It’s been a year since report #7; about time to move on to #8.
Yes, my bike is street legal but I still don’t allow it to get on the road. It has proven itself to be rideable but what I want is performance, letting out what’s potential inside. And that’s certainly more than 357 bhp at the rear wheel we measured in April 2012.
But before tuning I’d to take some time for minor and major repair: the bike suffered a lot from the salty conditions at the RDW test facility. Half an hour of testing was enough to corrode all aluminum and bare iron parts.
Even more alarming was that the bike spilled coolant just minutes after the tests had ended. Bad luck but still I consider myself lucky it happened after the tests…
Play the video to see where the leak is located: between the left cylinder head and intake manifold.
Easy to repair? Nope …
One moment you have a rideable bike, the next moment just a tiny engine.
Had to remove the wiring, the injection system and the intake manifold too. It was the only way to reach the sick spot (see red circle).
This is how the inside looked on October 30 2013, as I assembled the intake. The grey edges I’d filled with 2K filler: hard as aluminum, water resistant, oil resistant, fuel resistant, heat resistant, just perfect.
Well, nót perfect.
Appearantly this stuff is not glycol resistant, and that is a main component of my cooling fluid. Bad luck.
I made two fitting aluminum frames the oldskool way: with a fretsaw, key files and sandpaper.
While I was at it, I made some stainless steel spacers and washers for the manifold.
On a turbo bike intake air cooling is eminent for engine performance (see report #6). So I continued optimizing the water-methanol injection system.
I added two pressure switches for a two-stage injection system: the first one (SQ1) activates the first spray at 0.25 bar, the second one (SQ2) adds the second spray at 0.75 bar turbo pressure.
Because the battery moved to the rear of the bike (see report #5) I had free space in front of the oil pan. I made a 3D drawing to decide where to put all components.
Like so many times before I visited Technisch Bureau Holl. And like all the times before they patiently helped me find the connecting components.
From simulation to cardboard.
Because of the lack of space (what’s new?) I made some connectors from scratch, and some of them I modified.
Some silver soldering …
… and here you go!
Important part: the nozzles. They vaporize the mixture so the intake heat is cooled down to the max.
To pressurize the air bottle to 20 bar I tried several 12V electric pumps: their specs were promising but none of them met the expectation. One of them even caught fire. Crap.
I’m building a static pressure system on my bike so why not a static refill station? For half the money of a pump I bought a 200 bar diving bottle. Just perfect.
This strange picture was taken under the bike, looking up the oil pan. A lot of electronics and hoses are involved, as the drawing already showed.
Pressure switches SQ1 and SQ2 are adjustable; their scale turned out to be very inaccurate. So I made a set-up to configure them the way I want.
The swing arm needed some connectors as well.
Hans Eickeler welded them with huge precision.
In this case silver soldering was not an option because the connectors had to be welded in the swing arm.
Like this. Like this? ;)
Two at the lower end of the swing arm, one on top.
When a pressurized, toxic, blinding and invisible burning mixture like water-methanol is involved, you want to take as little risk as possible. So I did an under water pressure test.
In the mean time Peter Scheepers improved his dyno test facility. He added a huge air brake to delay acceleration: very important when tuning a turbo bike.
No wonder he looks content.
On May 17 2015 we had another dyno day. There were familiar problems with lambda sensor and the PWM valve (see multiple reports), and after they were solved we did a run.
The engine accelerated like hell: it took no more than 1.5 seconds to reach 6,000 rpm. And the graph looked smooth and promising too. At first.
Looking closer: only 156.5 bhp at the rear wheel. My GSX-R has that! Disappointing, and puzzling.
I hoped for wheelspin but that was not the case: it was the clutch that slipped, confirmed by the fast temperature rise during acceleration.
That is a common fact when tuning a bike, or a vehicle in general: the moment you improve the engine (and thus performance) the next weakest spot comes to light. In this case the clutch: it couldn’t keep up with the acceleration.
That didn’t mean it was a lost day. Peter optimized the software …
… and Niels the hardware.
So progress was made.
The engine did not overheat a single time …
… and nothing was destroyed. Always a good thing to celebrate after a test day.
Next steps to take: temporary pause working on the water-methanol injection and focus on ‘getting grip’. Goals: being able to measure the clutch slip and, after that, find a remedy.
I contacted my childhood friend Benno Krabbenborg; you remember him from report #3. Loaded with equipment he visited the bike, and we did some tests.
Designing electronics next to the bike turned out not to be very efficient so I built us a simulator: five bolts on a disk and a sensor fixed at my drill machine.
We measured the pulses with an oscilloscope.
On the lower right you see the first test print. On the top right you see beer cans. :)
Collecting data …
… finetuning the pulses …
… enhancing the print …
… until the final scheme …
… and the almost-final print.
In the mean time I started another small project: building a shed for the CBX and for the V8. First in a 3D program …
… tore down the henhouse …
… made a steady floor with reinforced concrete …
… used the car for what its main purpose is: moving heavy stuff from A to B ….
… and making indestructible doors.
It’s taking shape.
Almost more a safe than a shed.
Nice (and necessary!) to get some workspace in my garage.
The CBX, waiting for his big brother.
Talking about waiting: I returned the engine Koen Roemaat loaned me … thirteen years ago; see report #1. Time flies.
In the mean time Niels had been busy too: he finished the wiring scheme of the bike.
He added Benno’s clutch slip sensorprint to the motor management system.
As if it had always been there.
On October 5 2015 Peter and I had another dyno day. A facemask was more essential then ever before: you don’t want to inhale exhaust gas nor burned water-methanol as it might turn in to formaldehyde gas.
Sadly the slip sensor did not live up to its hopes and expectations: for whatever reason the results were corrupt and thus useless. Damn.
We decided to enlarge the centrifugal weights on the clutch. A week before the test I’d spent a full day to egalize the clearance of the clutch to 0.7 mm: getting it exactly right is crucial for grip.
Both changes resulted in good grip: one test gave us 341.8 bhp and 491.2 Nm torque at the rear wheel. Not bad but not good as well: these are not the numbers I’m aiming for.
We’re still not sure why the bike doesn’t break the 350 bhp barrier with ease: it has the potential of doubling it, and more.
Intake heat is still an issue: I just finished building a purge system for the water-methanol injection system.
That way all hoses are filled with fluid the moment we test on the dyno, ánd I can activate the system any moment I want to, to see and manipulate its influence on the intake temperature.
I tested the system with air. The lower switch activates left and right spray individually. The top switch activates both nozzles simultaneously.
Works just fine. I’m curious what it will bring on the dyno.
So, what’s next? Continue testing and tuning, that’s what’s next. I made a (ever growing) to-do list for the next dyno day with Peter and Niels. Parallel to that I recently contacted some experts on the combination ‘V8 + turbo’ to discuss the possible bottlenecks in the system.
Am I looking forward to that? Yes, of course. Discuss, plan, develop, mill, lathe, weld and test: that’s what R&D a.k.a one-off bike building is all about.
So, again: to be continued … here.