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.
This FAQ/sticky is being created to aggregate all the "tribal knowledge" around speed density tuning. Use this for posting in links to existing threads and for posting specific information directly to this thread. This thread is NOT to ask questions in. Questions looking for tuning help in this thread will be deleted. Post any tuning help questions in the forum.
We are taking delivery of 2014 LS7 Camaro soon, and I'm really interested finding all the extra power I can and still stay "stock appearing" for running in SS classes.
For max HP area under the curve, is speed density a good path to pursue for NA, stock cam engines? Or is it more for modified engines?
SD does not restrict you in your intake and air filter setup.
You are not constrained with the requirement to position the maf x inches from the throttle in a straight piece of pipe y inches in diameter of length z
SD does not have an airflow cap. Early ECM OS MAF tables were capped at 512g/sec which is only really in the 400rwhp range.
MAF flow response goes asymptotic at low and high flow rates, resolution is lost.
MAF is not good on transitions due to slow response.
Some of the issues are resolved or at least reduced on later OS and applications. Larger MAF's, Higher airflow limits etc.
I've been going thru some SD (3 bar) tuning using my HPTuner Pro. I've had some guidance from a good book and had some time in the seat getting it setup.
One topic that seems to be glaringly absent from the primer of this effort is getting the 'right' MAP sensor and putting in the Linear and offset values attendant with the MAP sensor. This was made evident to me on my car (5.3 purpose built for Turbocharging) when I naively bought 'a 3 bar map sensor' from the classifieds here, thinking a 3 bar sensor is a 3 bar sensor. Little realizing that, while they might plug, they don't necessarily play. In HPtuners, you can navigate to the 'engine diagnostics' menu item, then select the 'airflow' tab. This exposes a curious, non-intuitive dialog where you'll see a Linear and an offset value. Maybe it's the conspiracy theorist in me but this seems to purposefully left out of a lot of discussions because it has a major effect on what your engine management will do in terms of fueling administration.
So I plug in this sensor,...
While it plugs right into the harness, it has some mounting tangs that require trimming to allow it to push into the LS1 manifold.
Having converted my former MAF tune to a 3 bar SD tune, HPTuner defaults a 305 kpa into the linear and leaves the offset at 10.33. Nice, I thought. Should fire right up. And it did, but would immediately die. My wide band reported too lean to provide a number. After several lean backfires, it would struggle to idle, then die. If I ran to the front of the engine and sprayed starting fluid into the throttle body, It would rev up, confirming the lean fuel issue. Puzzled about this, I tried setting the injector constant lower, fiddling with the VE, etc. and I could barely get it to lean idle but it was weak and any gas pedal would make it weakly rev up unevenly.
Off to the internet where I came across these conversations about the linear and offsets.
Turns out some sensors are known as POS. Mine was one of them. Others are the old school 'brick' ones where they are bulky and won't mate to the LS1 manifold nor plug into the harness as-is. Then there's some that are plug and play. I ordered one of those and it's on it's way. While waiting for the new sensor to ship, I kept looking for the linear and offset that someone might have happened to find on my current sensor (AC Delco 12623671) and finally found one. It said the linear was 333.33 but only gave a calc for the offset. By playing around with different numbers in the offset while scanning the sensors readying at KOEO(Key On, Engine Off) I was able to attain 104kpa when I used an offset of 38 while the Linear was 333.33. I cranked it up and she idled fine. Plus, I could now give it gas and get a strong response when rev'ing it. I'll still wait for my new plug n play 3 bar sensor. It will have the Linear and offset of 305/10.33
A tid bit of experience for the penny jar of tuning....
When you lift from the throttle during cruise situations where peak cruise timing is used (often 36-44* btdc of timing region) it is generally acceptable to become leaner (15:1 or so). However, when you really REALLY lift, and hit that super high vacuum region, it is often best to give the engine a slight rich dip, around 13.8:1 in that region.
