On another note, how many of you think that spraying a 50/50 mix of water/meth will require an adjustment to the fuel table or a fuel trim?
The answer:
Depends on the volume of course Adam
Roger Rabbit?
I'm a little slow!
Your son will know M54's better than the bulk of BMW techs.
Sent from my E6782 using Tapatalk
1996 332IS
Built 3.2
CES/Steed TS Precision 6466, spraying a "$π!℅" load of meth.
Technique Tuning 80# tune.
1/4 mile 10.84 @ 136.72
Your 1 and only stop for all your BMW performance needs
WWW.CESMOTORSPORT.COM
So what does it mean? I know I turn down my water/meth flow in winter. If I don’t the engine will stumble sometimes. What works great in July may be too much in December, but you could also flow just a modest amount in July and be fine in December as a compromise. I’d rather flow more in hot weather and less in cold. With the Aquamist it is easy to adjust flow.
It means I have to go to a dyno. LOL
I'm interested in why an engine stumbles. Is it increased mass flow (From additional cooling) that isn't being fuelled properly? Or is it fluid coating the plug....so much so that it doesn't evaporate at TDC causing a miss-fire? I don't pretend to have an answer, but I'm interested in getting to that point with my setup to at least try to understand it....
This is the temps for pre-turbo injection:
What is really interesting is the fact that the outlet temp is cooler than the inlet temp under light load. Also, the inlet temp goes down as mass-flow increases....then goes up dramatically right after high load. My guess is that there is a significant amount of heat transfer from the exhaust manifold into the un-insulated turbo inlet pipe. Shortly, I'm going to add significant heat insulatation product to the inlet pipe to see how this changes things. Secondly, the outlet pipe is exposed directly to airflow under the car...and the temp sensor is about 18" after the turbo outlet. I'm guessing that under light load, the charge is slightly cooled by the pipe, and that the sensor itself is slightly cooled by the airflow. But once the pressure ratio goes up, this changes.
The only way to measure the effect of the pre-turbo injection is to do back to back dyno pulls with and without. However.....I think I need to add thermocouples to the turbo CHRA, and maybe the compressor cover to make sure there is a common starting point. (Not heat saturated on the 2nd run) Yes.....I have more thermocouples....and free inputs. (9 of them by my last count)
Manifold:
Another interesting situation.
The temp after the intercooler under light load, is lower (3.4 deg C) than the inlet manifold temp. I chalk this up to heat soak of the inlet manifold. Then under higher load, the situation changes as Water/Meth is injected. The Manifold WM Delta channel shows this difference, with a drop of 9 deg C under load. Then off throttle, the delta increases....which doesn't correlate with minor drops in the Intercooler outlet temps. It looks like at this point that there is residual Water/Meth in the system.....probably distributed across inlet pipe and manifold port walls?
I wouldn't read too much into temps that are +/- ~2-3 C from each other. Even with tight tolerance Type K thermocouples, it wasn't uncommon to see a 4-5 deg F swing between the highest and lowest with a large enough sample size of say 10-12 of them. The sensors are just not accurate enough to really say a temp measured 1-2 deg C different from another is actually any different at all in reality.
There could be some minor cooling, but I think sensor variance is a more likely explanation based on lots of thermal testing.
Where are those 3 sensors, intercooler temp, inlet air temp, and engine charge temp located approximately? What is Manifold WM Delta?
Intercooler temp - 12" after the intercooler outlet
Inlet air temp - Inside of the inlet manifold
Engine charge temp - Calculated by Motec using something called "charge cooling gain"
Change in temperature of the intake charge due to evaporation of fuel.
See Engine Charge Cooling Gain Primary or Engine Charge Cooling Gain Secondary for information on how to calibrate this compensation.
If primary and secondary injectors are active at the same time the value is linearily interpolated based on the contribution of each injector.
Manifold WM Delta is the difference between the intercooler outlet temp, and the inlet manifold temp. Basically....I'm aiming to measure the affect of water/meth injection, and nitrous injection to temps.
You'll find that most OEMs do something similar....I think GM even calculates the temp of the inlet valve.
