while playing about with a boost gauge last month i discovered that when my 99my uk turbo was not on full throttle @ about 2000rpm (starting to boost)the oe dump valve was being sucked/blown?very slightley open resulting in a loss of 4psi of valuable pressure.so i made a mounting plate out of 3mm plate and a bit of 1"pipe and fitted a bailey evo dump valve and can confirm the car now picks up much quicker than before.i was talking to a new scooby related friend last week and metioned my discovery and he said he felt his 99my ireland import to be doing the same,so we spent a sunday afternoon swapping my evo set up to his car with the same results so give it a try.

 

how can u see this as i a have a UK MY99?

 

explain what you mean

 

Sorry, what procedure should i do to see if this is happening to my car, i have a MY99 and HKS Boost Gauge

regards

 

Sorry, what procedure should i do to see if this is happening to my car, i have a MY99 and HKS Boost Gauge

regards

 

MTR, you are getting totally confused on how an Anti-Lag system works.

I run a Group A Cosworth rally car with Pectel P8 mountune mapped engine management, using ALS and water injection.

This is how it works.

Firstly, boost is controlled very acurately by using 2 fuel injectors as air valves, these feed the boost/atmospheric pressure to the waiste gate actuator, allowing minute and very fast acting boost control.

Secondly, the normal method of controlling engine iddle via an idle control valve, that bypasses the throtle butterfly is dispensed with, the static idle is set in the old fashioned way by a locked off bypass screw.

Thirdly, where the iddle control valve used to be, a new valve is fitted that has a solenoid controlled valve (looks like an engine inlet valve internally) that can open/close the throttle buterfly bypass orifice very qickly.

Forthly, a miram turbo shaft is used to withstand the massive increase in turbo temperature generated by ALS.

Fifth, there is no dump valve used on an ALS equipped engine, the whole point of ALS is to keep boosting the engine, even on closed throttle.

The way it works is that when you come off the throttle, for example braking hard into a corner, the throttle butterfly closes as per normal, once the boost drops to a certain level, the ECU starts to pulse the Throttle butterfly bypass valve, this has the same effect as you constantly stabbing the throttle on and off very quickly, this means that the engine is being driven by the ECU as though it wants to accelerate again, however this occilation of the throttle bypass valve is happening so fast, it has the effect of allowing the turbo to stay spinning yet produces no additional torque to drive the car forward.
The effect of all this is that massive heat is generated in the turbo exhaust side, which ignites all the excess fuel, there is no deliberate overfuelling going on to explode the turbo impellar area.

The most modern of ALS systems on the current world rally cars are far more efficient than a couple of years ago, which is why the ALS banging was less aparent this year on the RAC rally.

The turbo used in ALS equipped rally cars is smaller than you would see on a race car for two reasons. the first is that the FIA has a mandatory 34mm turbo inlet restrictor that limits air flow into the turbo, this means that it is pointless useing a large inlet turbine unit as the inlet could not flow any more air, secondly the lag would increase due to the inertia of the larger impeller.

The water injection system works by injecting a fine mist of water/air into the inlet plenum chamber when the boost/air charge temp reaches a certain level, the reason WI is used is to increase the effective octane rating of the fuel/air mixture to elliminate detonation, this allows far higher boost to be maintained, even in very hot climates.
This is not the same thing as intercooler sprays, these cool the external surfaces of the intercooler to help the cooling efficiency of the air/air intercooler.

A conventional dump valve uses a flexible membrane to vent to atmosphere, this membrane is held shut by two things, firstly there is a spring that pushes the membrane valve shut, secondly the back of the membrane is presurised by the boost in the inlet plenum chamber.

What happens when you come off the throttle is that the throttle butterfly closes, this causes a vacumb in the inlet plenum chamber, this causes a negative presure on the dump valve membrane forcing it open to vent the boost to atmosphere.

The problem with these standard valves is that the construction is cheap, the first thing to go is usually the spring looses its effectiveness, this has the effect of not closing the dump valve fully untill the boost in the inlet plenum starts to build, the presure on the membrane daiphram then forces the valve shut.

I suggest this is what you are seeing.

The second thing that goes wrong with these dump valves is that the membrane splits due to fatigue, this is more common on uprated boost or older cars. This has the effect of leaking away to atmosphere some of the boost, resulting in a loss of top end power in particular.

