Power Fc

[paulr33]

see diagram for relocated AFM

[Gav]

the problem doesnt stem from max power or how it performs on the dyno. the problem stems from how a map sensor system guesses the amount of airflow in the manifold. as its a guess based system it will never be as accurate as an airflow meter measuring it directly.

once you reach max boost as designed by your gate, controller, solenoid or what not the map runs across a horizontal axis, so you don't scale down as more and more air coming into the system increase. this is due to the way map guesses airflow based on pressure.

volume and pressure are two different thing

the map setup says

if i have 0.50kgcm2 pressure and my guestimation calcuation is:

pressure x magic-number (lets pretend its 10800) = amount of air we get;

0.50 x 10800 = 5400

so our current airflow equals 5400 units. now the AFM version would say lets pretend 5400 airflow units also. now as boost pressure increase more and more air is coming into the system so the AFM signal goes up.

so lets say we now have reached target boost (say 1.3bar)

1.3 x 10800 = 14040

and our AFM says there is 14000 units of air present

now we are at 4500rpm so theres lots more RPM to go

AFM = 14000 units @ 4500rpm

MAP = 14040 units @ 4500rpm

now lets pretend we have had full throttle nailed for some time and are at 6500rpm

using our same equations (we are still at target boost)

MAP

1.3 x 10800 = 14040

AFM says 18000 units of air present

the AFM has clearly shown more air has come into the system ,as RPM increase more air comes in, despite the same fixed pressure.

So what happens in turn is the djetro version or map sensor version shoots across the RPM axis as load never changes, as the MAP guesses air present based on pressure, which never changes.

The AFM version keeps scaling across both the RPM and Load axis as more air slowly is coming in and RPM is being increased. so it scrolls diagnoally if that makes sense.

Hi Paul

An interesting discussion.

I guess firstly, I can't explain why the D-Jetro does show a consistent gain in power over the std PFC, however as I have said, I've seen this first hand on 3 GT-Rs now. I don't think there is any magic involved in this, however.

With regards to MAP sensors "guessing" air quantity, I disagree. It simply uses basic gas laws to determine the quantity of air:

PV=nRT, where P = pressure (i.e. map sensor), V = volume (known from the engine displacement and the rpm), n = the quantity of air (what we need to know), R is the Universal Gas Constant (the "magic" number) and T is the air temperature (taken from inlet manifold temperature probe).

You can see that there is an inherent problem now with the fact that it is assumed that volume is related directly to engine revs only. For NA motors, this isn't a particularly good assumption as volumentric efficiency changes significantly across the rev range (indeed this is one reason for going to big lift and longer duration cams). For forced induction motors, the proportonal change in volumentric efficiency is comparitively much less, and therefore the equation hold much better.

In your example above, you assume that the y axis (or load axis) is actually pressure on the D-Jetro rather than a calculated quantity of air ("n") in this case. I am sure that this isn't the case as (to use your example again) when target boost is reached well before the RPM redline, I've seen that the map trace doesn't go horizontally across the x (rev) axis, but rather continues diagonally down. This shows that additional load is being calculated by the D-Jetro and fuel/timing changes made to suit.

There are certainly inherrant problems at low rpm and idle situations. Here the volumetric efficiency variations have a much greater effect and the manifold pressure jumps all over the place (particularly with "aggressive" cams!). For this reason the D-Jetro cross references against TPS and uses an average of the 2 MAP sensors in the manifold.

With regards to MAF determination of air quantity (i.e. factory ECUs and std PFC), this is also a "guess" if you want to put it that way. Just about all of these systems use what is known as a "hot wire" AFM that relies on the fact that electrical resistance of a wire changes with temperature. The amount of air passing the wire is calculated by the change in electrical resistance. Generally a very reliable and cheap to produce system. I suppose that this (and the fact that most cars produced are NA) may be a reason for OEMs turning away from MAP to MAF based systems of measuing load).

Some well known downsides of MAF systems are:

* when air that has been measured doesn't go into the engine (i.e. blow off valves venting to atmosphere)

* when the hot wire gets fouled by oil that may insulate the wire from the air

* the delay between measuring on the inlet side of a compressor and actually entering the engine (especially on big intercooler volumes)

* the fact that generally there is no account (or even measurement) of manifold air pressure or temperature. Both of these properties can vary even at a given AFM output.

[paulr33]

Hi Paul

An interesting discussion.

I guess firstly, I can't explain why the D-Jetro does show a consistent gain in power over the std PFC, however as I have said, I've seen this first hand on 3 GT-Rs now. I don't think there is any magic involved in this, however.

With regards to MAP sensors "guessing" air quantity, I disagree. It simply uses basic gas laws to determine the quantity of air:

PV=nRT, where P = pressure (i.e. map sensor), V = volume (known from the engine displacement and the rpm), n = the quantity of air (what we need to know), R is the Universal Gas Constant (the "magic" number) and T is the air temperature (taken from inlet manifold temperature probe).

You can see that there is an inherent problem now with the fact that it is assumed that volume is related directly to engine revs only. For NA motors, this isn't a particularly good assumption as volumentric efficiency changes significantly across the rev range (indeed this is one reason for going to big lift and longer duration cams). For forced induction motors, the proportonal change in volumentric efficiency is comparitively much less, and therefore the equation hold much better.

