h a l f b a k e r yCaution!
Contents may be not!
add, search, annotate, link, view, overview, recent, by name, random
news, help, about, links, report a problem
browse anonymously,
or get an account
and write.
register,
|
|
|
Regenerative Throttle
Recoup some electricity from your throttle losses with an axial turbine/compressor! | |
Ok... Here's the theory...
For a petrol engine (gasoline if you prefer) your engine output is controlled by the amount of air allowed into the intake manifold. This is usually done by putting a bloody great blockage in the ducting to drop the pressure, and therefore the density of the air breathed
in by the cylinders.
You are therefore asking the engine to do work, as you have a mass flow of fluid across a pressure drop. Why not make use of some of this by having an axial turbomachine runing/run off a DC motor rather than a butterfly valve, which could either produce electricity at low load to help out the alternator, or be fed energy to become a mild supercharger.
Granted the engineering challenges of designing the device to operate vaguely efficiently over the range of massflows and pressure differences required by the engine, probably requiring adjustable vanes on both the stators and rotor, but I imagine the amount of savings made when idling in traffic would be significant. Please see below... I imagine wrong!
Being an engineer, I like to do 'back of an envelope' maths...
a 1.0l engine idling at 1200rpm will absorb 1 * 1200/2 = 600l/m air from it's manifold, which is 10l/s.
Assuming it's operating at 0.9atm vacuum, that's:
Volume flow * deltaP = Power
10l/s * (1 - 0.9)# * 0.001 m3/l * 0.9atm * 100,000Pa/atm = 90W
# inlet volume flow scaled due to density change
available, about 40 of which will likely get to the device that's using it (air con/fan/...). I knew this idea was pretty half-baked when I decided to put it on here, but I didn't realise it was that bad... Quick, everyone drive cars with 10l engines that idle at 3,000rpm!
Example manifold pressures
http://www.aa1car.c...ary/map_sensors.htm
A table at the top of the screen shows the sorts of pressures my device would have to produce for the engine control/work with to produce power. [Skrewloose, Nov 13 2008]
[link]
|
| |
The engine's power output is controlled not by the airflow per se, but by the amount of fuel entering the cylinders. The restriction of airflow at low power demand is to ensure that the fuel/air mixture is at the right concentration for efficient combustion. |
|
| |
This is valid for both carburated and fuel-injected spark-ignition engines. Diesel engines typically have very little in the way of restriction in the induction pathway. |
|
| |
The energy "lost" by restricting the input airflow is not great, but could conceivably be partially recovered by a system as described. But the additional mass, cost, and above all complexity (likely to reduce reliability) would probably not be justified. |
|
| |
I know the usefulness to costs ratio is all wrong, as shown in my last paragraph after the work output was shown to be apalling, this is Halfbakery after all. All I was meaning was that the 'pumping losses' for an otto cycle engine could be put to use, rather than simply wasted as turbulence in the intake manifold, upsetting all the hours of work the engineers have put in on the flow bench tweaking the intake manifold and ports. |
|
| |
I am aware (having spent a year working for a Diesel engine manufacturer) that a Diesel engine inducts all the air possible and is then controlled by fuelling. |
|
| |
I think you'll find, however, that the control on a petrol engine is provided by the amount of air allowed into the cylinders, and the fuel (be it carburetted or injected) is adjusted to optimise the energy extracted from the given amount of air (within emmissions restrictions of course). All initial thermodynamics are based on a mass of air which has 'heat' added to it. |
|
| |
Being an engineer, I like my maths and my assumptions to be correct. |
|
| |
Correcting your maths only, the power goes to 900W, as there are in fact 100,000 Pascals in this atmosphere of ours. |
|
| |
Cue the assumption checker, you calculate power as though the volume flow rate is constant. It is not 10L/s near the mouth the turbine, in fact it is sod all. |
|
| |
Your turbine will produce very little power because the mass flow rate through it is so small. |
|
| |
the friction would be even greater! come on! |
|
| |
Apologies [texticle], I did put the idea up late at night. As it happens, the inlet adjustment (0.1 atm outlet gives a 0.1 factor to inlet volume flow) cancels out the missing 0 from the pressure, still giving power extracted about the same as a single traditional headlight bulb. |
|
| |
Hope you're happy with the alterations. |
|
| |
How cofident are you in the 0.1atm assumption? |
|
| |
// How did cofident are you in the 0.1atm assumption // |
|
| |
Annotation [marked-for-removal], gibberish. |
|
| |
