-
1
- #61
Tek-Tips is the largest IT community on the Internet today!
Members share and learn making Tek-Tips Forums the best source of peer-reviewed technical information on the Internet!
-
Congratulations cowski on being selected by the Eng-Tips community for having the most helpful posts in the forums last week. Way to Go!
water injection to gas engines 7
- Thread starter PISTO
- Start date
- Status
kiwinjuneer
Mechanical
Not exactly water injection, but Gunnarman (sic?) claimed significant dissociation of the water with his A-21/A-55 Naptha/Water emulsion fueled SI engines, thereby reducing the amount of air required to achieve stoichiometry.
I find this a little hard to credit but any can anyone clarify (preferably someone with hands-on expereince with the technology)?
I find this a little hard to credit but any can anyone clarify (preferably someone with hands-on expereince with the technology)?
By way of introduction this is my first post over here. I am the hotrod turboice mentions in credit on his web site, but had to change my handle slightly as it was already in use here.
Turboice: In ref to your comment.
The erosion of the blades discussion is very interesting, and may apply to others differently than me. I am running plain old T04, oil cooled turbos, and have been for many years.
Seems I recall a reference I think in the NACA papers that mentions that blade erosion is not a concern provided the water mist does not exceed 10% of the mass flow. In the military fighter aircraft they injected some of the water on some configuration directly into the "eye" of the compressor.
Given the effectiveness of high pressure water cutting jets it makes perfect sense that it would be highly dependent on droplet size. If you've ever been on a motorcycle in a rain storm you have an intuitive feel for the effect.
I will look around and see if I can find that ref and double check the configuration.
Years ago we used to trickle water directly into the carburator primaries at a fast idle to blow out some of the carbon and as mentioned above when the water passed through the venturi you could see the shear forces tear it apart.
Turbododges experience makes perfect sense with hub injection.
You could maximize his technique with a small modification (may not even be necessary). Years ago I worked in a ceramics operation machining raw ceramics prior to firing. They made extensive use of carbide tooling and we sharpened it on diamond imbedded wheels. The wheels were shaped like a shallow cup with the grit on the outer face of the cylinder. To provide cooling water a very small drip of water was applied to the interior of the cup. Centrifugal force spread the water to a very thin film and it followed the front face and the was slung off as a microfine mist. I presume the same effect is taking place on turbododges compressor hub. The radial velocity of the center shaft and nut is probably slow enough you get no erosion but simply blast the water into ultra fine drops with brute force.
On researching water injection I also stumbled on another reason why methanol (or any alcohol) mixed with the water will help performance of a plain water injection system.
If you go to the major spray nozzle vendors and look through their catalogs you will see that droplet size of spray nozzles depend in part on the surface tension and viscosity of the fluid. Alcohol has a significantly lower viscosity than water and lowers the surface tension so all else being equal a spray nozzle running a Alcohol/water mix should produce a smaller mist droplet.
The spray droplet size also depends on the density of the fluid the spray is being injected into. The mist from a nozzle into a supercharged intake at 2 BAR pressure would be significantly finer than the same nozzle sprayed at ambient pressure. In effect the nozzle effeciency should go up as boost pressure increases.
Larry
Turboice: In ref to your comment.
The erosion of the blades discussion is very interesting, and may apply to others differently than me. I am running plain old T04, oil cooled turbos, and have been for many years.
Seems I recall a reference I think in the NACA papers that mentions that blade erosion is not a concern provided the water mist does not exceed 10% of the mass flow. In the military fighter aircraft they injected some of the water on some configuration directly into the "eye" of the compressor.
Given the effectiveness of high pressure water cutting jets it makes perfect sense that it would be highly dependent on droplet size. If you've ever been on a motorcycle in a rain storm you have an intuitive feel for the effect.
I will look around and see if I can find that ref and double check the configuration.
Years ago we used to trickle water directly into the carburator primaries at a fast idle to blow out some of the carbon and as mentioned above when the water passed through the venturi you could see the shear forces tear it apart.
Turbododges experience makes perfect sense with hub injection.
