GS(X)1100E turbo EFI

[49er]

Difference in volume is small but it makes 0.2-0.3 difference in compression ratio. Not very dramatic though but if the engine is running on the edge of detonation it matters.



Yep, it should be interesting I'm planning for 1bar / 15psi maximum boost or maybe slightly more. I got the turbo from a guy who was using it on very similar engine. He got about 300hp at 15psi and 370hp at 24psi.

Water injection will be for charge cooling and detonation control. I'm not going to use an intercooler. I think that an intercooler that is large enough for intended power would be too large for reasonable installation. And it would be difficult to install so that it won't take too much heat from the engine and the headers when there isn't enough speed for proper air flow.

So the water injection will be only charge cooling method. I have an Aquamist set with high speed water valve. With that I can use the Megasquirt to control amount of water and thus keep water / fuel ratio constant for different load situations. That should allow larger amounts of water at top end without flooding the engine at mid range.

I think your choice of water injection is a good one, with the mods and boost you'll be using.

The pros and cons of intercooling vs water injection do need to be weighed carefully. One advantage of intercooling is that you can design the system that incorporates packing ice around tank area for "special events" such as turning up the wick on the drag strip. This is a way of adding considerably more cooling over short durations. But you're right, the cooler size is a major dis-incentive to this system.

One bonus that comes from using water injection is that the amount of carbon build up in the combustion chambers is minimal compared to normal aspiration. The superheated steam does a great cleaning job in that area.

 

[ArttuH]

Progress has been slow but finally it's back on the road.

Assembling the top end went fine except one snapped cam cap bolt. Just plenty of tedious tinkering like checking ring end gaps, squeezing the pistons to the cylinders, timing the cams and so on.

Because the combustion chambers were hinting that air distribution between the cylinders has been somewhat uneven I did a slight modification for the plenum. I installed a perforated plate between the entry connection and rest of the plenum. That should help to get even air flow through whole length of the plenum.

Then servo control for VNT vanes of the turbo. I'm using a RC servo that pulls VNT control lever through a cable. The servo is controlled by boost control output of the Megasquirt. Some additional electronics was needed to convert 0-100% duty cycle pulses from the Megasquirt to 1-2ms pulses that the servo excepts.

After these the bike was ready for the first test ride. Tuning parameters for the VNT control were naturally way off so boost build-up was far from ideal. Otherwise the bike was working fine, no oil leaks nor strange noise etc. So now it needs some tuning for the VNT and bedding in the piston rings. Then it's time to see if the overhaul improved the power department.

 

[49er]

That plenum mod is similar to the inlet design on a stock air box.
Have you continued the holes along the top of that plate, to allow even release of air into the top chamber? That should work quite well.

The challenge is keeping the inlet pressure equally dispersed to each cylinder. That boost control into the plenum sounds like it will work once you get the electronic maths and signalling sorted. You could even try twin blow off valves, one at each end of the plenum, to try and keep the pressure even across the inlets.

Water injection testing, as part of the final tuning sessions?

 

[ArttuH]

That plenum mod is similar to the inlet design on a stock air box.
Have you continued the holes along the top of that plate, to allow even release of air into the top chamber? That should work quite well.

No, the holes are just on the back side of the plate. The idea is to force air go around one more corner before entering throttle body inlets. So air enters to the plenum at lower front corner. At first it gets to a small tunnel formed of that L-shape plate and lower corner of the main plenum. Then air flows to the main plenum through these small holes that are spread over whole length of the plenum. Cross section of the small holes is roughly the same than the inlet tube. This should ensure that the flow will be divided quite equally over each small hole. Then air flows from bottom of the main plenum to upper part and to throttle body inlets. At this point the flow should be relatively slow and pressure should be equal for each inlet. Of course this is just theory but anyways situation should be far better now than without extra plate.

The challenge is keeping the inlet pressure equally dispersed to each cylinder. That boost control into the plenum sounds like it will work once you get the electronic maths and signalling sorted. You could even try twin blow off valves, one at each end of the plenum, to try and keep the pressure even across the inlets.

There is just one blow off valve on feed pipe of the plenum. I think that blow off configuration shouldn't affect much to air distribution between the cylinders. When the blow off opens the throttle blades are more or less closed.

Water injection testing, as part of the final tuning sessions?

