john banks

MY00 goes to 10PSI with no restrictor.

StephenDone

Just to see, I'll remove the restrictor - temporarily. I'll also remove the return line which may/may not cause an appreciable vacuum. We'll see what the boost falls to. Then we'll know how strong the spring is on my car.

Cheers

Steve

[This message has been edited by StephenDone (edited 27 September 2001).]

Bob Rawle

CSA ... cross sectional area.

Actuator with no spring .... suggest you get to the dealer nearest to you ... if you need to that is.

Leave you guys to it, it was an interesting discussion b4 we got into springless actuators.

BTW there will only be boost pressure against the actuator diaphram, there is only vac in the inlet manifold and unless the turbo spins backwards !!! ... maybe that is true to.

With the turbo connected directly to the actuator you should get circa 7/8 psi (ish) do NOT be tempted to try anything with the actuator disconnected and left to atmosphere, unless you fancy a possible engine rebuild that is.

[This message has been edited by Bob Rawle (edited 27 September 2001).]

CharliePsycho

Sorry to leave it so long for the reply... Believe it or not I was fitting my APS induction kit ~ very nice piece of engineering, I'll do a write up later with piccies.

Anyway back to the point...

What I was trying to say was that laminar airflow doesn't in fact flow more than turbulent airflow, but it does <I>look</I> like it does!!!

The airflow in the centre of a laminar flow within a pipe is much higher than in a turbulent flow, but the edges are the same speed... However althogh the flow in the middle of a turbulent flow is <I>much</I> less, it is consistent across much more of the width of a pipe, hence actually flowing more air... (Refer to the piccie above, the red line is flow rates)

The MAF sensor seems to have a leading opening to accept flow, rather than flowing over a surface (as you might think) so it certainly seems to be looking at flows in the centre of the pipe.

This works sucking or blowing (air works a little differently to water (!) but we are initially discussing flow rather than pressure drop...

Anyway I'll think some more as well, Dinner time

...©

CharliePsycho

Hmmm,

I've put on the APS induction and although I have no quantative results (i.e. Dyno) it feels exactly as you have described with your analysis of the Cone+induction hose.

The engine revs more freely at the top end, but something is definetly missing at 2500 rpm, by 3000 it's like a switch has gone and the turbo can be heard spooling and your off.

On carb cars exactly the same thing would happen if you took the airfilter off and used a trumpet instead... you would have to re-jet to get the low-midrange response back (sometimes you never did) but ultimately power was always improved at the top end.

it feels almost similar to and induction/manifold leak; almost to the point of misfire at 2500rpm, wich would indicate lean running (uh-oh.) I am going to ring a few old-school tuning buddies to get a reaction.

...©

P.S. APS = Whhiiiiiizzzzzz Whoosh pop

StephenDone

Hi Bob,

&gt;Leave you guys to it,
&gt;it was an interesting discussion b4
&gt;we got into springless actuators.
&gt;
I'm sorry you feel that way. Ok so that was wrong, but it doesn't invalidate the rest of the thread. I still have a problem and will do my best to sort it out.

&gt;BTW there will only be boost pressure
&gt;against the actuator diaphram,
&gt;there is only vac in the inlet manifold
&gt; and unless the turbo spins backwards !!!
&gt;... maybe that is true to.
&gt;
Nobody said turbos spin backwards - people have just talked about the wastegate solenoid either connecting the diaphragm to boost pressure or the intake hose depending on whether it is energised or not. The debate was whether there was a small amount of vacuum in the intake hose. Logic says yes, since you are pulling air through a filter, and that this vacuum is lower once you have changed the air filter to one of lower resistance. Ok, we've established that the level of vacuum is insignificant relative to the spring strength, but then most of us don't do this stuff every day, and it's fun to understand. So please don't go, but tell us when we're wrong and why.

Regs.

Steve

StephenDone

Hi Charlie,

If that's your first mod, go get a dyno run anyway. The standard curves for our cars are all a pretty similar shape, and you'll easily see if the shape is different. Ignore the height though, since overall power/torque varies from car to car.

Get them to plug on a select monitor on graph view. You can have a look at stuff like air flow, knock, and lambda.

btw, the mail from the guy at APS said lean running at part throttle. Did he perhaps mean lean running at mid revs ? I thought part throttle mixture was controlled the lambda sensor, only going open loop at near full whack.

Cheers

Steve

CharliePsycho

OK, spoken to some old tuning brains, yes once we replaced the air filters it was common to re-jet a bit richer (the non webber carbs on Triumphs/BL engines you could adjust the jet depth on the bottom on the carb) to get the mixture right, although we had been known to put restrictors at the bottom of the trumpets where the engine had too big a carb as well...