I have come to this conclusion after years and years of data logging. I used to leave the engine lean, 15.4:1 for example, coming off the throttle into vacuum from a cruise. However my data logs are showing me this is raising and holding my EGT up. In other words, as you cruise at 15:1 the engine is getting hot and lean, and if you stay there for 20-100 miles the engine will reach a steady state, which is quickly accentuated by the additional lean-ness of throttle lift which is often even leaner than the cruise state to begin with. This is facilitating the high EGT with no relief for the engine, such that when you TIP-IN back into the throttle, the engine is still lean and hot, and subtle spark knock can occur with these accented lean/hot situations.
By giving the last 2~ fuel cells near the ultra high vacuum state of the map this 13.8:1 rich dip, when you lift from the throttle nearly fully, instead of getting even hotter and leaner, the engine takes a cool dip in a spring for an instant. The same for when you get back into it, this will facilitate the "wall wetting" behavior, it will cost less accel pump enrichment to keep the engine happy and will help control EGT while cruising and reduce unwanted tip-in spark knock.
I cannot see how this "dip" as you call it which last for 1 second? or a millisecond is going to aid in and substantial cooling affect. and as you say this is when you lift off the throttle, so is it clear to say that someone should be constantly be lifting of the throttle at a steady state to cool the cylinder down?
I cannot see how this "dip" as you call it which last for 1 second? or a millisecond is going to aid in and substantial cooling affect. and as you say this is when you lift off the throttle, so is it clear to say that someone should be constantly be lifting of the throttle at a steady state to cool the cylinder down?
As you lift from the throttle the base map generally becomes leaner. An engine running closed loop will mask the reduced fueling because the computer will compensate to hold the A/F steady. However, the computer works slowly/gradually and there is a delay between adjustments and possibly a learned value for each cell (depends on the ECU). The learning value will change depending on whether the cell is experiencing an OFF throttle transition, or an ON throttle transition. In other words, if you always are coming off the throttle at 3200rpm and 15" of vacuum+ the 'learning value' will always be too lean in those cells, because the computer only has experience with throttle position reduction (as you reduce throttle position, the wideband will show rich condition even if the cell is perfectly fueled, and the computer can remember and compensate in the future, causing the cell to be 'too lean' for coming back into the throttle).
Now, look at what happens when we get back into the throttle, and watch the wideband change it's mind, it will go leaner for an instant. The cell to which the map moves when you step back into the throttle does NOT have enough fuel in it for the transition (otherwise it will be too rich for a steady state) thus the need for an ACCEL pump function, to compensate for the pressure drop. There is often a delay, or insufficient accel pump action for an instant where the wideband will read lean, especially when throttle position moves only slowly (very slow moving throttle = very little accel pump action because of noise filtering and sensitivity options). This is where the build of the engine comes into play. A 'big camshaft' or 'big intake/head' combo will yield a larger pressure drop for less throttle position change, which requires a much larger accel pump function, which the engine will rarely receive since the throttle position is not moving fast enough (we are still talking about slowly moving throttle position, not a quick STAB but a very slow gradual step down). This is a very difficult problem to solve since we cannot add fuel to the cells themselves (The engine would be rich during steady state if it was stoich during a transition) and we cannot increase accel pump function because the TPS is not sensitive enough, or the throttle body is too large, or both. So how can we protect the engine from this transition state, where the wideband shoots lean as we tip back into the throttle?
