Landtank MAF surprising scangauge results

This site may earn a commission from merchant affiliate
links, including eBay, Amazon, Skimlinks, and others.

I already cite site the 4in Lingenfelter MAF scaling in this regard back in post 444. I'm sure you just didn't take the time to review the content of the cite-site (pun intended) I referenced.

http://www.lingenfelter.com/sites/l...files/LPE100mmMAFsensorflowcurvesv1.0_000.xls

If you buy the SCT tuning software for the EEC-IV/V programming, they not only list the stock Ford MAF voltage slopes, but also about a dozen aftermarket MAF voltage slopes for MAF scaling purposes.

I would hope a designer and vendor of a modified MAF meter be able to clearly present how it actually works and be able to provide the voltage slopes if asked. It's not only common practice, I cite the site where a retailer has it available for download for a variety of apps.

About that FPR manifold reference line? Do purchasers of this LTMAF hook up that line to maintain Injector Differential Pressure constant or not?

Cheers

Scott J
94 FZJ80 Supercharged
 
“The definition of insanity is doing the same thing over and over again and expecting different results”. - Albert Einstein

....but he didn't know anything either (read sarcastically)
 
SUMOTOY said:
I already cite site the 4in Lingenfelter MAF scaling in this regard back in post 444. I'm sure you just didn't take the time to review the content of the cite-site (pun intended) I referenced.

http://www.lingenfelter.com/sites/l...files/LPE100mmMAFsensorflowcurvesv1.0_000.xls

If you buy the SCT tuning software for the EEC-IV/V programming, they not only list the stock Ford MAF voltage slopes, but also about a dozen aftermarket MAF voltage slopes for MAF scaling purposes.

I would hope a designer and vendor of a modified MAF meter be able to clearly present how it actually works and be able to provide the voltage slopes if asked. It's not only common practice, I cite the site where a retailer has it available for download for a variety of apps.

About that FPR manifold reference line? Do purchasers of this LTMAF hook up that line to maintain Injector Differential Pressure constant or not?

Cheers

Scott J
94 FZJ80 Supercharged
Click to expand...



nice to know for when I'm modding a FORD.
 
SUMOTOY said:
I already cite site the 4in Lingenfelter MAF scaling in this regard back in post 444. I'm sure you just didn't take the time to review the content of the cite-site (pun intended) I referenced.

http://www.lingenfelter.com/sites/l...files/LPE100mmMAFsensorflowcurvesv1.0_000.xls
Click to expand...

Scott that just further confirms my suspicions that you have nothing to back this up like you claim.

You posted an Excel spreadsheet from an aftermarket tuner that seems to show two values: cycles per second (what, sample rate?) and gallons per second. Where are the voltage slopes? Where is the factory or personally recorded data?
 
Okay, still trying to stay totally technical here ... this was written above in post 409:

"""The 'wing' designs, and all obstructions are inserted to change the airflow/voltage profile of a given sensor. Specifically to the two Stockers (LS1 and Toyota), the airfoil designs *increase* low flow resolution (low voltage for a given flow in the charts above shifts to the right - without necessarily shifting the high voltage flow). The ecu software has a given voltage curve programmed based on the Stock Toyota (B) sensor profile."""

... and it really isn't right at all. The "wing" was not inserted in an effort to increase low flow resolution or high flow resolution for that matter. In fact it does not and can not change "resolution" at all. Resolution always refers to sampling rate/sampling size or increments of measurement and these things aren't affected by the wings, present or absent. The real reason it is inserted is to "buffer" the wires within the MAF from "buffeting" caused by changes in airflow either from turbulence through the basic bore design itself (more turbulent air flow the closer to the cylindrical circumference you get) or from the throttle plate opening and closing (lots of forward and reverse "waves" of air as the throttle plate opens and closes). Think of it as an "acoustic" shield; with many less modern MAF designs an attempt at getting a super smooth stream of air within a super small cross section of the basic bore was made ... with the more modern MAF designs an attempt at getting the largest cross section of the basic bore is made and then they focus on filling openings with screens so that the turbulence is "uniform".

... again an acoustical analogy is about as good as I am willing to word this phenomenon ... it is very much like a directional and distanced ribbon microphone and the more modern designs with the wires tucked into tubes above an acoustically shaped cancelling channel is the more modern method for making sure turbulance isn't an issue. Just to correct that error there. Thanks. :cheers:
 
Last edited:
NG, take the excel spreadsheet and plot a graph with it. If it helps, I can put up a voltage slope, but I believe that if you don't understand what these values mean, it won't help.

