Atkinsoning motorbike engines

[renault_megane_dci]

The GPZ has also the luck to have a shorter cam derivative (minus 30°) which I intend to use as exhaust cam.

So I expect this setup to perform a little better MPG wise.

Do you plan on milling the head ?

Did you find the excel file useful ?

(there are some hidden column with other bike setups and including a 350 less short stroke but the bike donor bike (250 eliminator) is not that common in France)

[beatr911]

[QUOTE=beatr911;359925]
For what it's worth, I found a discussion on incorrect cam timing on the Concours.org forum that is interesting. The symptom was that it didn't rev past about 4000 rpm. Turns out the poster had mistakenly installed the intake cam 3 teeth retarded. No mileage indications but apparently it idled fine and ran fairly well under 4K.

Possibly going 2 full teeth retarded would be a significant step toward what we are after. I think the cam sprockets are 38 teeth for 18.95* per tooth or a retard of 37.9* for two teeth. Adding the 37.9* to the stock 54*ABDC intake closing results in closing at 91.9* ABDC. Wow, not much compression left there. The Prius retards from 72 to 105 ABDC so this is right in there to get a mild effect.

I normally get low to mid 50s mpg so am expecting to see mid to high 50s to consider it a valid improvement.
QUOTE]


OK, so I retarded the intake cam one tooth as a first test, before trying two teeth. Since compression is reduced when doing this, I increased the ignition advance from my +4* baseline to +6* which was as far as my ignition advancer would go. This to help optimize the combination somewhat.

Power was down a little but since it's pretty powerful anyway the effect was not great especially above 3500rpm or so. Throttle response was a little softer under 3500rpm or so. Overall it ran fine and idled a little smoother probably due to a little less overlap. Though smoother, the idle was more erratic, sometimes it would idle fast, other times slow. Also as the weather began to get warm, say over 75*F at very small throttle openings and under about 2500 rpm the engine would eventually load up as if it was extremely rich. It took less than a minute in the low speed/low load conditions to begin to see the loading up. It is likely that reversion is greatest at low rpm causing the rich condition as the carb meters fuel to the same air flow 3 times.

Conclusions: Fuel efficiency was the same over two fuel tanks. However, if the rich condition could be controlled by a low speed fuel jet reduction a fuel efficiency improvement would likely be seen. Because most motorcycles are designed with very long duration cams from the manufacturer there is likely little room to retard the cam and maintain accurate carburetion metering. Very short intake runners and no carburetor plenum to share air/fuel mixture between cylinders sends a very strong reversion signal to the carburetor, exacerbating the negative effects of the reversion signal to the carburetor. Electronic Fuel Injection or a carburetor with a large volume between the carburetor and intake valve are possibly the only ways to accurately tune an engine with a significant amount of intake reversion such as with an Atkinson cycle type intake valve timing.

As it had pretty similar power from say 3500rpm up, I think that Akinson valve timing has a greater effect at low engine speeds. A Harley Davidson fuel injected 883 with tall 1200 gearing and retarded intake cams might be a great combination.

For my (overpowered) bike, possibly the easiest solution would be to make a plenum and connect one or two carburetors to feed the four cylinders, or cobble together a Megasquirt EFI system. No way I could retard two teeth in it's current configuration as the carbs would be completely confused at low speeds.

For my purposes, I just sold the bike after returning it to stock. Now bikeless, I'm searching for a great deal on a bike with great potential.

[renault_megane_dci]

Thanks for investigating this !
It was my thought that carbs would complain and EFI would probably be required.

Let's hope your new ride brings you joy (go for an EFI bike ;-) )

[beatr911]

At this point I'm looking for a 650cc or smaller bike. Due to curtailed finances at the moment there is a Chinese 250cc V-twin (Yamaha Virago 250 clone) locally for <$500. I always kind of liked those little critters. I can wrench so I'm not too escared of a Chinabike.

[sendler]

I would consider trying this on my CBR250R if I had a local dyno operator that would give me a deal. The only place around is the Harley dealer and they get $400 per session. I wouldn't want to try it without a dyno to see where and how much power is left and to retune the mixture. Even though the bike is injected and has an O2 sensor, The adaptation is somewhat global and limited in range. I think a remap fuel controller would be required.