By keeping the high vacuum rows slightly rich, the engine will naturally transition into a 13.8:1 or 14.2:1 state as we roll off the throttle into the high vacuum situation (imagine coming off the throttle at 100mph or 120mph gradually to slow down) you want to see 14.2:1 there, not 14.7 or 15:1 in that state. In fact, for all high speed 'cruising' at 90mph+ you do not want to be stoich with turbo, or other high performance build, because this situation is generally coupled to a higher EGT than is typical for a cruise (at say, 70mph) and more temperature is being relieved by engine as fast as it is produced (hopefully). If the high vacuum row is slightly rich, then, when you tip back into the throttle, the cylinder was already 'wet' with fuel (slightly rich), helping to prevent the lean tip-in that nearly every engine experiences with slowly moving throttle position at high engine RPM as the fuel map moves back into the 'cruise region' which is tuned leaner to begin with. If you decide to stab the pedal down the engine would be fine; but since in our example the throttle is moving only slowly, right into a lean 'cruise' portion of the fuel map during a high RPM cruise (100+mph perhaps) this can be a dangerous situation where EGT(cylinder temp) is higher, the engine tip in is lean and there is very little if any accel pump action. What many do not realize is that many engines have a small reservoir of fuel in each cylinder that helps it stay 'wet' and that the cylinder can actually take a fair amount of fuel (over the course of a couple milliseconds, quite a few cylinder events can occur) before this 'wetness' is replaced; in other words, if you completely dry the cylinders up, then tip slowly back into the throttle right into the cruise portion of a fuel map the first couple injector events are not going to be as effective as they would have been if the cylinder was already 'wet'. The A/F will be even leaner for an instant and you will probably not catch it on the wideband. By keeping the high vacuum rows of the fuel map 'wet' we can help the cylinder stay 'prepped' while decelerating with our foot on the gas pedal; likewise if we close the throttle all the way completely and the engine goes into fuel cut mode (dry cylinders) then just barely touch back into the throttle, there is no pump shot action so our first couple rows of 'slight richness' will help 're-wet' the cylinder as you slowly step back into the throttle into the cruise portion. The worst thing you can do getting back into the throttle is go to 16 or 17:1 for an instant which many engines do since their final rows are programmed leanest (around 14.8:1) which translates into 17:1 or so when the cell is transitioning through to more open throttle positions (as opposed to steady state). This is not an issue with a mostly stock engine, we can see 18:1 and be fine with such a high vacuum situation (many do), however when the engine is more capable (high output parts such as increased static compression ratios and camshafts which improve dynamic compression ratios or boost which also increases compression ratio rapidly) of going from a very high EGT/vacuum cruise where the chamber/valves/etc is able to get very HOT (for economy reasons, there is often some desired temp increase during a cruise) to suddenly transitioning into a high output state (more powerful engines), the same temperature rise can become a disaster for a cast piston for which only a couple of poor/hot cylinder events may put a scar, weakening the piston or possibly fracturing it over time. I hope all this helps explain it better.
Everything you just wrote is completely irrelevant to tuning these cars with a stock pcm. The PCM utilizes a DFCO (and CFCO if applicable). This forces the injectors off and and severely decreases timing to help with fuel economy.
Secondly they use an X-tau transient fueling that models fuel to port wall impact and evaporation on the port wall at various throttle positions. It is also able to proportion this effect between accel and decel. If modeled appropriately neglects the need for a "pump shot" effect. This is a mapping of the dynamic environment within the engine. This also assumes your steady state airflow calibrations are done correctly.
Everything you just wrote is completely irrelevant to tuning these cars with a stock pcm. The PCM utilizes a DFCO (and CFCO if applicable). This forces the injectors off and and severely decreases timing to help with fuel economy.
Secondly they use an X-tau transient fueling that models fuel to port wall impact and evaporation on the port wall at various throttle positions. It is also able to proportion this effect between accel and decel. If modeled appropriately neglects the need for a "pump shot" effect. This is a mapping of the dynamic environment within the engine. This also assumes your steady state airflow calibrations are done correctly.
The fact that the ecu already contains these features tells me that transition tuning is extremely relevant, to understand what is going on. Just because the ECU already has some kind of 'helping feature' to assist with exactly the same thing I am trying to help with is no reason not to give the engine even more additional help with the exact same issue. Attacking the same problem from multiple fronts is how you work out the problem.
Secondly they use an X-tau transient fueling that models fuel to port wall impact and evaporation on the port wall at various throttle positions. It is also able to proportion this effect between accel and decel. If modeled appropriately neglects the need for a "pump shot" effect. This is a mapping of the dynamic environment within the engine.
