Yes - that would be my understanding of it. Basically, the more defined sites that you can program, the less interpolation that is required, and as such, the less compromise.

Even so, this ultimately can only make the best use of the area within physics. As such, once the optimum boost level is achieved at a given compression ratio, the only solution left irrespective of the map will be to drop the compression ratio to increase the power (given that we are not increasing capacity etc etc etc)

Nice site that one Gareth! The colours are quite useful on the maps to assist in visualising the map!

I have an S8 on the shelf somewhere upstairs - I had a quick play with it, and whilst I never got to the point where I felt comfortable using it on the car, I at least got a very general idea on how the map works.

I think that the mapping itself is a subject in its own right, which Stu recently did a great write up in the Fast Ford Techincal section, so I wont go over old ground.

That said, I note that the more modern engines use a combination of Mass Air flow meters as well as MAP sensors. I believe that some of the Scoobies use this setup? Or is it that the MAF is there in replacement of the MAP.

Having discussed the subject further, I can now see the benefits of the MAF over the MAP, as it is measuring the volume of air, as opposed to the blocked air (boost). I guess that this would lead to more accurate maping.

My confusion is that a number of people I know that tune their Scoobies have removed the MAF in favour of the MAP. Why is this?

JJ

 

MAF becomes a restiction at certian power levels..

 

Ah okay - that would seem sensible - out of interest, why is this? Is it just that they are unable to read beyond a certain yield accurately?

JJ

 

 

a small turbo escort uses both mas and map

right,,,, a map based system,,, if nothing changes to effect the VE of the engine, and the ecu was calibrated correctly in the first place great system

advantage to map,, theres nothing restricing the intake volume, disadvantage, will no adjust for slight changes in VE without a re-calibration

right MAF systems,,, these actualy measure the airflow though the engine, so can alter for slight changes in VE as these will be measured changes,, but sadly we have a slight restriction in our inlet system

 

think its the amount of air that can actually flow thru the sensor.

 

the std small turbo MAF sensor runs out of scale/ is maxxed out at 35lbs/min of air

 

all sensors have a range of accurancy,, and genrally if we extend the range, we loose accuracy DAM,,, we can never have it always

 

JJ, you are getting there now mate, I will correct a few things tomorrow if I get a chance before I leave for the ring, as TBH I cant be arsed getting technical at midnight, lol

If not, I'll speak to you on the phone tomorrow perhaps.

 

retard the ignition i think you mean

 

Yes that makes perfect sense!

So it all starts fitting together. The MAP is as you rightly say capable of measuring the boost with little inteference to the overall air flow itself, and so long as the map is set up correctly for the car, there is no reason why it should be changed.

The MAF by definition has to measure the flow of air and as such may in themselves reduce the VE!

Cool - I understood that in record time!

Right lets turn to fuelling.

In a different thread, I questioned why some cars need larger fuel injectors than other. for example, my 550cc injectors on my cossie more than adequately flow enough fuel for any power that I am currently running - ie about 400bhp ideally.

However, if you wanted to achieve that kind of power in other cars, then you would need larger injector sizes to get the same power. As I understand it, this relates to the pressure of the fuel rail. It kinda boils down to the same thing as the VE v Boost - it is a measure of how much flow against maximum size??

Given the above, I note that in cosworth land, people rarely change the fuel pressure. I wonder why this is - is it because ford just got it right - and if so, why is the jap world different?

JJ

 

ahhh right,,, here goes

firstly engine have different BSFC

"brake specific fuel consumption"

this is a measurement how how efficent the engine is at turning fuel into power, so fuel consution devided by power


right injectors differential pressures

the higher this is,,,, the more the injector will flow for a given injector opening time ,, if cause this is upper and lower limits to this

 

I knew I would get it the wrong way round!!!!!

Reference to fuel, yes I remember you explaining this to me before - IIRC the type of injector also has an effect here.

Going back to the BSFC - lets focus on this for a moment. given that we have taken our standard engine, and potentially changed the compression ratio, the cam timing, the flow of air through the head, the VE and the turbo etc, I am mindful to wonder whether we have now changed our BSFC. If not, then what exactly affects this?

Sorry if this is a dumb question

JJ

 

BSFC actually varies even based on different RPM points and different amounts of boost.