You could maximize his technique with a small modification (may not even be necessary). Years ago I worked in a ceramics operation machining raw ceramics prior to firing. They made extensive use of carbide tooling and we sharpened it on diamond imbedded wheels. The wheels were shaped like a shallow cup with the grit on the outer face of the cylinder. To provide cooling water a very small drip of water was applied to the interior of the cup. Centrifugal force spread the water to a very thin film and it followed the front face and the was slung off as a microfine mist. I presume the same effect is taking place on turbododges compressor hub. The radial velocity of the center shaft and nut is probably slow enough you get no erosion but simply blast the water into ultra fine drops with brute force.
On researching water injection I also stumbled on another reason why methanol (or any alcohol) mixed with the water will help performance of a plain water injection system.
If you go to the major spray nozzle vendors and look through their catalogs you will see that droplet size of spray nozzles depend in part on the surface tension and viscosity of the fluid. Alcohol has a significantly lower viscosity than water and lowers the surface tension so all else being equal a spray nozzle running a Alcohol/water mix should produce a smaller mist droplet.
The spray droplet size also depends on the density of the fluid the spray is being injected into. The mist from a nozzle into a supercharged intake at 2 BAR pressure would be significantly finer than the same nozzle sprayed at ambient pressure. In effect the nozzle effeciency should go up as boost pressure increases.
Larry
After searching around a bit I found the reference I was thinking about. It was not in the NACA reports (although they may also contain the same info)
These refs all pretty much say the same thing, that the WWII aircraft engines in many cases injected both fuel and ADI fluid into the "eye" of the supercharger compressor. The first three have some interesting additional info in them as well.
Here's a brief extract from the first.
Induction was by through a huge Bendix-Stromberg PR100B4 injection carburetor, originally used on the Pratt and Whitney R-4360 radial engines that powered planes like the B-36 bomber. The carburetor measured the air entering the engine, and provided metered fuel which went to fuel injectors directed into the eye's of the intake impellers. ADI injectors were also co-located with the fuel injectors.
A control increases maximum boost when ADI is active. The automatic boost control also incorporates an additional control which energizes ADI when a predetermined boost pressure is reached. The ADI injector is incorporated into the throttle body, ADI being directed into the eye of the supercharger. Finally, an injector incorporated in the throttle body directs fuel into the eye of the supercharger.
Turbocharging the Griffon:
Rolls-Royce carried our research into a turbocharged Griffon under a contract from the British Air-ministry in 1948. The RGT.30.SM Griffon used a GEC turbo-supercharger, and was intended for a projected Supermarine long-range flying boat and as an upgrade for the Avro Shackleton. At sea-level the engine had similar performance to the Griffon 57/58, since a wastegate was partially opened, limiting SL boost to 25 psig when used with ADI and preventing detonation. Altitude the wastegate closed however, routing all exhaust through the turbine. The interesting thing about this engine was the turbo was not staged with the supercharger, but instead the air pumped was used to cool the exhaust pipes and turbine, with the turbine also driving the supercharger through a free-wheeling (one-way) reduction drive. This engine configuration is really a compound engine -- the turbine is used to recover exhaust heat energy and deliver that power to the crankshaft. It was not used to increase the pressure ratio of the induction system.
<-------- Bendix Stromberg PR-100 Injection Carburettor
Larry
These refs all pretty much say the same thing, that the WWII aircraft engines in many cases injected both fuel and ADI fluid into the "eye" of the supercharger compressor. The first three have some interesting additional info in them as well.
Here's a brief extract from the first.
Induction was by through a huge Bendix-Stromberg PR100B4 injection carburetor, originally used on the Pratt and Whitney R-4360 radial engines that powered planes like the B-36 bomber. The carburetor measured the air entering the engine, and provided metered fuel which went to fuel injectors directed into the eye's of the intake impellers. ADI injectors were also co-located with the fuel injectors.
A control increases maximum boost when ADI is active. The automatic boost control also incorporates an additional control which energizes ADI when a predetermined boost pressure is reached. The ADI injector is incorporated into the throttle body, ADI being directed into the eye of the supercharger. Finally, an injector incorporated in the throttle body directs fuel into the eye of the supercharger.