Motec gives you a starting value for this, then there is a procedure for tuning to your application. I'm still using the default value....because I need a chassis dyno with load control to do the test......
Last edited by PEI330Ci; 12-22-2017 at 05:25 AM.
I run 70/30 meth/water. In cold weather you need to reduce flow if you are running a lot in hot weather.
So when you key on after the car has been off 24+ hours, every sensor reads exactly the same? If so, you've got yourself a great batch of sensors. Mine have always deviated by +/- 3-5 deg F before startup (typically coolant and intake air temp, maybe a diff or trans temp at times).
Nope, not exactly.
Motec now has drop-down menus in M1 Tune for various types of sensor, so that you get the correct CAL file directly. I input the raw values into the M1 ECU, either through direct connection, or over CAN, then the scaling is done by the ECU.
Every sensor reads different.
I had the car in the shop without heat for about a day to allow the temps to stabilize.
EGTs (K Type Thermocouple) ranged from 49.1 to 50.3 deg F. This averages to 9.7 Deg C.
Coolant temp = 10.0 Deg C
Oil temp = 5.7 Deg C
Turbo Inlet = 10.5 Deg C
Turbo Outlet = 10.3 Deg C
Intercooler Outlet = 10.3 Deg C
Inlet Manifold = 9.3 Deg C
This took a bit of work to acquire, as the M1 ECU code is written only to data log when the engine is running. I wrote a switch into the firmware that allows me to log continuously with the switch on.
The results:
The Oil Temp sensor I'm going to look at replacing with a spare to see the effect. Compared to Coolant Temp, this has been consistently lower over the past few years across many tests.....I just didn't think it was critical to address. Same for the Fuel Temp sensor. One thing to note is that the fluid temp sensors have a different calibration curve from the air temp sensors.
The OEM BMW Inlet Temp sensor does not have an "official" calibration curve. I literally had to build one myself with a heat gun....so it's always going to be a little bit different from the other sensors....but it tracks very similar across the range.
Last edited by PEI330Ci; 12-22-2017 at 11:17 AM.
Ok, doing a one point calibration between sensors does help, but it can vary off that setpoint by a bit. I still wouldn't buy too much into small deviations, especially in transient events. Sensor convective cooling can dominate your transient events more than actual fluid medium temp when the delta T is fairly low from fluid medium to sensor temp.
I'm very interested in those pre-turbo results, when you can get them. But of most interest will be the post-IC result with pre-turbo WMI on. The ability of this method to extend the compressor map to the RHS and seriously drop the pre-IC temps is very interesting. What gave me a sad face was the realisation that if the WMI condenses back into liquid again in the IC, you've just lost your cooling advantage with the reverse of the evaporisation. What pressure + temp this would occur at, I'm not across right at this moment. It would also probably depend on the saturation of water/meth in the air stream as well. Regardless of that scenario, it still holds true that the WMI pre-turbo has the ability to extend the compressor map, and or lower the shaft HP requirement on the turbine.
1989 E30 - M50B28 Turbo - ZF 8 Speed
I really need to get turbine speed working....
One of the points of view I have heard is that the water/meth might displace "X" volume of air at the turbo compressor, resulting in a reduction in total mass flow of oxygen through the engine. This would be provable on a dyno where there were no temperature compensations applied to the ignition or fuel tables, and you just measured power output with the pre-turbo injection on or off and stepped load points. Obviously the turbo would have the ability to consume more water/meth at the edges at the compressor map.....but you would want the ability to map that....with a turbine speed compressor.
The Vapor Pressure for methanol is very high compared to water. For example, if a person was to take a sealed bottle of air at atmospheric pressure and inject water into the sealed bottle and allowed to evaporate, the pressure inside the bottle would go up by 3.1 kPa at constant temperature. With methanol, the pressure would increase by 13.02 kPa. So yes, air (oxygen) is displaced due to evaporation of these liquids. Because of this, I changed my direct port meth injection to reduce evaporation in the intake port.
i could see were pre turbo water injection could benifit in hot, dry climates may benifit. Anywho, awaiting dyno results from this awesome data collection, build thread.
WOT
Bookmarks