john banks

Thinking of using a manual boost controller - Dawes.

1. What are the implications of running higher boost on part throttle - should the ECU (PPP) just find the point on the map using the MAF sensor data?

2. When my peak boost hits 18-19 PSI or holds there a bit long at present, the boost backs off a bit before rising again. This is NOT abrupt. This does not occur if the peak/held boost is 17-18PSI. I assume this is the ECU controlling excessive boost. How does it do this: a) fuel cut or b) altering solenoid pulsation to drop the boost a bit or c) some other way? I am wondering if those using EBC/MBC still get this safety feature if the boost is a little high for too long. Obviously you still retain the BIG fuel cut if you don't fool the MAP sensor - I am not intending to fiddle with this.

3. Am I better monitoring Lambda or Knock or both with this sort of system with PPP, DP, induction. I am only intending to run the 1.2bar/18PSI that the current system peaks slightly over and holds slightly under. Would Lambda monitoring highlight problems earlier - ie lean(ish) mixture before a Knocklink would light up and therefore be a better early warning? Of course if things were a bit lean I could drop the boost a bit or remove my induction kit.

I am hoping the (presumably) higher fuel cut of 19.3PSI on the PP ECU along with (presumably) better fueling will help me here as well as always running SUL and also considering adding OB.

Is the Select monitor a good way of ensuring that the fueling is adequate and there is no knock once I get things installed, saving me filling my car with monitors with flashing lights?

Dave R

Where are you sourcing your Dawes valve from?

Cheers
Daver

john banks

Group buy in general section or failing that I'll order direct.

john banks

?

john banks

BTTT

GavinP

John,

As understand it, the ECU calculates the correct settings using all the engine sensors and looks up the relevant part of the ECU map from this. Pressure is indicated to the ECU by the MAP sensor.
The MAF sensor measures air flow mass into the engine and is used for base ignition timing.

As for the ECU regulating boost, I would think it is the ECU pulsing the boost solenoid - the fuel cut is very abrupt like turning off the engine for a second. If this feature is retained, I would say is dependent on how you use the Dawes - if you remove ECU boost control, then it would not happen. If you use the boost controller on a low setting (under 7psi) and place it in the feed line to the wastegate solenoid, this would reduce the leakage.

What I would be wary of personally is if the ECU is using "leakage" as the method of regulating the boost on your car - which sounds a distinct possibility.

I would be interested as well if anyone has a definitive answer....

I have a Knocklink and Lambdalink and watch both (although I do spend most of the time watching the road! ) - they are both good early warning systems so worth having IMHO.

If I were you, I would seriously consider a trip to PE or somewhere similar to setup the boost on a rolling road whilst being able to check A/F etc at the same time. PE charge about £100 for this but compared to the potential bill for repairs, it is a bit of a no-brainer really.... Peace of mind would be worth the outlay.

As your car is running higher than normal boost already, I would think it is by far the best option IMHO.

Thanks

Gavin

john banks

Thanks Gavin.

With my current boost levels, the Select monitor detected no knock or lean running. PE is 350 miles away! There is a RR nearby, but I don't think they know particularly much about Scoobies.

It helps now I know my boost guage is under-reading by up to 2 PSI! Therefore when I got fluctuations I reckon the MAP sensor was reading 19.3PSI. Previously it hit 22-23PSI without the abrupt fuel cut out, so I almost believe it would never do it at any boost levels I would venture towards! I wonder if this is different on PPP ECU. Mike Wood did say there was still a cut out on PPP - I am just wondering if it is a little less savage than other Scoobies. This one is no more vicious than the rev limiter, which on my car is very soft indeed. On Peugeots it is a true brick wall moment in comparison.

If I set the controller at 7 or so PSI, I don't understand how it makes a difference if the wastegate spring is about this strength - so boost hits 7PSI and the Dawes pops open at the point the wastegate would have only just started to open anyway. How about more like 15 PSI and let the ECU control the last 2 PSI?

Hoppy

Good luck with this thread, John. Everyone in the Dawes Group Buy is waiting for you to get this sorted before we all blow our engines to bits Bob Rawle's been a bit quiet, which is worrying

Richard.