[renault_megane_dci]

Is the CBR a DOHC ?

[sendler]

Yes. DOHC. It looks like the gears are shrunk on with a key so only whole teeth adjustment would be possible. Vacuum is already quite low. It appears to run about 5-7 inches at idle and drops away toward 0 with any throttle. Unless I was "T"ed into a restricted port that is also connected to a steady vacuum consumer. I am actually more interested in lowering the rpm of the torque peak. Possibly by increasing the length of the tuned intake runner. But even this would tend to take the mixture off of the stock map so I think any real economy gains from engine changes would require an after market fuel computer with wide band mixture logging in lieu of a dyno with a probe. They are available from Bazzaz and I will probably get one eventually if nothing more than to play with some light throttle, lean burn settings.

[renault_megane_dci]

Originally Posted by sendler

Yes. DOHC. It looks like the gears are shrunk on with a key so only whole teeth adjustment would be possible.

Best to build your cam teeth swap on actual timing figures.
Let me explain further :
a good candidate for this concept needs to have fairly long duration cam (peaky engine) in order to keep some overlap when teeth are swapped.
For instance GPZ have 290° cam that allow it to be delayed up to the 105° retard for closure that is used on the Prius and yet have some overlap.
Not quite sure but I think overlap is required.
Usually, singles don't have big duration cams because the engine inherently doesn't spin fast and requires help to idle properly in the first place wich is not one of the strenght of long duration cam engines.

Originally Posted by sendler

Vacuum is already quite low. It appears to run about 5-7 inches at idle and drops away toward 0 with any throttle. Unless I was "T"ed into a restricted port that is also connected to a steady vacuum consumer.

I don't understand, you're being too specific on your setup.

Originally Posted by sendler

I am actually more interested in lowering the rpm of the torque peak. Possibly by increasing the length of the tuned intake runner. But even this would tend to take the mixture off of the stock map so I think any real economy gains from engine changes would require an after market fuel computer with wide band mixture logging in lieu of a dyno with a probe. They are available from Bazzaz and I will probably get one eventually if nothing more than to play with some light throttle, lean burn settings.

Honestly, I wouldn't face the hazard of wasting a properly working engine on a valuable bike.

To me, ecomodding a valuable bike (or car) is already on the edge of reason.

Tuned intake runner is interesting but packaging constraint could get in the way quite soon.

[beatr911]

Originally Posted by sendler

I would consider trying this on my CBR250R if I had a local dyno operator that would give me a deal. The only place around is the Harley dealer and they get $400 per session. I wouldn't want to try it without a dyno to see where and how much power is left and to retune the mixture. Even though the bike is injected and has an O2 sensor, The adaptation is somewhat global and limited in range. I think a remap fuel controller would be required.

Or you could always just try retarding a tooth and see what happens in the real world, something a dyno won't tell you. A one tooth retard is really not likely to damage anything. (Can't believe I just wrote that.)

If you are intending to advance the cams for increased lower speed power at the sacrifice of top end, first move the cams two teeth and rotate the crank gently by hand at least two revolutions to check for piston/valve contact, then move the cams back to only one tooth advanced. You'll be safe that way. Disconnect battery for 30 minutes or so to reset the ECU and give it some miles to learn it's new conditions before evaluating power and mpg effects.

[renault_megane_dci]

I would not advance the cam on any angled valve engine.

Remember the valve cutouts in the piston ?
They are here because of inlet valve advance and exhaust valve retard ...

Also, advancing the cam means more overlap wich cause lower rpm operations to be more temperamental.

Just sayin' ...

Reducing peak torque rpm is either shorter cam, increased compression ratio, more velocity in the inlet throught smaller throttle body, tuned lenght on the inlet, increased back pressure, agressive cam timing.
Honestly, I wouldn't spend time or money on any of these.
Bike engineers, they tend to maximise the potential of the engine in the first place anyway ...

There is this device that is fitted by motocrosser thought.
It is fitted after the throttle plate and is designed to increase velocity hence boost torque for moderate openings.
That is basically a flat piece of metal that improves the geometry of the air path after the throttle plate.
Worth a shot, especially when you have only one (single rules !) and you can maximises its use (unlike motocross where you are much least at a steady throttle opening)