A Labor of Love; aka building a Rocket III Speedster

Ok, now done and what I've configured so far uses every single input/output of the VCU except knock sensors, somewhat concerning but I may be able to free a bunch of them up using the IC7 Gauge Cluster...
Bi-Directional Quickshifter with autoblipping using a single bidirectional strain gauge (already on the engine)
Progressive torque reduction Traction Control by gear and speed (allows more wheel spin as speed increases and gear position increases)
Run Brake Turn lights all controlled via VCU
Reduced redline while stationary and a scaled table to limit redline until engine is warmed up
Starter Controlled by VCU
Short and Long term fuel trims enabled
Wideband Targets and Conditionals set by MAP/RPM no TPS/RPM, just an experiment to start off with
Base fuel map done using alphaN and uses all typical corrections: Barometric Pressure, Ambient Air Temp, Coolant Temp but also MAP sensor offsets
Injectors Programed EV14 450s @ 4BAR
Single press start sequence and stop using momentary switch, no need to hold in the button any longer, stop function disabled above 5kph
Kill switch will kill everything as will safety lanyard switch (VCU external wiring function)
Side stand configured in as "Emergency Brake"
Clutch Switch must be engaged to start bike as a safety
Both brake switches configured as trigger the brake light
Various limp modes based on severity of fault

Remaining to configure are:
Laser ride height wheelie control
Launch Control
Torque limiting by vehicle roll angle (predictive corner traction control), and if possible via CAN bus #2, talking to ABS controller to reduce ABS intervention threshold during cornering and prevent over braking in a corner.
Power/Traction Control Modes. I want to program a rain mode, easy riding mode, and hair on fire mode, like on my H2 with various throttle response curves and traction control curves. I will probably need to reach out to Haltech to see if they can help program the function as I want to do some sneaky conditional for it, something like Ebrake off, clutch switch in, engine over 1000 RPM, press starter button x times for x riding mode. Something not terribly visible and I don't want more hardware like a rotary knob/button panel.
Brake pressure switch (if I can find a free analogue input) so it can flash the brake lights during hard braking

Calling it a day.
 
Ok so, I've come up with a strategy to solve a few issues and sourced the required parts.

First up, oil pressure and oil temp. Since I no longer need the idiot light sender Triumph uses for oil, I can screw in a proper dual sensor into the back of the head. I'll be using this:
sensor.jpg
Bosch part number BOS0261230340
-40c - 140c temp / 10 bar pressure dual sensor

I'll also be using the same part number for fuel pressure/temp sensor, and coolant temp/pressure. For fuel, I'll be adding a dual banjo to the tank outlet and an adaptor to locate the sensor.

For coolant, I'm going to tackle two issues with one fix. I've observed my SAMCO water pump inlet hose collapsing at high RPM on footage I took a few years ago, so my solution at the time was to run the stock hose there with SAMCO everywhere else. This time, I'll be using a Mishimoto adapter like the below and an NPT to M10 adapter to locate the pressure/temp sensor for the cooling system. I will prevent collapse by replacing the area it was occurring with the aluminum adapter. Initially, I'll be sticking with the stock water pump.
mio-mmwhs-28-bk_xl.jpg

Why?
  • This then gives me the ability to observe if the engine pushes air into coolant via headgasket by cooling circuit pressure.
  • This gives me the ability to finally measure and observe oil pressure and temp. I plan to experiment a bit with the Amsoil 15/40 I normally run and Mobile 1 4T and get some hard data on pressures so I can tailor the oil weight to empirical data instead of anecdotal data.
  • The fuel data is mainly for early diagnosis of fuel pump health issues and provides a safety net to prevent detrimental lean incidence from fuel pump failure. The fuel temp will also let me provide a correction function to the fueling table based on fuel temps. The pressure will also let me get the correct flow rate on my 450 injectors, which become 531 at 4 bar, but... is it actually 4 bar as advertised? I'll know with the sensor installed.

Additionally, I've sent TTS a request for a centered narrow radiator. So that's in flight now.

Throttle Control:
I went on a little internet exploration adventure yesterday looking for suitable bar switch gear that has a built in e-throttle I trust. There is a lot of cheap crap out there, then there is a lot of uber-expensive overkill billet jewelry parts. I went and dug through Suzuki, Yamaha, and Kawi for parts since aftermarket failed to provide an option. Initially, I was hoping to find a set of Z H2 switchgear/throttle, mainly just so my bikes had the same controls and no mental adjustment would be required, that turned out to be cost prohibitive.

I'll swing by a KTM dealer nearby and check out the 1290 Adventure controls before buying, but this is the least expensive trustworthy e-throttle I've been able to find.

