ttg35fort

iTrader: (2)

There are two distinct aspcts to the benefits of a methanol, water, or H2O/meth injection system.

1. Reduction in temperature of the intake charge.

2. Increase in the net octane level of the intake charge.

When the intake charge is extremely hot, injection can significantly decrease the temperature of the intake charge, thereby increasing the amount of intake charge that can be crammed into the cylinder before it detonates. As the temperature of the intake charge is decreased, the effectiveness of the temperature reduction is decreased. However, the increase in net octane level is always a benefit. EDIT: when water changes from a liquid state to gas prior to or during combustion, it is endothermic (i.e. it absorbes energy), thereby reducing the ambient temperature of its environment. In the cylinder, this is helpful for staving off detonation. Methanol also is endothermic when converting from a liquid to a gas, but becomes exothermic what this state change occurs due to combustion.

For example, if you have a turbo setup without an intercooler, the use of H2O or H2O/meth injection will be much more effective at reducing the intake charge temperature in comparison to a turbo setup that does not use an intercooler. Also, the water component of the injection solution is more effective at reducing the intake charge temperature than the methanol component. For that reason, you see straight water as being very effective in some turbo systems (such as those that have a very small intercooler or no intercooler at all) and not in others.

When someone has a very good intercooler, straight water is not as effective as using straight methanol or an H2O/meth mix. In some FI systems (that use a very good intercooler), better results are actually obtained using straight meth injection because the added octane is much more effective at inhibiting the onset of detonation than the cooling effectiveness of the injected solution.

There is a NACA (the agency that evolved into NASA) document that I previously posted as a link in a post that provides quite a bit of technical information on this subject. I posted this probably over a year ago, but you should be able to find the post doing a search on this subject and/or on my screen name.

For my build, I had a meth system configured to inject the meth before the intake plenum (though I did not get a chance to bring it on line before I destroyed two rod bearings). This time arround, I will be injecting methanol in the intake runners. I do not want to risk methanol pooling in the intake plenum. My concern is based on Luie's (Eagletangreen's) experiences when he blew the end caps off of the Cosworth plenum. He was also using NOS at the time, but I'm being cautious.

frankie945

iTrader: (35)

Thanks for chiming in^. On a side note. Methanol injection has been around since World War II they injected it into there supercharged aircraft like the P-47, Spitfire and more. Reason was to get a faster speed and reach higher altitudes.

rcdash

iTrader: (18)

Luie did not use NOS. The theory for the end caps blowing off was a series of events (rapid throttle on/off with a high duration cam) in a setting of a quirky BOV that caused a backfire up the plenum (as I recall).

Terry the other aspect that I have not been able to find too much data on is the effect of W/M on EGTs and the interaction that has with tubular vs. log style manifolds and large vs. small turbines. My sense is that with a reduction in EGT the backpressure in the manifold, pre-turbine, will reduce and there will be less chance of reversion, thereby permitting use of higher boost levels. Have you seen any articles on this topic?

BriGuyMax

iTrader: (4)

Anyone using water/meth injection with the Cosworth intake manifold? This is the setup I will be running and I'm trying to determine what mix of water/meth I want to run.

rcdash

iTrader: (18)

Don't have the Cosworth Brian, but I know XKR ran 100% meth with the Cosworth and it held up fine. Many folks recommend 50/50 mix as it's not readily flammable at that concentration.

ttg35fort

iTrader: (2)

Hi Raj. Luie confirmed to me in PM correspondence that he had NOS running the last time he blew the endcaps off. I asked him about this before I bought my Cosworth. (This is assuming that my memory is not failing and it was not someone else who had the same issue of the end caps blowing off that I corresponded with. I don't think this is the case though, and I still beleive that it was Luie I exchanged PMs with.)

