Research Experimental 8
01-04-2012, 08:10 PM
#76
I chose an RX8 for this because I have an RX8 and was bored. Actually I had the pump and controller from a previous attempt at water injection which turned out to be unnecessary.
Update: Did annother data run tonight, I'll save you the useless data. MPG dropped by .3 when steam was injecting, but ambient temperature dropped by about 3 degrees between the run without steam and the one with steam. Not detrimental to fuel economy, but no benifits either. When it's light out I also need to make sure I haven't blown off a hose. I am not seeing the changes I saw before with coolant temperatures, but the thermostat was barely cracked most of the drive. I'd love to see what this does to coolant temps in warmer weather, but that's at least a week away. Need to check again with steam injected into the upper intake manifold to see if it's a heat transfer time issue. If that doesn't work I'm gonna write this one down to bad setup and observational bias.
Update: Did annother data run tonight, I'll save you the useless data. MPG dropped by .3 when steam was injecting, but ambient temperature dropped by about 3 degrees between the run without steam and the one with steam. Not detrimental to fuel economy, but no benifits either. When it's light out I also need to make sure I haven't blown off a hose. I am not seeing the changes I saw before with coolant temperatures, but the thermostat was barely cracked most of the drive. I'd love to see what this does to coolant temps in warmer weather, but that's at least a week away. Need to check again with steam injected into the upper intake manifold to see if it's a heat transfer time issue. If that doesn't work I'm gonna write this one down to bad setup and observational bias.
01-05-2012, 05:35 AM
#77
If there is determined to be a sweet spot when it comes to steam temperature then there needs to be a way to control the amount of heat going into the system and not only a way to meter the amount of steam going into the engine. You need both. I'm not sure exactly how I feel in regards to this experiment but I do know 2 things for sure. One is that you need to be able to control the amount of heat going into the system. The other thing I am sure of is that it's a good thing when people try to learn by doing things that others don't. I applaud that even if it fails.
Harlan, thank you for trying this, regardless of the outcome.
01-05-2012, 09:26 AM
#78
Registered
I'm not criticizing him at all. I'm just throwing out concerns and ideas. I actually have a huge love for steam engines since I'm a very big train nut. It's my main hobby. Steam has always fascinated me.
01-05-2012, 02:04 PM
#79
Yet more data, and this time it shows something! Turns out my pump was losing suction, so who knows how much previous data out the window. That said, here are the new results:
MPG Steam (obd2 instant averaged): 30.6
MPG Without steam: 31.1
Avg Coolant temperature Steam: 189.6
Avg Coolant temperature without steam: 195.8
Avg EGT Steam: 1602
Avg EGT without steam: 1600
What does this all mean? Not sure. Steam is definitely effective at cooling the engine. It does not appear to be interfering with combustion either. MPG and EGT are easily within margin of error. I guess it means I should move injection back to the UIM and test again.
The injection controller definitely makes readings wonky when it is plugged in/unplugged due to disruption of the MAF voltage, so I have kept it plugged in for all readings, also I am physically isolating water injection when not in use.
MPG Steam (obd2 instant averaged): 30.6
MPG Without steam: 31.1
Avg Coolant temperature Steam: 189.6
Avg Coolant temperature without steam: 195.8
Avg EGT Steam: 1602
Avg EGT without steam: 1600
What does this all mean? Not sure. Steam is definitely effective at cooling the engine. It does not appear to be interfering with combustion either. MPG and EGT are easily within margin of error. I guess it means I should move injection back to the UIM and test again.
The injection controller definitely makes readings wonky when it is plugged in/unplugged due to disruption of the MAF voltage, so I have kept it plugged in for all readings, also I am physically isolating water injection when not in use.
01-05-2012, 02:50 PM
#80
Registered
I'm curious, why do you feel that the steam is saturated based on the vacuum inside the engine? The temperature and pressure of the steam leaving the steam pipe would be of more concern to me as it's temperature and pressure could be very high within the pipe. In a running engine you just aren't giving it much time to cool down to a saturated state once it leaves that pipe and enters the engine. It is also greatly expanding as it enters the engine.
Superheated steam locomotives were vastly more powerful and efficient than standard non superheated saturated steam locomotives but remember steam was being injected into a cylinder at it's smallest point and moving it to it's largest point, not the other way around. Even with the cutoff set very low which only allowed steam to be injected over a very short time rather than the entire stroke, the expansion of steam once it got into the cylinder was still so great that it made power.