Its not just one fixed number that applies, its an entire graph.

Typically most changes will make it better in some places and worse in others.

 

Hi matey - I thought you were going to bed

Right - yes that makes sense. Given that very very little (if anything) is linear in engine design, why should BSFC be different.

Even so, if the BSFC accounts for the different amount of fuel required to create power, if we substantially change the specification of the engine as above, why is the BSFC effectively not requiring alteration of the fuelling pressure etc?

JJ

 

The fuelling on the engine is largely linked to the airflow, within 20% or so even comparing between full boost and cruise.

The BSFC though can vary greatly, but only typically if you get things badly wrong with the spec.

Any time you see massive heat its normally an indication you just wrecked your BSFC!

 

Okay, that makes sense - so I guess whilst we work to the right principals with our theoretical engine, then the BSFC should be constant.

Right - bed time - to be continued in the morining when my brain re-engages!

Thanks all who continue to contribue guys! Without the techhys, this would be just the rantings of a Building Surveyor!!

I will leave you with the following thought though.....

All we need is an engine that has variable compression, variable cam timing as well as lift etc, combined with our adjustable ports, and variable vein turbo, and possibly the most complicated map ever written - and we have the perfect engine!

JJ

 

JJ, not with pistons going up and down we dont have, thats a terrible idea, a decent form of rotary engine would be far better!

 

Chip, I think we had better stick with the normal four stroke for the time being PMSL!!

I am too easily confused!!

JJ

 

Hence the invention of the jet engine all four strokes of an engine in a line with no recieprecating parts - look at the power they make

 

With regard to modern ecu technology.....

There has been a bit of confusion here with regards the word resolution.

Yes, modern ecus have more mapping points as explained before but this
is nothing to do with resolution it just means you have more numbers in
the table(s) to play with to fine tune problem areas in the map.

Resolution is actually the accuracy and ability of the ignition and injectors
to be timed accurately and more at finer intervals.


All ecus are high speed (relatively ) computers with dedicated
timing hardware as part of the micro processor.
The hardware timer is continuously clocked at a speed depending on the
micro processor type and design.
(FYI This timer system is called input capture/output compare or TPU)
This timer is used to reference all measurements and outputs.
To some extent this timer can be rescaled to give different values but
the software ususally configures this for the following values
due to limitations of the design.

Here are the hardware timer values for some ecus....

Weber L1 - 8 Microseconds = 125,000 per second.
Weber L6 - 4 Microseconds = 250,000 per second.
Weber L8 - 4 Microseconds = 250,000 per second.
Weber P8 - 2 Microseconds = 500,000 per second.

SECS S8 - 0.05 Microseconds = 20,000,000 per second

As you can see above, as stuff gets newer, it gets faster and that means
better accuracy and stablity.

Ok, I now need to explain how this timing is used for the different ares....

FUELING

(keeping it simple and excluding corrections etc...)
(all calculations and figures are rough examples)

After calculating the map table point from RPM and LOAD (map/maf etc)
and then interpolating from the table points around it, a value is arrived at.
In Weber ecus this value is only a single byte which means it can only
have a data value of 0 to 255.
This means you ONLY have 256 possible injector times.

Now, as most injector duration ranges as from about 2 to 20 milliseconds
this means that you have a range of 18 milliseconds and...

18 milliseconds divide by 256 = 70 microseconds

As the hardware timer tuns faster than this actual reduced resolution,
we need another value in the software which is called the multiplier.
On weber ecus this is typically set to 32 or 64 microseconds to give us
our 256 possible injector values.

From this you can see that the weber system is only capable of a
reletively crude injector timing accuracy.

On small injectors this isnt too much of a problem as these large interval
steps only increase fuel slightly.
When larger injectors are added, you begin to loose accuracy.

Modern ecus have larger map table values and using the S8 as an example,
it has a word value (as apposed to a byte) to store the fuel value
which has 65336 combinations allowing the multiplier to be smaller
giving better resolution.



IGNITION

This is where modern ecus have the biggest edge.

The above map table data size problems apply to ignition table values aswell
but the real gain is that the higher timer speeds in moder ecus allow
greater ignition timing accuracy especially at higher RPMs.

