normal operating temperature. For water-cooled Ducatis, this is 85 degrees C. If the engine runs below 65 degrees the warmup circuit comes into effect, and the mixture will be richer. This will defeat the purpose of the exercise, as your readings will be wrong for normal operation.

We use several methods to control engine temperature, in order of preference:

1: - Use a dyno room with suction-only fans which move the entire room's air in a few minutes. The air inlet ducts are aimed at the front of the bike, but they have no fans in them, they simply feed air at atmospheric pressure. The exhaust suction fans are mounted behind the bike (or in the roof) and push all the hot air away from the bike. We use 2 x 11 Horsepower exhaust fans, with 2 rotor blades around 60cm across.
2: - Use total loss cooling on the bike. To do this you must run cold water into the bike via a control valve, and you bleed the system by removing the radiator cap rubber seal. This means the radiator caps blows water out of the overflow pipe, any time you feed cold water into the system. The entire system is running with no pressure. Lots of people will tell you that this will not work, but it does work. The coolant does not need pressure to keep the engine cool. The only reason you need pressure is to raise the boiling point of the coolant. If the coolant never gets close to boiling point, then it never develops pressure.
Anyway, we have used this method for years with no problems, as long as something (or somebody) regulates the cold water flow into the engine to maintain 85 degrees at all times. If you use this method you can test continuosly without waiting for the engine to cool down, and there is much less tyre and engine wear.
3: - Run the bike for short periods at a time, and then run the engine for five or ten minutes with no load to allow it to cool down again. We would still try to make the tests in the 85 degree region, but without adequate cooling a race engine can go from 85 to 105 degrees in about 60 seconds.

So, finally, this is what you do:

Set the bike up at the lowest of the useable RPM ranges, say 3000 RPM.
Adjust the brake control on the dyno to hold this RPM at any throttle.
Set the throttle position at the lowest point, then measure the CO.
Write this onto the test chart.
Open the throttle to the next position, then measure and record the CO again.
Repeat this for all throttle positions up to full throttle.
Check engine temperature, cool down if necessary.
Adjust the brake control for the next RPM position, say 3500 RPM.
repeat the entire throttle sequence, measuring from lowest to full throttle, and recording the CO at each position, then checking engine temperature.
Repeat the entire RPM sequence for each RPM up to the maximum, recording all throttle position CO figures.

This is pretty scary but once you get into a routine it does not take so long. We always work with a dyno operator to control the brake, and operate the bike. We simply make the readings and adjust the fuel using our diagnostic gear (which you may not have). But is very important to have one person to run the dyno, one person to make the measurements, and one person to control the engine temperature.

Here's a typical CO chart for a 916 Biposto, with each of the breakpoints using our hand terminal. Note that for voltage readings we have a different chart.

RPM /Throttle Breakpoints 1000 1500 2000 2500 3000 3500 4000 4500 5000 5500 6000 7000 8000 9000 10000 11000

83 . . . . . . . . . . . . . . . .

65 . . . . . . . . . . . . . . . .

53 . . . . . . . . . . . . . . . .

43 . . . . . . . . . . . . . . . .

38 . . . . . . . . . . . . . . . .

31 . . . . . . . . . . . . . . . .

27 . . . . . . . . . . . . . . . .

21 . . . . . . . . . . . . . . . .

17 . . . . . . . . . . . . . . . .

12 . . . . . . . . . . . . . . . .

9 . . . . . . . . . . . . . . . .

6 . . . . . . . . . . . . . . . .

4 . . . . . . . . . . . . . . . .

3 . . . . . . . . . . . . . . . .

2 . . . . . . . . . . . . . . . .

1 . . . . . . . . . . . . . . . .

RPM /Throttle Breakpoints	1000	1500	2000	2500	3000	3500	4000	4500	5000	5500	6000	7000	8000	9000	10000	11000
83	.	.	.	.	.	.	.	.	.	.	.	.	.	.	.	.
65	.	.	.	.	.	.	.	.	.	.	.	.	.	.	.	.
53	.	.	.	.	.	.	.	.	.	.	.	.	.	.	.	.
43	.	.	.	.	.	.	.	.	.	.	.	.	.	.	.	.
38	.	.	.	.	.	.	.	.	.	.	.	.	.	.	.	.
31	.	.	.	.	.	.	.	.	.	.	.	.	.	.	.	.
27	.	.	.	.	.	.	.	.	.	.	.	.	.	.	.	.
21	.	.	.	.	.	.	.	.	.	.	.	.	.	.	.	.
17	.	.	.	.	.	.	.	.	.	.	.	.	.	.	.	.
12	.	.	.	.	.	.	.	.	.	.	.	.	.	.	.	.
9	.	.	.	.	.	.	.	.	.	.	.	.	.	.	.	.
6	.	.	.	.	.	.	.	.	.	.	.	.	.	.	.	.
4	.	.	.	.	.	.	.	.	.	.	.	.	.	.	.	.
3	.	.	.	.	.	.	.	.	.	.	.	.	.	.	.	.
2	.	.	.	.	.	.	.	.	.	.	.	.	.	.	.	.
1	.	.	.	.	.	.	.	.	.	.	.	.	.	.	.	.

