Pumping losses

[theholycow]

Take a look, this guy might be able to end this discussion with real experimental data...
http://www.gassavers.org/showthread.php?t=7467

[monroe74]

Originally Posted by theholycow

I don't entirely understand what this discussion is about

As in many discussions, there are multiple overlapping questions that have been raised. But I think they're mostly about the relationships between these interrelated concepts: pumping losses, load, BSFC, lean burn and P&G.

Whatever amount of air is in there becomes a spring. As much energy is spent compressing it on upward strokes, is given back on downward strokes.

Exactly. An important point, but I don't understand why you're bringing it up.

You forgot a stroke.

I think you're making some kind of a point about the exhaust stroke, but I don't know what it is.

to clarify my understanding of what's being discussed, it would best be demonstrated with the fuel cut off, the engine not running but merely used for engine braking, and the question is the difference between closed and wide open throttle.

Yes. And I explained why closed-throttle creates more engine-braking. Are you saying that I forgot to account for the exhaust stroke? I don't think so.

Maybe what you're saying is that the extra work the engine does (in this engine-braking example, and with an intake restriction like a closed throttle or a sealed intake valve) is paid back during the exhaust cycle. That would indeed be true if there were a corresponding temporary restriction on the exhaust side. But there isn't. When there is a closed throttle on the intake side, during engine braking, there is no corresponding closed throttle on the exhaust side.

I didn't think that P&G necessarily required reducing pumping losses during the pulse

I don't agree. Here's why. "Pulse," by definition, means you're accelerating. When we accelerate, we generally do so by opening the throttle further. This reduces pumping losses. So it's hard for me to understand how it's possible to have a pulse without reducing pumping losses.

others report P&G working with high RPMs

It would interest me if you could show me some examples of people claiming that it's preferable to downshift during a pulse, rather than use a larger throttle opening in a higher gear. Keeping revs low is generally a good idea, and I think those who report success with P&G generally report that they are trying to keep revs low. Which means using a high gear and a relatively large throttle setting.

you are probably spending extra fuel just to spin the engine at enough RPM to NOT do any engine braking. This is especially true when you're going downhill.

I think you're onto something here. The essence of P&G has to do with the fact that for most cars (even a VX), the engine is much larger than it needs to be to sustain a steady cruise at, say, 60 mph. (We require cars to be built that way because we need that extra power in other circumstances.) So if you're driving the 'normal' way (i.e., not using P&G), you're using very little throttle. But then what happens, because it's very hard to hold a perfectly steady throttle, is that the car speeds up a little bit. But then we normally don't throw into neutral and coast. Rather, we ease off on the gas. But now what's going on is engine braking (and some engine braking can be taking place even if we haven't released the throttle completely). And you're right, this tends to especially happen going downhill.

So what we think of as a normal steady cruise is actually a series of these little fluctuations. And you can see how this is inefficient. Aside from lots of pumping losses due to a mostly-closed throttle, we also spend a lot of time inducing some level of engine braking.

And cruise control doesn't help that much. It doesn't make the pumping losses go away, and I think it also typically works via a series of very small fluctuations.

P&G solves these problems.

If your pulse is a little wasteful due to uneconomical gear choice by the automatic transmission, it's still better than cruising along using fuel just to spin the engine.

Auto trans with P&G is a complicated situation. I'm not saying it can't work. I'm just saying it's not a good idea to create the impression that downshifting is a generally desirable practice, when you're trying to do P&G.

this guy might be able to end this discussion with real experimental data

I don't think he's going to get very far. But I also don't see which of the various questions raised in this thread would be addressed by his experiment.

[monroe74]

Originally Posted by R.I.D.E.

if you don't understand something, just ask for a more thorough explanation

If you're available to provide more thorough explanations, a good place to start would be the questions I've already asked that you haven't acknowledged (let alone answered). There are many examples. Here's one. Earlier you said this: "At 100% load your throttle position is exactly where the rpm will neither rise or fall."

That makes no sense to me. I already explained why. If I'm traveling at a very moderate, steady speed, with a very moderate throttle opening, I can very easily be in a situation where my "throttle position is exactly where the rpm will neither rise or fall." But that doesn't mean I'm at 100% load.

I'm trying to grasp how you're defining the term "load." It seems different than the definition I understand, which I've explained.

You also said this: "WOT will make engine speed rise." No. WOT will only make engine speed rise if the engine, at current rpm, is producing more than enough power to overcome whatever level of drag (or brake force) is currently being placed on it.

