Water Injection for High Compression Naturally Aspirated
 

Hi guys,

A fellow working for me has a turbo VW Golf with an oversized turbo/intercooler and has been using water injection to control detonation for some time. He is totally convinced that it is THE answer...his system runs off of a boost pressure sensor and then sprays into the plenum. He uses stock windshield washer fluid (cheapest he can get ....basically water and methly alcohol) and a very simple atomization system. He takes the fluid from his windshield washer reservoir, which has a low level system and has set up his EMS to retard timing and turn on an "engine check" light when he runs out of water. He is going through about 4 litres of water per 3 weeks, putting on the average 2000 km per month and considers it to be a trivial expense and hassle. He took an endoscope to his intake system a while back and said his valves and combustion chambers looked brand new after something like 40k kilometers and driving it like he stole it.

Just wondering, considering that our cars have a detonation "issue", we have a huge honking washer reservoir...and in my case will have an EMS or similar....what about using that on the 911 normally aspirated? I figure making it activate on high load (>50% throttle) and off the rest of the time ought to cover the need....and we would not have any problems with detonation...right?

Any insight as to why this might be a bad idea? There is lots of stuff on the web about this...sounds perfect, other than the need to keep filling the water of course.

Thoughts?

Dennis

I don't think you would get the benefit with N/A cars, but water injection definitely works for turbo cars. I know many people with it installed on SAAB turbos. Huge difference in power, especially on very hot days when the intercooler is not cooling the air down as much. Turbo cars are breathing in very hot air. Even after intercooling you are at least 100 degrees over ambiant air temp. Water injection can bring the temps back down to almost ambiant temps.

N/A engines operate at ambiant air temps, so the benefit isn't as great. You would gain some horsepower by lowering intake temp below ambiant, but there are better ways to get more horsepower out of an n/a engine. Exhaust, fuel chips, cams, twin plugs, higher compression, etc...

The one nice thing about water injection is that it eliminates carbon deposits on the valves and pistons.

Dennis,

I don't see any reason not to try it. What would really be cool would be a knock sensor that would listen for the characteristic 5Khz 'ping' of detonation and then give it a little shpritz. You could even use a set of manifolds that provide for a second set of injectors. That said, to maximize the benefit would require a completely clean sheet of paper.

I read a good article by Patrick Bedard in CD that suggested that the true promise of Ethanol is its antiknock-- one could have a very small displacement engine that produced sufficient power for cruise at a 75% throttle setting, and then accelerate the vehicle from rest, pass, etc., using an electronically controlled wastegate turbo with direct cylinder injection of ethanol, taking advantage of its antiknock. So at 65 MPH, you'd be running zero boost and sipping ethanol, then when you put your foot down the wastegate closes and the boost goes to 4 bar and power goes up dramatically. . .

It works to control detonation in all types of combustion engines, just the effect and benefit is more noticeable in a turbocharged setup.

Rules of thumb in an NA setup is to NOT inject water over 8% of your total fuel used(turbocharged about 12-15%) IE if you're burning 800cc/min gas total, max water injection should be 8% of that or less. That's the limit on an NA setup where you'll start to see flame quench and torque drastically drop off.

Other important aspect is just to make sure you get a very fine atomization. Use proper nozzle(s) and high pressure pump(~120-150psi). I prefer to use two smaller nozzles vs one large unit to help with the atomization. Multiport is the best setup for a big turbo setup running lots of timing or higher compression but overly complicated really for an NA setup.

From what I can read the atomization will be the trick. Standard fuel injectors do not like water too much, so that pretty well rules them out. There are high pressure pumps on the market that have the correct volumes (generally too big is the problem....) but how to atomize properly.

I have not yet done a search, but I suspect that with 100+ psi in pressure, it would not take much of a nozzle to atomize a fair bit.

In terms of the proportion of water, there are a couple of calculators on the web, seem to be pretty good and the EMS ought to be able to manage the rate. I was thinking of a fairly simple algorithm, something like >3000 rpm and >75% throttle opening and then spray in a fixed volume of water (just turn the pump on).

In terms of where to inject, I am using the TWM throttle bodies with the stock Weber air cleaner set up, one thought is to just do a little fabricating and set up a nozzle per throat a la accelerator pump jets and off we go.

Just had a random thought for CIS....how about just putting water through the cold start manifold in the air cleaner and get some detonation control on a CIS car. Most would not care, but I have the 9.8 to 1 compression pistons in my 3.2 (Kremer engine built on a 3.0 case).

Dennis

Surflo pumps are the typical brand of rotary style DC motor pumps that 99% of aftermarket kits come setup with. Aquamist uses proprietary piston pumps. Shurflo pumps are cheap cost wise and very effective. If using straight water, you need about 120psi to get the water to mist right in most cases, which those pumps can easily handle(usually preset at 150psi). If you run a water/alky or water meth mix the pressure can be a little lower(100psi or so) and still get good atomization.

The simple setups that are not progressive can be a little trick to tune sometimes but it's all based on the setup. Often if the system just starts to dump injection fluid at X rpm and Y throttle position you can get a flatspot in torque but as you stated with most EMS's you can tune this out with some patience and trial/error in jet sizing.

