100% methanol seems preferable
 

Have posted some info on NASIOC about this, thought would also put it here at Ed's request.

Tried pump fuel tune, mid 11s AFR, 328 WHP on Delta Dash, 19 PSI at peak power.

Added methanol/water injection as screenwash or 50/50 mix and it didn't gain much and detted with high EGTs, tried various different mixtures/timing.

100% ethanol injection controlled the det better and took more timing and leaner AFR, but did not make more than about 335 WHP.

100% methanol injection about 10% of the fuel did 341 WHP.

10% methanol in the tank did 364 WHP.

I tried to optimise AFRs on each setup.

Anyone know how a Techedge wideband will read with a 10% methanol mixture in the gas tank? False high or false low? Seems nice at indicated 12.2 AFR.

Comments welcome.

Interesting
 

Just out of curiosity John.
Did you note any trend on the difference in ideal ignition timing between the pump gas, ethanol and methanol configurations?

I tried to chase down a value for the change in combustion speed as you add alcohol to fuel, but the folks over here at the National Renewable Energy Lab could only confirm that several sources note that the addition of alcohol slows the combustion but no one, gave any objective data so you could judge by how much.

My thinking is given that one of the "problems" WRX tuners have is the tremendous amounts of ignition advance the ECU dials in at high rpm, if you slowed the burn without getting detonation it might dramatically improve power output.

Larry

John, thanks for taking the time to come over and sharing your experiences here with those who visit. I still am not convinced that water is not preferable when richer than optimum AFR. But I do respect your contributions elsewhere and considered it worth sharing with as many as possible to consider your work and findings. And bring about a bit of discussion.

When using 50/50 mix - what was your timing like when you were getting det? Did retarding the timing at that point reduce EGT and det? and were you also losing power from retarding at that point? (i.e. is it all possible that you beyond MBT when you were getting det I guess is what I am getting at.)

As methanol is not naturally stoich to the sensor in the same way that water/steam is, I imagine the stoich of 98RON (I believe you are using) and methanol at a 9:1 ratio could be calculated mathematically from the chemical composition of the two. From there you could probably make a rough calibration of your lambda sensor and get a closer answer as to the air-mixed fuel ratio.

I've been playing in the range 25-32 degrees BTDC at 6000 RPM at 19 PSI approx.

Retarding the timing did fix the det with the 50/50 but the power was killed by doing so. EGTs easily got over 900C. Since the power was nothing like optimum despite trying a range of AFRs (limited by EGTs usually if I tried to lean it out) I don't think I got to MBT.

The problem with saying lean it out until there is no excess hydrocarbon is that the engine would have melted I believe. I really don't like EGTs over 900C that are achieved so quickly, on a very prolonged thrash I would tolerate up to 920C which is hotter than most would allow, and that is just before the turbo.

I don't think I can control conditions tightly enough to state the effect of ignition timing on EGTs although I know the theory. The things that affect my EGTs are boost, AFR and how much alcohol was used.

There are other variables that may make my findings different from the theory that water would be better. Probably lots of them.

I just plumped for what was easy to get power from without fear that I wasn't going to melt my engine or det.

Trying yo get my head around what my wideband is doing, I've ended up mapping to EGTs in the high 800s Celcius so far, "12.2" on the wideband seems to get there. Going leaner kills the timing and doesn't seem to add power.

I would agree that you had not reached or passed MBT just trying to bring out more of the details.

I also agree that getting much over 850* I would have stopped leaning you took it farther than I would have. Your results are just so much different from the results I achieved with the 2.0 wrx on both the td-04 at 19psi peak and the vf34 at 21psi peak. Trust me John not doubting your results just trying to understand them.

There is no doubt your boost, AFR and whatever supplement will be the primary determinant of EGT, was just making sure there wasn't timing induced det raising the EGTs but your timing was nowhere advanced enough for that. Your EGTs and det were definitely induction mixture level related for your boost level based on the additional information - but again at odds with my own exeperiences. Did you test the flow of your Aquamist system to make sure it was flowing as much as you thought?

You were using the 2d? What size are your fuel injectors and what was their max IDC? Another thing that has become common is that when using upgraded injectors with the 2d you are lowering your IDC which is also what the the HSV is referencing - therefore your injection mixture is going down too - as your EGTs are going up. Bad thing. The solution here has been with upgraded injectors to use two jets. I have even made a intercooler to throttle body hose with two bosses in it to assist with this without tapping the intercooler.

With those EGTs which I would have never gone near - I would actually say you were not flowing as much injection as you think. Did you have the manifold pressure sensor set to its max setting?

Sorry for the twenty questions. I want to either identify where something in this implementation didn't succeed as it should have not in theory but my actual use - or if there is something truely unusual with the results I and others have gotten with the EJ20 in the US.

At this point I suspect lowered IDCs at less than max Aquamist pressure and flow without a second jet may be the reason for your results. You just should not have been seeing those EGTs with 10% water/methanol mix.

Sorry for the rambling run-on sentences.