I'd be curious to see conventional water injection used instead where stays in liquid form and turns to steam and expands after combustion where the chamber is getting larger. A downside of course may be that all of the gasoline doesn't burn off as well but I suspect this could happen if steam was injected as well. Either way to experiments are interesting and everything is a mere hypothesis until test results verify the idea.
Superheated steam locomotives were vastly more powerful and efficient than standard non superheated saturated steam locomotives but remember steam was being injected into a cylinder at it's smallest point and moving it to it's largest point, not the other way around. Even with the cutoff set very low which only allowed steam to be injected over a very short time rather than the entire stroke, the expansion of steam once it got into the cylinder was still so great that it made power.
I'd be curious to see conventional water injection used instead where stays in liquid form and turns to steam and expands after combustion where the chamber is getting larger. A downside of course may be that all of the gasoline doesn't burn off as well but I suspect this could happen if steam was injected as well. Either way to experiments are interesting and everything is a mere hypothesis until test results verify the idea.
01-06-2012, 09:17 AM
#81
Wanted two--thanks --after over 8K posts there is one you agree with me on--lol. jk man.
Harlan--need to point out one thing on your math --and it may not matter---but remember that we dont run straight gas anymore. Ethanol is there ( and in varied amounts)--so the stoimethric(sic) should be around 14.1 rather than 14.7?
If you are getting 29mpg---thats pretty good
How do the sparkplugs look?
You may need to reduce the sparkplug gap a little?
RG--proven that w/m injection without tuning for it does cause a loss of power--so this amount of steam is probably doing the same?
Now if you can tune for the steam injection (leaning out the a/f's a little, advancing timing a tad) then the mpg should show some increase.
Curous about wot with this. Are your grams per sec affected? Fuel trims?
Harlan--need to point out one thing on your math --and it may not matter---but remember that we dont run straight gas anymore. Ethanol is there ( and in varied amounts)--so the stoimethric(sic) should be around 14.1 rather than 14.7?
If you are getting 29mpg---thats pretty good
How do the sparkplugs look?
You may need to reduce the sparkplug gap a little?
RG--proven that w/m injection without tuning for it does cause a loss of power--so this amount of steam is probably doing the same?
Now if you can tune for the steam injection (leaning out the a/f's a little, advancing timing a tad) then the mpg should show some increase.
Curous about wot with this. Are your grams per sec affected? Fuel trims?
01-07-2012, 08:26 PM
#85
I'm curious, why do you feel that the steam is saturated based on the vacuum inside the engine?
I'm assuming saturated steam because the contact area with the exhaust manifold shouldn't be big enough to transfer the heat required to dry the steam.
The temperature and pressure of the steam leaving the steam pipe would be of more concern to me as it's temperature and pressure could be very high within the pipe. In a running engine you just aren't giving it much time to cool down to a saturated state once it leaves that pipe and enters the engine. It is also greatly expanding as it enters the engine.
That would be true if there was a nozzle or other form of metering device, with no restriction pressure should be almost the same throughout. Temperature is again limited by pressure until you have superheat.
Superheated steam locomotives were vastly more powerful and efficient than standard non superheated saturated steam locomotives but remember steam was being injected into a cylinder at it's smallest point and moving it to it's largest point, not the other way around. Even with the cutoff set very low which only allowed steam to be injected over a very short time rather than the entire stroke, the expansion of steam once it got into the cylinder was still so great that it made power.
Very true, but the steam I'm injecting should mostly condense as it hits the colder air. Then when it hits it's smallest point the combustion should then boil it off again.
I'd be curious to see conventional water injection used instead where stays in liquid form and turns to steam and expands after combustion where the chamber is getting larger.
A lot of injected water does just that. It goes on to cool combustion chamber and acts as an internal coolant.
I'm assuming saturated steam because the contact area with the exhaust manifold shouldn't be big enough to transfer the heat required to dry the steam.
The temperature and pressure of the steam leaving the steam pipe would be of more concern to me as it's temperature and pressure could be very high within the pipe. In a running engine you just aren't giving it much time to cool down to a saturated state once it leaves that pipe and enters the engine. It is also greatly expanding as it enters the engine.
That would be true if there was a nozzle or other form of metering device, with no restriction pressure should be almost the same throughout. Temperature is again limited by pressure until you have superheat.
Superheated steam locomotives were vastly more powerful and efficient than standard non superheated saturated steam locomotives but remember steam was being injected into a cylinder at it's smallest point and moving it to it's largest point, not the other way around. Even with the cutoff set very low which only allowed steam to be injected over a very short time rather than the entire stroke, the expansion of steam once it got into the cylinder was still so great that it made power.
Very true, but the steam I'm injecting should mostly condense as it hits the colder air. Then when it hits it's smallest point the combustion should then boil it off again.