Lambda, or Air-Fuel Ratio, is a derived indicator of mixture. A lambda of 1.0 is equivalent to an air-fuel ratio of 14.7 : 1, that which is called stoichiometric, where optimum combustion, and minimum emissions, occur. This figure does not produce ideal acceleration, however. A richer mixture is required for acceleration. We use typically a Lambda of 0.85 ( Air-Fuel ratio of 12.5) for full throttle sustained power, 0.88 (air-fuel ratio of 13 : 1) for acceleration, and 0.96 (air-fuel ratio of 14 : 1) for cruise, with graduations between these points.

CO (Carbon Monoxide) is another primary indicator of mixture (richness / leanness). Typically we look for a CO of around 4.5% at full throttle, 2.5% to 3.5% at partial throttle, and 1.5% at idle. The figures we use are far more specific, but the trends are such.

Functionally Lambda and CO are the same. The major difference is in the type of equipment used to get the information, and the speed of obtaining readings at each point. CO meters suck the gas from the exhaust through a rubber hose, and measure the infra-red absorption of the gas in a reference chamber. This takes about ten seconds to stabilise, depending on the length of the rubber tube, and the electronics in the reference chamber. Lambda probes are a semi-conductor element which is inserted directly into the exhaust stream via a fitting in the header pipe, and they read within 100 milliseconds.

Lambda probes are commonly used in auto injection systems as part of a feedback loop which can optimise some parts of the fuel-injection programming. In the case of most current road cars, the probes are capable of accurate readings only at the lambda=1.0 point, and they are therefore only used to alter the overall trims in cruise mode. Any time the engine is in acceleration, the probe is unable to measure richer mixtures, and the probe is not functional in software.

This needs to be considered because many people think that a common automotive Lambda probe will deliver accuracy over a wide mixture range. This is simply not true. In fact the probes are designed specifically for accuracy only at lambda 1.0. To obtain accurate measurements at richer mixtures a true wide-range probe is necessary. These are manufactured by two companies, Bosch and NTK. The NTK product is called a UEGO sensor (Universal Exhaust Gas Oxygen) sensor, and costs around USD 500.00. The Bosch product is called the LSM11 wide range Motorsport probe and costs around USD 300.00. To obtain accurate results from these sensors an electronic box is attached which reads both the sensor voltage and temperature (via internal resistance measurement), and derives a specific air-fuel ratio from tables stored in the box. Without this temperature compensation and map look-up facility, it is impossible to obtain accurate and repeatable results.

Again, many people think that the common sensors will deliver this performance, simply by reading the output voltage and extending it's range around the lambda = 1.0 point of approximately 0.6 volts. Unfortunately, this is not possible, because as the mixture changes, so does the temperature of the probe (just think of EGT sensors for proof), and as the temperature changes, so does the output voltage. In other words, the two things are inter-related, and it is impossible to derive Lambda without reference to temperature and voltage lookup tables. This is not something that the inexpensive LED bar-graph indicators are capable of.

When these tests are performed on a brake dyno using our Hand Terminal, we can measure the initial mixture as delivered by the original map, then we can alter the fuel in real time using the hand terminal, to arrive at our ideal fuel mixture. Then we record the original Lambda (or CO), the final (or Target) lambda or CO, and the percentage trim used to achieve this target. These percentage trim figures can be overlaid on top of the original EPROM maps to derive new maps for each cylinder. This method is very precise, and very fast. We can map a bike in about 50 minutes, at over 150 points on each map, using these tools. Part of a typical trim chart is shown below, with typical readings. We make these corrections for each RPM range, so the entire trim chart is much bigger

RPM Throttle Initial Lambda Target Lambda Percentage Trim Inital HP Final HP

5000 83 0.92 0.88 +6% 78 82

5000 65 0.78 0.90 -14% 66 66

5000 53 0.92 0.92 0% 62 62

RPM	Throttle	Initial Lambda	Target Lambda	Percentage Trim	Inital HP	Final HP
5000	83	0.92	0.88	+6%	78	82
5000	65	0.78	0.90	-14%	66	66
5000	53	0.92	0.92	0%	62	62

Note that some points already show the ideal lambda, indicating that the factory map works well at this point.

Back to FIM Home page.

Want to let us know what you think?

Contact!