Turbocharging the Griffon:
Rolls-Royce carried our research into a turbocharged Griffon under a contract from the British Air-ministry in 1948. The RGT.30.SM Griffon used a GEC turbo-supercharger, and was intended for a projected Supermarine long-range flying boat and as an upgrade for the Avro Shackleton. At sea-level the engine had similar performance to the Griffon 57/58, since a wastegate was partially opened, limiting SL boost to 25 psig when used with ADI and preventing detonation. Altitude the wastegate closed however, routing all exhaust through the turbine. The interesting thing about this engine was the turbo was not staged with the supercharger, but instead the air pumped was used to cool the exhaust pipes and turbine, with the turbine also driving the supercharger through a free-wheeling (one-way) reduction drive. This engine configuration is really a compound engine -- the turbine is used to recover exhaust heat energy and deliver that power to the crankshaft. It was not used to increase the pressure ratio of the induction system.
<-------- Bendix Stromberg PR-100 Injection Carburettor
Larry
I didn't read through all of these posts. But as far as I always understood, water injection's main purpose is to lower combustion temps. In doing so it allows you to advance timing and compression ratio which produces more hp. Water injection can replace EGR with a similar result. thats the simple explaination as I understand it. Molecularly, your dilluting the amount of oxygen in the air kinda like EGR does. Water is fairly stable and doesn't add to combustion temps.....just takes away... therefore giving you lower combustion temps.. I dont know if Im way off here, but its what ive been taught at an engine builders level, not an ME level.
patprimmer
New member
JRW261
Your partly right, but there is more to it.
The liquid water displaces very little oxygen, so does not really act like EGR in that regard.
The water evaporating during compression and combustion will absorb some heat.
The steam generated during combustion will produce some pressure, and thereby heat. The guys good enough on thermodynamics tell me it gives a net increase in cylinder pressure.
At elevated temps during combustion, the water dissociates into hydrogen ions and hydroxyl ions.
These hydroxyl ions catalyse the reaction of carbon monoxide and oxygen into carbon dioxide, so accelerating the final phase of the conversion of hydrocarbon fuel into carbon dioxide and water.
A lot of this is from Turboice's website. I think he has a FAQ on the subject as well.
Regards
pat
eng-tips, by professional engineers for professional engineers
Please see FAQ731-376 for tips on how to make the best use of Eng-Tips Fora.
Your partly right, but there is more to it.
The liquid water displaces very little oxygen, so does not really act like EGR in that regard.
The water evaporating during compression and combustion will absorb some heat.
The steam generated during combustion will produce some pressure, and thereby heat. The guys good enough on thermodynamics tell me it gives a net increase in cylinder pressure.
At elevated temps during combustion, the water dissociates into hydrogen ions and hydroxyl ions.
These hydroxyl ions catalyse the reaction of carbon monoxide and oxygen into carbon dioxide, so accelerating the final phase of the conversion of hydrocarbon fuel into carbon dioxide and water.
A lot of this is from Turboice's website. I think he has a FAQ on the subject as well.
Regards
pat
eng-tips, by professional engineers for professional engineers
Please see FAQ731-376 for tips on how to make the best use of Eng-Tips Fora.
In the induction section of
"As the water absorbs heat the droplet sizes will decrease and the surface area of the water droplets will increase"
quoted from
This doesn't really make sense to me. As a substance absorbs heat, its internal energy increases causing higher vibration and expansion. that supports the surface area increase. I guess what I am saying is that the sentence contradics itself, or am I miss understanding it?
does it mean, the amount of droplets will increase, thus surface area increases, but individual droplet size decreases.... i guess that makes sense, is this what is ment?
I see where my explaination faltered.... water displaces air in the induction... which it does... however the water absorbs intake air heat causing a more dense intake charge, which overcomes any displacement the water originally cause and actually increasing intake air volume.
I know understand better, thanks. Did I figure out the answer to my first question regarding droplet size and area changes?