The bailey EVO valve works by replacing the membrane with a plunger valve, this is far more effective at coping with high boost and is made to a very high standard using alluminium rather that plastic.

The other thing you are confusing is the dump valve and the waist gate are not the same thing.

The dump valve vents the excess inlet boost on closed throttle to either atmosphere or more commonly on production engines back into the turbo inlet to be recirculated.

The bailey EVO valve vents directly to atmosphere, which is why you hear the woosh noise. the Dump valve is not controlled by the ECU at all, but is a purely mechanical device relying on a spring and pressure to operate.

The waiste gate is a valve in the exhaust side of the turbo which bypasses the turbo impellar. This is controlled by the waiste gate actuator. The waist gate acuator is a diaphram device that is controlled by the ECU to open the waist gate at a set boost level. This works by bypassing the gasses into the exhaust pipe and not into the turbine impeller, thus limiting the boost the turbo can produce. If you wish to increase the boost of your scooby, then you need to reprogram the ECU to not actuate the waist gate until a higher boost level is reached.

A cheap and very nasty way of doing this is to fit a bleed valve into the actuator air feed plumbing to reduce the effectiveness of the waist gate actuator, or you can change the preload on the waistgate actuator rod, both very nasty but do work.

I hope you find the above informative

 

put your car in 2nd or 3rd and build the revs up from about 1500rpm then hold them steady @2000rpm and make a mental note of boost pressure,also listen to the engine you will hear a hissing sound if your car has the same problem as mine.now put your foot down just a little further and you will hear the noise stop and the boost gauge should now read alot more than it did before with only a slight movement of the throttle. one explanation given to me was not enough boost and to much vacume overcomming the dumpvalve and opening it a tad.i dont know if this is correct but it is a better reason than i can come up with.perhaps my oe dump valve was just knackered,either way it gave me a good excuse to fit a cool sounding valve...hope that helps?

 

MTR

[This message has been edited by MTR (edited 17 September 2000).]

 

mtr, your comments are true the point i was trying to make was i think my dumpvalve was at fault and it seems a coincidence that another scooby not 400yds from my house and also a 99my had the same problem!the lancia by the way was called the 037.

 

i think i got one thing wrong above.

The standard bailey piston dump valve vents to atmosphere.

The EVO piston dump valve uses the standard method of dumping to the air box/turbo inlet.

The operation of the 2 units is identical, but the standard bailey piston valve is much louder.

 

On a standard car, the dump valve that vents to atmosphere is not such a good thing. The reason being is that ECU wonders where the air has gone and as such adjusts the fuel to maintain what it reckons should be the Air Fuel Ratio.

This means that the AFR changes and that can have a detrimental effect on the motor.

Ben

 

MTR

[This message has been edited by MTR (edited 17 September 2000).]

 

Ben,

All dump valves dump to atmosphere, that is the only way they can work. Atmosphere includes upto the inlet of the turbocharger remember, i.e. unpresureised.

What you might be refering to is that on some cars that use a air flow meter, like the RS Turbo for example. The fitting of a standard type dump valve used to set up occilations in the inlet tract and cause the air sensor flap to move irratically.

The Bailey piston dump valve does not cause this.

On the cosworth there is no metering of air flow at all, it relies entirely on the map in the ecu and all the temp/pressure/phase/crankshaft possition/MAP sensors to select the correct injection/ignition outputs.

On the scooby the air is metered by a thermister type sensor with no flap to occilate.

I dont understand how you come to your conclusion above for the skooby, but as i am no expert on the subaru, but know a great deal about the cossie engine config, then i am open to being educated.

Please explain your statement, as i just dont get what you are saying.

 

MTR the car you are refering to is the Lancia Delta S4, Group B car that Marku Allen and the late/great Henri Toivanen drove in the 1985 RAC onwards till the end of 1986 season when group B ended.

The engine was both turbo and supercharged.

The reason this config was succesfull then but is not usefull now, is because the engines had no inlet restrictors fitted, so a very large turbo was used for max power (500BHP). This caused massive lag, which is why the supercharger was introduced.

The engine management systems of the day were also relatively slow so an effective ALS system was not feasable at the time also.

The engine losses associated with this config now, with engines limited to approx 340BHP would be too great to make much use of the config.