I was using 1S and the pressure switch was at various levels between 10 and 16 PSI, I know it was working because when enabled the methanol and ethanol made an obvious change to the AFR - indicated leaner.

Theoretically I am not a fan of the 2D based on the effective map being the complete oppostive of what is shown on the Aquamist site - ie more in the midrange than at the top, not the other way around.

I am using 740cc injectors at 70% IDC at an extra 8 PSI base fuel pressure.

To be honest 900+ EGTs are par for the course at this sort of specific output on Subarus and not necessarily the concern everyone implies - there is of course a boundary layer with the piston and I have sodium filled exhaust valves - I ran over 920 on the original cast piston EJ20 and standard valvetrain and it was perfect, Bob and Harvey (with c. 500 BHP on 2.0 get up to 950C but they have forged pistons and sodium filled exhaust valves).

At the same time the static input of the 1s results in the pressure differential dropping fast at the jet head. While an ideally mapped WI system would peak in the midrange where more knock suppression is needed - I think the 1s is farther from that profile because of the drastic drop off mixture to fuel beyond the midrange. The input pressure is constant but through the jet is not and then the fuel is increasing during the same time the WI mixture is falling. The broader the boost range in which the WI is engaged the less well suited the 1s is the application.

For instance prior to the 2d I would frequently need to use a two or three stage system to get a good broad coverage of water to fuel ratio on higer than 12 psi of boost applications.

Were you hitting those EGTs and det in the midrange or top range?

The 2d will increase its duty cycle at the HSV going from mid to top range - but it will also is losing that pressure differential so it is not as far as the map example from the 2s document.

All my experiences are being related through my use of the 2d. Others may have comparison with a static input system like the 1s.

Det was everywhere, high EGTs were as always worst at the top where the EGBP builds on a small turbo.

From what I could see of the example Aquamist map the absolute amount injected reduced with increasing RPM, so the relative amount to fuel with reduce more dramatically.

The amount injected on the mappable system did not seem to be an RPM aware or scaled quantity but just a duty cycle.

So the 2d would be further from the example than the 1s which would be quite close?

Injecting methanol really complicates the tuning process. Complete combustion (stoich) for gas is about 14.7 A/F. For methanol it is about 6.3. Methanol has less fuel energy than gas, but adds considerable quantities of oxygen. Adding, 10% methanol, to gas creates a new unknown stoich value somewhere in between, maybe closer to about 13.8...who knows? The oxygen sensor is primarily measuring the very, very small amount of remaining oxygen from the combustion process. A sensor calibrated for gas will read leaner once significant alcohol is injected (more oxygen is available). The issue is calibration. "How do you know exactly what the A/F accuracy is with this new fuel mix?" Unfortunately, you don't.

It gets even more complex when using a rising rate of alcohol injection. My SMC hardware starts at about 6 psi boost with 70 psi injection line pressure, and ramps up to 100 psi by 16 psi boost. So, now you have not only a new fuel mix, it is also changing. Trying to tune using conventional wisdom with target A/F values is a bucket of worms that may have no practical solution. It's even tougher, when there is no access to an AWD dyno.

(con't)

After banging my head against this wall for too long now, I have come to the conclusion, "To hell with A/F." The answer is to tune directly from peak cylinder pressure (PCP), just as Top Fuel, NASCAR, Formula 1, and Indy cars do. Those cars generally are not dyno tuned. It is better done on the road/track. Just run the highest PCP the engine internals will comfortably survive, adjust the postiion of the PCP in the power stroke with ignition timing (knock free), and hold EGT within safe levels. The optimum A/F ratio will appear by default. Then, just monitor that A/F for safety purposes to detect any changes in intake air flow, and fuel delivery.

The trick is to figure out an affordable (DIY) cylinder pressure tuning system...

Peak cylinder pressure
 

I would tend to agree. I think folks are getting so wrapped up with the new technology they are just confusing the issue.

In effect you already have a peak cylinder pressure indicator.
Engine torque is directly related to the MEP in the cylinders. In theory all you need to do, is run a delta dash log and sit down with a calculator and figure out the engine torque at key rpms in a specific gear. Then you use those reference points as a measuring stick. If you make a change and the index for that rpm goes up you did good, if not go back one step and try something else.

If I had the money you could also use the same concept to do the same thing in real time. Put a pressure transducer in the right front motor mount and a second transducer with a free weight on it near by. subtract the vertical enertial signal from the free weight from the signal generated by engine torque loading of the right front motor mount. Poof instant real time engine dyno.

In concept it is trivial, if it would work in the real world is another question. It is essentailly what a chassis dyno does anyway so the hardware should already exist.

As long as you limit yourself to a specific test location and vehicle weight you could also do the same with either the time rate of change of engine rpm, or road speed, or with a G meter.

If your making more torque, (ie higher cylinder pressures) and your gear ratio and vehical weight is known, the rate of change of engine rpm should increase. From the delta dash logs I've seen I think there is fine enough time resolution to make that approach work. This of course assumes no wheel spin so would be best done in 3rd gear or higher.