I'd be curious to see conventional water injection used instead where stays in liquid form and turns to steam and expands after combustion where the chamber is getting larger.
A lot of injected water does just that. It goes on to cool combustion chamber and acts as an internal coolant.
If you are getting 29mpg---thats pretty good
29 indicated 19-20 actual. But still good for trending. I have a trip MPG indicator that is darn close, but does not instantly change.
How do the sparkplugs look?
Historically black and nasty. No major changes yet. Need to pull em again.
29 indicated 19-20 actual. But still good for trending. I have a trip MPG indicator that is darn close, but does not instantly change.
How do the sparkplugs look?
Historically black and nasty. No major changes yet. Need to pull em again.
01-09-2012, 01:24 PM
#87
It would seem to me that superheated steam would need to be directly injected, close to or at top dead center to avoid creating a resistance and detraction from power generation; but saturated steam would be more flexible, albeit less powerful.
Taking this to extreme, I was just spitballing and wondered if the heat exchanger could be used in combination with the catalyst to allow the catalyst to be able to shed head faster and thereby allowing it to be closer to the exhaust ports for faster light-off. That seems like it would be a good heat source for a superheating boiler, while providing an advantage in catalyst life and light-off time (by not allowing water flow until catalyst temperature reached a desired threshold and then being used to curtail excessive heat).
Taking this to extreme, I was just spitballing and wondered if the heat exchanger could be used in combination with the catalyst to allow the catalyst to be able to shed head faster and thereby allowing it to be closer to the exhaust ports for faster light-off. That seems like it would be a good heat source for a superheating boiler, while providing an advantage in catalyst life and light-off time (by not allowing water flow until catalyst temperature reached a desired threshold and then being used to curtail excessive heat).
01-10-2012, 02:25 PM
#88
Well, I tried to use a portable steamer to get results, but the results were not good. Engine Vac was down, running rough. Either too much steam or not enough at any given time. Too much condensation in the lines causing too much water injected into the engine. Engine kept stumbling efficiency went down vac went down. But it did help keep the engine cool. Starting to think I may want some superheat on the steam just to prevent it from condensing before it hits the engine. I think all the problems were related to water getting flung on the plugs, but I'm not giving up and my engine is probably clean as a whistle.
I've got some 3/16" brake line on order along with some 1/4" Gonna stagger the sizes. Use 20' of 1/4 to preheat and 20' of 3/16 to make steam. Should give me much more steam of a reasonable quality compared to <10' of 1/4 copper. Then the 3/16 is going to go into the maintenance ports with a Tee and two 3/16" compression fittings. May insulate may not. Definitely gonna put a thermocouple on the line as too much superheat is the same as an unmonitored increase in intake air temp, which could cause detonation. Then for pumping going to use a solenoid metering pump so that I can take direct control of feed water. If all this shows results it will be time to create a reasonable controller circuit and then tuning.
I've got some 3/16" brake line on order along with some 1/4" Gonna stagger the sizes. Use 20' of 1/4 to preheat and 20' of 3/16 to make steam. Should give me much more steam of a reasonable quality compared to <10' of 1/4 copper. Then the 3/16 is going to go into the maintenance ports with a Tee and two 3/16" compression fittings. May insulate may not. Definitely gonna put a thermocouple on the line as too much superheat is the same as an unmonitored increase in intake air temp, which could cause detonation. Then for pumping going to use a solenoid metering pump so that I can take direct control of feed water. If all this shows results it will be time to create a reasonable controller circuit and then tuning.
02-08-2012, 06:48 PM
#89
I've now got a new version of the steam generator. It's installed on the resonator. Pulled plugs they were fine. Nice light brown. Changed the oil, looked normal. No water or other surprises. Tried running a lean idle. Got to somewhere around 16:1 before the engine started running bad. Could probably get it into a solid lean burn if I had idle timing control. Gonna wait for lean burn on the road until I can tap into the knock sensor and hear the engine. Also need to get live tuning to work. Still progressing though. Will post pictures when I have something to be proud of.
02-18-2012, 10:05 AM
#92
To quote Dr Emmett Brown, " My God, do you know what this means? It means that this damn thing doesn't work at all!"
I still see possibilities, but steam's benefit over water injection seems to be non existent. It works great as an internal coolant, in fact it works so well that it prolongs the warmup period resulting in lower fuel economy until the car is completely up to temperature, although this is probably limited to colder days. As for preheating the fuel air, either it has no effect, or more likely the heat it takes away from housing/coolant balances any heat it adds. I still believe that with tuning it should show positive results, but without tuning it's pointless. It however does not decrease fuel economy other than the increase in warmup time.