In your example above, you assume that the y axis (or load axis) is actually pressure on the D-Jetro rather than a calculated quantity of air ("n") in this case. I am sure that this isn't the case as (to use your example again) when target boost is reached well before the RPM redline, I've seen that the map trace doesn't go horizontally across the x (rev) axis, but rather continues diagonally down. This shows that additional load is being calculated by the D-Jetro and fuel/timing changes made to suit.

There are certainly inherrant problems at low rpm and idle situations. Here the volumetric efficiency variations have a much greater effect and the manifold pressure jumps all over the place (particularly with "aggressive" cams!). For this reason the D-Jetro cross references against TPS and uses an average of the 2 MAP sensors in the manifold.

With regards to MAF determination of air quantity (i.e. factory ECUs and std PFC), this is also a "guess" if you want to put it that way. Just about all of these systems use what is known as a "hot wire" AFM that relies on the fact that electrical resistance of a wire changes with temperature. The amount of air passing the wire is calculated by the change in electrical resistance. Generally a very reliable and cheap to produce system. I suppose that this (and the fact that most cars produced are NA) may be a reason for OEMs turning away from MAP to MAF based systems of measuing load).

Some well known downsides of MAF systems are:

* when air that has been measured doesn't go into the engine (i.e. blow off valves venting to atmosphere)

* when the hot wire gets fouled by oil that may insulate the wire from the air

* the delay between measuring on the inlet side of a compressor and actually entering the engine (especially on big intercooler volumes)

* the fact that generally there is no account (or even measurement) of manifold air pressure or temperature. Both of these properties can vary even at a given AFM output.

i have seen the pvrt theory and i still believe the djetro PFC simply uses MAP sensor signal and nothing else to work out what load axis to jump to on the maps. i see no other supporting items to influence the LOAD axis on the djetro. datalgoit shows me a LOAD axis calibartion and for AFM

its purely AFM voltage and for map sensor its simply map sensor voltage

there is no calibration or table reference for TPS, volume, revs or anything else 3d'd onto the MAP axis.

* when air that has been measured doesn't go into the engine (i.e. blow off valves venting to atmosphere)

so use a correct blow off vavle system, its there by default so it must be there for a reason.

there is no need for ATMO bov. it didnt come atmo from the factory so nissan would have had to have been fairly certain the recirc idea was useful

* when the hot wire gets fouled by oil that may insulate the wire from the air

so clean the AFM. this would come under normal maitenance and would only take 5 minutes with some contact/carby cleaner. no different to cleaning your air intake filter or pod filter

* the delay between measuring on the inlet side of a compressor and actually entering the engine (especially on big intercooler volumes)

i dont understand this at all, once the air is in, its in and nothing more. i am unsure on this "delay" thing

* the fact that generally there is no account (or even measurement) of manifold air pressure or temperature. Both of these properties can vary even at a given AFM output.

the pfc ljetro supports airtemp correction. air temp correction with measuring how much air directly comes into the system is the ideal way to work out how much air is coming in.

there is no other better way than to measure how much air is coming in.

dont get me wrong im not against everyone or anything i just dont see why one would move to a map sensor system, there are no benefits

[rob82]

why avoid the wolf? they are quite similar in specs to the SM4

Dont compare a Wolf to an SM4 there is absolutely no comparison as the autronic has brains, end of discussion. Specs on a peice of paper saying they can do this and they can do that where in reality they are only trying to do it and failing miserably.

Paul before you get all righteous on AFM based ECU's try to remeber that there are MAP based systems that are far more complicated and ACCURATE than a MAF based PFC.

[paulr33]

i still fail to see why a djetro ecu would make more power?

what makes power, airflow. there is no magic to making more power.

the only way i would see it (the djetro) making more power if one of the following was true;

1) airflows meters in use were being maxed out, hitting ceiling limit and not able to show any more air to the ecu

2) airflows becoming a physical resctriction

in either case upgrade to larger airflow merters or fit an additional one

[paulr33]

maybe im reading this all wrong but heres how i see it;

a car tyre has 40psi of pressure

a truck tyre has 40psi of pressure

given the map sensor method both tyres have the same air volume in them. if you were to let all the air in each tire via a hotwire airflow meter i would bet the truck tyre would have more air in it

[paulr33]

croat: have a read of http://www.skylinesaustralia.com/forums/in...pic=80162&st=20

that covers relocating the airflow meter

[paulr33]

anyway im gonna give up

im in way over my head and i dont even know if im right or not

no one seems to be supporting so its likely im on some other tangent or have gotten it wrong. all good

map sensor it up croat

[paulr33]

fark rob's doing a BIG reply (im scared)

[rob82]

Delco ECU's aka Autronic are based on speed density based calculations where be the mass of air within the the chamber is calculated via PV=nRT as stated previously. The pressure is derived from the MAP sensor, the temperature however is far more complicated as it applys a thermodynamic model to the acutal intake manifold whereby the absorbtion of termperature from the engine is approximated. Delco and Autronic do this by using a 3d table that has map on 1 axis and ECT on the other axis, the acutal values in the table are percentaged based and is applyed to the intake air temp.

