How fast can water absorb heat from combustion gas to produce steam 2

[red123]

I have been working for many years on a true rotary engine project that will use injected water at 1500-psi and 263-F to cool super-hot combustion gas down to an intermediate temperature level of 1400-F. (combustion at all operating speeds is stoicheometric, approaching 4000-F). This commingled combustion gas and steam then goes through isentropic expansion until it exhausts from the cylinder at only 19.6-psi and 312-F. The engine is a 440 cubic inch displacement single cycle machine, that produces power on every revolution due to an innovative and proprietary mechanism that constantly exhausts the products of combustion from each previous cycle without interferring with the power being developed by the current cycle. The machine has a 40 inch long circular cylinder; 10 linier inches is dedicated to the combustion process and the remaining cylinder length is dedicated to expansion of the commingled gas/steam volume.

Although earlier calculations indicate that this process will have enough time to be completed, thereby cooling the circular cylinder metallurgy to a level that can be managed, I would like additional opinions about whether water sprayed through multiple ports directly into the combustion chamber immediately after combustion is completed, will have enough time to fully vaporize and absorb the necessary heat from the combustion gas volume.

Since developing a prototype engine is a major financial effort for which I must be fully responsible, I cannot proceed further until I am totally convinced that the cylinder cooling process is a sound means for protecting the metallurgy. I would like the help of engineers with expertise in the area of direct contact heat tranfer to express their candid opinion, citing their actual experience with similar activities, or reference materials to support their (agreement) that the process is sound, or their (dissagreement) that the process is not sound.

Design criteria: The engine operates at 1800-rpm electronically governed to plus/minus 1-rpm at any operating speed.
Fuel flow (87-octane gasoline)is 46.55/lb/hr. Combustion gas flow equals .2161/lb/sec., or (.0072/lb/rpm). Injected water vapour flow is .1258/lb/sec., or (.0052/lb/rpm).
Combustion pressure at the time of water injection is approx. 640-psi.

 

[IRstuff]

Cold water impinging on hot metal does not sound like a good idea to me.

TTFN

FAQ731-376

 

[25362]

Have you considered having some kind of thermal (ceramic) insulation? As suggested, for example, in

http://www.zircarceramics.com/featured_articles/howlowmass.htm

 

[hacksaw]

the short answer is no

you can calculate the evap rate given the droplet size, estimate the transport velocity of the combustion gases, and arrive at the mixing time required...

when you inject liquids into commercial furnaces, you really need large volume chambers, 10" is not enough

my two cents

 

[danberry]

Injecting water in an internal combustion engine is not new. It is done with some aircraft piston engines to lower flame temperatures and allows a power boost. Water can be injected in gas turbines as well for power boost or to reduce emissions.

Its best to inject water at the compression stage but injecting in the combustion chamber is also possible. I don't think injecting in the expansion stage is the best approach.

Water injection has a negative effect on efficiency because all the latent heat of evaporation is lost in the exhaust.

Very high quality water must be used and this is a problem with non-stationary machines because you need to carry demineralized water in addition to the fuel. Also, the water is not free.

As far as heat transfer goes, with sufficient atomization (no droplets here), the effective heat transfer surface is very high and, with high temperature differences and highly turbulent conditions, the water will vaporize almost instantly.

Can water be injected? sure
Do you want to inject water? not so sure.

 

[red123]

Hacksaw,

can you refer me to a particular reference material that provides formulas for doing the calculations you cited in your response.

Your observations are much appreciated, however, 10 linier inches (110 cubic inches) of the cylinder is dedicated strictly to the stoichemetric combustion process. the cylinder volume that is avaible to convert the 4000-F combustion gas into 1400-F commingled combustion gas and steam is actually 28 linier inches (308 cubic inches). The reference 1400-F is merely a target calculation for establishing the amount of water needed to bring the combustion gas temperature down to 1400-F. Since the commingled gas/water vapour will also be going through a constant expansion process as it absorbs the combustion gas
heat, the reference 1400-F is merely a transient number.