So I am now of the opinion that it is indeed running very slightly week at low revs because of a less restrictive air filter.

I am going to try an ECU reset first off, such a big change to induction probably deserves it anyway. If that fails then it'll be off with the kit and back with the standard filter (maybe without the resonator first) as I cannot think of a reliable way (apart from a Link) of getting the mixture right. (although maybe a restrictor in front of the MAF is worth a try)

Obviosly the Scooby ECU is very induction sensitive...

...©

CharliePsycho

By the way the loss at 2500 is at any throttle position (indeed the Dyno is at full throttle anyway)

I don't really know whether closed/open loop is just switched on throttle position or revs as well...

Reading back through Bobs comments (he is just the guy to help on this) I'm not sure I get the larger inlet pipe idea, certainly on the APS it is identical, besides the MAF is in it's own tube which does not get altered...

I'm racking my brains though on why the MAF would not be able to compensate for a more free-flowing filter... more airflow at any given throttle position, denser air, less pressure drop across filter, resonance?...

Seriously cosidering putting a piece of masking tape across the bottom of the MAF tube against the mesh now (temporary test only, if it worked I'd use polycarbonate sheet)...

I am begining to think of the MAF as the venturi in a carb now...

...©

[This message has been edited by CharliePsycho (edited 28 September 2001).]

Bob Rawle

Stephen I'm still watching the thread, its just that more and more these days a decent technical discussion seems to get hi-jacked (if thats the right word)and goes off at a tangent. Yes its well appreciated that there are loads of people who watch this bbs (and contribute) who don't have as much knowledge as some others, your problem is pretty common one, these cars can be very sensitive to induction and exhaust changes dependant on many factors.

"To keep the turbo on pressure at low revs, the underpressure between the airfilter and the compressor is used to keep the wastegate closed.

When you have a high restriction filter you will have more negatieve pressure, the down side is that on high revs it cost some air."

If I misunderstood the words airfilter and compressor seemed clear at least.

Back to the problem .... the golden rule is that if something causes a problem go back to the exact conditions that existed just before the change that triggered the problem. So if you do as you suggested and revert all or part of the setup you should see or feel the difference again. Then by single substitution you will be able to determine which aspect is causing the change.

Lambda is used as a closed loop system up to circa 0.2 bar and say 5500 rpm, the maf is in use all the time and it is used to "pick the point" on the map for fueling and timing (lambda then being used to hit stoich when allowed), map sensor is virtually only used for boost control and fuel compensations.

Not that this will help too much but the output voltage from the maf with ignition on and engine off should be between 0.9 and 1.1 volts. At idle it should be between 1.0 and 1.7 volts, may be worth a check. Outpuit can be measured on the ecu B136-1 (blue ecu connector nearest floor, just probe with a needle to get a contact. Pin 1 is the pin lowest to floor in top row of pins and the wire colour should be green (if i remember correctly)

StephenDone

Hi Bob,

Thanks for the tips. I'll check some voltage levels with the various mods on and off, and see what changes.

Cheers

Steve

Just as an aside, here is the reply I got from APS on their cool air induction system, and why it runs lean at part throttle...

From APS after I enquired about their kit:
&gt; &gt; A word of caution however. Our HFA should
&gt; &gt; not be used with the standard ECU. It is
&gt; &gt; designed for very high flow
&gt; &gt; rates and can result in lean air fuel
&gt; &gt; ratios at part throttle.
&gt; &gt;
What I then asked:
&gt; I would be grateful if you could explain
&gt; why the above happens.
&gt;
What George then said:

With the APS high flow air inlet installed the air speed at the sensor
is lower. The flow through the air intake and through the air mass meter
is more plug like in shape. Hence the air mass meter reads a lower
*mass* of air than is actually flowing through the inlet.

This unfortunately is unavoidable with an air intake system designed to
flow sufficient volume of air to produce 550 - 600 hp. - We need to make
every mm2 of cross sectional area count.

There are many after market products to correct the AFR - Unichip being
one of those.

I hope this answers your questions.

Regards
George Couyant

StephenDone

A bit of an update....

I attached a digital volt meter to my lambda sensor last night. I am running on the rich side from 2000 RPM upwards - the readings being around 0.7 volts. At higher revs, the value can get up to 0.8 - 0.9 volts.

I assume this means that I am not running lean, and that this is not the cause of my drop in power at low revs.

Any thoughts ?

Cheers

Steve

CharliePsycho

So what we are saying is that adding an aftermarket/free-flowing filter to a stock car will actually decrease performance?

I wonder if it makes any difference at the top end?

