GT 12 wastegate actuator

[Biggy72]

The wastegate actuator opens a little after 9 psi now, but I would like to get it to open up around 5 psi. Is there a way to make this change by putting in a new spring, or is there another place to buy a different actuator?

[Biggy72]

The wastegate actuator opens a little after 9 psi now, but I would like to get it to open up around 5 psi. Is there a way to make this change by putting in a new spring, or is there another place to buy a different actuator?

[Dan Deussen @ Weber Motor]

Derek,

first of all you can lower the spring preload by lengthening the actuator arm. Typically there is a long piece of hex stock that has an internal left and opposite right hand thread fixed with jam nuts. Just loosen the nuts and turn the hex piece in the direction that loosens it. I don't think you will get it set down to 5 psi, but you may be able to drop one, two or even three psi.
There are a few drawbacks to run either very little pretension or a soft spring. The boost buildup can be drastically delayed by the wastegate opening too early, even though you might bleed off the control air for the actuator. Just the exhaust gasses pushing against the valve could create a small leak with too little spring preload.
Another option to lower the boost pressure is to pressurize the opposite chamber of the actuator box. So basically you could use two solenoids, one to guide pressure to the chamber opposite of the spring and one to the side with the spring. With the correct setup you will be able to widen the adjustable range of boost above AND below the spring setpoint. Most likely the actuator that comes with your turbo does not support this type of setup.

[Biggy72]

That makes sense about the exhaust gasses pushing the valve open. I would like to lower the actuator activation pressure from 9psi at least some. Otherwise we'll be choked at the restrictor at around 10,000 rpm because I can't lower the boost pressure any more to prevent a vacuum from forming if any of my theoretical calcs are remotely accurate.

I'll mess with the preload and see if I can't get it to at least drop some. If I could get it down to 7psi I would be pretty happy with it since that should get us closer to 11500 rpm before we're completely choked again with theoretical calcs.

[Pete M]

If i may ask a stupid question, what's wrong with choking at 10,000 rpm? Is it that you're setting a fixed boost and you're worried what will happen above choke? Ideally you don't want your set boost to be unachievable due to the restrictor because the wastegate will shut and the turbo will scream its little head off, slaughtering the top end and potentially damaging the turbo.

If you can solve that problem though, choking lower is good. We choke just below 7,000 rpm. We use an all electric actuation system, which is more work to set up, but extremely flexible.

[Biggy72]

I would like to never have the restrictor choke. We will be dropping boost starting around 7000 rpm to achieve this though. I would like to end up at quite a bit higher intake pressure at 7000 rpm than what we will have at 11,500. I do not want to have the compressor spinning in a vaccuum because it only creates heat with no other benefit and while I haven't tested it yet, it seems like it would be much harder on the seals behind the turbine.

With that said we have an air controller to do this, but it can only mimick whatever amount of boost you would like to have going to the waste gate. So if I would like the system to have anything lower than 9psi at upper rpms then I'm stuck because the wastegate won't open below that as it sits now.

[Pete M]

None of the seals will care, they see positive pressure when you're at full throttle. The only real worry is overspeed. If you have a 600 cc engine at 11 grand, and you tell the turbo to run 9 psi (unachievable), the thing will spool its tits off trying to achieve that boost, until it either reaches an equilibrium (air flow is limited, therefore energy to the turbine is limited) or it comes apart. In our experience, the former usually happens first. We've run some fairly insane turbo speeds in the past due to mistakes (284,000 rpm is the record) and the turbo didn't fail (I don't think you'd manage to get a non-variable vane turbo to go that fast though). Even if it doesn't fail, the overspeed will slaughter your power.

There are other approaches though. Although i've never tried it myself, i've heard that using a dual port actuator and tying the other port to the restrictor near the compressor inlet works well. The idea is that as vacuum in the restrictor builds, it is effectively treated as boost and the wastegate will open. So say the restrictor chokes at 5 psi of boost at a particular rpm. If the actuator is set at 9 psi, then it will regulate it to a 4 psi vacuum in the restrictor.

[Biggy72]

That's actually a pretty good idea, but I'd like to find a couple more actuators to tear apart before I even tried experimenting with it.

But I know about the overspooling, that's what I'm effectively trying to prevent by the exhaust pressure so it doesn't try to spin the compressor any faster. How do you guys measure turbine speed?

[Biggy72]

So evidently there are actuators for the application I would want already being made. They're a little expensive, but they're also good down to 4psi. Thanks for the help Pete.