The major problem in arriving at a meaningful calculation is that as heat is absorbed by the injected water spray the temperature gradient becomes less and less, not only due to the ever reducing heat transfer rate, but also due to isentropic expansion that is constantly reducing the overall cylinder pressure and temperature. Perhaps you can direct me to a computer program that can do an exhaustive iteration analysis of the gas and water vapour as the temperature and pressure goes through this process of constant change.

Again, I do appreciate your views, as well as the views of anyone else that can offer help in this matter.

 

[red123]

danberry,

Your understanding of engine dynamics is very refreshing.

The very reason I inject water spray after the stoichiometric combustion process is to extract the very highest combustion efficiency possible. In this case the overall combustion/steam process is 45 percent efficient. By injecting water after the combustion process, the much higher heat level produced by the stoichiometric combustion process yields a higher pressure steam which collectively produces almost the same level of efficiency as straight stoichiometric combustion. The problem with stoichiometric combustion is its inherently unmanageable high temperature
level. By injecting water spray the temperature can be almost instantly brought down to a non-destructive level, and in so doing the extreme temparature level can be used to generate a higher steam pressure to develop greater engine torque output.

It is reasonable to assume that the injected water spray will instantly absorb the very high combustion heat and flash into steam, but, as we all know as Engineers, things do not always act or re-act as reason would indicate.

The cost of proceeding to a working prototype stage is just too expensive to leave anything to doubt, and since I cannot at this point act with a high degree of certainty that steam cooling will be as effective as my calculations indicate, I am asking the entire Engineering community to assist me in confirming or rejecting my proposed steam cooling process.

Water used for the injection process along with the water produced by the combustion process is condensed and re-used. In almost every application no make-up water will be necessary.

 

[Compositepro]

I'm not sure about this but 4000F sounds like a flame temperature you would get with pure oxygen rather than air.

 

[Bribyk]

Are you treating the water before re-injecting it? Your water is likely to increase in acidity as you run the engine. I have my doubts about not requiring makeup water or treatment...

 

[red123]

Bribyk

You are correct in your assessment that the water will probably need to be treated. It will also require filtering and periodic replacement. To what extent these requirements will need carrying out can only be guessed at until a prototype is in operation.

Since the combustion process is critically controlled for stoichiometric operation at all levels of power output, very little noxious elements are produced by the process. Because of the high temperature generated by the process the metallurgy will need to be an almost exotic grade of stainless steel, so contaminates such as metal oxidation that would otherwise be dissolved into the condensed recycle water would not be a major problem.

 

[chicopee]

You may want to take a look at direct fired steam generators primarily used by cast concrete plant. The output of such generator is a combination of saturated steam and the product of combustion which is used to cure cast concrete products overnight.

 

[hacksaw]

red123

the best source is for recovery boilers where they inject water or other fluids to keep the temperature down, it is tricky business anything but ideal

the paper industry is a good starting point (TAPPI)

 

[EdDanzer]

Having built a proof of concept 1 cylinder linear engine to do some of what you are attempting to simulate I understand the concerns.

Matlab and Simulink or Maple and Maplesim are the type software to do the simulations with if you can write the math and learn the geek speak to complete the task. $5000.00 is the entry level for these. It cost less than that to build the proof of concept and provided a reason and insight into how complex the simulation will be and what areas to focus the simulations on. Currently I have Matlab, Simulink and additional tools but need to test some components to have values for the blocks in the simulations. The testing of these components must take environmental temperatures into account.

The problem you will face is having correct values for simulation, Garbage in = garbage out. The combustion temperature, water droplet size, incoming air temperature, incoming water temperature, heat content of the fuel are just a few of the variables that could to corrupt the simulation results. Some of these values will require experimentation to obtain.

Ed Danzer

http://www.danzcoinc.com

http://www.dehyds.com