[Skrewloose], calm down .... it's [Texticle] we're poking fun at ... |
|
| |
Appologies - I'll remove the rant and get back to bending my head around L1 thermo (-83C sounds a bit cold for intake temperature!), oh how long ago it was that all the equations made sense. |
|
| |
Looks like 0.2 atm would be more appropriate (and conservative), please see link (hope it works!) |
|
| |
lets take a look at the logic here: I don't like the friction my TB is causing. hmmm. Maybe if I replace my TB with a complicated arrangement that does the same thing with many moving parts and a generator that still forces the air through a turbulent and heat producing passage and that will be nothing but parasitic on the engine when it is actually being used i can "reduce" the "frictional losses". Nothing you can do with a calculator is going to make this make sense in this universe. |
|
| |
So the compressor on your turbo is reducing your power is it? |
|
| |
My intention was that the device could be both a clever throttle (reducing idling emissions and fuel usage, which can make up up to half the output of an engine) and a mild supersharger. How many cars run at more than 1/2 throttle for a significant amount of time on the road anyway (unless of course the driver is a nutter)? |
|
| |
I agree this probably wouldn't be much use on a motorway (or highway if you like), but most cars are still at only half throttle in that situation. |
|
| |
Do you now see where I'm coming from... or do you drive everywhere in a chavmobile with a single bit digital throttle? |
|
| |
It's a good idea and one I've been thinking about for years. There are four solutions that I have come up with to solve the TB power loss issue: a CVT controlled supercharger, a variable displacement vane supercharger, heating the intake charge using waste heat to lower its density or a combination of the above. |
|
| |
The only downside is that the throttle body may not loose as much energy as you think as the sudden drop in pressure will produce cooler low density air. That lower temperature may gain you back some of the power you lost. |
|
| |
I like the CVT supercharger idea (not sure if I annotated on it) - although you'd need to be sure of at least a ratio range of 10 to get from idle to 'NA'. I don't know enough about vane pumps to make much of a coment, but I think you'd have response time issues with the waste heat idea. |
|
| |
After my magic 10:1 pressure ratio turbine (looks like it'll have to be multiple stage, and very variable to allow conditions other than idle), I'm going to need some waste heat if I've got an intake manifold temp of -83C (assuming 0.2 atm pressure, 30C ambient, and isentropic expansion). |
|
| |
[WcW], Skrewloose is right - there is significant energy wastage through a closed throttle as the air is expanded through it without doing useful work. However, as the maths seems to show, there's not a lot of recoverable energy to be had. |
|
| |
the inlet and exhaust valves, and lets be honest the whole combustion system is rife with thermodynamic ineffiency. The only way to address the speed/volume/load issue is to have a 100% varying valve action. Then we realize that we have simply moved a lot of that thermal inefficiency to the port/valve. As to the compressor wheel of my turbocharger, yes it is a huge blow to thermal efficiency and thus I must use a charge cooling device which would also have a net negative effect were it to not do an important job. Bleeding air filter falls into the same category. I suspect that you want some sort of diesel engine..... |
|
| |
Diesels are much better at most things than petrol (gasoline) nowadays, but there will always be diehards who won't accept the fact, meaning something done to increace the real-world effeciency of the vehicle is always a good thing. |
|
| |
Handling lowish to high speed air at pressures over a range of 20:1 without significant losses is always going to be difficult! My hope here is to re-coup some of the inherrant losses in the system. |
|
| |
By the looks of the calculations I've been doing, my idea would give temperatures too low (below -50C for idle) to evaporate any fuel until after at least most of the compression stroke, by which time you want it to be fully mixed and burning. Maybe a post-throttling charge warmer running off the engine's coolant might work (although I don't anvy anyone having to cold start an engine with this on!). |
|
| |
The best way to throttle a S.I. engine from what I can see is to use EM valves controlled directly from the ECU/ECM, giving you the correct amount of air all the time. The problem with this solution is the failsafe... how do you guarantee it won't fail open and cause a piston/valve crunch! |
|
| |
// how do you guarantee it won't fail open // |
|
| |
Some sort of varaible desmodromic coupling ? |
|
| |
Spring return seems to work Ok, you just need to make the actuation mechanism variable. |
|
| |