An example would be at idle as the air within the manifold is stagnent and therefore alot of heat energy is absorbed by the intake charge hence the volume is less, as manifold pressure rises or more air is consumed by the engine the manifold temp has less time to transfer heat to the charge so the actual chamber charge is more dense. They call this table a charge temp table and within the delco ECU's it changed with different manifolds that were used from factory ie standard 5L to twin throttle body GPA A.

Once the acual mass of air is known then its as simple as comanding a desired AFR. You want 13.0:1 at 5psi of boost you get 13.0:1.

The early map based fords used the same approach the later model fords however are alot more complicated as they acually apply a linear equation for each phase angle(cam angle). The more flow at the same cam angle the more fuel they inject. While you may think this sounds innaccurate I had a supercharged BA that was making about 350rwkw on about 10 psi of boost. At this level we knew the intake pipe feeding the supercharger was a restriction as soon as we took it off the car gained 30kw on the same boost pressure and you know what the the AFR were exactly the same as I was commanding beforehand on the lower power run. Expalin that!!

[rob82]

fark rob's doing a BIG reply (im scared)

You should be scared I'm drunk and rambling!!!! HA HA

Dont get me wrong I'm not tring to say one is better than the other as 95% of the time I would recomend a MAF based system over MAP as there is still merit to both with the right ECU!!! And I cant stress that last point enough THE RIGHT ECU!!!!!!!!!!!!!!!!!!!!!!!!!

[paulr33]

whilst over ecu's certainly would handle the map sensor implentation better the powerfc simply uses (that i can see) the map sensor signal and nothing else to decide "load" on its load axis.

still a great topic of dicussion either way

[R31Nismoid]

Why do you think so? I've seen a number of GT-Rs gain power when converting to D-Jetro from std PFC.

That doesnt prove a thing.

the original tune could has been piss poor, and changing the ECU might have resulted in a better tune... but thats not the ECU's fault at all

Thats tuner

[Gav]

That doesnt prove a thing.

the original tune could has been piss poor, and changing the ECU might have resulted in a better tune... but thats not the ECU's fault at all

Thats tuner

Your logic is correct - I can't think of a solid reason why the D-Jetro should provide more power over a standard PFC. But I've seen the same tuner achieve just that. Perhaps he's just better at tuning D-Jetro's??

My main reason for going to D-Jetro was that I wasn't happy with the Wolf ecu I had (couldn't get the ATTESSA reliably working) and I didn't want to fork out the bucks for new AFMs, adaptors etc.

[R31Nismoid]

Your logic is correct - I can't think of a solid reason why the D-Jetro should provide more power over a standard PFC. But I've seen the same tuner achieve just that. Perhaps he's just better at tuning D-Jetro's??

Thats what i was heading towards

[rb26s13]

sweet debate..

and yes you just cant compare a wolf to a autronic.. its like comparing a piece of gold to a piece of shit.. it just doesnt happen

Edited June 9, 2006 by rb26s13

[The Mafia]

where can i get the d-jetro from? can this be tuned everywhere as the AFM version? is it a better version of the power FC (better mapping?).

i personally thing the 4" pipe into the 2.5" pipe looks stupid but i could prob deal with the 3" Q45 deal..

thanks..

a q45 is much bigger than 3"

Its about 3.9 inches.

I'm running this setup (powerfc, 4 inch bellmouth and q45) and it worked 100%.

taken from here -

http://www.skylinesaustralia.com/forums/in...showtopic=89613

I suggest you have a look at my thread.

Edited June 10, 2006 by The Mafia

[MerlinTheHapyPig]

load is purely AFM or MAP

I think we've had this argument before.... but anyway..

different ecu's (such as powerfc) may have a different implementation, but speaking for the stock ECU

the stock ECU uses TP for load, which you have confused with throttle position, TP in this context is not throttle positon, it's Theretical Pulsewidth

This calculation is based on RPM, AFM voltage and a few other constants (injector k value, number of cylinders etc.)

On stock ECU's, this is the X axis, whilst the Y axis is RPM and the Z axis is fuel enrichment/enleanment

the theoretical pulsewidth is the injector pulsewidth required to deliver 14.7:1 afr, the values in the table are enrichment or enleanment from this based on the RPM

From reading a few other posts in this thread, it seems that other ecu's use different sort of calculations for the "load axis", it's not "just afm voltage or map sensor voltage"

don't worry about throttle position, though most ecu's use this for acceleration enrichment (e.g. quick bursts of full throttle)

Edited June 10, 2006 by MerlinTheHapyPig

[paulr33]

it sounds like you are spot on, silly me assumed it was either MAP or MAF signal and nothing else.

can anyone else confirm this is how the load axis of the FC is driven?

would like to add to the FAQ and just wanted to clear it up

[StockyMcStock]

look ma, no AFM!

plus you get better response the closer it is to the throttle body, otherwise the engine has to eat 15L of air before it begins reading the correct signal. trust me the response is MUCH crisper when it's close to the throttle body like mine is.