LT, a ford MAF is a 0-5v MAF just like the Toyota 0-5v MAF. Are you saying that the 80 voltage slope is somehow calculated differently other than the actual slope? That would require that you actually have those voltage slopes. We can speak to your MAF voltage measures when you get them.

TC, you can speculate on MAF 'turbulence', but I believe a basic EFI reference wouldn't support that. Doesn't any insert change Bournueli's Law of Flow = voltage slope? If you want to use an audio reference, think of this as signal to noise. If your (MAF)Pioneer Elite PD95 CD transport has a S/N of 125db, but your (ECU)Radio Shack amplifier only has a S/N of 85, using a class A component doesn't matter?

MAF resolution is defined within the software by processing speed and sampling rate.

Current instructions on FPR reference to Manifold Pressure line with the LTMAF mod? Rick?

Cheers

Scott J
94 FZJ80 Supercharged
 
SUMOTOY said:
TC, you can speculate on MAF 'turbulence', but I believe a basic EFI reference wouldn't support that. Doesn't any insert change Bournueli's Law of Flow = voltage slope? If you want to use an audio reference, think of this as signal to noise. If your (MAF)Pioneer Elite PD95 CD transport has a S/N of 125db, but your (ECU)Radio Shack amplifier only has a S/N of 85, using a class A component doesn't matter?

MAF resolution is defined within the software by processing speed and sampling rate.

Cheers

Scott J
94 FZJ80 Supercharged
Click to expand...

Scott, I'm not speculating, these were the design considerations which are well documented and no offense but as an audiophile your analogy was really wrong. Besides that then you went and worded exactly what I was saying by saying that MAF resolution is defined within the software, not within the wing as you originally offered. Still, as a good natured gest, this is terrific in the sense that you are actually starting to argue with yourself! This will save so many mudders so much time! :D :flipoff2: :D :cheers:
 
SUMOTOY said:
NG, take the excel spreadsheet and plot a graph with it. If it helps, I can put up a voltage slope, but I believe that if you don't understand what these values mean, it won't help.
Click to expand...

Scott, it is *very telling* that you don't realize how useless that would be. Where are the output (reference) voltages for these MAFs relating to the mass air rates. You've been banging on endlessly about voltage curves... this chart has nothing to do with voltage.

A set of reference voltage readings for each flow rate for each specific MAF application is entirely germane to your argument. All you're doing right now is a bunch of hand waving to try and direct attention elsewhere... and all that leads me to believe is a) you don't know wtf you are talking about and b) you don't have any of the requisite proof.
 
TC - The wing design within a MAF changes the voltage slope as the air velocity is higher at the lower flow. Which will change the programming within the software to change the low flow scaling to reflect a more accurate low flow. The presumption presented by the vendor to date is exactly the opposite. WRT audio, the example seems quite valid. Isn't audio S/N and class rating based on the worst component in the sytem? As a class A audio guy myself, me thinks so? So a Class A (MAF) feeding into a Radio Shack reciever, will only be as accurate to true sound as the RS component (80 ECU).

NG - the example I gave you was a GM unit, which is calibrated by frequency. I used that example to demonstrate that with frequency OR voltage, the slope is not linear, and they follow the same slope shape (a Bournuli thing), and the software scaling is defined by the software programming for the engine application. If you are having a difficult time with the concept of scaling, maybe this article will help you.
How and why to build a MAF table  First lets start with a little theory

LT - I would hope any vendor of a larger MAF, would not only present the voltage slopes when asked, but completely understand how a voltage slope and transfer function of a MAF meter works. It also appears Christo found a calibrator for you to send the MAF's to. I would also reasonably expect that if the instructions installing the MAF dictate removing the manifold reference line, a force induction motor should be instructed to do something else. Or do Romer and TC have this wrong?

This could be a good mod if it follows MAF function, theory and application correctly. Without voltage slope of stock or modded, it's just (still) different.

Scott J
94 FZJ80 Supercharged
 
SUMOTOY said:
NG - the example I gave you was a GM unit, which is calibrated by frequency. I used that example to demonstrate that with frequency OR voltage, the slope is not linear, and they follow the same slope shape (a Bournuli thing), and the software scaling is defined by the software programming for the engine application. If you are having a difficult time with the concept of scaling, maybe this article will help you.
How and why to build a MAF table* First lets start with a little theory
Click to expand...

Scott it is comical you're lecturing me on sensor theory. I've spent the last 5 years of my life studying electrical circuits and control systems in rigorous detail. I'm working on a Masters in electrical engineering, so spare me the trite and trivial explanations of electrical sensor operation...