I must admit, the area where it has dipped the torque curve is the area that the car is most used in traffic and is quite bad, if there is no real difference elsewhere then these kits are £150 worth of noise (not unlike BOV's )

This really says it all, I love the induction noise the APS system makes, not too loud in other words, but I suspect I shall have to make it more restrictive somehow to get back the performance lost in adding it!!!

I did wonder if putting a better downpipe would solve it, but your experience (Stephen) would suggest not. It would look like the only route to getting the best out of this fairly trivial mod would be to change the ECU, a somewhat less trivial mod...

Oh Well...
...CharlieP

GavinP

Hi

I have tried connecting the wastegate directly to the turbo "nipple" in the past and maximum boost I saw was 7 psi (which is what Bob stated).

One thing to bear in mind, the further the AFM is from the throttle, the worse the throttle response will be - adversely affecting driveability.

A couple of Autospeed articles about "Negative Boost" which you may find of interest:

StephenDone

Hi Charlie,

&gt;I wonder if it makes any difference at the top end?

I think you'd better look at the first post of this thread - I have dyno runs ! I gained &gt;10bhp top end.

&gt;I did wonder if putting a better downpipe would solve it

A downpipe improves bottom end. What I have done has subtracted about the same amount by adding the cone.

&gt;It would look like the only route to
&gt;getting the best out of this fairly trivial
&gt;mod would be to change the ECU, a somewhat &gt;less trivial mod
&gt;
I don't think this has by any means been proved to be true yet. The ECU argument was that I was running lean at low revs - checking the lambda sensor says that I am not. Without refitting my previous bits (which I will do soon), I cannot say whether I am running _leaner_ than before. I.e. before the mixture may have been even richer. Lambda=1 is not the best power - slightly richer than this is higher power. Perhaps I am now at lambda=1 rather than a little higher as previous.

Hi Gavin,

&gt;One thing to bear in mind, the further the
&gt;AFM is from the throttle, the worse the
&gt;throttle response will be
&gt;
I am keeping my AFM the same distance from the throttle. I may distance the filter from the AFM though. I don't think this will cause the problem you mention - haven't done it yet though :-) What d'you reckon ?

Thanks for the URL. There's some good stuff on that site.

Cheers

Steve

GavinP

<BLOCKQUOTE><font size="1" face="Verdana, Arial">quote:<HR>Originally posted by StephenDone:
<B>I am keeping my AFM the same distance from the throttle. I may distance the filter from the AFM though. I don't think this will cause the problem you mention - haven't done it yet though :-) What d'you reckon ?

Thanks for the URL. There's some good stuff on that site.

Cheers

Steve[/quote]

Steve,

I understood what you meant about extending the distance between the filter and AFM - I was considering doing the same. Leaving the AFM in the stock location will not make any difference. If you can get the AFM even nearer then it will improve throttle response - MRT sell parts to do this.

I had an exposed Pipercross filter for about 2 years and recently changed to a Green "Twister" filter which is encased in an alloy cone with 80mm ducting running right up to the front of the driver's wing - the car "seems" to be quicker and inlet temperatures have dropped by about 5 degrees C on average (more at low speed).

I expect that relocating the filter into the wing will bring similar benefits but will be a bit more tricky to clean....

Thanks

Gavin

Bob Rawle

Steven , you really need to be into the second decimal place of lambda output b4 you can start to differentiate in terms of this problem. The difference between 0.8 and 0.9 volts in afr terms is immense. Even 0.7 volts is not even close to stoich. I use a dvm reading to three decimal places and with a fsd of 2 volts for this.

StephenDone

Hi Bob,

What range of values should I be looking for ?
I can log them on a DSO, so voltage accuracy isn't a problem. Converting it to something meaningful is !

Like I say, readings are 0.8-0.9 volts higher up the rev range, but about 0.7 volts a little lower.

What does the sharp transition between 0 & 0.7 volts represent ?

Cheers

Steve

john banks

Steve - are you any further forward yet? My car is quite different to yours as I do not have the PE manifold, have a PPP ECU and a vanilla K&N 57i (without the trumpet or longer hose to the wing), with a Magnex DP. I cannot find this flat spot on the road. If you are running "leaner" but not actually "lean" on part throttle and therefore losing a bit of power low down, then it would all fit in with the explanations given about reduced turbulence leading to MAF misread. Could this be due to the longer hosing of the APS kit along with the BPM intake pipe you have? Could your PE manifold be having any effect?

I am trying to work out if cars with just a full exhaust and induction kit are likely to be suffering the same problems, and as yet have no clear answers. Sorry if my questions are a bit numpty....

Bob Rawle

The sharp transition is more between 0.4 and 0.7 volts, stoich is at about 0.6, the ecu rocks the mixture about this point which is why you see the readings changing, try this, take the readings at 0.5 bar steady boost and then 1 bar steady boost (or whatever is close) and see what readings you get, thats when the lambda sensor will produce steady output. if you hold the boost long enough you will see the readings falling as the sensor gets too hot.

