924Board.org

[ideola]

tuurbo, I'm not arguing with you, really. And I'm not making any claims. I'm simply expressing concern. And a WWII aircraft engine is a different thing than a car engine, particularly if it was designed to run with Methanol injection. All I stated was that you can't be certain (at least based on what you've presented so far) that those engines weren't prepped specifically for Methanol handling, nor can you be certain that they were made with the same grade of alu as our heads. That's all I'm saying.

I'm not trying to disabuse you of the idea of running methanol injection. I'm simply saying, in my case, I'd choose to run water only just to be safe. The rest of the crowd can do as they please based on their own judgment of the relative risks.
_________________
erstwhile owner of just about every 924 variant ever made

[tuurbo]

It's cool - I meant 'argue' in a collegial way - debate - which is usually a good thing. I have no complaints about lack of collegiality.

The wartime data don't test for reliability - though they do point out that cylinder head temperatures were way down using methanol and water - so they probably thought reliability would be enhanced with cooler cylinder head temperatures - but I'll take a closer look and see if reliability figures can be obtained. It's a good point you raise and since reliability is a good thing I am sure I can keep an eye out for it in my research into this topic.

The researchers did use existing gasoline engines and I suspect they may well have had heads made out of heavier material - but I'm not sure - I do know that weight was always an issue in aviation and manufacturers were cognizant of weight concerns, but I have no idea what materials were used in the test engine heads as the materials weren't specified in the write ups. So did they use aluminum heads in their test engines? Dunno. But I do know that whatever they used was readily available and they didn't make a head just for the test.

I have concerns about methanol and water injection too, and they go back to the basic idea of shoving something into the cylinder chambers that can destroy oil viscosity. The last conclusion here points to this issue:



It's the concluding remarks of Rothrock, Krsek and Jones (1942) Langley Memorial Aeronautical Laboratory. The caution here for Item 5 is all about worrying over throwing too much methanol and water into the system. Hence, I started with 60 ml/min and only tossed in another 60 ml/min once I saw that the existing dose was insufficient to get rid of detonation.

Now, I do the finger test whenever I check the oil to see if the viscosity is ok and to ensure that there isn't anything else in the oil besides oil.

I'm still open to any signs that methanol will degrade the head, or water, for that matter. If I have to choose between tightly gripping my theory and destroying my Porsche engine, versus letting go of my theory and keeping the car running right, I'll keep the car. I can not afford too tight a grip.
_________________
1980 924 turbo, MSD, Meth. Inj, otherwise stock.

[ideola]

[bruni]

I started this tread because I am after practical information regarding water methanol injection

[ideola]

[tuurbo]

In one of the studies I have here, the researchers did notice too another issue that might be relevant. Although cylinder head temperatures were generally lower with water methanol injection, the exhaust guides were a little hot! The researchers speculated, if I read them correctly, that the reason for this was that boost was considerably higher when they used water and methanol injection.



Note that the top row of the Table is with no water methanol injection and this is compared to the ideal 1.5 water fuel ratio. However, you need a Farnsboro indicator (What the hell is that?) to determine the precise water fuel ratio. Rothrock, Krsek and Jones (1942) go on to say:



You can find more reports like this at NASA:

http://ntrs.nasa.gov/search.jsp?N=4294752405

Note that the solution the researchers suggest is more water and methanol into the mixture, as they note "Higher water-fuel ratios than 1.0 caused a very rapid decrease in the exhaust-valve guide temperature".

I don't understand everything about this article. And neither can the reader because it's only parts of the article. But I think what you can carry away is that:

1. If you're going to measure cylinder head temperature, it's really going to matter 'where' you place your measuring instrument. I'm going to keep a lookout for one I can place as close as possible to the hot areas of the car.

2. In tuning the engine, one should keep in mind the cylinder head temperature just as one keeps in mind the IAT and other measures that can spell trouble.

3. Make sure you have good or new exhaust valve guides.
_________________
1980 924 turbo, MSD, Meth. Inj, otherwise stock.

[Scorpio]

[tuurbo]

Hey Scorp,

Yeah I am not sure about the Figure 12 (e) data - I've been dealing with curves like that lately and in my own work I'm wondering about measurement errors. But the 'right' thing happened and that is that the 1.5 water:fuel ratio showed the lowest flange temperature, as expected. But the other curves I don't get. Naturally when I see Figure 12 (e) I'm wondering whether they had trouble with the temperature mount or some other incident that obscured the data. The overlap of curves suggests some third factor impinging on the results and I don't remember the authors making mention of it. Maybe someone who knows more about air/fuel mixture and temperature can shine some light.

I prefer Figure 12 (f) because, for lack of a better word, the 'orderliness' of the four water fuel lines makes sense and is patterned and predictable. The lowest curve seems the most desirable to me and that's the magic 1.5 water:fuel ratio (which needs clarification) with the diamond.

I might chase down that water:fuel ratio some more. The curves for 1.0 water:fuel look pretty appealing too, for other reasons. I just have no idea right now what that frigging ratio 'means'.
_________________
1980 924 turbo, MSD, Meth. Inj, otherwise stock.

[bruni]

Hi those graphs are really interesting, it's good to have what you think will happen confirmed eg the weaker the mixture the hotter the engine runs and the more water added the cooler, it's unfortunate that the exact meaning of the various ratios is not clear. I think trying to measure temperatures within the cylinder head is very problematical and even today would prove quite a challenge. I'm sure the the temperature variations will be enormous, given the different materials, thickness, proximity to water galleries, and the all the other variables. I believe that EGT will be a reliable indicator as to what's happening during combustion. My own feeling is that you are not necessarily looking for the lowest combustion temperature, just low enough to prevent detonation, but you could only ascertain this on the dyno. For me (like I think yourself) I am just looking to make my road car a little more reliable with perhaps just a bit more boost with the the gas i can buy at the pumps.
I have decided on a home build system for a budget of around $200 (I like challenge) On this budget I can't achieve proportional water delivery, however I believe I can probably make a crude 3 step delivery, taking boost pressure and throttle position as the triggers. Nozzle size and position are the two variables that i'm not to sure about at the moment. I will post a bill of materials and layout as soon as I have confirmed suppliers and costs

PS keep researching
_________________
1982 931
1975 harley sportster

[tuurbo]

Doh!