Larry

Re: Peak cylinder pressure
 

Yes, computing torque and HP from average cylinder pressure (MEP) is a very accurate method to get "at the flywheel" values. Peak cylinder pressure can also be infered. However, a grapical plot of actual PCP vs crank position is what is needed for tuning purposes, especially to adjust timing. There is lots of software available that will do just that, also plot torque, HP, etc. A possible complication is how long (total degrees of crank) that the cylinder pressure is at work. Fuel burn time comes into play. Slower burn, such as with higher octane fuel, and especially when injecting alcohol, will spread out the energy (as much as 100 degrees of crank rotation). As far as computing wheel HP, it is easy enough to do because drive train losses are reasonably constant over a wide range of RPM and load.

Re: Peak cylinder pressure
 

That's a novel concept, and a very interesting idea. It might be doable with a horizontal boxer 4 engine. Extraneous vertical accelerations should be absent. Don't know if it would work with a convential straight 4 cylinder engine, since there is a net upward force from the rotating parts. That's why a balance shaft is required to counter-act the extra upward acceleration forces. A transducer system that requires measuring the vertical force component might be fooled.

Wow great reading.
Do you guys have any more information on this discussion?

Maybe I'm confused ... but ,
 

If you are getting det from pulled timing , the straight methanol would naturally help prevent the detonation , due to it's being 117 octane. However you could be getting the detonation from timing pulled due to excessive intake temps. Rule #1 of course , is it's the water that kills detonation , which should allow you to run more timing.
You guys know WAAAAY more than I do , but I would think you are chasing your tail by depending on the methanol to stop the detonation. Why wouldn't you use a higher octane fuel in your gas tank to prevent the timing being pulled , and garner the benefits of a 50/50 water/meth mix which should also lower your intake temps more ?
Sure , straight methanol will make more power due to it's higher octane rating than when you mix it with water , but running high octane gasoline AND using water injection to lower intake temps should make you waaay more power.

You did notice that this thread started two years ago, right? :smile:

He he he .....
 

Ughh ... Oops :roll:

re-kindling an old ppost is always better than starting a new topic about the same thing... :D

Plus i've been out for a while

After a few months of running methanol i noticed that my tune was going crazy. AFR's were all over the place. I was also getting huge flames from the tail pipe :lol: .

This is the chain of events.

Tuned for methanol as WI. tune went like this. 9 deg adv at 0.78 Lambda running 38 gph of petrol and 5.5 gph of methanol.
The car seemed really strong in this tune. After every hard run the inlet plenim was so cold. Like it had just been removed from the freezer. I will attach a pic of the plugs.

[/img]http://upload4.postimage.org/532610/25052006089.jpg

On the way down to a race i decided to have a mild remap due to a change in weather conditions. ( it got colder ) The ignition was breaking up like crazy whilst adding and subtracting fuel. It seemed to have a very fine balance between to rich and too lean. On race day the temp went back up again and the car drove well again.


We decided to add a few degrees to the ign a few weeks later. ( i basically got a EGT for my datalogit )
We found my egts to be very low. about 700 deg C post turbo. I eventually managed to get 17 deg of advance in before the power started to drop off from the tune. egt was in the 750-800 mark.

As well as this we leaned the car out at peak tq to 0.80 L and up to 0.84 after ( anybody that knows rotaries will be violently sick at the thought of going this lean )

So on to race day again. I had a disaster by crunching my LSD changing in to second gear. How eva the datalogs from the run show the car at 0.70 Lambda across boost. A 14 % change in 24 hours... how is this ?

I have now striped down the system as well as the tranny and found the nozzeles to be giving poor atomisation in free air. I have now cleaned them and the spray pattern is much better. I have also found bits sitting in the bottom of the tank. I know of 2 other people running a similar set up using differant Meth suppliers who have had similar problems.

Any ideas. I thought meth would have kept the nozzeles clear if anything.


Scott :?:

i thought so too .... could there be a problem with your meth supplier?
Or maybe your container?
When was the last time you cleaned your water filter?

The meth supplier is spot less. I am just gonna keep an eye on the situation.

I have just orderd some aquamist nozzels. i plan to run 2 x 0.8 jets but with this amount of flow i need to re think there placing.

Any ideas ?

Scotthttp://upload4.postimage.org/584209/DSC01549.jpg

Fuel methanol I've purchased will develop a "snow" like deposit in suspension with the fluid after it sits for a while.
I have been led to believe it is a lubricant or scent they add to the fuel but am not sure. I saw it in my WI system on a couple of occasions.
and started running a fuel filter in line with the lines to screen the crud out.

Not sure it your seeing the effect of the same thing but just a thought.

Have you tried putting a sample of your methanol in a clear jar and seeing what happens over time?

Larry

Argh ... That makes sense. I thought i was crazy when it had a smell to it. Usually it doesnt.
What you describe sounds spot on. I have filtered the methanol into a plastic container and a glass container so i can see if there is any differance.

Scott