"As the water absorbs heat the droplet sizes will decrease and the surface area of the water droplets will increase"
quoted from
This doesn't really make sense to me. As a substance absorbs heat, its internal energy increases causing higher vibration and expansion. that supports the surface area increase. I guess what I am saying is that the sentence contradics itself, or am I miss understanding it?
does it mean, the amount of droplets will increase, thus surface area increases, but individual droplet size decreases.... i guess that makes sense, is this what is ment?
I see where my explaination faltered.... water displaces air in the induction... which it does... however the water absorbs intake air heat causing a more dense intake charge, which overcomes any displacement the water originally cause and actually increasing intake air volume.
I know understand better, thanks. Did I figure out the answer to my first question regarding droplet size and area changes?
JayMaechtlen
Industrial
water vaporizes in the the cylinder. I'm not sure you can talk about surface area of a vapor.
The vapor certainly dilutes the charge, exactly as egr does.
The only issue there is reletive amounts.
The catalyzation? no idea.
jay
Jay Maechtlen
The vapor certainly dilutes the charge, exactly as egr does.
The only issue there is reletive amounts.
The catalyzation? no idea.
jay
Jay Maechtlen
patprimmer
New member
JRW261
All other factors being equal, as the water evaporates, it will form smaller drops, but I expect no more drops, therefore the surface area of the drops will decrease in absolute terms, but will increase in relative to mass of water terms. The outside factor will be shear on the drops due to motion in the inlet tract and compression and early parts of the power stroke. Also, water will collect on the walls and floor, making very large drops or puddles, so who knows the net result.
The cooling increases the density of the air, not the volume. The volume is virtually constant as it is defined by the dimension of the manifold, which only changes a few thou due to temperature changes of the metal. As you have higher density and similar volume, you have greater mass.
Water has a significantly lower evaporation rate than petrol, therefore, I would expect most of the evaporation to take place as the heat rises during the early stages of combustion, with very little occurring before the inlet valve closes.
The volume of liquid fuel and water vs volume of air is insignificant re volumetric efficiency, unless you are using nitro methane in high proportions.
JayMaechtlen
I agree pretty well with what you say, and any water vapour in the inlet tract will displace oxygen, but I feel there is very little volume in the liquid form, and only a small amount of evaporation before the inlet valve closes, then it cannot displace any oxygen.
Regards
pat
eng-tips, by professional engineers for professional engineers
Please see FAQ731-376 for tips on how to make the best use of Eng-Tips Fora.
All other factors being equal, as the water evaporates, it will form smaller drops, but I expect no more drops, therefore the surface area of the drops will decrease in absolute terms, but will increase in relative to mass of water terms. The outside factor will be shear on the drops due to motion in the inlet tract and compression and early parts of the power stroke. Also, water will collect on the walls and floor, making very large drops or puddles, so who knows the net result.
The cooling increases the density of the air, not the volume. The volume is virtually constant as it is defined by the dimension of the manifold, which only changes a few thou due to temperature changes of the metal. As you have higher density and similar volume, you have greater mass.
Water has a significantly lower evaporation rate than petrol, therefore, I would expect most of the evaporation to take place as the heat rises during the early stages of combustion, with very little occurring before the inlet valve closes.
The volume of liquid fuel and water vs volume of air is insignificant re volumetric efficiency, unless you are using nitro methane in high proportions.
JayMaechtlen
I agree pretty well with what you say, and any water vapour in the inlet tract will displace oxygen, but I feel there is very little volume in the liquid form, and only a small amount of evaporation before the inlet valve closes, then it cannot displace any oxygen.
Regards
pat
eng-tips, by professional engineers for professional engineers
Please see FAQ731-376 for tips on how to make the best use of Eng-Tips Fora.
The velocity of the induction charge makes it not only unlikely but improbable that there would be any accumulation on the surfaces worth mentioning. When you start talking about 150 g/sec of air on the low end it is unlikely that a few percent by mass of water will accumulate in the induction track. Keep in mind that water injection is not a spicket of water sprayed into the intake tract it is a couple or a few percent of the induction charge by mass or volume.
I would expect water droplets to change in size and number as they under go shear and also balance out in equilibrium temperature and humidity.
I would not be surprised if several sentences are not worded entirely correctly - something I will seek to correct in edits.