KTM:
60711012000
RIDE-BY-WIRE WITHOUT THROTTLE TUBE
$83.12
60702012000
GRIP TUBE (NO HEATING)
$61.45
60702039100
GRIP LHS W/O HEATER
$71.14

Left side switches will come off a 2015 H2 (mostly because I like the minimalistic design and two extra buttons which I can use for cruise control or other functions):
46091-0300 HOUSING-ASSY-CONTROL,LH
$178.15

The other end of the throttle, the throttle body. I had initially planned on converting to the single throttle body, but there's a major issue with that. The engine has huge cams with a lot of overlap, it has massive reversion at idle, and that is fine with independently breathing ITBs. It becomes a massive issues when behind a single throttle body with a shared plenum, typically only solved by very high idle RPM, which I don't want.

It looks like the best solution is a ride-by-wire actuator that can control the existing throttle body setup by modifying the linkage. I've found a single suitable design, but am looking for alternatives still, I suspect this one is very expensive.
Throttle-actuator-1-18-2021_R4_max.jpg
I have an inquiry in with them for a setup with an integrated TPS. This will give me two TPS sensors, as is desired on e-throttles. One on the actuator tracking the arm position, and the stock Triumph TPS, which will report the actual blade position. It will be nice to have the ability to identify deltas between them as a safety net and go into limp mode if they differ by more than a set percentage.
 
Last edited:
For throttle body, I think I've found how I want to modify the stock setup to do this. By removing everything in red, and drilling the green circle to remove the stock secondary actuator contact arm and fit an M5 screw, it creates a lever by which the e-throttle can actuat the primary blades directly using a heim linkage rod and the below actuator:
Remove.jpg

Actuator:
Drive-By-Wire-Actuator-DBW.jpg
Linkage:
drop-link-kit-including-rod-ends.jpg

This meets the actuator requirements of maintaining a stock TPS sensor, return springs, and no throttle stop since the actuator must have full 90 degree movement to not be damaged.

Video demo of it adapted to ITBs (sorry for insta link...it's where they hosted it):

Remaining will be finding a suitable method of mounting the actuator solidly to the throttle body assembly such that it cannot flex/move in relation to the primary blade rod except to rotate it, and I think this will be the first custom piece I need to have cut by cutandsend.
 
I think I've come up with a better plan, first I'd still be removing all the red stuff in the above picture because it's no longer necessary. Then, I can have a 1/4" steel plate cut that would mount the TB. By removing 1/4" of material at yellow, below and then sliding the actuator mounting plate there, it would prevent torsion from rotating the actuator in relation to the primary blade. I can then mount directly to the blade shaft on cylinder 1 (green line) using a coupling as seen below, preserving the ability to balance #2 and #3 in relation to #1 whilst getting solid direct actuation:

rev2.png
This is how it would function, just the actuator motor would be rotated so the motor runs parallel with the throttle blade shaft, but rotated almost 90 degrees anticlockwise from the picture to fit the bike's space constraints and turning at #1 shaft:
128b6f75d83c0e9d3a43811ee1f4b88f4b7872d4.jpeg

I like this solution, I think this is the most elegant method of adapting I've found. The ATpower actuator can mount the TPS on the back of the actuator shaft and give redundancy on the TPS, as I want and has a 110 sweep range, so can be calibrated to account for the Rocket TB design, and hit allows the mounting plate on either side of the actuator area, so I can adjust/space it out so it ends up perfectly perpendicular to the throttle shaft.

 
Last edited:
I think I've come up with a better plan, first I'd still be removing all the red stuff in the above picture because it's no longer necessary. Then, I can have a 1/4" steel plate cut that would mount the TB. By removing 1/4" of material at yellow, below and then sliding the actuator mounting plate there, it would prevent torsion from rotating the actuator in relation to the primary blade. I can then mount directly to the blade shaft on cylinder 1 (green line) using a coupling as seen below, preserving the ability to balance #2 and #3 in relation to #1 whilst getting solid direct actuation:

rev2.png
This is how it would function, just the actuator motor would be rotated so the motor runs parallel with the throttle blade shaft, but rotated almost 90 degrees anticlockwise from the picture to fit the bike's space constraints and turning at #1 shaft:
128b6f75d83c0e9d3a43811ee1f4b88f4b7872d4.jpeg

I like this solution, I think this is the most elegant method of adapting I've found. The ATpower actuator can mount the TPS on the back of the actuator shaft and give redundancy on the TPS, as I want and has a 110 sweep range, so can be calibrated to account for the Rocket TB design, and hit allows the mounting plate on either side of the actuator area, so I can adjust/space it out so it ends up perfectly perpendicular to the throttle shaft.

Maybe make the plate from aluminum, lighter and quicker to cool.
 
So a part of this project is clearing out my garage to make space for the work.

I'm also being gifted a true classic, 1984 (I think it's an 84) Kawasaki GPZ900R. It too needs space for a future restoration following the Rocket build.

