When you click on links to various merchants on this site and make a purchase, this can result in this site earning a commission. Affiliate programs and affiliations include, but are not limited to, the eBay Partner Network.
why does nobody measure brake specific air consumtion? this would give an indication of an engines efficiency, as opposed to bsfc that is mixture dependent.
Imo you measure air consumption with fuel consumption. You know the stoich of your fuel, you know where you're at on lambda. Air is just math from there.
So, if you port an intake port and get 40 cfm improvement in flow, is that roughly 20 hp total? That 40 cfm multiplied over 8 cylinders is 320 cfm of additional airflow into the engine. We aren't gaining an additional 160 hp. So the 2.2 cfm per hp doesn't seem correct.
It's really complex. I went through the ringer trying to understand some of these things, and I don't think I ever got to the bottom of it. What I did learn is this:
1. Peak torque corresponds with peak cylinder fill / volumetric efficiency. You can chart torque vs RPM and VE vs RPM and see the curves overlay nicely. it makes sense, because torque is about cylinder pressure. Peak cylinder pressure maximizes force on the piston and makes peak torque.
2. Peak power corresponds with peak air usage. This is your CFM number. But there is a catch, in the sense that not all ingested air gets used. VE models how much air was trapped in the cylinder. Mass Airflow Sensor measures how much air went through the air pump, but is not an indicator of combustion efficacy.
3. Peak power is almost always at an rpm higher than peak torque, because even though each combustion event is "less efficient" vs events at peak torque, there are more of them per second, which compensates for the reduced output per combustion event -- UNTIL the combustion event loses more efficiency than the increased frequency can make up for. At which point your power peaks and falls off past peak.
4. Overlap is power. Overlap is also air getting through the cylinders without getting used. Badly confounds the measurements and the calculations.
5. CFM is unreliable, because air is compressible. CFM is a volumetric measurement. g/sec air is more reliable. I typically use g/sec air flow divided by 0.75 to get a rough bhp estimate. This works for peak HP / peak airflow numbers as a rough calculation, but I take it with a grain of salt until it is backed up with a dyno. Example, an engine uses 300 g/sec, calculates out to 400 bhp, and the dyno shows 325 rwhp. That works. Another example, engine uses 600 g/sec, calculates out to 800 bhp, chassis dyno shows 570 hp. either the airmass numbers are way off or the driveline is causing serious power losses. All this assumes your fueling and timing is dialed in very well and your O2 sensors are functioning correctly.
6. This is a callback to both numbers 3 and 5. Say you trap 1.00g air in a single cylinder fill at 4800 rpm. That's 320 cylinder fills per second (4800/60*4) for a total of 320 g/sec. That works out to pretty near 430 HP at 4800 rpm or 470 lbs of torque. Pretty good number for a 408. Now that same 408 is now at 380 g/sec at 6300 rpm, making 510 hp. Torque has dropped to 420, and cylinder fill has dropped to 0.90 g/sec. Make sense?
After going down this entire rabbit hole, I realized that the perfection of the airflow numbers didn't matter as much as I thought. You cannot adjust actual airflow, except via throttle position. You can tell the computer the airflow is different and all it does is change your fueling, because the whole purpose of the airflow numbers in the computer is to assist with fueling calculations. Basically, the fueling has to be right. Then, based on O2 feedback, you adjust the airflow model. Once this is done, you're really not reading actual airflow, you're reading modeled airflow - especially in a speed density controller. But you are reading ACTUAL fuel consumption numbers, because you know your pulse widths, injector size, and fuel pressures.
"4. Overlap is power. Overlap is also air getting through the cylinders without getting used. Badly confounds the measurements and the calculations.":
i have been thinking about this. so if air gets flushed through the chamber despite the high pressure in the exhaust port, would the oxygen sensor not report "lean"?
"4. Overlap is power. Overlap is also air getting through the cylinders without getting used. Badly confounds the measurements and the calculations.":
i have been thinking about this. so if air gets flushed through the chamber despite the high pressure in the exhaust port, would the oxygen sensor not report "lean"?
Yessir. The term “false lean” is commonly used by tuners at idle because of this condition, so making the AFR gauge read 15:1-16:1 at idle is very common, especially with larger camshafts that can have 30+ degrees of overlap. This condition will dissipate at RPM due to higher velocity in the runners, but obviously overlap is overlap and is always there.