Adding 2 deg of advance during cruise has had a much more appreciable increase in fuel economy. Increasing the CL Exit for cruising cells (for 75MPH plus.) has also increased fuel economy. I would love to run a back to back test showing the difference, but without live tuning the results would be meaningless as the ECU would reset all learned data when reprogrammed. All I can say is that my fuel economy has improved slightly but noticeably. (About 1 mpg)
Here is what else I've learned so far:
Premixing does directly effect idle vacuum. (I know others have covered that before)
Heavier weight oil does cause a noticeable drop in fuel economy, but for the peace of mind it's worth it.
Adding zinc to oil does increase idle vacuum. In about 2 weeks I notice idle vacuum go from about 23" to 24.5" Might also have to do with heavier weight oil.
After coolant has warmed up it still takes 10+ minutes to reach best fuel economy.
Royal purple is the devil. It should be avoided at all cost. Syncromesh is wonderful stuff.
Water free coolant is great.
Ok, as to my previous results:
The reasons I get bad fuel economy are simple. Most of my driving is 20 minutes to and from work at highway speeds (and then some!). This barely gives the car time to warm up and get to best fuel economy. Before modification past about 75mph I go into open loop which drops my fuel economy like a stone. The short 5-10 minute trips like picking my son up from school have more effect on fuel economy than anything else I could do. Also I'm missing the engine underpanel which would also be dropping my fuel economy by increasing drag.
My initial results were flawed in two ways. First the Devils Own injection pump controller shifted my MAF value slightly which caused the ECU to indicate about 25% better fuel economy. Second, there is one stretch of road which appears level but is at a slight decline. Going 60-65 down that stretch with the pump controller connected I can read well over 30MPG.
Where to go from here? Well I'm not done yet. I'm working on a bit of electronic wizardry that should give me a useful knock sensor, and if it finally stops raining I'll try again to get the AP drivers to work and let me live tune. I can't wait to have that working and try lean burn! Also I'm waiting on parts for a real wild experiment.
I still see possibilities, but steam's benefit over water injection seems to be non existent. It works great as an internal coolant, in fact it works so well that it prolongs the warmup period resulting in lower fuel economy until the car is completely up to temperature, although this is probably limited to colder days. As for preheating the fuel air, either it has no effect, or more likely the heat it takes away from housing/coolant balances any heat it adds. I still believe that with tuning it should show positive results, but without tuning it's pointless. It however does not decrease fuel economy other than the increase in warmup time.
Adding 2 deg of advance during cruise has had a much more appreciable increase in fuel economy. Increasing the CL Exit for cruising cells (for 75MPH plus.) has also increased fuel economy. I would love to run a back to back test showing the difference, but without live tuning the results would be meaningless as the ECU would reset all learned data when reprogrammed. All I can say is that my fuel economy has improved slightly but noticeably. (About 1 mpg)
Here is what else I've learned so far:
Premixing does directly effect idle vacuum. (I know others have covered that before)
Heavier weight oil does cause a noticeable drop in fuel economy, but for the peace of mind it's worth it.
Adding zinc to oil does increase idle vacuum. In about 2 weeks I notice idle vacuum go from about 23" to 24.5" Might also have to do with heavier weight oil.
After coolant has warmed up it still takes 10+ minutes to reach best fuel economy.
Royal purple is the devil. It should be avoided at all cost. Syncromesh is wonderful stuff.
Water free coolant is great.
Ok, as to my previous results:
The reasons I get bad fuel economy are simple. Most of my driving is 20 minutes to and from work at highway speeds (and then some!). This barely gives the car time to warm up and get to best fuel economy. Before modification past about 75mph I go into open loop which drops my fuel economy like a stone. The short 5-10 minute trips like picking my son up from school have more effect on fuel economy than anything else I could do. Also I'm missing the engine underpanel which would also be dropping my fuel economy by increasing drag.
My initial results were flawed in two ways. First the Devils Own injection pump controller shifted my MAF value slightly which caused the ECU to indicate about 25% better fuel economy. Second, there is one stretch of road which appears level but is at a slight decline. Going 60-65 down that stretch with the pump controller connected I can read well over 30MPG.
Where to go from here? Well I'm not done yet. I'm working on a bit of electronic wizardry that should give me a useful knock sensor, and if it finally stops raining I'll try again to get the AP drivers to work and let me live tune. I can't wait to have that working and try lean burn! Also I'm waiting on parts for a real wild experiment.
02-18-2012, 02:48 PM
#93
24 inches of vacuum at idle? damn thats the highest I have ever heard anyone having on this engine.