Obviously MAF sensors do not directly provide a linear response curve, that'd be ridiculous to assume and AFAIK absolutely nobody is making that assumption. You bringing that up as a point of contention continues to demonstrate your fundamental lack of understanding of what you're even arguing or your attempts to misdirect as your argument is unraveled.


I like the idea of getting to the fundamentals, though. In that vein, why don't you demonstrate your understanding of an nth order state space control system and how the ECU uses such a system to respond to MAF sensor inputs. Once you've done that go back and try and justify your claim that the ECU is unable to adapt to moderate and even considerable drift in ideal MAF operation. Specifically address why you think Toyota, given an obviously powerful ECU capable of discrete sampling and considerable processing power, would ignore such a relatively well understood control design and opt for something completely archaic.


While we're at it, address how you think the ECU properly responds to progressive degeneration of the hot wire, either through moderate external contamination (filter oil, dust, etc) or intrinsic deviation due to repeated cycling of thermal states, etc. How exactly do you think your interpretation of the ECU (read: extremely rigid expectations of MAF readings fed strictly into lookup tables) would possibly work given the entirely nontheoretical reality that these trucks (and virtually every modern car on the road) routinely adapts to these scenarios and performs without fault.

Good luck!
 
NaterGator said:
While we're at it, address how you think the ECU properly responds to progressive degeneration of the hot wire, either through moderate external contamination (filter oil, dust, etc) or intrinsic deviation due to repeated cycling of thermal states, etc. How exactly do you think your interpretation of the ECU (read: extremely rigid expectations of MAF readings fed strictly into lookup tables) would possibly work given the entirely nontheoretical reality that these trucks (and virtually every modern car on the road) routinely adapts to these scenarios and performs without fault.

Good luck!
Click to expand...

I think this is one of the points people just aren't grasping. The system is designed to adapt to these changes. And while my MAF isn't 100% accurate to a brand new stock MAF, it doesn't have to. It really only needs to fall into the acceptable range of what the ECU wants. As long as the difference between the two is within that range it is seen by the ECU as being normal.
 
forget progressive degeneration of a hot wire, how does the earlier ecu adapt to the much less accurate signal from a VAF with an acceptable voltage fluctuation of more than 1v? quite obviously, it is looking for a familiar voltage curve, and the particular voltage is not that important provide the curve itself fits certain predefined parameters. equally obviously, the system must rely on other sensors, including engine rpm and tps, and particularly on o2 sensor feedback to "centre" and "correct" the voltage curve, particularly as the vaf/maf sensor ages and the signal degenerates.

it is painfully obvious that landtank's mod generates a familiar voltage curve well within the parameters and range that toyota has built into the ecu, and it is equally obvious that, if the sizing of the housing allows greater airflow for a given voltage point, that extra airflow is readily identified and re-centred by the resulting higher rpm and other engine signals, or else it is well within the scope for which an o2 sensor can correct. this only makes sense when you consider the minor nature of this mod compared to the various operating conditions toyota would have needed to plan for in designing an ecu, and particuarly considering toyota sold a forced induction supercharger to work with this ecu. in fact it makes sense based solely on the fact the engine is able to correct the static idle!

and based on the reported results to date, i would bet anyone $100 that the curve is within normal ecu curve parameters, and not a case of constant o2 sensor correction.
 
NaterGator said:
Further required reading for Scott: PID controllers, Stochastic control, dynamic sliding mode control, etc.
Click to expand...

Here are a few voltage slopes of actual MAF's as tested by the manufacturer. NG, please note that each of these sensors has an accuracy of <3% of measured airmass to actual airmass from min to max voltage, as tested by the MAF manufacturer.

I labelled the "D" (internal pipe) dimension of the various voltage curves as sloped by the manufacturer. As you can see, the 71mm MAF 3 would show half the airmass at 4.3v compared to the 82mm MAF 5.

The sensors themselves then are extremely accurate to the incoming air mass. The 80 software contains the voltage slope to accurately calculate Load, timing and fuel tables. Changes in fuel tables with 'larger' or 'smaller' injectors does not change the Load or timing tables affected by the changes in MAF values. Summary to the LTMAF mod vs stock: Without the voltage slopes, or software scaling, the accuracy of all tabled MAF voltages is actually worse, not better. Whatever one 'thinks' the ECU is doing, it's not doing it 'better' with an uncalibrated MAF sensor.

YOMV, but application of this concept seems pretty clear.

Scott J
94 FZJ80 Supercharged
scan0002.webp
 
Last edited:
Scott, you're running around in circles. Nobody disagrees that a hot wire with cool air blowing over it (and a proportional temperature-resistance dependence) is going to have a nonlinear response curve. Make sure you read that so you don't belabor the point in your next response because I'm sick of hearing it over and over.

