A Labor of Love; aka building a Rocket III Speedster

[Claviger]

Sensor adaptors for the Bosch dual sensor, not as easy to find as one might expect, most m10x1 to m10x1.25 are fine threads on male end, which doesn't help.

 

[Claviger]

Ah the value of a good night sleep.

Looking at the options out there for electronic throttle actuators, and the associated prices, I realize $1,100 is absolute highway robbery for an actuator.

It dawned on me that I could buy a $200 actuator Chop off the built-in throttle blade and throttle body, and have the same thing, a motor and a mounting bracket that uses a shaft output.

That got me searching, and then I found what will likely be the part I use, BMW have been using drive-by-wire for more than 23 years, I think the first BMW with it was a 740i and the '90s.

Enter the pierberg M3/Z3 S54 actuator. It has an invertible arm, so it can be flipped to open in pull or push, and has a built-in throttle position sensor, and the best part, you can find them used cheap. Additionally, Haltech Will be familiar with them because they're an OEM part on a popular tuner car. Further, the throttle position sensor is inexpensive, and rebuild kits for the actuator are less than $100.



As much as some people like to hate on BMW, Bosch, and ,Pierberg, I have never once had an with their electronics that wasn't user induced failure.

 

[Claviger]

As some may be wondering. I've not picked up the ECU yet because I'm not totally sure the Haltech Nexus R3 is my best choice.

The Motech GPR130 and MaxxECU Race Hydro or Pro are both equally or more capable on paper, I just need to play with everyone's software and figure out which one I like.

The MaxxECU Race Hydro is particularly attractive due to size, price and built in EGT controller and built in WB controller.

 

[Tripps]

Did you ever bore your throttle bodies? Like told them a long, dull story?

 

[Claviger]

Did you ever bore your throttle bodies? Like told them a long, dull story?

I did. 58mm ID/56mm ID top to bottom instead of 55.5/52 stock, then hand blended the 56 bottom into the rubber mount. I netted no meaningful result from it at the time, but, I now believe I've been hitting a flow limit at the exhaust valve, completely negating the gains from the TB/Nev's Cams/etc I'm working on solving that, I have a head at the shop getting some work done now for 40 intake / 34 exhaust valves.

One reason I suspect this to be the case, is a dyno sheet I found from a few years back. With 4" stacks and with stock stacks. The 4" stacks gained 7lbft everywhere below about 6500, and the two setups were within 2 hp up top, the short stacks should've pulled ahead up top, they did not indicating there's a flow restriction elsewhere and valves are the only remaining possibility.

On ECU Choice
I've tentatively settled on the MaxxECU Race Hydro as my ECU. Having an EGT sensor per cylinder is something I do want and having it built into the ECU is nice. The MaxxECU Race has more flexibility and features than the Motech or Haltech while sneaking in at about $1k less money and can play nicely with almost any CAN device on the market since it has it's own CAN sniffing/decoding functionality built in to the ECU.

So the full setup will be pared down to mo.unit blue for chassis harness control and ignition circuit control, MaxxECU Race Hydro for engine management, and Haltech IC-7 for Display. Bosch 2in1 Temp/Pressure sensors for oil/water/fuel. Triumph sensors for trigger/speed/wheel speed/ etc, KTM throttle tube, R35 GTR smart coils, and the BMW ITB Actuator controller. Cam position sensor is going to be attempted using a cam lobe as the trigger, if that doesn't work, I'll go to a single tooth cam gear modification.

A less ambitious conversion could be done with all stock sensors, gauges, vehicle harness and simply swap the MaxxECU, KTM throttle tube, coil igniter, and BMW ITB actuator for a DBW conversion for around $2750 to give 1 gen Rockets full traction/launch/engine control/cruise/multimap etc. Without the cam gear it will run in wasted spark, but if my cam lobe as the trigger mod works, it's trivial to get full sync control.

Edit:
I just found a key feature applicable to R3s... Cam position trigger is the biggest hurdle for this engine configuration, well..MaxxECU have a MAP based home trigger and configurable MAP pressure threshold. When Cylinder 1 is on the intake stroke it will spike that MAP signal and the ECU knows that #1 is on intake, then compares that to crank position and it gives a means for full sequential engine control without an extra cam sensor. BAM, Maxx wins the best for rocket 3 feature set with that alone. This is the Keihin home signal strategy used by the stock ECU and maxx the only fully compatible standalone with the stock hardware.

 

[Claviger]

So a turbo engine cylinder head should have better heat dissipation ability than an NA head

Step 1: source Manganese Bronze replacement valve guides. Got these today from the machine shop so I can source the correct replacements:


Hmm...

 

[Gerald]

So a turbo engine cylinder head should have better heat dissipation ability than an NA head

Step 1: source Manganese Bronze replacement valve guides. Got these today from the machine shop so I can source the correct replacements:


Hmm...

Why the different lengths of the guides, I assumed the guides would be same lengths, and are you thinking of shortening the length of the stroke, changing the rod angle and maybe a few mor rpm's?

 

[Claviger]

Since I couldn't find this info anywhere on the net:
Triumph Rocket 3 Valve Guides






I'd imagine the guide length disparity is due to port angle, the exhaust is a much shallower port angle so sits lower in the head.

Trying to get under 26 m/s mean piston speed at 9k and you nailed the other main benefit, a little less angle. Spool won't be an issue (in fact I'll have to delay it with WG solenoid) so extra length is in no way beneficial to a turbo rocket.

 

[Gerald]

Since I couldn't find this info anywhere on the net:
Triumph Rocket 3 Valve Guides





I'd imagine the guide length disparity is due to port angle, the exhaust is a much shallower port angle so sits lower in the head.

Trying to get under 26 m/s mean piston speed at 9k and you nailed the other main benefit, a little less angle. Spool won't be an issue (in fact I'll have to delay it with WG solenoid) so extra length is in no way beneficial to a turbo rocket.

Thanks for the explanation, I just assumed the lengths of the guides would be the same. You know what they say about "assume". I didn't realize you were possibly going turbo. A lot more opportunities for power, but also some pitfalls. As always, want you to succeed.

 

[Claviger]

First bike will stay NA.
Second Rocket will get boosted.

Bit of bad news -
KPMI, Supertech, Ferrea: none list a matching guide for either size.
Bit of good news -
KPMI makes guides with the correct sizes in all locations so just not all at once, meaning it shouldn't be an issue to have a set cut from Manganese Bronze.

With that I'll be shipping a set of valves, keepers, retainers, springs, and guides out to Ferrea on Monday so John can get accurate dimensions. Then he and I can discuss getting modified versions made.

Looking to swap to shimless buckets from Toyota 1zz using 1.6mm longer 39mm Stainless Intakes/34mm Inconel Exhaust valves, getting the seat above the interference fit but below the seal on the guides shortened by 2mm, and a shoulder added as a positive stop, then some 12lb/mm double springs made with matching tool steel retainers.

The idea is, the shorter guides will create space for a 12.7mm+ lift cam, the longer valves will allow for shimless buckets without sinking the valves in the head, and the last bit is the seats. I'm fairly certain an off the shelf 1.575OD beryllium copper seat will work but waiting depth of the stockers from my machinist to confirm.

Should wind up with a bulletproof head assembly that can spin very...very high.