These are a couple of examples of things I don't understand, and I am indeed asking for more thorough explanations, since you've kindly offered to provide them.

Another misquote

If you're claiming I misquoted you, then you should be more specific. Otherwise you're making yet another statement that makes no sense to me.

you avoid explaining the partial load throttle positions

If you asked a question that was clear, and I ignored it, I hope you'll remind me what it is.

applying a load beyond the engines 100% capacity

I think maybe a source of confusion is that you're using the term "load" as if it means the same thing as 'drag' or 'braking force' or 'burden.' But that's not my understanding of the term. Earlier I described my understanding of the term.

By the way, it would help me if you acknowledged your misstatements, that I've pointed out. Like this one: "Further down there is a map of the insight engines Brake Specific Fuel Consumption, which clearly illustrates the effects of lean burn at low engine speeds." When you fail to do so, it becomes harder for me to take you seriously.

[dosco]

Monroe:
I'll post my calculations a little later. My problem is that my mileage didn't improve much at all by pulsing to 2,500-3,000 RPM (according to the BSFC chart the optimum rev range assuming high enough load) and then coasting. After using the SFC chart and a dyno run, it was clear to me that my fuel flow rate was killing me. And if I can get to the same highway speed with lower flow rate, why would I worry about trying to optimize the thermodynamic efficiency of my engine when I'm actually burning fuel at a higher rate?

This past weekend, I drove to upstate NY from my place in MD, and tried "creeping up" (slowly accelerating) to my cutoff speed. I was also able to take advantage of long downhill coasts as well. I drove over 400 miles with less than a half a tank of gas(!) yielding about 39 MPG (my best ever before that was in late summer and got about 31 MPG - look at my gas log).

After some thought, it seems to me with an automatic transmission, P&G is more about minimizing fuel flow by reducing a trip's average RPM than maintaining optimum BSFC. In other words, if over a trip I can manipulate the transmission so that the car is idling for half of the trip, the average engine RPM is much lower than the same trip driving "conentionally."

[R.I.D.E.]

!00% load is the point where if any more load is applied the rpm is not sustainable. A dyno can be adjusted to create loads that exceed the capacity of almost any engine in order to determine their max torque and horsepower.

50% load is where you have to adjust the throttle to sustain half of the above load.

25% load is the same as 50% with less throttle.

Beyond 100% is where you can not sustain the load reagrdless of your throttle position which would be WOT, but you could not maintain the rpm or the load.

That's the example where you are climbing a grade that you can not maintain in your highest gear and you must downshift in order to maintain speed on the grade.

If you have another question lets work on them one at at time without all the bandwidth wasting snide remarks, inneundo and freudian crap.

I like to develop methods of testing my hypotheses that is the essence of scientific experimentation.

The intent of the original post was to see how aware we all are of the many different causes of pumping losses.

The term WOT as I understand it means the throttle position is as far as you can open it.

In my experience when you use WOT you are telling the ECU I want everything this puppy can produce. The ECU responds by reducing the AF ratio below the stoichiometric ideal in order to give you whay you are demanding.

Its obvious in most BSFC maps.

The statement about compression forces being returned if you eliminate combustion is appropriate and basically true, and I believe you are convinced it cancels out the cost of creating compression.

If that was the case the "air spring" created by disabling 4 cylinders in a V8 engine should provide more than a 10% gain in fuel economy at least theoretically.

The problem with theoretically and real world testing is its much more complicated than that.

I guess the best way to understand what I am trying to say is to understand Carnots Law, which (without exception) states that any change in the state of energy incurs a penalty without exception.
This basically says, there are no perpetual motion machines.
If you get back all you put in you can not be correct because you are claiming perpetual motion.
Compression forces are greater than vacuum forces. If your engine has 10 to 1 compression forces they are greater than the sm of all vacuum forces.
You dont get all of the compression forces back because the piston is being pulled away from the compressed air, the connecting rod angles are changing the leverage exerted on the piston in converting rotation of the crank into reciprocation of the piston, friction is robbing you of energy at every friction point in the engine.

Your vacuum forces are greatly reduced by combustion chamber area, the part that is not eliminated by the stroke of the piston. Another cause of losses is the emission controls that don't allow manifold vacuum to exceed a certain pre determined point, otherwise when you closed the throttle the engine would suck all the air out of the combustion chamber (or at least almost all of it) and start to suck the oil past the rings, which creates more emissions.

In the graph of throttle position the throttle position is fixed at the percentage listed on the graph and the dyno measures the power until the engine can produce no more.

So in a load graph you adjust throttle for load, while in a throttle position graph you adjust the dyno to measure the load produced by the engine.