Derek.P

John,
At part throttle/cruise the ECU is working in closed-loop and
uses the Lambda (oxygen) sensor to adjust fuelling to keep the
mixture at the stoichiometric point. Increasing the boost pressure
slightly with the Dawes MBC will be compensated for by the
Lambda/MAF sensors and as long as you don't defeat the ECU
overboost protection you "should" be ok.
I would recommend using an A/F meter to keep check on the mixture.

Cheers,
Derek..

john banks

Cheers Derek.

Sounds like an A/F meter is on the cards and sticking to 17PSI maximum for me. This is what some of the guys with EBC's were running - Nito Morelli and Andy Tang.

But if my findings are anything to go by, don't trust any old boost gauge.

I'm still looking at some bits to wire up my own A/F meter based on Bob's data in http://www.scoobynet.co.uk/bbs/thread.asp?threadid=30133&Page=3

Derek.P

----------------------------
If I set the controller at 7 or so PSI, I don't understand how it makes a difference if the wastegate spring is about this strength - so boost hits 7PSI and the Dawes pops open at the point the wastegate would have only just started to open anyway. How about more like 15 PSI and let the ECU control the last 2 PSI? If I set the controller at 7 or so PSI, I don't understand how it makes a difference if the wastegate spring is about this strength - so boost hits 7PSI and the Dawes pops open at the point the wastegate would have only just started to open anyway. How about more like 15 PSI and let the ECU control the last 2 PSI?
--------------------------

John,
The wastegate starts to creep open (loosing boost)as the pressure
rises and is fully open at around 7psi at the wastegate actuator
- with the other 8-9psi being bleed off by the wastegate solenoid
valve.
The Dawes MBC stops most of the creep by not cracking open until
it reaches a set pressure, say 16psi, then the pressure passes
through the MBC and opens the wastegate to regulate the boost
pressure. As the boost falls the MBC shuts again causing the
wastegate to shut and the boost pressure starts to increases
again...and so on.

If you read the articles I sent to you previously, they explain
it a bit better.

Going back to the RST days, one problem that can occur is
the wastegate poppet valve getting sucked open by the exhaust
gasses going through the turbo, due to the "weak" spring on the
actuator, this was resolved by fitting a bigger wastegate actuator
and connecting it straight on to the inlet manifold and adjusting
the actuator rod to set the boost pressure.

I have fitted a Dawes type MBC on an old Subaru XT Turbo and it
works far better that an ordinary bleed valve.

Cheers,
Derek..

john banks

Thanks Derek. So the wastegate is FULLY open at the spring pressure. I understand how 7PSI setting can help now.
Those earlier articles were very helpful.

Derek.P

Gavin,
The "standard" car ECU modulates the wastegate solenoid to
bleed pressure away from the wastegate actuator thus allowing
the boost pressure to rise higher than the actuator pressure,
i.e. the ECU can control the boost from approx 7psi upward.

The MAP sensor tells the ECU what pressure the inlet is at and
it uses this infomation to tell when the car is overboosting
& applies the cut-out.

The Dawes type MBC should have one side connected to the
wastegate actuator and the other to the inlet (note it is
critical to fit this device the correct way round!). For
best effect, the wastegate soleniod valve is not used but
left electrically connected to the ECU.


Cheers,
Derek..

GavinP

Derek,

Thanks for that - I already have a Dawes fitted between the turbo "nipple" and the wastegate actuator. Like this diagram on the Dawes site:
Here

The MAP sensor is used for the fuel cut but is also used by the ECU to calculate the correct ignition and fuelling for the given boost pressure.

Most of the people using aftermarket ECUs "taper" the boost to dip slightly at high revs for safety reasons which is only possible with ECU control (i.e. the method we(?) are using would not do that).

The other consideration is that if the stock ECU detects a serious fault and limits boost to 7psi - this would also not be implemented. You would however see a "Check Engine" light.

It's good to hear other people experiences with the Dawes..

Thanks

Gavin

john banks

So are we saying that the "soft" boost reduction I get is a fuel cut then? Are we saying the ECU uses the MAP to detect overboost and fuel cut is the only way it deals with it? Then the ECU controls held boost using the solenoid in a feedback circuit?

john banks

Gavin we cross posted. I'm glad we're improving the collective knowledge base re these MBCs as we are all getting into them (most of us theoretically so far!)

Interesting point about boost limit/limp home Gavin.