One thing i am very interested in at the moment is the centrifugal(turbine) superchargers that have recently appeared on the market. These are very similar to a turbo on the compression side, but are belt driven on the drive side. They have the advantage of less lag and less heat but i have seen problems with tolerances causing low boost and supersonic stalling on the units i have seen used to date. I am sure there is some mileage in developing these units further.

 

The problem initially described is substantially correct but it is not caused by a faulty dump valve. As John has stated two things keep a dump valve (bov) closed, spring pressure and boost pressure. The spring has therefore got to be able to hold the valve closed when vacuum is present above the diaphram instead of pressure. On cruise, up to 0.5 bar pressure can build up in the intercooler, contrary to popular opinion the turbo is spinning quite fast even under cruise conditions. So we have say 0.4 bar pressure in intercooler, say minus 0.3 bar of vac in the manifold total pressure difference is then 0.7 bar. As most dump valve springs only hold 0.3 bar max then the dump valve will continually bleed potential boost away.

To overcome this all that needs to be done is to increase the spring rate to suit the car. WRC cars do not use a dump valve, as an experiment I blocked off mine and immediately found better pickup especially when changing gear, the down side is that pressure bleeds back through the turbo impeller causing shock loading to the turbo bearing system. So a compromise is a spring that vents higher boosts, above 0.6 bar say but stays shut at cruise. My dump valve (which is the early plastic cheapy) cracks open at 0.25 bar, I have dismantled it and shimmed the spring to increase that to 0.6 bar, the result is as described above. I have actually taken the trouble to check all this out using twin boost guages one each side of the throttle butterfly.

The EVO Bailey bov is actually a twin piston type to overcome the tendancy to leak as described, it still vents to atmosphere though. Bailey do a closed loop one but it will not fit an Impreza yet. The EVO valve was developed to overcome the idle problems a traditional open bov caused to Vauxhall's and Ford's.

Bob

So the improvement comes from increasing the spring rate, not changing the dump valve although changing to one with a stronger spring obviously provides a fix. It is not good to allow a bov to vent to atmo as it screws the mixture up and can confuse the ecu.

 

MTR

[This message has been edited by MTR (edited 17 September 2000).]

 

Bob, i still dont get why venting to atmosphere would screw up the mixture, as the only time the valve should open is when you are off the throttle and the ecu shuts the fuel down then anyway.

Are you saying this because there is an inherant fault in the standard dump valve?

The main reason production cars are fitted with a dump valve is not to reduce lag. It is to improve turbo life by preventing the shock pulses bob explains.

On my 350BHP cossie road car i have just sold, i had no dump valve at all, this is because the turbo i was using had a 360 degree thrust bearing fitted that could cope with the shock loading, you actually get MORE lag with a dump valve, not less.

Most aftermarket dump valve advertising is pure bull, the sad thing is that lots of mags like CCC etc backup the bull***t.

 

In a car that uses a mass air flow meter all the air passing through it into the engine is measured, when you use a recirc bov the fact that it recircs means it stays within the system and when it gets through the engine the mixture will have been correctly set according to the maf and lambda. If you then vent a pulse of air to atmo its lost and does not pass thro the engine, this means that the ecu can mis-read the mixture as being too rich, it then leans it only to find that its now gone too far the other way and you get a big flat spot and hesitation. This does not happen all the time or an all cars. I know of cars running an external bov that are fine, I know of others that were terrible afterwards. Certainly any Vauxhall (Bosch engine management) can suffer from this unless the fueling is modified to compensate. Programmable ecu's can be mapped to compensate as the accel enrichment is changeable to suit.

Don't forget that the bov vents whenever you lift off, it does not need you to be off the throttle, ask anyone with a Blitz or HKS filter to tell you about the lovely hissing and wooshing noises that occur as you move the throttle (not lift off). So trailing the throttle loses air momentarily and sends the mixture rich, compensation starts and the ecu can take a few seconds to sort it out. Its during that short period that you get hesitation etc.

I suppose the inherant fault is that the bov's normally come equiped with a weak spring to minimise the turbo shock. Its a trade off against good throttle response. The trick is to modify the spring to suit the car.

Bob

[This message has been edited by Bob Rawle (edited 02-01-2000).]

 

Thanks for that bob.

I only have a sport impreza so cant check this out myself.

