: Air intake and airflow



Aurora40
03-23-06, 11:15 PM
Well, the thread on the eBay cone filter was interesting. So I thought I'd start a new one with more of a discussion than about a specific eBay item. Some people said that without a larger TB, or without substantial mods, a different intake won't matter.

I have to wonder what that is based on?

I've been monitoring the PCM data lately and it's kind of neat, and also got me thinking. The MAF will tell you the mass of air that flows through it. So this would be a valid way of telling how much an intake helps, and will account for heat, as hotter air has less mass per the same volume and pressure.

The only problem is I don't have a cone, nor any other intake, to try out.

But here is the airflow vs rpm. This is through 1st and 2nd gear, though I wasn't WOT in 1st as I was taking a right turn. Note that the airflow falls pretty level by about 5,000 rpm. To me this would seem to say the intake is restricted. The MAP falls by a whole "Hg also, though it seems to have pretty rough granularity so it's hard to say how accurate that is.

http://members.aol.com/aurora402002/misc/scan/auroraairflow2.JPG

That doesn't mean the intake itself is the restriction, but would seem to indicate something is. It could be the TB, the air filter, the design of the box, anything. And this is only to 5,800 rpm, not up to 6,500.

Anyway, I thought it would be interesting to have some intelligent discussion of it. So fire away. :)

weister42
03-24-06, 12:04 AM
Nice to see someone taking a scientific apporach to the N* intake discussion, some guy called me an idiot for my intake post but whatever. So you're saying the factory air intake system starts to max out by 5000 rpm?

96-deville-man
03-24-06, 12:10 AM
Nice to see someone taking a scientific apporach to the N* intake discussion, some guy called me an idiot for my intake post but whatever. So you're saying the factory air intake system starts to max out by 5000 rpm?

yes if im reading that graph correct then yes it maxs out at about 5000rpm. you can get a cheap cone filter from auto zone and use the stock tube. to test.

N0DIH
03-24-06, 12:59 AM
I just put a K&N on my 96 Burb today, it DOES has improved SOTP feel in the low end. It was replacing a fairly NEW Wix filter (the larger of the 2 I can run on my Burb, so more surface area).

My theory is that the engine is more sensitive to air flow when air flow is slow. I can't prove it without a flow bench and some testing that way. Even then it would likely be subjective.

So if a paper element can flow say 350 cfm, and the same size K&N can flow at least the same, that flow is at a certain depression of mercury. Say 28 inches. BUT, if we are at 2 inches of mercury, such as low air flow times when the car is at slow speeds or at idle, I am curious what the air flow differences are. So it might just take more throttle to overcome the filter restriction of the paper vs K&N to get the same amount of air to start actually MOVING through it.

Does anyone follow?

davesdeville
03-24-06, 04:23 AM
Do another test, immediately pull the filter out, do it again. That'll tell you if the filter is the restriction.

eldorado1
03-24-06, 08:57 AM
Finally, intelligent discussion. :bouncy:

Your airflow curve shows about what you'd expect from a 4.0L.

4L*1000*(1/2.54^3)*(1/12^3)*(1/2)*80%VE*5800rpm = 330CFM

That's about 25#/min

Note that just because the airflow levels off, that it doesn't mean the air intake is becoming restricted. More than likely, your VE is falling off, which is more a function of intake runner length, and cam specs. In other words, as you increase your rpm's, the engine fights more and more with itself to get air into the cylinders. VE peak on the northstar (don't know about your 4.0) is 4400 rpm. As somebody pointed out, at this point it's 92% efficient. So for every 2 revolutions, it's pulling in 4.2L of air. However, as you pass 4400rpm, the VE falls off, as it starts to fight with itself. So at 6000rpm, it might only be 80% efficient, it's pulling in 3.7L of air every 2 revolutions. But since it pulls in less air more often, it's about the same amount of airflow.

Now that's not to say that there is no restriction. IMHO, any time you see the air pressure fall off in the intake, there's probably a restriction somewhere. It would be nice to see your MAP sensor reading in kPa... Just for a comparison, I have a 80mm throttle body and a k&n filter. And before anybody floggs be for not practicing what I preach - it's the only intake that would fit. I'm working on getting a "real" filter. Anyways, at wide open throttle, there is only a 2kPa drop between 1500rpm and 5800rpm. I believe that's probably close to 0.6" Hg.