To add to that, if you have one of those cams and it is idling at stoich, it is very likely very rich. Rich mixture contributes to idle surge, so you hear tuners discuss "idling lean" to smooth it out.
This is why the old way is still the best way. Tune idle by the spark plugs.
Flow numbers on heads are at 28" of vacuum. The engine at wot pulls more than that. Flow numbers on heads are just a glimpse into what you could potentially see when the head is on the engine. It's much like a chassis dyno, you can use it to compare changes but is hardly useful when you change the testing parameters or don't benchmark the combination. The 2.2 hp per cfm of Flow doesn't seem to translate well in the real world. If it did, then Flow would be all that mattered because more flow would make more power. Truth is, it's way more complex than that.
With two cams that are exactly the same other than say 10 degrees of overlap how much power would be on the table in an LS1?
Say two cams like a 227/235 on a 115 LSA vs 110 LSA? Would we expect the 110 LSA cam to make roughly ~10 to 15 more hp?
...10 degree over lap...similar to Tick's SNS2...1 degree overlap...similar to the LPE GT12
It's never that simple. Imagine a rubik's cube, or if you have one in front of you, even better. Now, change the blue face of the cube without changing anything else about the cube. Not possible. Every time you move the cube, you change the whole cube. That's why it's such a good puzzle. The whole reason the rubik's cube was invented (by an engineering professor named Rubik) was to teach his students that in complex systems, there is no such thing as "only changing one variable", because of variable interdependency. If you change a variable, you change the system.
In your question, you "only changed the LSA". Nope, you actually changed all four valve events and dynamic compression. It's almost impossible to perfectly compare overlap at equal durations, because of the number of variables that are affected. There is no possible way to just say something like "ten degrees of overlap is worth 15 hp" with any degree of accuracy. That's without getting into the difference between peak power and average power. Here's what I can say:
* Your 110 LSA cam will peak slightly lower in the RPM range, likely by a couple hundred rpm, but hang on longer after peak power
* Your 110 LSA cam will make more torque, and probably a good bit more torque owing to the earlier IVC / increased dynamic compression.
* Your 115 LSA cam might have a higher rpm peak power, but have a more anemic torque curve and fall off faster.
* Bottom line is both cams might make identical peak power, but I would expect the 110 LSA cam to make better average power from 3000 to 7000 rpm.
* To show it's real, here is a dyno chart from a holdener cam LSA test from back in 2015. All three cams were 232/242 with .624 lift. Blue line is the tightest LSA with most overlap. Highest torque curve, best average power, and you can see it carried the best past peak. But all three made nearly identical peak power numbers. You'll have to trust me that I went and found this after typing all of the above, but I don't blame you if you don't believe me...
Now, think this one over -- what if a better comparison is 227/235-115+2 vs 232/240-112.5+2? I know, I know, the duration is higher, but also in this case, the two most important valve events match - 46.5 IVC and 54.5 EVO. And dynamic compression matches. We traded one set of variable changes for another, and in this case, it's no question which cam will make more power. Both likely peak around the same rpm, but the bigger cam will definitely carry further and make more average power.
Last edited by Darth_V8r; 08-04-2022 at 06:48 AM.
Reason: Added dyno sheet
To add to that, if you have one of those cams and it is idling at stoich, it is very likely very rich. Rich mixture contributes to idle surge, so you hear tuners discuss "idling lean" to smooth it out.
This is why the old way is still the best way. Tune idle by the spark plugs.
Completely agree. I don’t care if the AFR gauge says 25:1…plugs don’t lie.
Want another monkey wrench? On a salt flat car drive lines are often 4 to 6 sections all different lengths to kill harmonics, even a mildly harmonic drive shaft can suck 50+ hp at 200 MPH..
Longer overlap will show up as vacuum loss. And driveability loss. But detrimental to power? Have you seen some of the really big cams on the big power cars?
Want another monkey wrench? On a salt flat car drive lines are often 4 to 6 sections all different lengths to kill harmonics, even a mildly harmonic drive shaft can suck 50+ hp at 200 MPH..