And for multi nozzle setups I use small nitrous fuel jets. Perfect size and work the nuts. Determine your flow requirement and divide by 6. Factor in a little bigger jet size for running straight water. Cheap too and definitely the way to go with TWM's as it makes the fluid distribution near perfect. Run the pump into a distribution manifold(block of aluminum works good) and branch off the with equal length line to the jet's.

I set the jets up into 1/8NPT brass fittings, brazing the nitrous jets(usually brass) into the back side of the NPT thread section. The other end is either AN style connection for braided line or a compression fitting style end for soft airbrake or hardline back to the manifold.

Makes for a nice, clean, and functional setup when all said and done.

I used a Holley kit back in the day, on an El Camino my cousin and I built.

It was a Chev 350 with 11.5:1 CR.

It worked.

Please don't ask me to rememmber all the details. I do remmember it had a throttle swithch. When it reached something like 85% of full throttle it would kick in.

(Plus the used it on a P-51 Mustang........oh just the most beautiful fighter ever built.)

BLASPHEMY!!!!! P-38 Lightning :D P-51D nice too. Fun

Focke-Wulf FW-190 A-8 used the H20/methanol injection and turbocharging on the BMW 801D-2 Sternmotor- 1,700 PS Hochleistung!:D
http://forums.pelicanparts.com/uploa...1214353466.jpg
http://forums.pelicanparts.com/uploa...1214353481.jpg

Flieger: What finally make the !90A run at altitude was the nitrous injection that they developed, and even then it couldn't keep up with a 47 above 20,000 feet for long. Most people don't know that the BMW radial was a Pratt-Whitney 2800 made under license before the war. The 190D however, was a different cup of tea, I truly think it was probably the best piston engine fighter of the war To get back to the subject, water injection was used by F-6's, P-47's, F-8's, F-4U's, to prevent detonation under war emergency high boost conditions. The benefit in NA applications is minimal and not worth the effort unless substandard fuel has to be used. In climates where high ambient temperature are the norm, water alcohol injection is much more effective than air to air intercoolers. Years ago I was using water-methanol injection on a 1977 930 in Phoenix Arizona to much better effect than a air to air intercooler. I did have trouble with compressor impeller degradation due to the water droplets in the injection mist after 10,000 miles.

aws

You are right about the N2O. The FW190 had to stick low to out fly a P-47 but could perform better at altitude than the Spitfire V.

I know that the engine in the prototype FW190 was a licenced development of the P+W, as was the BMW 801C, IIRC.

I was under the understanding that the BMW 801D was a new design, though it is always difficult to tell how much knowledge is carried over.

It is interesting that Focke-Wulf tried the low drag fairing for the magnesium cooling fan for the engine. They reverted to the P+W fairing for production for ease of maintainance but kept the fan. Much like a 911:)

I think N2O is supposed to lower engine temperatures in a similar way to alcohol fuel. As long as not too much more gasoline is burned, there is more efficient combustion due to the better oxidation. The injection of the N2O going from liquid to gas also sucks out heat from the intake charge.

Go with the N2O, not the methanol/water;)

Oh for the love of the forked devils! P51 is nice, but I can't count the number of P38's I built as a kid...

-Michael

I like the Doppel-Schwanz Teufel but the P-51 is superb:)
The snarl of that engine as the aircraft screams by is unique and thrilling.
http://forums.pelicanparts.com/uploa...1214540560.jpg
http://forums.pelicanparts.com/uploa...1214540576.jpg

On a turbo, even better is a 50/50 mix of distilled water and methanol, as water alone will decrease the btu output, while methanol will cool in addition to adding to the burn. About 15 years ago, I had a SAAB 16v turbo that could run a max of 20 lbs of boost on pure fuel enrichment alone before the knock sensing APC system would pull back the boost. I then piggybacked 4 Bosch fuel injectors behind the stock injectors which sync'd with the factory injectors after 10 psi boost, which allowed me to dial in over 35 psi boost without any detonation ... but then I kept shearing off all the teeth in third from all the torque...

I think it was Klaus of Innovate Motorsports who told me of his experiments with running water injection in a very high compression NA motor. He said it worked very effectively and it did not need much water to make it very effective. He also told me about one of their clients that was building extremely high compression NA motors that used water injection which effectively gave the motor high torque output like a diesel, with outstanding fuel efficiency. Remember that a diesel motor produces its high low rpm torque because of its high compression. Also, the high compression significantly increases the effective compression of the motor when the engine is under light loads such as cruising, which makes a big difference in fuel efficiency - again like a diesel.

Working out the details of a system will be challenge. To overcome intake manifold boost pressure, you need a high pressure pump. Tap water will clog up a Bosch fuel pump or injector within minutes. I had used a Flojet diaphram pump, but you need to be aware of the rise and response time to full pressure. A Volvo tuner I knew back in the day, that used a baffle tank to contain the water, which he kept in the trunk, and pressurized to about 100 psi, and activated the injector with a Hobbs pressure switch.