[Pete M]

Commercial turbocharger speed sensors exist these days. The older style relied on a notch in the shaft or something like that. The newer style sense compressor blades, so all you have to do is drill and tap your compressor housing. They are not exactly cheap though, we got ours sponsored.

[Dan G]

Yes, we use both shaft and blade speed sensors. The shaft sensors read a flat machined in between the bearings. The blade counters read through the compressor housing and divide by the number of blades.

Unfortunately we don't have the resources to offer these parts with our program. Our aftermarket group did just release a blade counting speed sensor, so you should be able to order one of those through a Garrett distributor such as ATP or Limit Engineering.

[Kirk Feldkamp]

I did my own personal layout for the GT12 a while back, so I know it's possible. I don't know if the aftermarket group did the GT12 for their speed sensor project, but it's certainly worth a look.

-Kirk

[Dan G]

Here's the blade counting sensor for reference...

http://www.turbobygarrett.com/turbobygarrett/products/A...ue_speed_sensor.html (http://www.turbobygarrett.com/turbobygarrett/products/Accessories_continue_speed_sensor.html)

The aftermarket "Pro Kit" skips the gauge (gauge doesn't read high enough for a GT12 anyway) but includes the sensor, installation and signal instructions needed. It outputs a fixed voltage/duration square wave whose frequency is proportional to turbo speed.

[Pete M]

That's the sensor we use. We don't use the gauge though, we just run it straight to our ECU.

[R/TErnie]

why don't you make a dual port actuator and run a solenoid? If your EMS has provisions for boost control you should be able to control ramp rates, gear dependant boost, etc.

To the OP...if you want to lower your boost and not do any hard work... Move your boost signal line from the compressor cover of your turbocharger to your Intake manifold. The pressure drop across your IC/IC piping will give your actuator a lower pressure reading.

(PS you should have a restrictor in the line to the actuator)

[Biggy72]

why would you put your boost control line at the compressor cover??? That's not the boost the engine's seeing, so that seems pretty obvious to me.

I've already got a pretty good idea of what I'm doing, but I did look at the dual port actuators. I'm just curious how you propose to actually make one work with only one solenoid? Where does the other side of the actuator plumb into that makes it any better? If I wanted to make my setup more accurate, but wayyyy more complicated I would have gone with it, instead I just got a 6lb actuator which should be good to about 12k for me and that's pretty good as I don't see us being up there very much.

[Drew Price]

Derek,

I have never seen an OEM that did not put the wastegate signal line right off the compressor housing. All the turbos I have seen come plumbed with a nipple on the compressor housing just for that purpose, to run right into the actuator if no controller, or into a solenoid if regulated.

Good point about not being the same pressure that actually enters the engine though, and my guess would be that the slight additional pressure at the compressor would be able to do a better job at holding the wastegate shut when running higher boost levels than we do. You could run into fluctuation problems with not enough pressure at the signal line. Does that make any sense?

Best,
Drew

[Biggy72]

Well all of the fluctuation problems occur from choking the restrictor and over spinning the turbo. I imagine that you could solve the problem pretty easily by simply messing with the tune and having it open the actuator a little sooner if the pressure does start to fluctuate on the top end.

[Drew Price]

Exactly, for the restricted application you're going to be really unlikely to have control problems because the pressure is much lower than what the turbo was intended for.

I was only speculating as to why the signal line usually comes off the compressor housing, and not the intake manifold.

Do you see fluctuations when you hit choke and the turbo starts making lots of noise and heat, and not any more air flow? Share how much?

I have no experience doing forced induction with a restrictor yet, so good things for me to be absorbing.

Best,
Drew

[Biggy72]

Unfortunately I don't have any experience with it yet either. The car was originally supposed to be running this week.... but it got pushed off. Hopefully we'll be up and running in another week or so once everyone gets back to school.

[Pete M]

My understanding is that the wastegate signal is taken as close to the compressor as possible because it's a control system. You generally want as little delay as possible in the feedback circuit as it makes stable control harder.

On the other hand, it may be done simply for neater packaging. Either way, unless you have a super restrictive intercooler or something, i don't think it'll make much difference.

[R/TErnie]

Pete M - "Either way, unless you have a super restrictive intercooler or something, i don't think it'll make much difference."

If you cool your charge air... you will have a pressure loss. Typically it will be 1-3 psi depending on IC core design. If you're looking to change it 1-3 psi..its a quick easy way.