However, since you have a 4.0, which probably has the same intake as the 4.6, it's likely the 4.6's would have a much larger pressure drop completely stock. In other words, if you started with 250hp and 95kPa at WOT, and got larger/better everything and gained 3kPa at WOT, to 98, in theory you could pick up 8hp.

That was a lot of rambling... hopefully it makes some sense

Aurora40
03-24-06, 09:58 AM
Note that just because the airflow levels off, that it doesn't mean the air intake is becoming restricted. More than likely, your VE is falling off
Yes, I'd thought about that. However, I wouldn't think it would fall off so much that the airflow at 5,000 rpm is the same as at 6,500 (not on this graph, but assuming it stays level). But I don't know for sure or have any past data to go by. It's possible that is all that illustrates.

The MAP value would definitely point to a restriction of something, though it could be the TB and not the air filter. I have a K&N air filter already. I have thought of trying it with the filter pulled, or possibly taping the filter to the top end and leaving it open, to elminate the box from the equation. Oh, and yeah, I believe the airbox is identical to the 4.6L one.

What seems odd to me is the rigid steps the MAP takes. I can try selecting fewer parameters to scan to see if it updates quicker. But it stayed at one exact level, then fell and stayed at another exact level. That just seemed odd, and makes me wonder how much value the readings have at any rpm.

Your car falling off only 2kPa, that seems excellent. That to me would suggest you are not limited by your intake. If you have a cone, though, it wouldn't account for any inefficiency caused by hot air (if that is even a problem with cones, it may or may not be), as it just measures pressure not mass. Though watching the IAT would account for that.

As an aside, I have an airbox wrapped in reflectix. I got some flack about that here, and the idea was the temps will drop in an instant due to the fact plastic holds so little heat. What I've seen is temps when stopped (even for short periods) go up to over 20F higher than ambient, and they only drop back about 1/2 of that through 1st gear. So if the temp would go up even higher without the reflectix, it would take even longer to come back down. It certainly isn't instant (even if the actual air temp change was, the sensor is what will feed into the fuel delivered).

Aurora40
03-24-06, 10:07 AM
Your airflow curve shows about what you'd expect from a 4.0L.

4L*1000*(1/2.54^3)*(1/12^3)*(1/2)*80%VE*5800rpm = 330CFM

That's about 25#/min

Oh hey, I had a question about this. How can you take CFM and turn it into lbs/min? Doesn't the lbs/min depend on the pressure and temperature of the air passing through at 330CFM? Not trying to be critical, I just wasn't sure how to accurately compare cfm with mass. Actually if I could accurately compute cfm, that would help to spitball whether certain sized openings are adequate for the motor, like the 3.5" entrance to the airbox, the 75mm TB, etc.

eldorado1
03-24-06, 10:49 AM
Doesn't the lbs/min depend on the pressure and temperature of the air passing through at 330CFM?

Well you know the lbs/min from the mass airflow meter. I just assumed the WOT run was made at 70 deg F, at sea level. It's probably not an accurate assumption since it's winter, I have no idea where you live (it IS 70 here ;) ), and you didn't provide any IAT readings. But yeah, just multiply lbs/min by the density of air at whatever temp/pressure you're at, and you can get volume/min.

As for the airflow falling off and all... Find a dyno curve. If the HP is level from 5000 to 6500, you can assume the airflow stays roughly the same.

eldorado1
03-24-06, 11:02 AM
If you want some more numbers to play with... from McCowden's 275hp 4.6 datalog:



MAP Airflow RPM
(kPa) (lb/min)
99 4.967 1998
98 8.388 2443
98 11.292 2527
98 11.806 2557
98 12.087 2748
98 12.631 3461
97 16.631 3695
96 18.179 3875
96 19.489 4017
96 20.588 4177
95 21.971 4299
95 22.539 4441
95 23.173 4586
94 24.091 4733
95 25.019 4898
94 25.848 5032
93 26.037 5173
93 26.488 5304
93 27.019 5451
93 27.384 5553
93 27.506 5638
92 27.466 5660

Hopefully it's alright to post these.. BTW, try to see if you can change the units on your MAP output.

Aurora40
03-24-06, 11:35 AM
Yeah but it is the pressure inside the system, not ambient pressure. The MAF is behind the air filter. And the temperature is not ambient either. Even the MAP pressure won't tell you the right pressure since that is on the other side of the TB, which could also have a pressure drop. :)

Aurora40
03-24-06, 11:37 AM
If you want some more numbers to play with... from McCowden's 275hp 4.6 datalog:
Is there a thread that came from? Looks like his airflow increases from 5,000 to 5,500 rpm. It isn't flat.

