Twin Engine Twin Turbo
danwagonboy
Band Wagon
This idea has been rattling around in my head for some time now. Just to give a quick history for kicks decided to put a second identical DPFI (auto) drivetrain in the back of a wagon just to give it a little more umph. The only problem ( I knew this going into the project) is that if I'm lucky it gets low 20's (21-22 mpg) on the highway which is less than a modern, similarly sized, v6. as far as additional losses go, this is all I can think of.
(1) transmission losses - should be more or less proportional to the net power being lost. For instance, say at 70 mph it takes 20 hp to keep the car going. Two transmissions handling 10 hp each should create roughly the same amount of thermal waste as a single one at 20 hp. Say a net of 3 hp (15%).
(2) thermal losses - having two separate engines doesn't seem as thermally efficient as a single one, you want the greatest change in temperature possible between intake and exhaust to get good cycle efficiency.
(3) pumping losses - running two engines at 70 mph at 1/8 throttle vs a single one at 1/4+ throttle incurs much more pumping losses.
To summarize, thermal efficiency can't be gained except by removing an engine (tempting some times). Pumping losses, however can be addressed by running a small turbocharger to reroute exhaust energy to overcome pumping losses and increase low end torque while tapering off in the high end to avoid bent crankshafts etc. Is my logic right here? this is purely hypothetical at this point. Its alot of complexity for something to gain, say 2 mpg and 50% bottom end torque
(1) transmission losses - should be more or less proportional to the net power being lost. For instance, say at 70 mph it takes 20 hp to keep the car going. Two transmissions handling 10 hp each should create roughly the same amount of thermal waste as a single one at 20 hp. Say a net of 3 hp (15%).
(2) thermal losses - having two separate engines doesn't seem as thermally efficient as a single one, you want the greatest change in temperature possible between intake and exhaust to get good cycle efficiency.
(3) pumping losses - running two engines at 70 mph at 1/8 throttle vs a single one at 1/4+ throttle incurs much more pumping losses.
To summarize, thermal efficiency can't be gained except by removing an engine (tempting some times). Pumping losses, however can be addressed by running a small turbocharger to reroute exhaust energy to overcome pumping losses and increase low end torque while tapering off in the high end to avoid bent crankshafts etc. Is my logic right here? this is purely hypothetical at this point. Its alot of complexity for something to gain, say 2 mpg and 50% bottom end torque
Comments
A single v6?
(1) would change weight distro from 50/50 to like 70/30
(2) would have the same problem under light load (depending on gear ratios)
A double v6?
(1) curb weight would be over 3000 lbs
(2) would be running with 1/16 throttle and even more pumping losses
(3) >4 liters combined displacement is vulgar and not sporty (IMHO)
Reasons for keeping d15
(1) d15 is a simple, cheap, reliable engine
(2) California emissions
(3) stubbornness / twin engine setup is working fine now
I don't want a power boost out of this really, but if I could use wasted energy to increase efficiency it makes sense to me. Has anyone looked at the Brake Specific Fuel Consumption graphs for partial throttle conditions (reference: http://autospeed.com.au/cms/title_Brake ... ticle.html)
What I see here is that below 1/4 throttle the efficiency drops off significantly because the engine is acting as a big pump while outputting very little total horsepower. If you boost under these conditions, the car will be less likely to downshift and that loss associated to pumping almost goes away (energy is taken from exhaust heat rather than engine output). This same thing might be applicable to H22 swaps, etc where displacement is increased slightly.
.
wagodizzle, his twin engine has already been done...
quite bad ass actually
viewtopic.php?f=5&t=9397
im no expert either so this is over my head too. but you sound to be on top of it.
i looked over your link a little... lol
very complex. especially since ive never heard anything on this before.
if you want my opinion i say install a turbo simply for the reason that you can tear some 4wd ass in it!
is this your daily driver?????
otherwise i wouldnt be worried about mileage
could you run only 1 engine and keep the other in neutral?
this would give a gain right? the weight of the extra engine would lower mpg but these little engines are so light. it would be like you had a fat friend riding with you at all times haha
ive looked up the engine weight before... wet its only like 240lbs i think. dont know about trans
Whoa. I humbly retract my previous statement.