I am simply amazed that you can habitually ignore everything I've posted and harp on the same stupid trivial point. Get back to me when you understand what I'm saying about control systems and thoroughly read my posts.

In the meantime I will quote from Won et al. from the IEEE Transactions on Control Systems Technology article Air-to-Fuel Ratio Control of Spark Ignition Engines Using Gaussian Network Sliding Control
The mass air flow rate into the manifold can be measured by
hot wire or film sensors, and the sensor characteristics changes
with temperature and aging. Therefore, the measurement is bi-
ased with temperature and aging. The bias in the measurement
of the mass air flow rate into the manifold is also adapted on-
line by the oxygen sensor and a Gaussian neural network. The
advantage of this method is its robustness to engine aging and
individual engine characteristics by using on-line adaptation.
Click to expand...

That should pretty much settle the issue since it comes directly from published IEEE members (who are obviously established electrical engineers). Kthxbye.
 
NG, isn't the Gausian concept actually a function of the MAF bridge design in the sensor circuit of the sensor itself? It certainly is in the MAF I put above. This concept applies to Hot Film and Hotwire sensors that do not have a self cleaning burn-off function, because it is not needed. Simply put: There is a trimmer resistor in series with the measuring resistor, which compensates for operating temperature and age. With this compensation circuitry in the bridge, the voltage output of airmass is <3% variation to actual airmass. It needs to be accurate to properly table, load, fuel and timing tables.

In your quote presentation, 'beyond' the 3% accuracy of the airmass, the 02 fuel trim tables will compensate for those changes in accuracy in fuel. These voltage variations do not change the 'Load' or 'Timing' values. As such, when the timing values are not within specification, Mr. T instructs to replace MAF with known good, or compare voltage outputs with known good.

Jumping a bit ahead of the game here, since the output voltages between the LTMAF and the Stock MAF have never been compared. Easy to test your theory with enough 'n' we should see the variation of age and temp. IME, it's a lot less than you believe it to be. And MAF value is not a learned software function in the 80 ecu. The MAF output voltage is just tabled.

Scott J
'94 FZJ80 Supercharged
 
Last edited:
WOW... I haven't been reading or posting much lately, too busy, and I return to exactly the same discussion. Different topic, same discussion.

Goes like this, no?

Someone posted information.

Scott / Sumo questioned the information.

No matter what answers are given to the questions Scott's not satisfied.

People start getting irritated, attacking Scott and his attitude, and his lack of real knowledge or data, and his inability to see real facts and physics.

Some people give up.

Some people say put him on your ignore list (which I've done in the past, and it makes threads hard to read).

Scott declares victory by being the last one posting.


Sound about right? I didn't read all 17 pages, didn't have to. Oh, and he's not short, at least not as I remember. Sold him the engine out of a totaled '94 some time back so had the chance to meet him in person. Was buisy with kids so didn't really converse with him much at the time.

Here's an idea - start the thread over. Start the thread with an ALL CAPS WARNING - that Sumo will try to derail the topic, and that all those who want to see an honest discussion on said topic, will not enguage Sumo in any way shape or form. Might actually be able to get through a topic that way. Two things I can guarantee with that approach - 1. Sumo will try harder and harder to get someone to enguage. 2. It will bug the hell out him if no one responds to him.
 
SUMOTOY said:
NG, isn't the Gausian concept actually a function of the MAF bridge design in the sensor circuit of the sensor itself? It certainly is in the MAF I put above.
Click to expand...

The modifications made to the wheatstone bridge are not implicitly designed to compensate for aging or contamination since both are expressed with some degree of variability. The bridge is definitely optimized to provide the best transient response of the hot wire for the most common operating parameters, but that's about it.

Here is a great whitepaper on the subject from all the way back in 1991:
https://eprints.kfupm.edu.sa/42430/1/42430.pdf
Though it is a bit dated (back when logic hardware was expensive (predating windows 3.1) the principles of Gaussian neural networks are the same.


WE: You're pretty much right on the money, but since this thread has been sufficiently derailed and I believe the past bit of discussion pretty clearly demonstrates why Scott's claims are unfounded, I'll pursue his recognition of the flaws in his claims until I tire or he relents. Odds are very good I'll tire, but at least there will be a record of Scott's unwillingness to acknowledge the flaws in his claims in light of contradictory evidence for people that get this far in future readings (or are currently reading this thread). :beer:
 
You must log in or register to reply here.

Share this page

Back
Top Bottom