As for mass displacement here is a good link that goes over that topic -
Net/net you should not expect any "displacement" of air or oxygen from your induction charge.
Also due to other improvements that the cooling of the charge and valves brings to VE there is likely more oxygen available at spark ignition than there would otherwise be.
While water in combustion will enter and leave other phases it is water on the way in and water on the way out. Other than lower EGTs I would not expect the water to have any other effect on the energy release of combustion. The real benefit of water in combustion is that it helps complete the combustion process by improving the oxidation of CO, the complete oxidation of the CO will release more energy during combustion translating into increased torque.
Sorry that my language is not always technical as I am not an engineer by training and my recollection of physics and chemical terminology is not complete. Though I am pretty confident I have a grasp of the concepts behind the terms and that is what I try to communicate.
Ed.
I would expect water droplets to change in size and number as they under go shear and also balance out in equilibrium temperature and humidity.
I would not be surprised if several sentences are not worded entirely correctly - something I will seek to correct in edits.
As for mass displacement here is a good link that goes over that topic -
Net/net you should not expect any "displacement" of air or oxygen from your induction charge.
Also due to other improvements that the cooling of the charge and valves brings to VE there is likely more oxygen available at spark ignition than there would otherwise be.
While water in combustion will enter and leave other phases it is water on the way in and water on the way out. Other than lower EGTs I would not expect the water to have any other effect on the energy release of combustion. The real benefit of water in combustion is that it helps complete the combustion process by improving the oxidation of CO, the complete oxidation of the CO will release more energy during combustion translating into increased torque.
Sorry that my language is not always technical as I am not an engineer by training and my recollection of physics and chemical terminology is not complete. Though I am pretty confident I have a grasp of the concepts behind the terms and that is what I try to communicate.
Ed.
JayMaechtlen
Industrial
Pat:
"... and any water vapour in the inlet tract will displace oxygen, but I feel there is very little volume in the liquid form, and only a small amount of evaporation before the inlet valve closes, then it cannot displace any oxygen."
In terms of limiting the amount of air-fuel mix, that sounds right. We can probably assume that the water is mostly liquid (in small droplets) untill it gets into the cylinder and the compression stroke starts.
As the mix is compressed and the water finally vaporizes, then it will act to dilute the charge (like EGR), won't it?
regards
Jay Maechtlen
"... and any water vapour in the inlet tract will displace oxygen, but I feel there is very little volume in the liquid form, and only a small amount of evaporation before the inlet valve closes, then it cannot displace any oxygen."
In terms of limiting the amount of air-fuel mix, that sounds right. We can probably assume that the water is mostly liquid (in small droplets) untill it gets into the cylinder and the compression stroke starts.
As the mix is compressed and the water finally vaporizes, then it will act to dilute the charge (like EGR), won't it?
regards
Jay Maechtlen
patprimmer
New member
JayMaechtlen
Good point, it will dilute, but not displace. I was presumeing you were useing dilution for displacement, as this is a very common missuse of dilution.
I still don't think that the water vapor dilution has the same effect as EGR dilution, as wth water disociates and takes part in propagating some of the oxidation of the fuel, but then again, exhaust contains a fairly large portion of water. I need to think about what the carbon dipoxide in the exhaust will do to the reaction. I expect that it just gets in the way at a molecule to molecule level, and slows the reaction a little.
Exhaust gas will not undergo a phase change, where as injected water will, and the heat absorbed and the potential volume increase will both affect the temperatures and pressures.
Regards
pat
eng-tips, by professional engineers for professional engineers
Please see FAQ731-376 for tips on how to make the best use of Eng-Tips Fora.
Good point, it will dilute, but not displace. I was presumeing you were useing dilution for displacement, as this is a very common missuse of dilution.
I still don't think that the water vapor dilution has the same effect as EGR dilution, as wth water disociates and takes part in propagating some of the oxidation of the fuel, but then again, exhaust contains a fairly large portion of water. I need to think about what the carbon dipoxide in the exhaust will do to the reaction. I expect that it just gets in the way at a molecule to molecule level, and slows the reaction a little.
Exhaust gas will not undergo a phase change, where as injected water will, and the heat absorbed and the potential volume increase will both affect the temperatures and pressures.
