electronic wastegate

[Davezj]

i had a bit of an Epiphany when i was at work today whilst sat on the bog. well i thought it was an Epiphany.

i will let you decide,

why do we use actual boost pressure to activate the wastegates, in a mechanical way, i know it has always been done like that, but the standard wastegates fitted to the vr4 do seem to have there limitation.

they are prone to creeping, and leaking.
this is what i mean, when one of our turbos is off the car you can easily see the wastegate valve next to the exhaust turbine wheel and all the turbos i have seen weather it is supposed to be like this or not they have been all the same, the valve flap does not seem to be slammed shut. the flap disk seems to be able to rotate quite freely, which implies it leaks all the time. not be much but it does leak and that is with no exhaust pressure behind it.

let me explain my dilemma, the exhaust gas come from the cylinder and into the manifold and these gases get forced against the exhaust turbine wheel and the wastegate valve, now i believe that the wastegate valve closes onto the flow of exhaust vales and the only thing that holds this valve shut is the spring in the wastegate actuator.
hope i have everything correct up to this point.

ed80_1.jpeg
this is the exhaust mounting side of the turbine

DSC02935.jpg
picture of the wastegate actuator if you are not familier

now to the nitty gritty of the rambling, how can you tell if the wastegate value is completely closed, up to the point where you want the valve to open to regulate the boost pressure.
with out watching the wastegate actuator rod whist the car is boosting i don't think you can.

i believe the wastegate spring force is about 4-5lb (this is a guess as it has to be less than the stock boost pressure and that is about 8psi) which equates to 4-5psi pressure on the diaphragm in the wastegate actuator. for this scenario lets remove the boost control valve from the equation and assume the wastegate is connected directly to the manifold pressure.
so with that 4-5psi being applied to the actuator the valve is fully open, so as the pressure builds the valve starts to open a bit then open a bit more and more until fully open. this i believe is called wastegate creep.
am i right in thinking that the ideal wastegate valve operation would be slammed sult until you reach the desired pressure then fully open to vent the pressure, then closed again to build pressure again. or would this lead to boost spiking. there could be a bit of a soft open and close on the valve to avoid boost spiking.

anyway my thought is to remove the wastegate actuator (spring, diaphragm, boost pipes) and replace it with a servo motor that has a 90 deg travel to open the valve and a driver linked to a pressure sensor in the manifold, the pressure converted to a voltage would be compared to a reference signal that would be set in the car via a switch. so the boost control would be via the rotary switch in the car. no need for any other manual or electronic boost control, which is a bonus.

i hear you say this is all a bit much to effectively get to the point we are already at. but there is a big plus point with the servo activation is that the servo will keep the wastegate valve slammed shut all the time you want it to be shut.

i understand that most people that already have some kind of external boost control making and changes like this would not be cost effective. but i am looking at this from first principles.

do i have some assumption i have made wrong or do you think there are major flaws in my thought.

don't be shy in posting your comments even if it is just to ask me to explain a bit more about it.

[foxdie]

One word; Temperature.

A solenoid / actuator / stepper motor / whatever isn't going to be reliable at the extremely high temperatures that a turbo can get to. If it fails, it could fail with the wastegate stuck open meaning no boost, or even worse get stuck closed which could damage the turbo and possibly the engine.

I think they're mechanical in line with the KISS principle. Less things likely to go wrong.

[Nutter_John]

diesels use electronic wasteagtes quite a lot so you may be able to cobble something off one

You do not want the wastesgates to just open and close as the you will end up having a very choppy ride going from no boost to full boost to no boost to full boost , it is far better to allow the gases to bleed off rather than snap on/off

This is why a boost controller does not keep the full pressure hidden from the wastegates and bleed it off

[jayp]

It's a good idea in principle, but like others have said mechanical actuaters are simpler cheaper and more reliable over the long run.

I would hazard a guess the reason the valve rotates is the same reason an exhaust valve rotates, to stop hot spots and the valve seat burning away.

If you could get components that will withstand the heat and work out a way to get that bit of valve creep to keep the boost curve smooth rather than lots of Sharp peaks and trough it would make adjustments simpler, and I would assume would cut out spikes either completely or as near as dam it.

Sent from my HTC Desire S using Tapatalk

[lancerevo3]

next time bring some reading material to the bog with you.

[Nick Mann]

I'd imagine reaction time will need to be considered too.

[Subaru ETA]

As John said - diesels use electronic waste gates. An example is Hyundai who are using them now to allow the ECU to control it.

The reason behind it was that it is hard to get a turbo to have the torque when you want it - this is where variable veins are handy.

Anywho, hyundais problem was that they want a shot load of torque to get then off the mark - which they did. Only problem was that they had cars not making it home when purchased brand new - the were over boosting and bend conrods and throw legs out of bed.

To fix this they went to electronic waste gates so the they could control the delivery of boost pressure better and not have a tonne of boost in 1st gear, causing the engine to go pop

[Davezj]

nice bit of thought gone into the responses, cheers guys.

dave (eta) you have forced my next line of enquiry.

my next thought was making the response of the turbo programmable depending which gear your in, you could use some kind of pic to control that, so you could have as much or as little boost in any gear you wanted, as long as the turbo can produce it. once you can get the control of the boost electronically then you can start doing all kind of weird stuff, programming wise.

the heat for me is not really an issue as the servo can be shielded from the heat. also remember the servo can be placed anywhere you can get a mechanical line of sight connection to the wastegate valve, the servo does not have to be located in the same place as the standard wastegate actuator.

this is just theory, and i am wondering what people think of the idea.

[paulg23]

I guess you could use a bowden cable to connect the servo, in which case the servo could just about be anywhere.

[Wodjno]

A decent electronic boost controller will do away with Boost creep.

[Davezj]

A decent electronic boost controller will do away with Boost creep.

Would you say a good electronic boost controller would cost about £150-£200.

Yes a boost controller would do it, but the follow on setup would not be possible, tuning the boost for each gear and throughout the rev range.
As you could use the rpm, and other signals to influence the desired valve opening as well as manifold pressure, so the shape of the boost curve could be set to anything below the maximum possible boost pressure.
This just theory at the moment, but I am thinking about it.

[Davezj]

I guess you could use a bowden cable to connect the servo, in which case the servo could just about be anywhere.

He goes off to google Bowden cable!

Ah, a bike brake cable or car accelerator cable, type of thing.

Yes that would do it and readily available for not much money.

It would also allow you to run both wastegate valves off the servo, which would be nice.

[paulg23]

Ebay item number 140624238587. Possibly use something like this as the actuator, then "just" need a controller.

[The Vee]

Surely though a cable will hold it shut when under tension but when released you'll need some sort of spring mechanism pulling the valve lever in the opposite direction for it to open properly.

[Davezj]

Surely though a cable will hold it shut when under tension but when released you'll need some sort of spring mechanism pulling the valve lever in the opposite direction for it to open properly.

i think the proper cables paul was taking about have a very stong inner cable and are as good pushing as they are pulling. the movement of the valve is only about 10mm tops so i don't think the cable would not kink in that amount of travel. a servo pulls and pushes with the same force.

as i same it is only a thought, but i am getting more interested in trying it the more i ponder it.

[Jesus-Ninja]

i think the proper cables paul was taking about have a very stong inner cable and are as good pushing as they are pulling.

No apologies for the thread revival

Yes, a decent cable will push and pull - look at the VR4 gear linkages