The ignition angle is calculated from the map and the current time it
takes between 2 crank teeth is measured.
This crank time is then divided by the number of degrees advance
required and a second timer is loaded with this new value.
When this second timer has gone to zero, the ignition coil is fired.

The higher the RPM the less time there is to divide into ignition advance
degrees which reduces accuracy.
As the timer runs faster, you have more resolution to be accurate in
a modern ecu.

As an example the L8 has 0.2 degrees map resolution but at high rpm
at around 5000 rpm the actual possible error value from the desired angle
is could be as high as 1.5 degrees !!!!!

As (for example) the S8 has a timer that is running 40 times faster, the
error is reduced by the same factor size.
This means it has a mappable accuracy of 0.1 degrees right upto its
12000 rpm designed rev limit.




SUMARY

WIth regards to injector accuracy, this makes only a minor problem for
large injectors on systems mapped by people who pay attention to detail
on older ecus.

With ignition accuracy, this means the det limit can be mapped too more
precisely and a reasonable accurate comfort zone added.

Although high timer resolutions give advantages, the real advantage of
modern ecus are accesability, extra features etc.....


*** I have used the S8 as an example as I have no direct experience of
the internal workings of any other high speed modern ecu that I am able
to publish without upsetting someones copyright LOL.

 

Fantastic info there simon

 

Mind you - its gonna be a stretch to get the cosworth to match that configuration with general tuning! ROFL

Okey dokey - shall be get cracked back into it - I know that some of our tech chaps will be Ringbound, so lets take advantage of them whilst they are still here!

Can we focus in a little on the cam timing? I want to see if we can shed a little light on this area of tuning.

Lets create another hypothetical. We have agreed now the following;

1. The bottom end is now configured for mildly low compression (i dunno say 7.2:1 - Does that sound about right?) All else is std - stroke, piston size, cc etc.

2. The clyinder head has been mildly ported - lets say equivalent to a Norris Stage 1?

3. The Turbocharger at the moment is to be say a T38 (I guess with a 0.63 zorst housing IIRC). I know that the popular choice would be the T4, but bear with me

4. The fuelling and mapping is a given, and will be correct for whatever setup we finally work with.

Now you will note that we havent referred to the cams as such. Having built my virtual engine, I now have a Boost threshold of shall we say 4000rpm or there abouts, and it is proving to be very unresponsive off boost. The car owner (Mr Blobby) has asked that Jacobs independent car tuners (patent pending) set up his car to if possible reduce the boost threshold, and improve the off boost running.

So where do we start?

JJ

 

Yes, thats correct for a given bottom end, the way to calculate it is:
(Swept + Unswept) / Unswept

Unswept is the bit above the piston (including piston dish and the gasket as well as the volume of the head)
Swept is the volume displaced by the piston as it moves from bottom to top, so 500cc in the case of a 2000cc engine (YB is actually a touch under 2000cc)
Swept is calculated as Stroke * Bore * Bore * Pi

Thats correct, just to add a little more detail though, the reason that a lower CR will lead to a greater permissable VE is that if you are squeezing the mixture less hard, you can have more mixture to squeeze essentially.
So the CR does not in any way effect the VE (slightly untrue actually, there is a TINY reduction in VE when you drop the CR, which is most effected at the top of the rev range)
So your comment of allowing other things to be changed would on a turbo engine typically mean that more boost would be safe without risking det, and its from this increased boost that the VE improvement comes.

Its RETARD the ignition, ie fire it sooner. (easy to see why doing it sooner sounds like you have advanced it though, quite confusing that really!)

Braodly speaking yes, but as with most things in life its an "X squared" type of curve in terms of power versus compression ratio, ie the power will go up and up until you get to a point and then start to come down again, this point is the optimum CR for your setup at those particular revs.

Yes, thats correct, there is a direct relationship between air consumed per cylce and torque, so as we have learnt that we cant over fill the cylinder too much in one cylce for risk of detonation, it means that we are effectively capping the amount of torque we can ask the engine to reliabley produce for any given fuel.
So if we want big power figures within a fixed amount of rpm (ie we cant get it by carrying the torque on till 10,000 rpm as there are mechanical limits) then it means that we need to alter the CR to allow us to generate more torque reliabley per cycle so that we can get the power we want within the fixed number of cycles per minute we are limited to.