I welcome anyone here to jump in if my explanation is incorrect.

Monroe, I would appreciate it if you just try to stay on the specific point where my explanation has failed to make my position clear.

regards
gary

[theholycow]

Originally Posted by monroe74

I don't agree. Here's why. "Pulse," by definition, means you're accelerating. When we accelerate, we generally do so by opening the throttle further. This reduces pumping losses. So it's hard for me to understand how it's possible to have a pulse without reducing pumping losses.

My point wasn't that you could or should avoid minimizing pumping losses during the pulse, only that it's not a requirement; it was my guess (based on my ideas about cruising, which I described later) that P&G would still be more efficient if you had the same pumping losses during the pulse as you otherwise would at steady speed.

It would interest me if you could show me some examples of people claiming that it's preferable to downshift during a pulse, rather than use a larger throttle opening in a higher gear. Keeping revs low is generally a good idea, and I think those who report success with P&G generally report that they are trying to keep revs low. Which means using a high gear and a relatively large throttle setting.

In my manual car I of course pulse in high gear. In my auto truck I haven't had enough P&G experience but so far I've been pulsing in high gear, since its transmission is very cooperative -- I can give it a LOT of throttle and not have it downshift unless I really stab the pedal (though it's always willing to jump as soon as I ask for a hurry; altogether an extremely satisfying automatic transmission). However, in another thread there is someone reporting success with 4000rpm pulsing:
http://www.gassavers.org/showpost.ph...4&postcount=14
http://www.gassavers.org/showpost.ph...9&postcount=20

Auto trans with P&G is a complicated situation.

Perhaps, but it is at least an exceedingly easy experiment -- it's way easier for me to do in my truck, even with that balky column shifter. Those with floor shifters will find it even easier. All that rev matching and clutch work (in addition to moving the shifter) gets tiring in my VW.

I'm just saying it's not a good idea to create the impression that downshifting is a generally desirable practice, when you're trying to do P&G.

I did not mean to imply that. If I did, it was a mistake. I do believe that lower RPM + wider throttle == better economy, my own experiments have demonstrated it to me quite clearly. My point was merely that P&G can yield good results even if you are unable to accomplish that ideal.

[dosco]

Monroe:
As promised, some calculations, etc.

LONG POST.

I used the charts from this site: http://www.autospeed.com/cms/A_110216/article.html ; specifically the 2nd and the 4th (I also used a slightly different version of the 4th that came right out of the report referred to in the writeup).

I also found a dyno run for a 2.2L Toyota Camry here: http://www.dragtimes.com/2000-Toyota...aphs-8975.html . Of course there are some issues, it is the hp at the wheel, it's a 2000 model and I have a 1999, etc., but it's good enough for me to get an idea of what's happening.

So, according to chart 4, the optimum efficiency occurs at a BMEP range between 80 and 120, between ~1,500 and 3,100 rpm. For giggles, lets start at 2,000 rpm. The dyno chart doesn't have data for that RPM, but let's use linear interpolation and assume it is about 65hp.

Using chart 2, I calculate fuel flow as 29.9 lbs/hr for 100% load, and 33.8 lbs/hr at 50% load.

At 2,500 rpm, I calculate fuel flow as 32.4 lbs/hr for 100% load, and 37.4 lbs/hr at 50% load.

At 3,000 rpm, I calculate fuel flow as 35.7 lbs/hr for 100% load, and 39.1 lbs/hr for 50% load.

Here's another way to look at the data:

2,000 rpm: 100% load: 29.9 lbs/hr 50% load: 33.8 lbs/hr
2,500 rpm: 100% load: 32.4 lbs/hr 50% load: 37.4 lbs/hr
3,000 rpm: 100% load: 35.7 lbs/hr 50% load: 39.1 lbs/hr

Right off the bat, if I operate at 2,000 rpm at 50% rather than 3,000 rpm at 100%, I'm using 1.9 lbs/hr less fuel.

So you say, if you can use 100% load at 2,000 rpm, you'll use even less fuel. Indeed. However I can't do that in my automatic transmission without triggering a downshift. Without more instrumentation, I also can't be sure what my throttle angle is, nor can I be sure that the engine is in (or out) some sort of phase where it commands more fuel than normal (i.e. open-loop).

The difference in fuel flow rates - particularly at highway speeds - immediately translates into more miles per gallon for me. If I had a stick shift, that might be a different story. Bear in mind, though, that the BSFC chart is telling me that I'm still in the "optimal efficiency zone" which can be misleading, since as the rpm increases so does the fuel flow rate. In my Camry, I can easily travel at 60 mph at 2,000 rpm in top gear.

