The story of tuning a turbocharged jetski

So here's a story of the journey from the current year. I've got a few friends who competitively race on jetskis and they were running a few different setups but we'll generalize for this post so the discussion will be easier.

Need to knows:

Engine

800 cc

2 cylinder

4 stroke

Compression ratio 8,5:1

Bore 85mm

Turbocharged running from 0,5 to 1,5 bar depending on class

The way I got it

So a buddy approached me if I can help them improve their jet skis for competition as they felt there was something being left on the table in their endavours and I made no promises but I said I'd have a look at it. Applying the Andre principle of EFI is EFI no matter where you put it™ I installed the software and opened the maps from the Autronic SM4 ECU. WOW.

The background - this map is provided with the Autronic SM4 by a "top level tuner" and the map is "ready to be used at any boost level" - just set the boost in the map and it will control it closed loop and run it correctly.

The maps

Fueling map (top is the fuel quanitity, red rectangle is Target A/F Ration and green is "charge temperature estimate")

Yes... You're seeing that right. It's TPS based - no other AlphaN maps or something similar. This is what controls the fuel. AFR on gasoline was anywhere from 10:1 to 13,5:1 around the map.

Ignition map (top is the spark advance, bottom left is Boost compensation retard map and bottom right is intake temperature compensation)

I'm not even going to comment on this...

So this is what you get for 800$ at some parts of the world. (I'm against naming and shaming so I'll keep any names out, we're here to improve each other, so don't even ask).

Fast forward

Through the season, I fixed their maps to the extent where the jet was actually drivable, faster and had a stable lambda, but I kept the "original" map as a benchmark of what the jet is capable in the first place, as they don't have knock sensors. I plugged my lambda and did the best with what I was given - and mapping on water sucks.

Now... I convinced them to toss the Autronic and I'll be supporting them in the next season with a MoTeC M130 ECU by my side. I'm sure we'll make a great team :)

The problematics I see for preparing the jet for next season:

- quanitifying changes (apart from an engine dyno, I find it near impossible to actually estimate any gains throughout the map)

- revlimiting (the jet jumps on every wave, hits revlimit, bumps back in to the water etc. etc.)

- developing a more ridable jet - combining the gyroscope with a form of traction control (this is just a thought for now)

How would you go on about tuning a jetski where you only have water at your disposal ? I can only pretty much estimate an absolute improvement by testing topspeed but that's really just the top RPM/Load cell of the map, rest is hard to estimate a gain. Ofcourse we'll be using knock sensors etc. to find the engine limits and be able to run a reliable race. And well - the revlimiting again - is a bitch here... You do hit cuts OFTEN (as many as once per second really) and the races last 15 minutes + 1 lap. So what would be the least damaging way to try and limit the RPM?

I know it's a long post that is "two different things at the same time" but let's try to keep it a discussion where you refer to the part of the post you are replying to and then justify your comments!

have a nice day everyone

I do like your post!

Abandon Autronic is a good call. As a solid technological basis I recommend high speed telemetry. From there on, use the data along with on-board datalogging to reach MBT and optimal mixture.

There are culprints, Andre please chip in, to determine where best power is being achieved in your particular situation without having the luxury of a dynamometer bolted on.

DynoDom: yes, I forgot to add the MoTeC Will ofcourse be upgraded with the 200Hz logging option.

You are on the right track. You may even want to consider a GPS based logging device on-board to sort of emulate a dyno. I am not sure, nor have I ever used this aporoach, but it seems a practical method to find MBT on water under real world conditions.

If you have room for it, you could mount a torque sensor on the shaft. Like this one: Texense Wireless Torque Sensor

Onboard dyno :)

Hey that is an awesome idea !

Could torque be measured by a spring scale.

Tie to a jetty, drive against the spring. Maybe you could steady state tune then

I've had the opportunity to tune a few jet boats, including a competitive jet sprint boat that makes around 1200 hp, and the techniques I use for these are applicable to your situation. Obviously the dyno is the best place to do your tuning, however you can still take some pointers from your data logging. A jet unit takes a certain amount of power to spin at a given rpm, so if we see the rpm increase after a tuning change then this indicates the engine is now producing more power. This does require a reasonably smooth and consistent log though and you're going to need to do this on a pretty smooth surface otherwise the rpm may look a little erratic as the ski goes across waves.

One thing to keep in mind is that the jet unit needs to be selected/sized based on the engine power. What I mean by this is that there is a fairly narrow rpm range where the jet unit is working at optimal efficiency - For the competition jet sprint boat I tune we target 6000 rpm for example. If we make more power then the rpm will increase beyond 6000 rpm, however the jet unit is less efficient and provides less thrust as a result.

I'm not convinced that strapping the ski to the wharf would prove useful for tuning purposes. I feel the water velocity in and around the jet unit intake would be unrealistic in comparison to what would be actually seen at perhaps 100 kmh. I'd assume this would put the engine rpm in a different place stationary than what you'd see in real world conditions.

Personal Watercraft definitely need to have good logging to be tuned properly on the water, if you can get the engine on the dyno, that can be helpful, but still doesn't match real world conditions, especially coolant temps.

With the M1, you can set the rev limiter up to be very smooth, so that when the impeller is out of the water, the limiter kicks in, but doesn't kill the engine so that you still get good drive when the pump is running again.

We take the PWC's to the local water-sport centre, this typically has smooth water and is about 2km long. This means that the rider has the time and room to get up to full speed and hold it, so that we get a good log set, we then look at the lambda, top speed, acceleration ramp and knock to make any adjustments.

I agree with both of you Andre and Rex. Andre - the part about looking for the "sweet spot" of the jet propulsion system getting inefficient and cavitating (actually not having enough water to provide the thrust - hence losing efficiency) - I guess the only way to check where the "sweet spot" in the rev range is with checking the top speed. I have the option of using a small lake that has flat water so we can do top speed runs uninterrupted, but setting it up for real world conditions and jumping in and out of the water is another story. Also the part about the matching coolant temps is impossible to do on an engine dyno as there's really no way to provide the same amount of cooling that a jet gets running in cool fresh water that's being pushed through the engine at insane speeds all the time. I guess GPS and fast logging Will be key to determine positive / negative changes to mapping.