I think to change the units on mine, I'd have to change all of them to metric, not just the MAP. So it'd be rotations per metric second and stuff like that... ;)

eldorado1
03-24-06, 12:32 PM
Yeah but it is the pressure inside the system, not ambient pressure. The MAF is behind the air filter. And the temperature is not ambient either. Even the MAP pressure won't tell you the right pressure since that is on the other side of the TB, which could also have a pressure drop. :)

If you REALLY wanna get technical, humidity will affect the density as well. You can use the barometer function of the MAP to check the current atmospheric pressure. Or just check it with the engine not running.

There's no thread the data came from, he just sent me a datalog a while back...

chevelle
03-24-06, 12:49 PM
Keep in mind that air flow "restriction" at max RPM might have a slight affect on the advertised or dyno horse power when the engine is held at that point but it will have minimal to no effect on the absolute performance of the car. The engine spends very little time in the RPM range (peak RPM) where the air flow might start to get restricted so the time the engine is being "strangled" is very small. Since the acceleration of the vehicle is across the broad range of the operating RPM that peak power reading and the airflow at that point is not as important as you might think...except for bragging rights.

eldorado1
03-24-06, 01:59 PM
Since the acceleration of the vehicle is across the broad range of the operating RPM that peak power reading and the airflow at that point is not as important as you might think...except for bragging rights.

Exactly why I'm always pointing out that a larger throttle body is better than a k&n filter - say they both make 6hp. (I don't know what a k&n makes) The larger throttle body makes more hp everywhere in the power band, where the k&n only makes more power at the hp peak. Same peak hp numbers, only one will out accelerate the other. ;)

It's all about area under the torque curve.

N0DIH
03-30-06, 08:24 PM
Here is one of my airflow runs (peaks around 225-255 gm/sec on the datalog, this one was around 228.25 peak)

Interesting is I see peak airflow at 5400 rpm, not at 5900 rpm. It shows that something else is the restriction (cam, heads, exhaust, intake, etc). Likely the heads and cam combo, as it didn't peak and hold, it peaked and fell off.

My VE chart from TunerCat is also attached. Interesting is the LT1's (Corvette/Z28/TransAM with alum heads and larger cam) and the B/D Body (Caprice/Roadmaster/Fleetwood with smaller tigher lobe sep cam and iron heads that outflow the alum heads) all have the exact same VE map.

How do I convert CFM to Lb/hr or min?

My LT1 is running around a peak of 255 gm/sec, or 33.73 lb/min if I calculate that right.

To make 6 hp, they probably dyno'd it with a 50K paper filter that ran in the Baja 1000 when it was new and has a collection of dust and dirt from around the world.... and then put on a new K&N. The car mag's do it all the time to support their claims. Like 60 hp on a LT1 from bolt ons. Bolt ons replacing crap old parts that are worn out.

eldorado1
03-30-06, 09:03 PM
My VE chart from TunerCat is also attached. Interesting is the LT1's (Corvette/Z28/TransAM with alum heads and larger cam) and the B/D Body (Caprice/Roadmaster/Fleetwood with smaller tigher lobe sep cam and iron heads that outflow the alum heads) all have the exact same VE map.

My LT1 is running around a peak of 255 gm/sec, or 33.73 lb/min if I calculate that right.


Did you remove the screen from your MAF sensor? The high rpm values are very noisy. Also note that the VE table doesn't tell you actual VE. It just tells you the shape of the VE curve (assuming it's tuned for one AFR, if it's not then all bets are off). You are not at 92% from 4500 up, your VE peak is actually at 3600rpm, and it's probably around 85%.

Aurora40
03-31-06, 12:23 AM
Hey, just FYI, but with Excel you can use two vertical axis'. It will make the smaller numbers like RPM/100 and TPS volts a lot more readable.

I've been playing around with TTS Datamaster more, the laptop stuff really puts scan tools to shame. Though it's a real pain putting all that laptop junk in the car. My LT5 drops from about 101Kpa to about 98Kpa from low rpm up to 7,000. It's pretty non-intake-restricted. It's not a MAP car, though, so no airflow to compare with.

eldorado1
03-31-06, 10:09 AM
My LT5 drops from about 101Kpa to about 98Kpa from low rpm up to 7,000. It's pretty non-intake-restricted. It's not a MAP car, though, so no airflow to compare with.