Based on what you are saying, the dump valve must be venting back into the inlet tract, after the air flow meter, and not into the air box before the filter.

Is this correct?

 

I'm learning some interesting stuff today and have some questions and comments:

Firstly for John:

I didn't know that WRC cars didn't use blow-off valves (dump valves).I always thought the fluttering sound you heard on TV was the valve releasing the air. What is creating that sound? Also, since when has this been the case? I have a photo of the engine of an EVO3 GRPA car that appears to have a blow-off valve.

Secondly when you were talking about the ALS, when the valve pulsates and let's air into the engine when the throttle is closed, I assume spark is killed. Is this so? Also the Mitsu Evo 4-6 road cars have an ALS system, do you know how they work?

How does the blow-off valve contribute to lag? Thinking about it and as the adds state, whne the throttle is closed it would seem to make sense that the air attempting to flow back through the compressor would slow it done as well as shock loading the bearings. However, as a test, I did disconnect my blow-off valve for a short run and did not notice any increase (or decrease for that matter) in lag, so I would tend to agree that the valve is more for turbo life.

Does your cosworth not use an oxygen/lambda sensor then, seeing as al fueling is done off of maps?

What bob says about the ECU getting confused if you blow the air off to atmosphere confirms my experience. The other problem I had is that with my stock valve (Lancer EVO I) it would open at idle because of the weak spring, creating a big vaccum leak that caused the car to stall when trying to idle. This does not happen with the HKS, BLITZ etc because they have a much stiffer spring. The problem when changing gear or backing off th throttle is still present with those valves though.

 

John, the bov vents back into the air inlet tract between the mass airflow meter and the turbo.

By blocking of the bov so that air cannot recirc you prevent the bleed off of useable pressure that can build up even when there is vac in the manifold. As you then further open the throttle this pressure is released into the engine prviding a swifter pick up. It does very much depend on your driving style as to how much benefit you see. Its particularly noticeable on cars running enhancements. I thought that the Cossie boys had been doing this for some time.

Bob

 

Blimey guys, i thought this was a simple thing

What should i be doing, if anything?

Regards

Simpleton of Surrey

 

As mentioned by Bob, the BOV on the Subaru EJ20 engine is a recirculatory design, venting back into the inlet after the MAF/AFM, a few inches away from the turbo impelller. The standard plastic one is very easily opened.... it will leak when there's manifold vacuum, ie pretty much any time other than WOT. I ditched mine for a rather nice (but not Bailey) non-recirc BOV. The effect is to a) improve throttle response and b) cause the car to backfire under certain conditions... the ECU has injected enough fuel for the amount of air metered but part of it is missing making the mixture rich. When stepping on the go-faster pedal again the excess fuel in the exhaust ignites and you might even get a flame out the back!

The effect on turbo-lag of the BOV depends very much on your driving style and the inertia of the spindle. With a "light" spindle and slow-ish gear changes the BOV will make a big difference. If it's a big turbo and you're a God at changing gears, a BOV won't make any difference to the power delivery (but may be slightly kinder on the turbo).

If the BOV is open under "normal" conditions, it will have to close before pressure can build up... this would manifest itself as lag. One way to fix the problem is as Bob has done, to dismantle the BOV and adjust the spring pressure... unfortunatley the design doesn't easily lend itself to being dismantled and re-assembled :-( An alternative would be to stick a vacuum reducer between the manifold and the BOV; the trick is to find one :-)

Personally, I quite like the non-recirc because it's quite obvious when it's open (you can clearly hear it!) but I've yet to get a chance to see what affect it will have on the ECU's idea of the AFR.... thus far it seems quite happy.

Cheers,

Pat.

 

bob, that explains your statement on confusing the ecu then, thanks for that.

On the cossie motor there is no air flow meter at all, the map does everything.

If you talk to ahmed bayjoo about dump valves he will tell you that the only cossie engine that actually needs one is an RS500 motor running high boost, producing above 450BHP, this is because the turbo is so large and the boost is so high (32PSI) that the shock loadings are huge on the turbo bearings.

On smaller turbo's, even my hybrid T04/T3, which would produce 28PSI at 7000RPM, which is large for a road car, does not see a large enough reversal in flow to slow the compressor wheel, which on boost is revving at 70,000 RPM.