After a remap of course to take account of the new airflow characteristsics

"run properly" is a misleading term, the correct way of wording it would be more like "in order to make the same amount of torque as before" or similar

Yes altering the cams can help you to increase the low speed VE of the engine to get back the torque we lost by going lower CR potentially, but it will of course have a negative impact further up the rev range if you do this, so great care is needed so as not to end up choking the engine at high rpm just to get back the (from a performance point of view) off boost drivability of the engine, so again, its a compromise sadly.

Its because of the increased boost, the fact that the engine has got a new turbo on the side, its because of all of it, including the CR

THIS is one of the most important factors, but its not by any means the only important factor, getting the right cylinder head modifications can be equally as important for example, particuarly at higher power levels, and the correct selection of a turbo unrestrictive enough to flow well at high rpm and still small enough to make boost readily is also vital.

Im quite surpised how much you seemed to have picked up in the last week, you obviously arent as thick as GarethT looks

 

First of all we need to accept that any change we make to effect low speed running positively will nearly always result in a detrimental effect on high speed, and particuarly high boost running.


So, after our sombering acceptance, we look first at altering the lobe centre angle of the cams so that we can improve the bottom end response with the existing hardware (well, other than the addition of a set of vernier pulleys), a couple of degrees can make a big difference here.

with regards to where we go next with cams, well, it depends what cams are fitted, because if its the standard ones, you are pretty stuffed really, as you arent going to get the top end response if you go any more mild than that.
So i would imagine in this instance, if the cams are "wrong" its because mr blobby has read that BD16s make good power and has slipped a set of them in or similar.


But to answer the question you are trying to ask, the vital thing to getting good turbo spooling is to hit it with short sharp pulses low down where you have limited airflow, so a short duration camshaft is the order of the day here as a longer duration one will dampen the effect.

 

Firstly - Simon - What a fantasitc little write up on ECUs! Top top write up! That one came up whilst I was still posting!


Chip - Thanks mate - those clarifications are important, as it allows anyone else to see where my understanding is erroneous!

As far as my ability to learn is concerned - I am a great believer that the ability of an individual to learn is far more dependent on those that are teaching - so thanks to you and to all others for taking the time!

So I have two ways to go now :-

1. further into cam timing and shaping (I will need to visit this again at some point)

2. I read an interesting article in EVO about the new Koenigseg (add the necessary letters as required) - which is designed to run on BioEthanol. I would like to discuss the pros and cons of alternative fuels in our high power engine

Your choice guys

JJ

 

Sadly mate, with regards to cams, you arent going to get much further on here talking generally, its only by actually getting stuck in and trying different cams back to back on the specific engine spec you are interested in and making slight changes in LSA etc that you are going to learn more about that.

 

Thanks matey - acutally your previous post goes a long way, but it was obviously posted whilst I was writing!

I think what I will do now is sit down, consider everything that has been said, and see if I can write a summary - this will help to see if I got it, and also should prevent us from repeating ourselves.

JJ

 

sorry simon, i was using the word rselution as defined as "amount of inmage that can be displayed" and not in the electronic sense

 

I was also referring to map resolution.


But I agree its much more than that, just being able to use high impedance very quick injectors these days for example makes for better options than when it was almost a case of greys or nothing!

 

eaten too many eggs - they need to have more fibre in their diet

 

 

 

 

Just a quicky

Why is it that when the turbo charger is operating at boost that the Stoichiometry figure for AF is disregarded for a richer mix? (ie 14.7:1 becomes 11:1)

Is it because of the temperature increase that occurs?

Thanks

JJ

 

Stoich is still 14.7, thats still the "chemically correct" mixture.

But that would be too hot a burn, so is no use, and would not give good power, so is no use.

12.5/12.6 is the point of peak power, this is what you want to be running if you are able to without heat issues, on a YB though, high 11s is as close as most people go for safety reasons.

 

Though 14.7:1 is satisfactory for low/small load situation (ie closed loop situations)?

Cheers

JJ

 

14.7 is fine for light loads, 15.2 is perfect for economy.

 

Thanks matey -

JJ

 

a picture speaks a 1000 words again