Engine Water Injection Information – Scientific Research Sources
4 min read
This article compiles together some of the best sources on engine water injection information and gives my key takeaways. These articles cover things like where to inject the water, how much water to inject, the change in power and more.
Water Injection – Key Takeaways
My understanding of the articles is as follows.
- Water injection can be an excellent way to reduce exhaust emissions including NOx & HC
- Fuel consumption can be reduced by 10-15% when using water injection at higher engine loads.
- Water injection lowers the lambda/AFR, it makes the air fuel mixture read richer. Not a problem on a car using a wideband lambda sensor but tuning may be required on narrowband equipped cars
- Water injection can massively reduce the tendency of an engine to knock.
- Water injection lowers EGT temperatures and combustion chamber temps in general.
- Water injection can lead to a reduction in power if the engine ECU is not re-tuned.
- Depending on the fuel/water mix, ignition timing can be advanced more and advancing ignition timing can overcome the loss in power of injecting water and lead to an engine making more power with the same fuel.
- Water can be mixed with methanol.
- Pre-turbo water injection makes the turbo behave like a larger turbo. It moves the compressor efficiency islands to the right.
- The water:fuel ratio needs to be chosen. Generally between 10 and 25% water, seems to be a common range.
- Port injection appears to be the best way of introducing fuel into the cylinder.
- If using port injection and generally injecting water anywhere where the water can pool in the combustion chamber, it is essential that the components being used are 100% reliable. Hydrolocking will destroy an engine immediately.
- There should be some indication when the system is working properly and not working properly.
- Automated fail safe is also preferable. If engine is tuned to run with water injection then I would say an automatic fail safe interfacing with the engine ECU to pull timing when there is a problem is essential.
Water Injection – Information Sources.
Those are my key takeaways, here are some more sources of information
Aquamist are probably the most respected manufacturers of water injection systems, a key part of this is their fast acting valve and fail-safe output. This forum post on their website goes in-depth on pre-turbo water spraying.
Eng-tips have a post here which deals with the effect of increasing the humidity of air entering the engine.. Not exactly water injection but interesting nonetheless.
AutoSpeed has an article highlighting research projects here & here.
This article deals with, among other things, effects of different injection locations and the effect of injecting water of different temperature. Same website has a test on pre-turbo water injection.
Water Injection Information – Summary
What struck me about my journey into researching water injection so far is how little is known about it’s effects. A lot of the key/famous research was done in the 1920s by Ricardo and in the 40s by NACA. After that, scientific research is hard to come by, especially on engines which are used in road cars.
Sure people understand it’s benefits but when it comes to hard numbers such as injecting X amount of water into intake mass of X ignition timing could be advanced by D and power increased to E aren’t there.
Reading through the forum post above on the Aquamist website, even for Aquamist themselves water injection seems a bit of a dark art, there are still plenty of ways of using water injection that have not been properly analysed.
For example, there are many really popular turbos out there, for example a Garrett GT28RS, but hard data on the effects of using water injection on this turbo are hard to come by.
What does this all mean?
Yes it’s understood that water injection is a massive boon for the internal combustion engine on all levels but there seems to be a big vacuum when it comes to the specifics. It seems at the moment our knowledge stops at “research has shown it makes these things better, tuners are getting big increases in power using water, that pure water to a 50/50 mix water and methanol mix is good and that every engine responds differently”.
I guess what I’m saying is it would be nice to have some ball park numbers so people know what they are getting into. At the moment, water injection looks amazing but perhaps the reason everyone is not running it is that they don’t how much better it will be and if the expense is worth it.