I saw this article today (http://autospeed.com/cms/A_109931/article.html) in which they claim a similarly spec'd car with some (well engineered) low end boost got better mileage. However, the transmissions between the turbo'd car and the n/a car weren't identical, and I was wondering if the gearing might be the reason highway mileage increased.
My hypothesis is that any car with a large displacement swap (say a 2.2L or bigger) could benefit from a small amount of boosting to compensate for the fact that under low load, the throttle will incur great pumping losses. This will not compensate for the fact that there are more moving parts, increased friction, etc due to bigger displacement - meaning a turbo 2.2 won't perform better than a na 1.8, etc. How many mpg does a Porsche get?
(1) run external hydraulic pump to lubricate transmission - could engage on the fly
(2) put some type of clutch/pull differential apart in transmission housing
(3) adapt some kind of locking hub from another car
I was curious if anyone did a similar low-boost turbo setup and saw a significant change (in either direction) in mileage - although I'm guessing most setups are focused on peak power output and not running too lean.
however i feel that they dont actually understand what pumping loss really is, they blame the the throttle butterfly for the pumping loss yet they dont look past it. bmw tried to break this by using a variable valve engine, it improved fuel economy but not much. i dont know if they noticed that theres still vacuum in the cyl though it wast showing in their manifold. just imagine how hard it is to pull the plunger in a blocked syringe. now think how hard it is to pull the connecting rod and 75mm piston 3000 times in a minute. if you put a turbo its not the exhaust restriction thats cause the bsfc to rise but the pressure thats being introduced inside the engine therefore eliminating vacuum. ever notice the lower volume the engine gets, the higher the fuel economy gets? remember geo metro? 1000cc 3 cyl 2 valves and carburated? it gest like 60mpg.. its not the engines efficiency or complexities, maybe just a phenomenon
it is true that the peak torque is also the highest bsfc... that is in wot, in partial throttle its a different one. the bsfc curves is mostly affected by air that is being introduced in the cylinder, the dip in the curve is most likely the intake runners freq. the honda insight has longer and narrower runners while the quad rotor le mans engine surely should have a shorter/bigger runners/valves,
honda perfected this with their vtec-e, stratified charge, lean burn, low vacuum, low friction. much like a diesel, thats why it competes with the fuel economy of a diesel. its so good that they are still using it on their i-vtec. i guess what im saying is run lean even if its not a vtec-e, surely that will lower the vacuum and will also run like shit.
a vacuum gauge is an essential tool i never realized before also maybe an egt and data logging if you were to tune for lean burn
yes there would be parasitic loss from the pulley system but you would gain a better power curve.
koji- thanks for the comments, the project took longer than I ever thought it would. To clarify I think what your saying is:
(1) vacuum being pulled in the cylinder is what the engine sees as parasitic loss can be overcome with the "free" thermal power a turbo provides.
(2) a small engine is more thermally efficient because a given amount of heat generated has less heat transfer through the cylinder walls during a cycle.
(3) a turbocharged engine running under boost is more thermally efficient say at wot because of greater charge density
(4) if you look at the BSFC graph, the low load condition could be 30 psi bmef or 2 atm and the vacuum of .15 atm could become a higher % factor
Leaning out the engine and changing gearing are probably the most practical things to do but turbocharging has fun potential. It would be interesting to see if the engine has some potential for leaning out under load without causing damage (egt sensor would help here, as well as a mass airflow sensor - damn obd0). A simple test in my case would be to look at vacuum with singe engine, then double engine, and determine if a problem exists.
kylerwho - I think that the supercharger pulls power from the crank which is equivalent to the engine having to suck on vacuum efficiency wise, although under power it will probably be more thermally efficient. It would be interesting if a smaller supercharged engine would produce power more efficiently and overcome the parasitic losses associated with the blower belt. another complication is turbos with low RPM boost are hard to design because it's exhaust gas driven, so the power available may be very little at cruise conditions.
likewise, dunno how i missed this :?
obviously i fail at life.