:holycrap:

somebody's addicted to dohc v8's.......

N0DIH
03-31-06, 02:20 PM
No, I won't remove the screen, no matter what, it is there for a good reason. I have a Vortec MAF I can put it temp (my wife will kill me if I steal it...) and I have the tables on it to make it correctly setup.

I was wondering if VE changes from WOT to part throttle, no one has ever seemed to have a clue. Most stuff is only measured at WOT.


Did you remove the screen from your MAF sensor? The high rpm values are very noisy. Also note that the VE table doesn't tell you actual VE. It just tells you the shape of the VE curve (assuming it's tuned for one AFR, if it's not then all bets are off). You are not at 92% from 4500 up, your VE peak is actually at 3600rpm, and it's probably around 85%.

Katshot
03-31-06, 02:27 PM
Max VE should be at WOT and max torque. RPM and throttle angle are definately variables.

N0DIH
03-31-06, 02:38 PM
So if VE on the chart is say (arb number) 90% @ 4000 rpm @ 90 kpa, and I am at WOT and only getting 90 kpa, but if I am at part throttle, and getteing the same values, is it the same? The airflow through the MAF, is it the same or higher?

If I am just asking dumb questions, feel free to slap me. But no one has ever been able to answer, I get I don't know's to, we don't test that, yada yada....

I am guessing all is just plain offset by the MAF values actual read. But is the VE still the same?

Reason I ask, is peak torque (lowest BSFC) typically is best fuel economy too. BUT, peak torque is read at WOT. We don't drive at cruise at WOT. So, if we get peak torque @ WOT @ 2800 rpm, I can agree we get best efficient work (towing, etc) at 2800 rpm @ WOT. But cruise is different, is the VE more efficient at lower rpms NOT at WOT?

I feel like I am chasing my tail.....

eldorado1
03-31-06, 03:15 PM
So if VE on the chart is say (arb number) 90% @ 4000 rpm @ 90 kpa, and I am at WOT and only getting 90 kpa, but if I am at part throttle, and getteing the same values, is it the same? The airflow through the MAF, is it the same or higher?

If I am just asking dumb questions, feel free to slap me. But no one has ever been able to answer, I get I don't know's to, we don't test that, yada yada....

I am guessing all is just plain offset by the MAF values actual read. But is the VE still the same?


I'm not entirely sure what your question is... There are a couple issues with it... First off, the MAP sensor measures manifold vacuum, or engine load. A value of 90-100 kPa means your engine is putting out all it can. It would be very difficult to see 90 kpa at part throttle. It would probably be impossible without a really steep hill (like 90 degrees straight up), or an engine dyno. It may be impossible period, because 90kpa means the engine needs a LOT of air to sustain that load. That can only happen with the throttle wide open. Actually, thinking about it more, you have to have the throttle basically wide open for that condition to exist.

Now volumetric efficiency is just how much air the engine is taking in vs how much it could theoretically take in at that RPM. So any time there is a throttle restriction (i.e. not at WOT), the engine is going to be less efficient than it would if it were unthrottled.

Volumetric efficiency is directly related to torque output. If you have a VE curve, you can stick that in a graph with different numbers for a torque curve.

Now, if you're on an engine dyno, and can vary the load however you want at any rpm... If you hit the same point on the VE curve, the airflow through the engine will always be the same. In other words, if you turn up the load to 90 kpa, and run it to 4000rpm, you will always have xx amount of airflow through the engine.

VinnyT
03-31-06, 03:25 PM
Here:

http://www.caddyinfo.com/airboxvscone.htm

Aurora40
03-31-06, 05:49 PM
Here:

http://www.caddyinfo.com/airboxvscone.htm

That always seemed pretty questionable. Run your car 3 times with no changes and see how close the 0-60 times stay. I'm not saying cones work, just that that doesn't really prove they don't (to me). Also, the power computed by the device climbs from 215 to 219 for the cone (or falls to 212 depending on the run you look at), yet no mention or explanation of that.

N0DIH
03-31-06, 06:42 PM
I agree, statistically 3 is insignificant. You need to make more like 18 back to back runs with no changes, then make a change, run it for 18 runs, then change it back and run 6 more runs to verify the baseline to honestly prove that a change was effective.