Or use other means: P.O.Box 851, Apollo Bay, Victoria 3233, AUSTRALIA
Telephone +61 3 5237 4206 / Fax +61 3 5237 4209

RPM /Throttle Breakpoints	1000	1500	2000	2500	3000	3500	4000	4500	5000	5500	6000	7000	8000	9000	10000	11000
83	.	.	.	.	.	.	.	.	.	.	.	.	.	.	.	.
65	.	.	.	.	.	.	.	.	.	.	.	.	.	.	.	.
53	.	.	.	.	.	.	.	.	.	.	.	.	.	.	.	.
43	.	.	.	.	.	.	.	.	.	.	.	.	.	.	.	.
38	.	.	.	.	.	.	.	.	.	.	.	.	.	.	.	.
31	.	.	.	.	.	.	.	.	.	.	.	.	.	.	.	.
27	.	.	.	.	.	.	.	.	.	.	.	.	.	.	.	.
21	.	.	.	.	.	.	.	.	.	.	.	.	.	.	.	.
17	.	.	.	.	.	.	.	.	.	.	.	.	.	.	.	.
12	.	.	.	.	.	.	.	.	.	.	.	.	.	.	.	.
9	.	.	.	.	.	.	.	.	.	.	.	.	.	.	.	.
6	.	.	.	.	.	.	.	.	.	.	.	.	.	.	.	.
4	.	.	.	.	.	.	.	.	.	.	.	.	.	.	.	.
3	.	.	.	.	.	.	.	.	.	.	.	.	.	.	.	.
2	.	.	.	.	.	.	.	.	.	.	.	.	.	.	.	.
1	.	.	.	.	.	.	.	.	.	.	.	.	.	.	.	.

RPM /Throttle Breakpoints	1000	1500	2000	2500	3000	3500	4000	4500	5000	5500	6000	7000	8000	9000	10000	11000
83	.	.	.	.	.	.	.	.	.	.	.	.	.	.	.	.
65	.	.	.	.	.	.	.	.	.	.	.	.	.	.	.	.
53	.	.	.	.	.	.	.	.	.	.	.	.	.	.	.	.
43	.	.	.	.	.	.	.	.	.	.	.	.	.	.	.	.
38	.	.	.	.	.	.	.	.	.	.	.	.	.	.	.	.
31	.	.	.	.	.	.	.	.	.	.	.	.	.	.	.	.
27	.	.	.	.	.	.	.	.	.	.	.	.	.	.	.	.
21	.	.	.	.	.	.	.	.	.	.	.	.	.	.	.	.
17	.	.	.	.	.	.	.	.	.	.	.	.	.	.	.	.
12	.	.	.	.	.	.	.	.	.	.	.	.	.	.	.	.
9	.	.	.	.	.	.	.	.	.	.	.	.	.	.	.	.
6	.	.	.	.	.	.	.	.	.	.	.	.	.	.	.	.
4	.	.	.	.	.	.	.	.	.	.	.	.	.	.	.	.
3	.	.	.	.	.	.	.	.	.	.	.	.	.	.	.	.
2	.	.	.	.	.	.	.	.	.	.	.	.	.	.	.	.
1	.	.	.	.	.	.	.	.	.	.	.	.	.	.	.	.

RPM /Throttle Breakpoints	1000	1500	2000	2500	3000	3500	4000	4500	5000	5500	6000	7000	8000	9000	10000	11000
83	.	.	.	.	.	.	.	.	.	.	.	.	.	.	.	.
65	.	.	.	.	.	.	.	.	.	.	.	.	.	.	.	.
53	.	.	.	.	.	.	.	.	.	.	.	.	.	.	.	.
43	.	.	.	.	.	.	.	.	.	.	.	.	.	.	.	.
38	.	.	.	.	.	.	.	.	.	.	.	.	.	.	.	.
31	.	.	.	.	.	.	.	.	.	.	.	.	.	.	.	.
27	.	.	.	.	.	.	.	.	.	.	.	.	.	.	.	.
21	.	.	.	.	.	.	.	.	.	.	.	.	.	.	.	.
17	.	.	.	.	.	.	.	.	.	.	.	.	.	.	.	.
12	.	.	.	.	.	.	.	.	.	.	.	.	.	.	.	.
9	.	.	.	.	.	.	.	.	.	.	.	.	.	.	.	.
6	.	.	.	.	.	.	.	.	.	.	.	.	.	.	.	.
4	.	.	.	.	.	.	.	.	.	.	.	.	.	.	.	.
3	.	.	.	.	.	.	.	.	.	.	.	.	.	.	.	.
2	.	.	.	.	.	.	.	.	.	.	.	.	.	.	.	.
1	.	.	.	.	.	.	.	.	.	.	.	.	.	.	.	.