There is a lot to this question. To start, here is the Ideal gas law: pV=nRT, where p is the absolute pressure of the gas; V is the volume of the gas; n is the amount of substance of the gas, usually measured in moles; R is the gas constant (which is 8.314472 JK−1mol−1 in SI units[4]); and T is the absolute temperature. (Thanks Wikipedia) . Since temperature and pressure are directly proportional, increased temperature = increased pressure on the intake to the turbine = greater turbine rotational velocity = greater boost.

Now, remember that the pressure in the exhaust manifold is what spins the turbines of the turbochargers. Ignoring other factors, the rate at which a turbine spins is roughly proportional to the difference between the pressure on the intake side of the turbine (i.e. the exhaust manifold pressure) and the pressure on the back side of the turbine (i.e. the pressure in the downpipes/cats). The greater this pressure differential, the faster the turbine will spin. Accordingly, reducing EGT will reduce the transfer of energy to the turbine. So, although reducing EGT has a number of advantages, transfer of energy to the turbine is not one of them.

Reversion also is based on a pressure differential, this time between the pressure of the intake charge and the pressure in the exhaust manifold. You are correct that lowering the EGT will lower the exhaust pressure, and thus reduce the amount of reversion that takes place for a given level of boost. But, remember that as the exhaust temperature increases, so does the boost, which increases the pressure of the intake charge. So, in the particular case of a turbo charged motor, I don't see that there is an advantage to be gained by lowering the EGT in the exhaust manifold from the perspective of reversion. As EGT decreases, so too will the boost, and thus the pressure of the intake charge. Again, there are a number of other issues I am glossing over here.

That said, note that turbines reach a point of maximum efficiency past which the gains in boost in relation to the previously described pressure differential on the exhaust side will diminish. Here is an example from chart from Garrett that reflects turbine flow vs. pressure differnetial:

http://www.turbobygarrett.com/turbob...71R_turb_e.jpg

Once the pressure differential reaches a certain point, the exhaust flow will stop increasing, as will the boost pressure. At this point you will see the exhast manifold pressure start to increase without a corresponding pressure increase for the intake charge, and reversion will become a greater issue.

As you increase the size of the turbine, you generally decrease the back pressure, which will lower the amount of reversion. The downside of this is that the turbo will take longer to spool up.

Now, with respect to tubular vs. log style manifolds, generally speaking, the greater surface area there is, the greater amount of thermal transfer between the exhaust gas and the ambient air (in this case the engine compartment) will occur. Steps can be taken to reduce the thermal transfer, for example heat wrap and ceramic coating. Given an equivalent thermal conductivity, the exhaust manifold with the smaller surface area will retain the greatest amount of heat, and thus retain more energy with which to spin the turbines.

On the other hand, log style exhaust manifolds tend to be more restrictive, which also can reduce the energy transfer to the turbines. I guess it is a balancing act, and which style is better will depend on the application. Also, note that as you increase the exhaust manifold volume you will increase turbo lag. In other words, as you increase the size of the exhaust manifold, it takes more exhaust to build up pressure.

BriGuyMax

iTrader: (4)

Ok, good I was worried for a bit. I'll probably go with a 50/50 mix first and see what kind of results I get before I go any more aggressive.

Thanks!

r0mey

iTrader: (35)

I got the cosworth brian and 50/50 is what Im running on my new setup

Sylvan Lake V35

iTrader: (23)

I tried but the cozi pissed meth out of the end caps....my own fault the bolts backed out and I never checked them once the seal was broke I couldn't get it to seal again by tightening the bolts alone.

BriGuyMax

iTrader: (4)

Does Cosworth recommend the use of loctite on those bolts?

BriGuyMax

iTrader: (4)

rcdash

iTrader: (18)

Terry, thanks for the analysis. It's a similar exercise that I went through though not in so much detail! My theory is that once you've passed the point of maximum turbine efficiency (for whatever reason: small turbine, high EGT, restrictive manifold, or just too much boost), then water injection at that point and the resultant drop in EGT will provide a bit more headroom. I have small JWT log style manifolds and tiny GT28rs turbines and I don't believe I will gain much from pushing beyond 17-18 psi. So I was thinking of upgrading (of course!), but I thought I might give W/M injection a try before giving up on the turbos.