I am slightly anxious about the higher boost at high revs hence my trying to understand the Lambda sensor output at WOT/open loop as this looks to be the zone where the damage could be done. Hopefully there will still be plenty of fuel with a health duty cycle on the injectors from what the really mad modifiers say. Also of note from i-club is that there is little benefit in putting held boost much over 17PSI on a TD04 at high revs even if it does it as it is inefficient and does not seem to liberate more power.

[Edited by john banks - 10/28/2001 10:02:56 PM]

GavinP

John,

[Cross posted again... ]

It is not using a fuel cut in the OE sense as by their very nature, will be abrupt. On my car, it feels like you have just hit a brick wall!

As Derek says above, the ECU is probably manipulating the wastegate solenoid into taking measures to keep the boost within limits.

Possibly the Prodrive ECU program is more sophisticated than the OE ECU - it doesn't use the "hard" fuel cut at all just moderates the boost by other means ?

Mind you, I thought that that part of the reason of a fuel cut was as a safety net if the wastegate failed to open???

Sorry, more questions than answers...

Thanks

Gavin

[Edited by GavinP - 10/28/2001 10:34:44 PM]

GavinP

John,

As you say, it's good to keep the discussion going.

If I were you, I would see if your local tuner can hire you a wide-band air/fuel meter to check your fuelling either on the road or test it on their rolling road if they don't have a portable unit.

The lambda sensor on our cars is apparently not really accurate enough to be depended on - not something I have tried first hand but seems to be a very common opinion shared.

I don't know what MY car you have but I have heard that the U.S. /new age models use a larger turbo than the MY99/00 cars - even though they are both "TD04"s.

I haven't seen the efficiency maps of the different turbos but it is possible that the U.S. guys are nudging the limits. You could measure your intake temps which would provide some insight.

Life is never simple is it ?

Thanks

Gavin

john banks

I have a MY00 UK spec. Some of the US guys get twitchy with 16 PSI on the new age machine, but I could not see why.

My plan so far is to sort some sort of AFR meter (see other thread in drivetrain) and run the Dawes at 15,16 then finally 17PSI and check the fuelling at high revs.

The only remaining concern I have is what the extra boost could do at high revs as this is where the difference is greatest on my car. But the guys with OEM ECUs and EBCs seem OK holding big boost to the red line.

john banks

This is on a US MY02 but it really shows the difference in boost at the top end.

http://wrxworld.com/Datalogging/boostmeasurements.htm

Derek.P

John,
As Gavin says in his post, our cars use narrow band 02 sensors
which are only really accurate around the stoich point, the
ECU doesn't use the 02 sensor at WOT because it is not an
accurate reading that could be used to control the rich fuelling,
so it goes into "open loop".
The professional AFR meters use expensive wide band sensors and
also apply exhaust gas temperature correction to the reading. The
temperature causes quite a variation on the output of the 02 sensor.

If you want to check the WOT AFR then it would be best to do it
on a rolling road with a proffessional AFR meter, since you will
never get anything near the correct reading with the narrow band
sensor and a simple meter. I suppose its a bit like your boost
gauge - how accurate do you want to be? - the more accurate the
higher the cost!. In saying that the LambdaLink is just a simple
meter which gives you an indication of what your AFR is & doesn't
apply EGT correction to the reading - and everyone swares by them!!.

Cheers,
Derek..


[Edited by Derek.P - 10/29/2001 12:08:32 AM]

dowser

I guess it all depends how you want to use the Dawes - for me, I just want to set it at 7psi to see what difference removing wastegate creep has on the boost curve. For this, it'll be connected after the t-junction to the boost solenoid still giving the ecu control of boost above 7psi.

Shirley, anything else is just a regular bleed valve...with all the associated risks?

Cheers
Richard

john banks

Sounds like I can get as good as a Lambda Link then with a reasonably accurate DVM. That's maybe the route I'll go down initially - so does 0.90 to 0.95 V sound the right sort of ballpark then under WOT?

StephenDone

Hi John,

I believe (anyone correct me if I'm wrong) that a lambda sensor becomes inaccurate as it gets too hot - e.g. when you are driving hard. The sensor becomes rather 'pessimistic', reporting that you are running lean when you in fact are not. As I understand it, this is safe, since the sensor won't tell you that you are running rich when in fact the mixture is lean - it is always the other way round due to the effect that EGT has.

Steve