Furthermore...I did a little test this morning on the way to work. I drive on MD State Route 50 to get to work. It is relatively flat with some gentle rises and falls. At 60 mph and top gear, the engine operates at ~2,100 rpm. In the same conditions at 70 mph, it runs at ~2,500 rpm. At 50% load, that's a difference of 3.6 lbs/hr!!

I also tried the P&G using a gentle acceleration and a more brisk acceleration. With the gentle acceleration, I started at about 60 mph and 2,100 rpm, and over about 18 seconds pulsed to 70 mph and 2,500 rpm. Then I tried the same scenario using a larger throttle angle. The pulse started at 2,100 rpm at 60 mph, and ended in after 10 seconds at 2,900 rpm (I think the trans may have downshifted b/c of the throttle angle, I don't recall exactly and didn't make a note of it during the test). So for an 8-second improvement in pulse I ended up with a much higher rpm. But note that even operating more efficiently, by ending at about 3,000 rpm my fuel flow rate is also commensurately higher than in the other test case.

Evidence of effectiveness was clear on my long highway drive to NY - almost 40mpg - more than 400 miles on less than half a tank! Now that I'm back to local driving back-and-forth from work, I'm certain my mileage is going to go down again substantially for a reason similar to what you mentioned. At lower speeds, less miles are covered ... but the fuel flow rates are the same (or very similar), since they primarily are a function of rpm.

My conclusion is that for my car, it is better to keep the rpm way down, use every opportunity to glide, and use engine braking at stoplights by shifting to "2" and "1" as needed; also realizing that mileage is going to be far better with highway drives as opposed to city traffic.

Lastly. It would be interesting to create some sort of computerized modeling ad simulation tool to model and understand many of the possibilities.

[dosco]

Here is another thought: by using P&G, we're trying to maximize the time in idle. What is the engine SFC at idle, especially compared to cruise?

Another: what if you could magically reduce the coefficient of drag on your car? If that happened, the engine load would decrease (particularly at highway speeds), which theoretically would reduce the fuel efficiency of your vehicle. Do we see that in real life? (look at some of the guys here on this site)

[monroe74]

Originally Posted by dosco

My problem is that my mileage didn't improve much at all by pulsing to 2,500-3,000 RPM (according to the BSFC chart the optimum rev range assuming high enough load) and then coasting.

You're correct that this (2500-3000 rpm) is a desirable place to be, according the graph (http://www.autospeed.com.au/cms/gall...&a=110216&i=6). Actually, 3000-3500 is an even better place to be. But this assumes 100% load. Despite what Gary says, that means WOT.

But how are you attaining, or maintaining, that rev range with WOT? The problem is you have an automatic. Optimal P&G is very hard with an automatic. Let's say you're trying to do P&G. Let's say you're on a road with a 50 mph limit. You've just coasted, in neutral, down to 35 mph. You put the car back in gear and floor it. I don't know your exact gear ratios, but on your car that probably means your transmission will select 2nd gear. That's fine, at first, because it means you are probably at about 2500 rpm, and at 100% load (WOT). That's a pretty efficient place to be. But as you maintain WOT, and accelerate toward your target speed (50 mph), your tranny will linger in 2nd gear. By the time you reach 50, your revs will probably be up around 4000. Trouble is, 4000 rpm at WOT is a lot less efficient than 3000 rpm at WOT (and, as you noticed, it's also less efficient than partial throttle at lower rpm). What you really want is to select a higher gear, so your revs can drop. But you can't select a higher gear, and also maintain WOT, because you have an automatic. The machine is selecting the gear, not you.

Optimizing BSFC calls for a large throttle opening. But with an automatic, a large throttle opening puts you in a lower gear than where you really want to be. That's the problem. The best you can do, while in the pulse phase, is provide just enough throttle to keep the tranny from downshifting. But this usually implies a throttle opening that is quite far from WOT.

This past weekend, I drove to upstate NY from my place in MD, and tried "creeping up" (slowly accelerating) to my cutoff speed. ... I drove over 400 miles with less than a half a tank of gas(!)

Exactly. You applied the technique I just described (use the largest possible throttle opening that will not trigger a downshift), which is the best you can do with an automatic. But you would do even better with a stick because it would allow you to use an even larger throttle opening, and still select a gear that will force revs to be low. Because here's what we want: low revs combined with a large throttle setting.

if I can get to the same highway speed with lower flow rate, why would I worry about trying to optimize the thermodynamic efficiency of my engine when I'm actually burning fuel at a higher rate?