Like you said, it's a balance and I feel I may inherently decrease turbine efficiency by lowering EGT, unless I'm past the point where increased pre-turbine pressure is helping. So basically, I got to just try it I suppose.

ttg35fort

iTrader: (2)

Hi Raj, I'm not quite following your analysis. It's not clear to me how dropping the EGT will provide more head room. Unless your pressure ratio is above certain point where maximum turbine flow is acheived (for example about 2.25 for the green line in the below chart), lowering the EGT will serve to lower your boost (lower EGT = lower exhaust pressure = lower pressure ratio accross the turbine), which seems to be opposite of what you want.



So, instead of the effects caused by lowering the EGT, we should look at the effects on the intake charge.

One note, as you cool the intake charge you should be able to detect that with lower EGT readings, but in this context, the EGT readings are merely an indicator of what is happening to the intake charge, not a factor that effects the intake charge temperature (assuming the same level of boost). In other words, the goal is to lower the temperature of the intake charge, not to lower EGT. It just so happens that as you lower the temperature of the intake charge, the EGT generally will decrease as well.

If you have a very good intercooler, the data shows that straight water injection will provide a little more head room at staving off detonation, which will allow you to advance your timing a little and/or increase boost if your turbos are not already maxed out, but it won't buy much. A 50/50 methanol and water mix will provide more head room, and straight methanol generally will provide even more head room in our cars.

When you don't have a proper intercooler, that is when the benefits of water really come through.

Raj, in your case, I only see meth/H2O injection being a significant benefit if your turbos are NOT already maxed out. If this is the case, injection will enable you to crank up the boost a little more. If they are maxed out, injection will allow you to advance your timing a little more, but I'm not sure how much HP this will get you. Again, this assumes that you have a good intercooler, which I believe that you do.

BriGuyMax

iTrader: (4)

That's my plan. I have a standard APS twin kit and my power falls off hard above 5500rpm. I know a lot of this is because I run the stock intake manifold and stock cams, but I'm hoping to squeeze a little more out of the APS turbos and just enjoy a ~600rwhp car.

rcdash

iTrader: (18)

Terry, I think I'm not doing a great job explaining where I'm going with this. Or I could just be overlooking some key concept and you can save me the headache of a futile exercise. One thing that graph above on turbine airflow does not show is the effect of temperature on airflow. It is only pressure vs mass air. I think you can more accurately chart true airflow via a turbine map (just like a compressor map) rather than the simplified diagram that manufacturers use. Of course I think it would be difficult since few folks attempt to change EGT to improve turbine efficiency (as opposed to IAT and compressor efficiency). If I want to optimize/maintain the pressure differential across the turbine to maximize power with a fixed volume manifold, the only way I can see increasing mass airflow is by reducing temperature (EGT) since the number of moles of air is inversely related to temperature as per the ideal gas law you posted above.

You are correct that reducing temperature may decrease the pressure pre-turbine, but I think it may be possible to maintain the pressure ratio while decreasing temperature and maximize mass airflow. So the plan is to maintain the pressure ratio by increasing boost. With the decreased temperature to increase moles of air (lb/min), I'm thinking the turbine efficiency should be improved, allowing it to spin faster than it otherwise would with hotter air and the same pressure ratio.

One of the reasons this is relevant to Brian's question earlier about what mix to use is because I believe water alone is much more effective in reducing EGT vs. methanol alone. So finding some hard data would be great, but otherwise it's always possible to run different mixes and re-tune for max power and see what the car does...

I hope this is making some sense!

I'll admit that I don't know how significant a difference I can make with W/M injection if what I postulate is true anyway. I now wish I had had Sharif stick that EGT sensor in there - it's the only sensor I'm missing.