So I will have to fit Mufasa, the new rocket with room to work on them, my Daytona, the GPZ and have the H2 accessible to ride. Plus I have the extra rocket frame and engine. A full two car garage worth of bikes for about a year while I work through the restorations.

I need space.

A step for this, is being able to haul frames and big stuff around, maiden tow validation done. The little hauler did pretty well, Id guess it's a 500lb trailer empty. Tomorrow a load gets cleared out and the first "heavy" tow test, probably around 1100lbs.

20230826_162039.jpg

It's almost like Mini had it in mind. Sport Mode kicks electric water pump to max full time and delays shift points to higher RPM, same as most tow modes do on SUV/Trucks. Brakes get sharper and power steering gets heavier, all beneficial when pulling weight.
 
Last edited:
First towing event complete, 900lb load, no issues, validated enough for me to tow a rocket (with reduced weight) with, 300lb for an aluminum trailer and 700 for the bike. EGTs fairly normal for a tuned turbo 4, 700-750c while climbing steep hills in boost, I see 850 on track so well within safe limits. Long high altitude passes, like in Montana, may be an issue and need to be avoided.

So back to electronics planning, I need to make a full on wiring diagram but...

PDM-60+mo.blue+Nexus should give me everything I need for chassis controls (lights/horn/etc) and engine control without any fuses or relays needed.

All high current devices; starter, headlights, fan, fuel pump, coils, injectors get dedicated and configurable circuit breaker protected outputs. mo.blue provides configurable voltage cutoff to save battery if the bike is left in the on position but not running. Can configure phone as key, or NFC key, or physical key, or a combination of them all.

The mo.blue + the mo.button seems a brilliant accessory for RIIIs in general. Allowing you to correct a whole host of design deficiencies of the Standard/Classic and further improve the Roadster/Touring. To be honest, it seems to provide better keyless than the factory triumph system on the R3.
 
Ok so step one of all this is freeing the beating heart from Mufasa and validating it's healthy.

I started stripping the carcass down today. I was trying to get to the cam cover so I could check clearances and verify the cams didn't fracture from the impact. Sadly...1 fing cam cover bolt (the last one, and sadly the hardest to reach) striped so.... I can't do that stuff until the engine is free.

Que me removing a bunch of stuff. Engine mounting bolts are all cracked loose but still in place as I'm not quite ready to pull it.

I did remove the crank cover and turn it over by hand, spins freely I can hear/feel the compression events so I suspect the rotating stuff is all ok, but there may be damage thats not apparent.

What I do know is I'm going to have to pull the head for inspection. I see some stuff unspecified for now suspect stuff...

Some pics:

8mm bolt for scale on the scratches the case took from the work of art foot peg rail.
20230827_165034.jpg

Exhaust side, all ok over here. Red color on ports is RTV. Red color on valves is MMT not rust.
20230827_170437.jpg

1 stubborn bolt stops this from coming off
20230827_182705.jpg

Throttle Bodies are undamaged and removed
20230827_182713.jpg

I think I could pull a rocket engine in about an hour at this point if cheese metal bolts don't fail....

Radiator, somehow, is sound still but it's bent. Won't be reusing it.
 
There's a very real possibility I end up gutting the black cases, re-ringing the pistons, and putting it all into the silver engine I have.

In that event if the Carpenter head is still good, I'll either paint it or Powdercoat it to match the custom frame color.

Speaking of colors.....

Color 1 will be this, it will go on:
Tank, will have some stuff airbrushed on.
Fenders will get color matched using automotive paint since they can't be PCd.
Everything small, metal, and black/chrome that can take the heat (triple trees, brackets, etc).
Stock Wheels (the 16" is done already).

It's a very dark metallic purple that looks black in anything not bright and direct sunlight.
Screenshot_20230827_190512_Opera.jpg

Second color will be, and it will go on:,
Frame.
TaylorMade wheels.
Calipers.
A metallic red with gold flake that pops in the sun.
Screenshot_20230827_190823_Opera.jpg

Shooting for a single chrome piece on the whole bike...the cam cover.
 
I'm a bit torn on the fork issue. I'll have a set of nice and clean silver forks. I have Mufasa's black forks, but they have a lot of marks on them from various configurations the bike has been through and don't look nice.

So ... If I use the silver forks I'll have to figure out how and if it's possible to affix the ABS sensor spacer.

If I use Mufasa's forks I'll need to figure out how to refinish them, assuming they're still straight.

OR...use Mufasa's stanchions/fork lowers and the silver uppers ... This is what I'm leaning to right now. It keeps compatibility with no issue for ABS and removes the visual issues and question about the viability of the crashed uppers.

What year will the bike be, well I'll just say "yes" because...it's going to be a huge mashup lol.
 
Back
Top