However, 18 is not necessarily practical, and hard on the car, and can be $$ with the fuel used and tires worn.


That always seemed pretty questionable. Run your car 3 times with no changes and see how close the 0-60 times stay. I'm not saying cones work, just that that doesn't really prove they don't (to me). Also, the power computed by the device climbs from 215 to 219 for the cone (or falls to 212 depending on the run you look at), yet no mention or explanation of that.

Aurora40
03-31-06, 07:15 PM
:holycrap:

somebody's addicted to dohc v8's.......
Only ones that set endurance records... :) :cheers:

In regards to the VE thing, here's my stab at it. May be wrong, maybe not. I'd think at, say 4,400 rpm, the VE is the same no matter what. It is say 85% efficient. But at part throttle, the pressure in the intake is maybe 50Kpa. So the motor will pull in 85% of that, and thus make less power. But it's still the same efficiency.

I wouldn't think VE has much to do with economy, though. Your car wouldn't get better mileage if you geared it up so that it was turning 4,400 rpm at 60mph all the time. The friction/motion/etc losses would be a lot greater. If at 1,500 rpm cruising, the motor is 60% efficient, that just means how much air it pulls in. Not efficiency of burning fuel. So you, say, have the throttle open such that it is 40kpa at 60% efficiency, and the car makes the 30hp it takes to cruise (all total guesses). Say you were at 4,400 rpm and 85% efficiency to cruise (ignoring bigger losses), you'd just close the throttle more so it's at 28kpa and 85% efficiency. You'd pull in the same amount of air, burn the same amount of fuel, and make the same 30hp needed to cruise.

Actually that last part doesn't account for the increased number of rotations, but I think you get the idea. I can't picture VE having much to do with cruising efficiency. But I could be wrong.

Katshot
04-01-06, 09:08 AM
I think there's some confusion here. VE (Volumetric Efficiency) and ME (Mechanical Efficiency) are two different things. And VE is only a comparison of how much air the engine is moving at any given moment compared to a theoretical maximum that it can move. So since that "theoretical" maximum is calculated at the point of maximum torque, the highest VE will also be at that point. It will be lower at any other point.

N0DIH
04-01-06, 06:09 PM
How is each calculated or determined?




I think there's some confusion here. VE (Volumetric Efficiency) and ME (Mechanical Efficiency) are two different things. And VE is only a comparison of how much air the engine is moving at any given moment compared to a theoretical maximum that it can move. So since that "theoretical" maximum is calculated at the point of maximum torque, the highest VE will also be at that point. It will be lower at any other point.

Katshot
04-02-06, 07:54 AM
How is each calculated or determined?

Here's a good site concerning VE:
http://www.installuniversity.com/install_university/installu_articles/volumetric_efficiency/ve_computation_9.012000.htm

As far as ME, I'm not sure there's any way to "calculate" it. Mainly, it's things like frictional loss as I recall, but it has nothing to do with VE. They're two different animals.

N0DIH
04-08-06, 11:14 PM
What would cause the high rpm values to be noisy? vibrations? I have the MAF connected to a K&N cone (stock Vortec 350 truck, like 9inches x 6in round), which isn't strapped down, so the filter/MAF can vibrate some.

If I should secure, let me know.


Did you remove the screen from your MAF sensor? The high rpm values are very noisy. Also note that the VE table doesn't tell you actual VE. It just tells you the shape of the VE curve (assuming it's tuned for one AFR, if it's not then all bets are off). You are not at 92% from 4500 up, your VE peak is actually at 3600rpm, and it's probably around 85%.

eldorado1
04-08-06, 11:46 PM
No, vibration wouldn't do it. Having the filter so close to the MAF without an intermediate tube of some sort might induce some kind of turbulence though... If you've got the stock parts, swap back and see if it goes away.

N0DIH
04-09-06, 12:00 AM
I do have it, and can give it a try.

Peak airflow the other day was 233 gm/sec. I will check the other data logs and see how well they show, see if it is settled down or not.

What is the common datalogger people use? I am using Freescan as my 20 runs on TTS Datamaster are expired. I do like TTS better, but I need to evaluate which works best for me. So far, there is additional things like EGR on/off, Canister Purge on/off, and trans stuff that Freescan doesn't have, which makes me lean that way.

Is there any other dataloggers out there that I should look into?