Oops, sorry got it confused for when he blew his Crawford plenum up!

https://my350z.com/forum/5332548-post3.html

ttg35fort

iTrader: (2)

It goes back to the ideal gas law: pV=nRT, where p is pressure and T is temperature. Pressure and temperature are dirctly related. In other words, if the temperature decreases by 50%, the pressure will decrease by 50%. n is constant based on the number of molecules in the air/fuel mixture that is combusted.


The map I posted is Garrett's published turbine map. It is labled "Corrected Gas Turbine Flow". If there are other styles of turbine maps, please provide a link and I'll look at it.

I think this is where some of the confusion lies. What spins the turbine is the pressure differential accross the turbine. For a given amount of air/fuel mixture that is combusted, assuming the same efficiency of combusion, the number of molecules in the exhaust will remain constant, regardless of the EGT. In other words, the value of "n" in the Ideal gas law is not going to change based on temperature. It might increase very slightly due to the additional water content if you are injecting water, but that will be relatively minimal. Accordingly, if you lower the EGT, and "n" remains roughly the same, you are going to see a net decrease in the exhaust manifold pressure.

Unless you are in the flat portion of the turbine map (e.g. above a 2.25 pressure ratio for the green line), as you decrease the pressure in the exhaust manifold, your turbine will slow, thereby slowing the compressor wheel and reducing the boost at the output of the compressor wheel.

If you are wastegating the boost supplied by the compressor wheel, you can adjust the amount of compressed air that is wastegated, and perhaps this is what you are talking about.


EDIT, you are forgetting about the net octane of the air/fuel mixture. When the intake air temperature is low, pure methanol will stave off detonation better than pure water. As the intake air temperature decreases, the effectiveness of water at allowing greater boost/timing decreases, especially in comparison to methanol. This is all laid out in the data from the NACA document I mentioned. In their testing, a 70% meth/30% water ratio worked well, but their intake air temperatures were still higher than what we see.

(Your going to make me go find the NACA document now, aren't you. ) If I dig out this document, then you owe me dinner at the next Zdays event, assuming I make it.

An EGT sensor is not bad to have, but I don't think you will achieve the results that you are looking for.

Raj, please, look at the Garrett turbine map and visit their website for further explanation. Then use the Ideal gas law and step through the analysis in a precise, logical manner. Also review the NACA document with regard to the effectiveness of water vs intake charge temperature.

Raj, if you do these things, you will find that the information I have provided is correct. I mean absolutely no offense, but if you don't apply the proper equations and the proper data to reach your conclusion, your conlusion is merely wishful thinking.

rcdash

iTrader: (18)

Thanks Terry for your analysis. I will look into it. I get what you are saying. For academic reasons, I'm just trying to think outside the box a little here. On a more practical note, if you find me a link to that document, steak is definitely on me at ZdayZ, with a few beers for good measure!

I really need to see that document in order to understand how methanol could be a better detonation suppressant than water. Regardless of AIT, the latent heat of vaporization is so much higher for water, that internal cylinder temps should be much lower with water than methanol.

In regards to your analysis above, the one thing you are assuming is that water injection will occur alone. It will be accompanied by an increase in boost pressure (via boost controller, so "n" is NOT constant). So yes, the pressure will decrease with WI but the plan is to take advantage of that inherent decrease by increasing boost to maintain the same pressure ratio. If you assume that a particular turbo is limited by turbine size (rather than compressor air flow), this analysis is feasible. For example what do you think will happen if you make your exhaust manifolds little straws? The pressure ratio will be high but you won't be moving those turbines! I'll think it through a bit more, but I think it may be worth exploring for my setup!

(But would still very much appreciate that NACA document - thanks!)

EDIT: I found these, but it doesn't speak to it exactly (interesting though):
http://www.turbotuning.net/Artikel/naca-wr-e-264.pdf
http://www.rbracing-rsr.com/downloads/naca_H2O.pdf

ttg35fort

iTrader: (2)

Hi Raj,

Remember that the pressure differential is that accross the turbine blades. If you make the exhaust manifolds like straws, the pressure on the blades will be very, very low due to the relative volume change between the exhaust manifold and the turbine chamber.