No, I'm afraid. Installing the water injection system itself wouldn't be too big effort since I have all parts already but water tank is a problem. At the moment there isn't any reasonable place for it on the bike. In future I will probably modify the rear fender to make more room or something like that. But I guess that I won't have time for that during this summer.

 

[ArttuH]

Done plenty of VNT tuning. Now it begins to be drivable but transition to boost isn't yet as smooth as it could be. So some more tuning in order.

Last Friday I did my first drag racing attempt. Results weren't anything too special but it was great fun. My best ET after several passes was about 10.8s, 60ft times were around 2 seconds and terminal speeds around 210km/h. I guess that Metz Z4 isn't the best rear tire for this purpose It was quite difficult to get good traction at launch.

But I guess that piston rings should be properly bedded in now :-\\\

 

[rapidray]

Is that a GPZ turbo over in the other lane? Ray.

 

[ArttuH]

Yes, my friend's bike. GPZ750 with home made turbo kit, bored to 810cc and also with Megasquirt engine management.

 

[Sunburn]

How did you handle your fuel return to the tank post fuel rail?

 

[TheCafeKid]

Arttu what wheels did that bike ORIGINALY have on it? They arent STOCK are they? Never looked like that over here anyway... Id be interested in finding a set like that however.

 

[ArttuH]

How did you handle your fuel return to the tank post fuel rail?

I have replaced the fuel tap with a flange with two outlets. The return line has extension pipe inside the tank so the returning fuel exits at top of the tank.

 

[ArttuH]

Arttu what wheels did that bike ORIGINALY have on it? They arent STOCK are they? Never looked like that over here anyway... Id be interested in finding a set like that however.

Original wheels can be seen on this picture. Current wheels are off from GSX-R1100 '91-'93.

 

[Sunburn]

I have replaced the fuel tap with a flange with two outlets. The return line has extension pipe inside the tank so the returning fuel exits at top of the tank.

Did you find a flange with the extension tube or did you have to modify one? Did you consider a simple check valve or would that not be as effective?

 

[ArttuH]

Did you find a flange with the extension tube or did you have to modify one? Did you consider a simple check valve or would that not be as effective?

The flange is home made. Actually I got it along some other engine parts that I bought. Anyways, it's very simple piece of hardware. Just a flat piece of iron with two tubes going through it. And I think that return line doesn't actually need that long extension. As long as it doesn't exit right next to outgoing line inlet it should be fine.

I'm not sure if I understand what you mean with that check valve. Where it would be installed?

 

[Sunburn]

I'm not sure if I understand what you mean with that check valve. Where it would be installed?

Check valve only allows flow in one direction. Intalled on the return fuel line at the point of entry of the tank. I thought this would elimnate the need for an extention tube but I am asuming the need for the extention tube is to allow the fuel to re-enter the tank without the risk of fuel backflowing.

Can the extention tube still be below the fuel line and work correctly? Am I over thinking this? LOL

 

[ArttuH]

There is no need for a check valve on the return line. There shouldn't be any back flow since the pump is pushing fuel always to the tank.

Basically there are two important requirements for the fuel return:
-return point should be far enough from pump intake line inlet to prevent heated fuel and gas bubbles coming from return line entering right back to the pump circuit.
-there shouldn't be any significant restriction on the return line that could disturb fuel pressure regulation.

So exit below fuel level shouldn't be any problem.

 

[Sunburn]

Good info, thanks

 

[Sunburn]

Pictures from your 1st EFI project looks like you used the GPZ pump as well. How did you figure what relay to use, or did that come from GPZ to.

 

[ArttuH]

Yes, I used the GPZ pump in first EFI setup. The pump was fed from a relay built in the Megasquirt box (my own circuit board version). Anyway, any general purpose automotive relay should work fine. There isn't any special requirements for the pump relay except current capacity (>10A).

 

[Sunburn]

The pump was fed from a relay built in the Megasquirt box (my own circuit board version). special requirements for the pump relay except current capacity (>10A).

I can see why you did it that way. There are tons of relays under 10A that are for direct fit to a board. There are not many options so far for auto style terminals. Most are 30A or more. Funny how the little things turn into big things.



Mark

 

[ArttuH]

I can see why you did it that way. There are tons of relays under 10A that are for direct fit to a board. There are not many options so far for auto style terminals. Most are 30A or more. Funny how the little things turn into big things.

Hmm, I meant that the relay must be able to carry at least 10A. So typical automotive 30A relay is just fine.

I fitted the relay on the board just because I wanted to save space and external wiring.