This means that the rate of change of the engine pressure, or engine mass flow, will reflect whether the ecu is adding or removing fuel to whichever cell the fuel map is reading from. It is simply some number in additional, if the injector pulse were to be 1.000 now it is +/- some number, all fancy talk aside. So when I am suggesting we add this directly to the fuel cells themselves, it will do the same thing. It will not be as economical because there will be no reduction in fueling when the rate of change of pressure/mass is negative(higher vacuum) and small, and the engine will run too rich, but that was my point: let the engine stay rich when rate of change is negative and small. The next step that direction is fuel cut; the next step the other direction is power. The stock settings reflect stock engines, pay more attention to this setting if you are modifying the engine. Perhaps many settings are left untouched in capable computers because nobody knows exactly what they do or why they exist.
I'm not gonna waste my time arguing with you. You think that because you've done tuning on different platforms with different controllers that is applicable. While it should be applicable you consistently show that you do things in the wrong manner. The advice you give is generally bad advice, and it's certainly bad advice in the context of tuning the stock PCM.
I'm not gonna waste my time arguing with you. You think that because you've done tuning on different platforms with different controllers that is applicable. While it should be applicable you consistently show that you do things in the wrong manner. The advice you give is generally bad advice, and it's certainly bad advice in the context of tuning the stock PCM.
After tuning a 10~ different kinds of engines and perhaps 10-15 different types of computers I have developed a system that works quickly, enables the user end base map to do the job of multiple sub routines within the computer without having to make tiny adjustments to each option over the course of several hours. Lots of people try to say my advice is POOR but have no evidence nor can they provide even the slightest REASON why; I think that speaks for itself, your post is a perfect example. "Hey you are wrong! I dont know why but , it just is" most likely because you are unfamiliar with the concepts I am providing, and unable to connect the dots between the ECU thought process and the actual combustion events that occur inside the engine, else you would see what I propose is an entirely safe alternative to traditional bells and whistles "window" chasing. For god's sake, i am trying to get folks to add fuel to high vacuum regions! There is nothing un-safe or un-reasonable with this concept.
At the end of the day it doesn't matter what computer is in the car, all that matters is the A/F and timing is correctly applicable to whatever the application is. My method is foolproof through a vast majority of applications/computers and WILL WORK regardless of whether you are tuning a stock PCM or aftermarket Haltech style computer, and it will work QUICKLY to get the car driving safely, which is perhaps the single biggest benefit. And there is no way to prove that one method is any better over the other; the only thing I can prove is my chemistry/tuning background and the theory behind what I am suggesting is backed up by hundreds of thousands of miles across multiple platforms who's engines spark plugs come out looking almost new while the engine does it's job.
I'm not gonna waste my time arguing with you. You think that because you've done tuning on different platforms with different controllers that is applicable. While it should be applicable you consistently show that you do things in the wrong manner. The advice you give is generally bad advice, and it's certainly bad advice in the context of tuning the stock PCM.
After tuning a 10~ different kinds of engines and perhaps 10-15 different types of computers I have developed a system that works quickly, enables the user end base map to do the job of multiple sub routines within the computer without having to make tiny adjustments to each option over the course of several hours. Lots of people try to say my advice is POOR but have no evidence nor can they provide even the slightest REASON why; I think that speaks for itself, your post is a perfect example. "Hey you are wrong! I dont know why but , it just is" most likely because you are unfamiliar with the concepts I am providing, and unable to connect the dots between the ECU thought process and the actual combustion events that occur inside the engine, else you would see what I propose is an entirely safe alternative to traditional bells and whistles "window" chasing. For god's sake, i am trying to get folks to add fuel to high vacuum regions! There is nothing un-safe or un-reasonable with this concept.
At the end of the day it doesn't matter what computer is in the car, all that matters is the A/F and timing is correctly applicable to whatever the application is. My method is foolproof through a vast majority of applications/computers and WILL WORK regardless of whether you are tuning a stock PCM or aftermarket Haltech style computer, and it will work QUICKLY to get the car driving safely, which is perhaps the single biggest benefit. And there is no way to prove that one method is any better over the other; the only thing I can prove is my chemistry/tuning background and the theory behind what I am suggesting is backed up by hundreds of thousands of miles across multiple platforms who's engines spark plugs come out looking almost new while the engine does it's job.
Blah blah blah, who cares if it's right! It works and is quick.
03-26-2014, 10:04 AM
Speed Density Tuning - FAQ 