StephenDone

Hi John,
========

No, I'm no further forward at the moment. I'm spending most of my time rewriting dyno.scoobynet at the moment. My brain can't cope with too many things at once :-) I am hell bent on fixing it though, don't worry !

Like Bob says, I think it's MAF reading related. The readings may be lower for the same volume of air, so less fuel is injected. I _will_ prove/disprove this, but I need to build some gubbins to read RPM, and this involves pulse widths rather than voltages, so I can't use my DSO for that. Then I can plot lambda & MAF against RPM, before and after the K&N swapout.

&gt;Could your PE manifold be having any effect?
&gt;
I think Bob said in another thread that he had to add more fuel when he mapped a new manifold, so I don't doubt it is contributing in some ways. Seeing as you have to jack up the engine block to fit my manifold, I don't think I'll be swapping it back to check though :-) I will swap back to my original air box and compare readings though. But before I can do this, I need to wait for some 89mm Samco hose to turn up - my intake pipe won't reach a standard air box - it's shorter to allow more room for a nice beefy air filter.

&gt;Sorry if my questions are a bit numpty....
&gt;
Not as numpty as my answers probably.

Hi Bob,
=======
What difference am I looking for between the two sets of readings taken at 0.5 & 1 bar ? Will 0.5 bar be closed loop lambda, whilst 1 bar goes open ? I.e. if the 1 bar readings are lower than the peaks of the 0.5 bar ones, then I am running leaner than I would on closed loop ?

Also, if I am seeing 0.8-0.9v at high revs, this is extremely rich at high revs isn't it ?

I'll get those readings.

Cheers

Steve

CharliePsycho

I think I am very close to solving it now.

I have place an obstruction before the MAF, against the mesh. basically I have blocked off the bottom 5mm of the tube (bottom as in the side that the MAF enters the induction tube...) I have experimented with between 3mm and 15mm in 4 steps, with comparisons against the original airbox... All seat of the pants I'm afraid, but I have a relitivly educated ****

The dead spot at 2500 revs has all but gone and the occasional roughness when hovering at these revs (60mph in 5th) has all but gone as well.

I've used extra thick insulating tape as a temporary baffle, I'm going to replace it with a polycarbonate sheet as a permanent solution, I am also going to play with the shape before the baffle, hopefully making the airflow vector up rather than just hitting the baffle and tumbling (although this may actually be what is giving the benefit!!) but the current solution is very nearly spot on... I'm at the point of deminishing returns now...

The idea was to increase the airflow over the MAF, hence making the mixture richer. Now I don't know if it has affected the top end much, it doesn't feel like it and there is still an improvement over the standard airbox. Oh and the noises are still just as good (which makes me feel better )

Without spending lots of money on a rolling road it is about as good as I am going to get it. So now I will make a permenant solution and write it up so I can post it along with the fitting guide for the APS. When I get enough money together I'll do a rolling road comparison, but it will need 3 runs on a warmed engine (never the best to work on ) 1 for standard, 1 for APS on it's own and 1 for APS with baffle.

In the mean time I can now enjoy the induction kit (and noises)without the annoying hesitation or worry of it blowing my engine... You never know it might have improved power as well! (well maybe )

...©

dowser

Steve

You don't think that's a bit extreme? You wrote software that can reprogramme (remap) a standard JECS ecu just so you could get around problems with your new induction kit, lol!....good on ya'

I'll happily put some cash towards the time you spent coding......

Richard

BugEyed

Awesome work Steve! I'd be with Richard about the donations .. if only the MY01 ECU wasn't so different.

Duncan

StephenDone

Solved !

The problem was lean running. Due to the induction mods I have, the air flow sensor voltage was low for revs between 1500 and 2500RPM. I proved this by data logging the air flow, lambda and RPM values. Lambda voltage dropped to near 0v for a short period.

Remapping the fuel map in my 99 ECU got rid of this problem completely. My turbo is now spooling up (and giving the same torque) 500RPM earlier than before. 50% of this is due to the fuelling, whilst the other 50% is due to a few degrees additional advance at low revs.

Thanks to all who gave suggestions on the problem. It's been 4 months since the last posting to this thread (it took that long to write the software), but I thought I'd reply back and let you know the result.

If anyone is interested, I've put a graph on my web site...

http://www.steve.ukmail.org/car/induction.gif

292 BHP @ 5610 RPM
300 LBFT @ 4510 RPM

All that remains to do now is a little bit of boost remapping to reduce the risk of any overboost.

Cheers

Steve

[Edited by StephenDone - 3/6/2002 2:55:00 AM]