You get much better response without a dump valve fitted as the boost cannot leak away, dont forget also that even on closed throttle the presurised air is still going into the engine via the idle control ports, it is not an air tight seal, which would have a major effect.

Wreckleford, the noise you hear is the waiste gate chattering as the gate opens and closes rapidly to limit the boost to the ECU controlled level. this is the valve in the turbo housing that bypasses the exhaust gasses down the exhaust pipe and not to the turbo impeller, thus limiting the amount of gas the exhaust impellar sees and therefore boost produced. The very early turbo designs didnt have a waiste gate( talking upto F1 years) but they were unreliable as they kept overboosting themselves and disintigrating.

The EVO3 car you have seen will not be running ALS if it has a dump valve, ALS simply cant work with one fitted. There are lots of Group A rally cars out there that are not genuine group A on the engines, you have to spend a lot of money to get a pukka ALS system on the car, the turbo alone is in the £1500 mark, with the uprated Miram shaft to withstand the colosal heat generated.

The spark is not killed during ALS operation, the combustion process is still needed to drive the exhaust impellar. I know it is hard to believe that you can have an engine producing power and yet not driving torque but it does, that is why ALS only apeared relatively recently, as the control software is complex.

Any road car that says it has ALS is not producing the same levels of boost off throttle that a pukka rally car ALS system does, not even close. On road sections between stages we have to switch the ALS off as the effect of running this in public road driving is unbelievable to see, with massive heat generated. The heat is so great on these ALS systems that the exhaust manifold usually lasts about 15 rallies before fatigue cracks it. We have to run massive heat shileding in the engine bay area as everthing just melts/burns otherwise.

The scariest thing on these ALS systems is when the occilating valve sticks open, giving the effect of about 50% throttle, it makes braking into a 90 right from 125MPH in the forest very interesting.

We now remove this valve every service, approx 20 Miles, and clean it to remove sut on the valve surfaces and we have never had a stuck valve on stage since.

No uprated cossies use a lambda sensor, the map does everything. If i was uprating a skooby myself i would junk the air flow sensor and lambda sensor and fit an fully mappable ecu. I am not really sure why subaru are using an air flow sensor as it is not needed on a fully mapped engine, but having no real experience to speak of on the skooby tuning i am open to an explanation of this.

 

John,

the reason why production vehicles use an AFM/MAF is that in order to ditch it and rely on MAP alone would require individual mapping due to differences in VE due to manufacturing tolerances. It's cheaper to fit and AFM/MAF than it is to setup each engine on a dyno before it leaves the factory :-)

You are right, if you do map the engine yourself then you can ditch the AFM/MAF. IIRC when fitting the Link you run it without the AFM/MAF... also, things like the HKS PFC-FCON ecu supply their own MAP sensor and run the injection and ignition ignoring the AFM/MAF reading...

Cheers,

Pat.

 

Pat, you are wrong on that one.

Every single sierra/sapphire/escort cosworth was sold with no airflow meter on a standard factory map. They produced between 205 and 220 BHP depending on model.

The only reason i can see to using an airflow meter is to allow a cheap, low speck engine management system to be used. The cossy had a very high spec Magneti mareli/ weber system that was continually updated, untill they did the last small turbo escort cossie that used a ford EECIV ECU that every junks when they upgrade the motor.

 

John,

I stand corrected on the AFM/MAF issue w.r.t. Cosworth engines. But if you look at, say, the EEC-IV you'll see that it runs an AFM with an option to attach a MAP sensor...

I'm not sure what you're getting at w.r.t. cost on the ECU. The JECS ECU already has a MAP input and could, theoretically and with a different control program, run the engine only off the MAP reading, so I doubt that it was a cost constraint.... could be wrong, though.

Cheers,

Pat.

 

you have all being putting you views accross on the subject and i have learnt lots,back to the problem i first mentoned,, go to the sidc and click on common problems,symptoms&remedies,does the first problem look familiar???

 

sorry,you have to select FAQ first.

 

Pat, you are getting your sensors mixed up.

The MAP sensor is used to measure boost in the plenum chamber, all turbo engines use a MAP sensor.

It stands for Manifold Absolute Pressure.

No cossie engine uses an air flow meter.

The ford EECIV unit is used on most of the 90's ford cars and can run with an airflow meter in normally aspirated engines.