I think there's a basic misunderstanding. There's no free lunch. Optimizing the thermodynamic efficiency of your engine is what it's all about. There's no way around that.

An effective FE strategy is not just about trying a achieve a lower flow rate, or something like that. It's not about simply using less fuel. It's about using less fuel per unit of work that's being produced. That's what the BSFC concept means.

We burn fuel in order to move a certain weight, over a certain distance, at a certain average speed. It's always possible to reduce fuel use by shaving one of those parameters. Trouble is, there is a point where we don't want to shave the parameter any more. For example, you could further reduce weight by taking the back seats out of your car. You can further reduce distance by getting your boss to relocate the office closer to where you live. You can further reduce speed by deciding that it's OK to do 40 mph on the interstate.

But when you've decided that you've taken all those options as far as you can, and you still want to reduce fuel use, then there's only one choice: improve the thermodynamic efficiency of your engine. There's no getting around that.

It's possible you have had the following experience. You decided to implement a P&G technique where you accelerate very moderately. Good. And you discover that mpg was high. Good. But to really assess what's going on, you have to ask yourself the following question: was mpg high because I optimized thermodynamic efficiency, or was mpg high because I lowered my overall speed, for this trip?

Either answer is OK. I'm just saying it's a good idea to be clear about what's really going on. And optimizing thermodynamic efficiency is almost always going to be a good idea, because it's essentially free (aside from some extra effort and skill on the part of the driver). In other words, when you optimize thermodynamic efficiency, you can save gas without making further sacrifices with regard to weight, distance, and time.

it seems to me with an automatic transmission, P&G is more about minimizing fuel flow by reducing a trip's average RPM than maintaining optimum BSFC

That's what's known as a false choice. You don't have to choose one or the other. They go hand-in-hand. Reducing rpm tends to be a key part of optimizing BSFC.

Don't be misled by the fact that 3000 rpm at WOT is more efficient than 1500 rpm at WOT. You need to remember that the former setting produces probably about twice as much power. Yes, it produces that power very efficiently (in other words, let's say it provides double the power without using twice as much fuel), but you're wasting fuel unless you're in a position to put that power to good use.

It goes back to the fact that our engines have a lot of excess power. It would be wise (in FE terms) to stay at 3000 rpm and WOT all day long, provided that we were permitted to travel at 90 mph and provided that the laws of aerodynamics were also suspended. Then your mpg would be very high, because you would be operating in a zone of optimal BSFC.

P&G is so effective because it lets us temporarily run the engine in a very efficient way, where it is producing a lot of extra power for not a lot of extra fuel. And we store that extra power in the form of momentum, which we then use to coast.

[monroe74]

Originally Posted by R.I.D.E.

!00% load is the point where if any more load is applied the rpm is not sustainable.

I think it would be better if you could provide a definition of "load" that didn't include the word "load" (and I did exactly that, in one of my posts above). A basic feature of a helpful definition is that it doesn't require you to already know the meaning of the word that's being defined. But let's put that problem aside for a moment.

I've already explained why what you're saying makes no sense. If I'm traveling at a very moderate, steady speed, with a very moderate throttle opening, I can very easily be in a situation where my "throttle position is exactly where the rpm will neither rise or fall" (I'm quoting the words you used when you offered essentially this same definition, earlier). But that doesn't mean I'm at 100% load. What you've described is a state of equilibrium, but it has nothing to do with 100% load.

Let's look at it another way. I'm driving my 400 hp Corvette. I'm doing about 60, and I'm in top gear, and I'm using a very moderate throttle setting. 5%, let's say. The engine is turning over at about 2000 rpm. Maybe less. It's probably producing, in this moment, less than 10% of its rated output.

But I'm in equilibrium, which means my speed is constant. All of the parameters are constant: vehicle speed, engine speed, and throttle position.

Now imagine that the weather changes, and I'm driving into a storm, and suddenly a 40 mph headwind appears. Assuming I don't change the throttle position, what's going to happen? My speed is going to drop, right? Because given my constant throttle setting, my prior rpm (and vehicle speed) is not sustainable, if there is an increase in the drag forces that the engine is working to overcome.

By your definition, I was at 100% load, when I was loafing along at 2000 rpm. Because you said "!00% load is the point where if any more load is applied the rpm is not sustainable." Really?

If you have another question lets work on them one at at time

Good idea. Your latest post contains lots of other statements that make no sense to me, but let's put that aside for now.