To understand this point, once again, examine pV=nRT. The product pV will be the same in the exhaust manifold as it is in the turbine chamber. i.e., p1V1 = p2V2. However, when the exhaust gas exits the manifold into the turbine chamber, the volume will rapidly expand, i.e. V2>>V1. Therefore, the pressure will rapidly drop, i.e. p2<<p2.

I'll look up that reference. I look forward to hanging out and having a few beers.

Best regards,

Terry

rcdash

iTrader: (18)

Yes, of course, you are right. Your analytical mind finds flaws quickly! But it was just an example to get an idea for what I'm talking about more generally. If you take it a step further and consider the entire pre-turbine section as restrictive, then you can see that mass airflow (always key since you need to eject what you're putting in to prevent reversion) is dependent not just on the pressure ratio as the Garrett chart would lead you to believe. It is dependent on temperature and volume as well since they all interact per the ideal gas law. Volume is usually constant (unless you have a twin scroll turbine with a SP valve on there!). For the most part, temperature is usually as well (so no point for Garrett to include it as a variable in their turbine chart like they do in their compressor chart, expressed as efficiency islands). Unless of course, you do something to alter temperature, such as WI.

Anyway at next ZdayZ we can discuss it over a beer and perhaps I'll have some hard data on hand! After a couple of beers, we'll be in perfect agreement.

ttg35fort

iTrader: (2)

Hey Raj:

I found the report, NACA report No. 812. Here is the link:

http://naca.central.cranfield.ac.uk/...report-812.pdf

The really funny part is, look at post #31 in this thread where I posted the link to the report: https://my350z.com/forum/forced-indu...ml#post6197228

Now, look at who made post #32.

We have been discussing this **** for over a year now. (It's still fun though. )

One thing to note in the report is that the tests were performed with inlet-air temperatures of 150 deg. F. and 250 deg. F. The effectiveness at straight water in comparison to the 70% meth/30% water mix diminishes as the inlet-air temperature is decreased. Compare the top of Fig. 5(a) (250 deg.) to FIG. 5(C) (150 deg.).

Note that the X-axis is in fuel-air ratio (FAR), not air-fuel ratio (AFR), so the numbers are reciprocals of AFR. E.g., a 14.3:1 AFR = .07 FAR, a 12.5:1 AFR = .08 FAR, a 11.1:1 AFR = .09 FAR, 10:1 AFR = .01 FAR, a 9:1 AFR = .11 FAR, etc.

At 250 deg. the effectiveness of water at preventing knock is lower, but nonetheless fairly close to the 70/30 meth/H2O mix. The difference at a 12.5:1 AFR (.08 FAR) is about 21%.

At 150 deg., the effectiveness of the 70/30 meth/H2O mix increases more that the effectiveness of water at preventing knock. The difference here at 12.5:1 AFR (.08 FAR) is about 30%.

If we assume that this trend will continue as the inlet-air temperature continues to decrease, the difference in effectiveness between the 70/30 mix over water will become even greater.

Unfortunately, the report does not discuss the use of straight methanol. But, understanding the materials used gives some indication of what is going on that we can use to extrapolate the results.

The heat of vaporization for water is about 40.65 kJ/mol. The heat of vaporization for methanol is about 43.5 kJ/mol. http://en.wikipedia.org/wiki/Enthalpy_of_vaporization. When these liquids vaporize during the combustion process, water will absorb about 6.5% less energy than methanol. Accordingly, water will not be as effective as methanol in cooling the air/fuel mixture during combustion. EDIT: This is based on the number of moles of liquid. Water has less than half the atomic weight of methanol, and thus has more than twice the heat of vaporization as measured per weight (which corresponds to what you said above).

The heat of vaporization alone does not account for the differences in knock prevention indicated in the report. Methanol adds additional fuel with a very high octane, whereas water does not. I believe that it is the additional octane that enables the meth/H2O mixture perform better at preventing knock. Moreover, I suspect that straight methanol may perform even better than the meth/H20 mixture.