It takes the oil a while to heat up to 190-200F. I am wondering about a water/oil cooler from motul to help with that--but I am not sure.
I do wonder what you will be working on now.
If you change the gearing the mpg will go up.
It takes the oil a while to heat up to 190-200F. I am wondering about a water/oil cooler from motul to help with that--but I am not sure.
I do wonder what you will be working on now.
If you change the gearing the mpg will go up.
02-19-2012, 04:07 PM
#94
Ok, so here is what I accomplished today. Using the writeup here:
http://www.k20a.org/forum/showthread.php?t=27258
I was able to tap into the knock sensor and record some sound.
I had to use the Black and yellow wire, as the black and red is power supplied from the ECU and held constant. Also the 100K ohm resister was way too big, I used a 7k because it was what I had on hand, but the output was a little on the loud side. Somewhere in between would be perfect.
Here is a zip file with the sound I recorded. There is some interference as the laptop's mic in sounds like crap. I was also able to take a listen to the Eccentric Shaft Position sensor, which will be useful for future projects.
I had something else I wanted to try today, but after looking for a half hour for a 10mm wrench I decided to call it quits.
With filtering and a VU meter this could be very useful as a knock meter.
http://www.k20a.org/forum/showthread.php?t=27258
I was able to tap into the knock sensor and record some sound.
I had to use the Black and yellow wire, as the black and red is power supplied from the ECU and held constant. Also the 100K ohm resister was way too big, I used a 7k because it was what I had on hand, but the output was a little on the loud side. Somewhere in between would be perfect.
Here is a zip file with the sound I recorded. There is some interference as the laptop's mic in sounds like crap. I was also able to take a listen to the Eccentric Shaft Position sensor, which will be useful for future projects.
I had something else I wanted to try today, but after looking for a half hour for a 10mm wrench I decided to call it quits.
With filtering and a VU meter this could be very useful as a knock meter.
Last edited by Harlan; 02-19-2012 at 04:10 PM.
03-03-2012, 11:00 AM
#96
Longpath, I'm manually controlling injection, so I can choose when to start or stop injection.
However, I have found out why my idle vacuum is so high and part of my fuel economy problem. My front o2 sensor was fouled and lagging. I'm sure it happened with the old engine, blowing all that coolant into the exhaust plus the idiots who worked on it flooding it repeatedly causing all sorts of gas and oil to go through the tailpipe, not to mention it was already at 72k with previous idiot owners. Now with a new o2 sensor my idle fuel trims are hovering around zero, and the idle is so smooth. Haven't been through a full tank, but it's gonna be better. Next step is to replace the engine under tray.
I had suspected the o2 sensor, but it never threw a code or was otherwise screwy. The only thing was relatively high fuel trims and that was only at idle. Finnaly noticed a wierd idle loop and took an expensive long shot. It paid off. Will post more detailed results when I have them.
However, I have found out why my idle vacuum is so high and part of my fuel economy problem. My front o2 sensor was fouled and lagging. I'm sure it happened with the old engine, blowing all that coolant into the exhaust plus the idiots who worked on it flooding it repeatedly causing all sorts of gas and oil to go through the tailpipe, not to mention it was already at 72k with previous idiot owners. Now with a new o2 sensor my idle fuel trims are hovering around zero, and the idle is so smooth. Haven't been through a full tank, but it's gonna be better. Next step is to replace the engine under tray.
I had suspected the o2 sensor, but it never threw a code or was otherwise screwy. The only thing was relatively high fuel trims and that was only at idle. Finnaly noticed a wierd idle loop and took an expensive long shot. It paid off. Will post more detailed results when I have them.
03-04-2012, 03:44 PM
#100
Checked out the thread. Pretty cool. Tried to do lean burn durring cruise again. Failed. Managed a quick drop in short term trim then the commanded AFR dropped to normalize my tweaked AFR indication. I think I have it figured out. The Wideband O2 is checked by the narrow band. If the wideband reads rich and the narrow band reads lean then commanded AFR is adjusted untill the narrowband reads right. Think it only applies at stoich. I know someone on this forum, I think MM said that the narrow band has no effect on mixture, but there is something causing the commanded AFR to drop when I tweak the o2 to read richer than actual.
Damn Subbie ECU is too redundant to easily trick. I'm gonna try tweaking both o2 sensors at the same time. Maybe then I can get it into lean burn at cruise.
Damn Subbie ECU is too redundant to easily trick. I'm gonna try tweaking both o2 sensors at the same time. Maybe then I can get it into lean burn at cruise.