I'm thinking about trying to build a flat-ish torque curve from 2-5k and then start tapering off boost to reduce high speed stress on the stock parts. Most sites I have seen suggest the D15B2 is capable of taking 6psi of boost, so cutting boost off in the higher revs will produce slightly less power but prevent things from breaking as much. Has anyone found a significant decrease in economy associated with swapping in a larger OEM engine?
ive done a turbo vtec-e, whenever i get on lean burn i actually notice 5-15^c drop on my coolant temp, leaning the mixture in high loads(not throttle)doesnt cause preignition(my knock doest detect it) instead the combustion is slower and sometimes will not ignite so i bucks and shakes if over leaned. my wideband only reads 10-20afr i know ive gone much leaner. i cant say much about egt since its my guide for ignition timing other than my knock sensor so i may not consider it as factor for fuel econmy. although i cannot justify my fuel consumption since my commute doesnt take me very far, so most of the time the fuel is used during preheat and mostly momentum waste during accel/decel since i live in the philippines. a third world country roads are not meant for fast commute and we have to share the narrow road w whatever you have to dodge to get around. but im confident to say i can pull off a 14km/l from a 10(no lean burn)
if i saw your car..man id be giggling. then you tell me its a 4ws.. i wouldnt know what to tell after that
wagon..........................4ever...........................
look into getting twin busa motors (1300cc) since that would put the cruising speed in a better efficiency island. not to mention you would lose a couple hundred pounds of weight.
http://www.youtube.com/watch?v=X2Tyi3XFt5g
or this?
http://video.google.com/videoplay?docid ... 003719934#
It would be interesting but I doubt they will be good for 300k miles
koji - your saying the boost causes the engine to run lean and dump more heat into the radiator because there is less fuel to vaporize and draw heat out of the engine? But your also saying when you lean out your engine the coolant temp drops indicating less heat is being wasted through the radiator. A drop in coolant temp would mean a slightly more efficient system (although if the thermostat closes that could introduce other pumping losses) -because the delta T is higher?
I was thinking about layout if I could flip the intake manifold upside down and run the turbo behind the exhaust manifold, then I could get a very short circuit path which would not allow for an inter-cooler. The higher intake temperatures, under slight boost pressure may actually encourage fuel atomization to some degree.
Another question I can't seem to answer is to look at the highway case when there is a small boost present, wouldn't that result in the driver closing the throttle even further for a given speed, and then incur more pumping losses? Although those should be taken care of with the free turbo energy. Get to low enough throttle and the locking torque converter kicks out too.
The other think I noticed is my freeway driving habits with single vs double engine are much different, ie accelerate from 60-85-60-85 (double engine) vs 65-65-65-65 (single engine) and that might have more to do with hwy mileage than anything lol.
if i were to build boost it results from 20:1 afr to 12.5 meaning more pressure in the cylinder from boost and more fuel from 12.5 meaning more heat against the cyl walls. this area doesnt concern cruising this is all out power. in my perspective people say they gain fuel economy with turbo for me is because the turbine slows the engine down(i cant believe how restricted the turbines are esp .48) where you have to increase the load therefore reducing pumping loss, everyone says you need back pressure to increase fuel economy, the same principle applies. upgrading a free flowing full exhaust hurts fe because of this. sorry if it seems that im talking out of my ass but observation they say is good teacher. im not an automotive engineer or anything, only mechanical background i have is turbojet mechanic. sorry again if i sound confusing
Would a larger turbo be better at doing this? I still don't understand for a given load/rpm a boosted air to fuel mixture would be richer (g air/ g fuel) why? wouldn't the added charge lean out the mixture, or because of the 02 feedback loop the ecu would compensate by dumping in more fuel than needed? This seems like an engine management issue rather than something inherent to boosting. I know the car's definitely a boat at 2750 lbs curb.
yeah block the engine compartment first, if you get good mileage then try the radiator shroud. downside is youre going to loose some power. compromise i guess.
Kudos on that btw.