Regards
pat
eng-tips, by professional engineers for professional engineers
Please see FAQ731-376 for tips on how to make the best use of Eng-Tips Fora.
A couple of questions:
Why is the simulation performed at atmospheric pressure as opposed to some elecated pressure? What effect would that have on it's results?
Just how optimistic is 100% vaporization? This thread makes it sound quite unlikely, yet that link is highly dependant on that.
Why is the simulation performed at atmospheric pressure as opposed to some elecated pressure? What effect would that have on it's results?
Just how optimistic is 100% vaporization? This thread makes it sound quite unlikely, yet that link is highly dependant on that.
DaveKillens
Electrical
For most applications, water is injected to cool the intake charge, and thus, aid in increasing power. Water is used to suppress detonation caused by high temperature and pressure developed within the combustion chamber (usually) when the effective compression ratio has been taken beyond the auto-ignition point by either a turbo or a supercharger.
Water, with its high latent heat content, is extremely effective for controlling not only the onset of detonation but also the production of oxides of nitrogen in the modern leanburn engines.
Water, with its high latent heat content, is extremely effective for controlling not only the onset of detonation but also the production of oxides of nitrogen in the modern leanburn engines.
Mitzibitzi
Aerospace
I've read this thread with great interest and would like to add my tuppence-worth.
My late father was a big fan of water injection on his Mk1 Honda Prelude track cars. One of these ran a turbocharged 2.2 Accord engine (very expensive to build and damned unreliable) and the other a supercharged 2L that he got from the breakers and bolted an Eaton Blower straight on without much in the way of anything else. (Much more fun to drive than the turbo, pulled like a train from idle)
Both were fitted with modified Edelbrock Vario water systems, though I can't specifically remember what we did to them at this point.
The turbo knocked like hell if the water tank ever ran dry, leading us to believe that we had a touch too much compression for the boost we were using, but was spot on with (about) 1.8% water/methanol.
The supercharged one didn't seem to care knock wise, but definitely pulled a little more strongly with the water on.
I've also had a 2.3l Saab 9000 Turbo (someone set fire to it when I was working away!!!) which is controlled by an Automatic Performance Computer - APC, that limits the boost with reference to a knock sensor (and other stuff.) I had a Spearco setup on this and can definitely attest that the electronics would give you 1-2 Ibs more boost with the water on than without.
Although I agree with Patprimmer that the real benefit of water injection is to make up for lack of fuel octane and excessive compression, based on my experience with the turbos, I do wonder whether there might not be some other factor at work based on the supercharged engine. I'm building a hotrod at present with a Supercharged Mitsubishi 4G63 2.0, so I intend to slap on a Spearco or similar and check the effect on that.
My late father was a big fan of water injection on his Mk1 Honda Prelude track cars. One of these ran a turbocharged 2.2 Accord engine (very expensive to build and damned unreliable) and the other a supercharged 2L that he got from the breakers and bolted an Eaton Blower straight on without much in the way of anything else. (Much more fun to drive than the turbo, pulled like a train from idle)
Both were fitted with modified Edelbrock Vario water systems, though I can't specifically remember what we did to them at this point.
The turbo knocked like hell if the water tank ever ran dry, leading us to believe that we had a touch too much compression for the boost we were using, but was spot on with (about) 1.8% water/methanol.
The supercharged one didn't seem to care knock wise, but definitely pulled a little more strongly with the water on.
I've also had a 2.3l Saab 9000 Turbo (someone set fire to it when I was working away!!!) which is controlled by an Automatic Performance Computer - APC, that limits the boost with reference to a knock sensor (and other stuff.) I had a Spearco setup on this and can definitely attest that the electronics would give you 1-2 Ibs more boost with the water on than without.
Although I agree with Patprimmer that the real benefit of water injection is to make up for lack of fuel octane and excessive compression, based on my experience with the turbos, I do wonder whether there might not be some other factor at work based on the supercharged engine. I'm building a hotrod at present with a Supercharged Mitsubishi 4G63 2.0, so I intend to slap on a Spearco or similar and check the effect on that.
- Status