I just realized why the authors of the Technical Report, above, failed to explain the Farnsboro instrument, and the precise water:fuel mixture, and why, generally, it's not as informative as would be useful for our purposes: official secrecy. It's a shame really that the people essentially pay for this research, but the researchers bury the essential details that would make their research benefit everyone.

Anyway, the nozzle position used by 15 inches upstream from the inlet port and 9 inches upstream of the fuel injection valve - injected only during the inlet stroke, and nozzle, it was thought, could be moved 5 more inches upstream without affecting the results.

They don't say the pressure of the pump. Mine is 150. Pump pressure severely affects spray pattern and atomization. When I turned my pump down to about 75 to 100, the spray was fine but not nearly like the case when the pump was fully at 150 PSI - it simply blasts a mist of water that makes you jump. You need a full power pump.

If you want to adjust things, adjust nozzle, placement, boost onset, or whatever, but don't adjust the pump - just put it at full power and 'fuggetaboutit'.

I think I understand where you're at in the process and I recall being there, I think, the spring. At that time I wanted to make sure I was using the right nozzle for the flow rate I wanted. And for this I had to get the fuel pump to the right speed. I dug up these tables and a schematic. But, as usual, there are unanswered questions.

I think for a 931 the 100 ml/min nozzle should get someone up to speed with the stock vehicle, and to allow the owner to push the car to the limits of the CIS at 10-12 PSI. Basically, if the fuel system can give the car enough fuel to reach 12 PSI without running lean, the methanol and water injection should make it run like it has an air-air/water intercooler, and better. I figure if my CIS isn't giving enough fuel between 10 PSI and 12 PSI, my methanol and water system is compensating just fine.

The 175 ml/min nozzle I can part with if you like. I also have nozzles for the higher horsepowers too, like, I think 250 HP 350 HP and so on. All you'd need to buy is the nozzle holder from Snow. But if you want better metal lines, you can use a brass fitting from a water cooler or air conditioning parts supplier. Maybe even Home Depot. Then you can use plastic or metal lines from the pump and tank.

The following comes from RSR and I found this using Search of our 924 website:

http://www.rbracing-rsr.com/waterinjection.html

It features a methanol-water calculator - I used this to get a list of factors that might be relevant for my instance. My car scores 14.8% coolant to fuel ratio, and 11:1 air to liquid mass ratio.

Snow has a formula they use suggesting for nozzles and HP, but I had the figure 100 ml/min suggested to me by a Snow Tech Support representative who said 100 ml/min should be enough. Mine is 120 ml/min which gives me some room to grow the car:



I chose the combination that suited my car for HP and for desired atomization, but I'd bet a 100 ml/min nozzle couldn't give the atomization I'm seeing with the 60 ml ones. I just like it - it's totally intuitive, and I have no idea if it truly makes a difference.

924 Owners Should Share Data in some Central Source

The 'Search' feature is nice but terribly time consuming. If anyone wants to send in some figures, I think we could actually determine ourselves the precise water:fuel ratios ourselves.

Vehicle 1: Stock 931
Vehicle 2: Stock 931 with water injection Approach 1
Vehicle 3: Stock 931 with water injection Approach 2 - Intercooled
Vehicle 4: Stock 931 without water injection Approach 3 - Intercooled

It's possible to make comparisons if we just had the raw data sorted in to the right four groups. Sure everyone's car would be a different setup, but as long as the extra mods are known it's still useful.
_________________
1980 924 turbo, MSD, Meth. Inj, otherwise stock.

[bruni]

Fantastic at last someone talking some sense I am planning to use 2 pumps and 3 nozzles, 2 triggered by pressure sensors and 1 by throttle position. All the nozzle sizes I have seen only state ml/min or some other unit of flow rate and make no reference to pressure, I assume that the flow rate is proportional to the pressure! My gut feeling is that the nozzles should be placed close to the exit of the Turbo, the air should be at its hottest and most turbulent at this point. I am going to use the pump described in the following autospeed article http://autospeed.com/cms/A_110368/article.html there is a question mark over durability but you can pick them up on ebay for about $15 so has to be worth a shot. I will try to post the BOM and the schematic over the weekend and look forward to some response to your post
_________________
1982 931
1975 harley sportster

[alexvex]

You guys sure have complicated setups. I've got a 1 pump, 1 nozzle and it's controlled via a hobb's switch. works great. would allow me to run ~6psi more on my old escort before i would get ping.

Also, I originally ran 50/50 methanol and water with no ill results. I then got lazy/cheap and I've been using wiper fluid with great results. Just find the cheapest, coldest rating you can find as those will have the most methanol and the least detergents.
_________________
--Alex--
1992 Porsche 968 (VW 2.5L I5 Turbo swap in the works)
1995 Audi 90 Quattro Sport (4.2L V8 Swap)
2010 Touareg TDI
Past Pcars: 80 931, 87 924S, 87 951

[Min]

[ideola]

Stumbled across this today while looking for other stuff:
Water Injection Combustion Chemistry
_________________
erstwhile owner of just about every 924 variant ever made

[ideola]

Here's another relevant article that talks about how to figure out how much to inject (scroll down):
Some Charge Air Cooling Thoughts
_________________
erstwhile owner of just about every 924 variant ever made