I do see Freescan pause from time to time while I am driving, is this common? It does seem to duplicate the entries while it is paused. So I am starting to up the process priority to help with it.

eldorado1
04-09-06, 12:06 AM
If you can find an ALDL definition file for your car, you can try Tunerpro RT. Otherwise, try playing with the timing in freescan. I think I have mine set to either 10 or 100mS.

chevelle
04-09-06, 01:08 AM
Some of these comments are confusing wide open throttle conditions and part throttle conditions and unthrottled conditions.

It is entirely possible to be at part throttle...i.e..say 50% throttle...and have the engine unthrottled..i.e..the MAP is very nearly equal to the barometer or vacuum in the intake is zero.

At 3000 RPM the engine will pump much less air than at 6000 RPM. So, at 3000 RPM it may be necessary to open the throttle only 50% to achieve an unthrottled condition while at 6000 RPM the throttle will have to be wide open or 100% to achieve an unthrottled condition.

This is very easy to see if you hook up a vacuum gauge to the intake manifold and drive the car. Any time the vacuum falls to zero the engine is unthrottled. You'll see zero vacuum or very low vacuum just cruising down the highway and climbing a relatively steep hill on the interstate. With the engine cruising at 2500 RPM it doesn't take much throttle opening to unthrottle the engine or go to zero vacuum.

Any time the vacuum is at zero or near to it the volumetric efficiency is going to be determined purely by the pumping efficiency of the engine. As the throttle is closed and vacuum increases the volumetric efficiency is going to drop as the engine is being throttled.

The volumetric efficiency information the PCM uses is programed into the calibration chip based on dyno measurements at each speed and load point. The calibrator provides that info to the PCM via the calibration. It is determined empirically and then loaded into the cal. It is actually possible to determine the volumetric efficiency of most part load points by just allowing the closed loop control system to run the engine at 14.7:1 and then keep changing the VE value for that speed and load until the closed loop system is not having to do any correction. Then the VE provided in the calibration will be correct.

The speed density calculation calculates how much air is being pumped by the engine based on the RPM, manifold pressure (or vacuum), displacement of the engine, volumtric efficiency of the engine at that speed and load, and the temperature of the intake air. The volumetric efficiency of the engine is going to be nearly the same (about 90% or so) at all RPM's if the engine is unthrottled. The amount of air the engine is pumping will change as the RPM changes but the volumetric efficiency will not change much at all. Some but not much. The biggest changes in VE values are due to throttleing as the throttle is closed further and further and the engine pumps less and less air due to the throttling effect of the throttle blades.

The MAF signal can be grossly affected by changing the air cleaner and duct configuration before and after the MAF. The MAF is VERY sensitive to stratification of flow inside the sensor. Each application has specfic MAF correction curves built in depending on the design of the airbox and air ducts and how the designs affect the flow inside the MAF. Taking the screen out exacerbates this problem. Not that it won't run without the screen but the signal to noise ratio on the MAF goes all to hell ususally when the induction system is modified and/or the screen is removed.

N0DIH
04-09-06, 01:23 AM
So installing the K&N in place of the factory air box could be causing some MAF instability at higher airflows? If I was to have say 6 inches of ducting before the MAF (after the K&N) would that stabilize it more?

So back "in the old days" when racers would use a air horn or some sort of air smoother into the carb like the Snub Stack, would be smoothing the airflow in. So this likely would be a good thing on the MAF, assuming you could calibrate for it. Or would you need to?

I have noticed on my data logging times where I am at say, 1500 rpm, and see MAP values around 80, which seem odd that it would be seeing that much load or lack of vacuum. So at this point in time, the VE is actually low (as seen on the VE maps).

In my driving of my old 455 in my T/A, I would see relatively high vacuum (low MAP if I had one) readings almost all the time, except when I would get on it. Larger engines I guess aren't as volumetric efficient due to being throttled almost all the time, where as a smaller engine would be much less throttled more of the time in comparisson.

Is best mpg or lowest fuel consumption at high MAP or low MAP? Or can you really even make a statement like that?

i.e. If I am running my engine in 4th gear at 30 mph, TCC locked, and seeing around 1000 rpm, my MAP would be high (low vacuum), but if I was running it in say 1st gear, I would be throttled, seeing a high vacuum (low MAP), even though I am at a higher rpm.

Which would use more fuel? Or is it technically the same? With respect to the friction of course.