I thought of this idea earlier, though I had thought of it in terms of cruising-speed operation (vacuum-powered alternator) since I would guess there's more useful power available there. Any power that once can harvest from the pressure gradient will represent heat that isn't being added to the intake air. Since one of the limiting factors on the efficiency of internal combustion engines is the ratio of absolute final temperature to absolute initial temperature, reducing the initial temperature should improve the efficiency of the engine itself, in addition to providing whatever amount of free energy the alternator itself yields. |
|
| |
My worry, [8th of 7] is that if you get a short somewhere providing power to the solenoid in the valve, it's going to be held open ==> BAD!! |
|
| |
A spring return would certainly help with normal operation though. |
|
| |
[supercat] I agree that you'll be getting more returns on the motorway, but my theory is that thet's when your engine is in it's 'sweet spot' for effeciency anyway.When idling, you're burning an awful lot of fuel doing nothing! |
|
| |
Although the lower intake temp would increace efficiency by absorbing more heat from the cylinder walls, I doubt trying to evaporate fuel in a -50 odd C environment is going to go too well! And that's before you go to a cold country!! |
|
| |
//I agree that you'll be getting more returns on the motorway, but my theory is that thet's when your engine is in it's 'sweet spot' for effeciency anyway.When idling, you're burning an awful lot of fuel doing nothing!// |
|
| |
The amount of fuel burned per hour at idle is much lower than the amount burned per hour traveling at motorway speeds. Further, many vehicles spend more time running at motorway speeds than at idle. Consequently, a 10% reduction in fuel usage at motorway speeds would generally be more useful than a 99% reduction in fuel usage at idle. |
|
| |
Further, the only cars whose engines achieve optimal efficiency cruising at highway speeds are apt to be those with barely enough power go that fast, and insufficient power to to provide good acceleration at such speeds. |
|
| |
//Although the lower intake temp would increace efficiency by absorbing more heat from the cylinder walls, I doubt trying to evaporate fuel in a -50 odd C environment is going to go too well! And that's before you go to a cold country!!// |
|
| |
My idea would be that a vehicle would use a vacuum-powered turbine generator/alternator in parallel with a conventional throttle valve. The turbine would only be effective when current out of the generator was limited and the conventional throttle valve was closed. In the event that the turbine circuit fails free-wheeling, air intake would be unlimited but the engine controller could cut the fuel injectors to prevent a runaway condition (it might be possible to provide a means to manually cap off the turbine to allow the car to be driven in such case until it can be serviced, but I think the risk of that failure should be slight enough that I'm not sure it would be worthwhile). |
|
| |
If such a parallel-throttle arrangement were used, an engine controller could effectively disable the turbine when the engine was too cold to allow its effective operation. |
|
| |
I like the parallel idea [supercat]! That concept could also be used to make the system more efficient by having multiple channels to cope with the varying flow rates due to engine speed changes. |
|
| |
I agree that the idea could (in theory) save a large amount of fuel on the motorway (possibly more, as you say), but my point is that a lot could be saved on cars that tend to live in the rush-hour, rather than the open road (where best fuel economy is reached anyway, due to the manufacturer being able to tune a sweet spot around 70-85mph, whilst trading off economy for emissions elsewhere in the torque curve). |
|
| |
Basically, what I'm saying is that there's energy to be saved if you can expand your intake air usefully, then replace the heat lost using engine water heat (water-air intercooler in reverse) to give a gain. |
|
| |
50K * 1kJ/kgK gives 50kJ energy extracted per kg air, with around 15 times the air as fuel (a/f ratio), giving 750kJ, or 0.75MJ per kg fuel. |
|
| |
As fuel has ~45MJ/kg, that's a 0.75/45 = 1.5%. In real terms though... 1.5%/30% gives a 1.5% saving in fuel economy when you're at motorway efficiency, or 1.5%/10% = 15% gain when in town with an unloaded engine most of the time. |
|
| |
Obviously these kind of gains could be gotten from a hybrid drive or stop-start of the engine, but this system avoids that sort of complexity or potential engine wear. |
|
| |
Pumping losses are a good percentage of the
thermodynamic loss of power in any engine. Asking
the intake air to do more work on its way in is simply
going to add to the engine pumping work, in the
same way that adding some turbulence to swirl the
air and get a better combustion event also increases
pumping work. But replacing that brick wall throttle
with a spinnable turbine of some sort really isn't a
terrible idea, imho. |
|
| |