Biggy72 - "I'm just curious how you propose to actually make one work with only one solenoid? Where does the other side of the actuator plumb into that makes it any better? "

The bottom port is connected to your boost source, the top port is connected to your solenoid which is then connected to your boost source. like so...
http://i221.photobucket.com/albums/dd45/eurness/EBCsetup.jpg

There are several reasons why this is MUCH better than running solely off spring pressure...
Springs open at a linear rate. So as your boost pressure increases the valve creeps open, bleeding off exhaust pressure necessary to spool the turbo (pressure and temperature differential across the turbine wheel). So you're making the turbo "laggier" than it should be. with teh twin port actuator you can keep the gate closed with FULL Spring pressure (Upper Port pressure+ spring Pressure - Lower Port Pressure= pressure applied on wastegate valve) so if your upper port is the same as your lower port...your spring pressure is holding your valve shut. Your EMS controls the solenoid and keeps the unmodified boost signal to the top port 2-3 psi prior to the desired boost level. At that point the pressure at the top port is blocked via the solenoid (0psi + spring pressure (maybe6psi) - Boost pressure (say 10psi)= -4... your valve is going to open very quickly... maintaining your boost level.

Beyond that...there are gear dependant maps, RPM based boost control, time based boost control, etc.

Single port actuators in comparison are quite inferior and harder to accurately control the wastegate valve compared to dual port gates.

[Biggy72]

Ernie,

You're right, except do you know what sort of boost pressure these engines run? With the restrictor and 12k you can't be running a very high boost pressure. With that said you would then need a dual port actuator with a very very very light spring in order to set the whole thing up the way you proposed, because otherwise there would never be a large enough pressure differential to open the thing.

[Pete M]

Maybe i'm wrong here, but why would cooling the gas necessarily change its pressure? This isn't an adiabatic process (by definition). It was my understanding that the only reason you get a pressure loss is due to friction from the relatively high velocity air going through all the fins.

We've seen about a 2 kPa (0.3 psi) drop across our intercooler at peak boost. I'd call an intercooler with a 3 psi drop an absolute disaster. 20 kPa of pressure loss is worth over 30 degrees of intake temp (assuming a bar of boost and ~50 degree intake temps). Maybe that kind of loss is typical in street cars, but it sounds terrible.

[R/TErnie]

Pete M,
Well when your intercooler is 24x12x5 and you're running 40lbs of boost on a 67mm turbo 3psi of pressure drop "isn't bad".

"Typically it will be 1-3 psi depending on IC core design." The more restrictive the IC the more pressure loss. As you were saying the pressure loss is due to the friction in the core. yeah you're right.

What kind of intercooler thermal efficiency are you seeing from your IC then? Whats your compressor discharge temp, post IC temperature, ambient temperature, and the pressures at those same points.

[Pete M]

I don't have the data in front of me, but 120-140?C compressor discharge, 40-50?C post IC for 20-30?C ambient is in the ballpark. Loss is about 2 kPa at peak power, somewhat higher by rev limit. This is all on about 16 psi of boost, although boost is heavily rpm dependent.

It's hard to actually get useful numbers for some of this stuff, because we're rarely on throttle long enough for anything to get to a steady state situation, and the core is sized far too small for steady state dyno loading. Thermal inertia tends to dominate on track.

[Biggy72]

I'm really impressed with 16 psi. We're not up and running yet, but from all of my calcs I'm hoping we can hit 12psi at peak. You guys are running the gt15 aren't you?

[Pete M]

16 psi is the approximate boost at peak power. It's more like 18 psi 1000 rpm lower. We rarely hit that on track though, except on a launch or if the driver exits in the wrong gear. We let it run the boost there because it doesn't seem to mind and it was just a number in a table. It's actually far more prone to detonation higher in the revs, even though the boost is lower.

One of our drivers was briefly stuck in 4th at aus last year. We were on full boost then. Our record last year in testing was 25 psi of boost when one of the drivers hit upshift at 6,900 rpm with the turbo fully spooled. Was firmly on overboost cut by that point though, so doesn't really count.

And yes, last year we used the 15v. Haven't made our final decision yet for this year.

[joshiepoo]

just curious...the teams that are running with turbos, are you using head shims at all?

[Pete M]

<BLOCKQUOTE class="ip-ubbcode-quote"><div class="ip-ubbcode-quote-title">quote:</div><div class="ip-ubbcode-quote-content">Originally posted by joshiepoo:
just curious...the teams that are running with turbos, are you using head shims at all? </div></BLOCKQUOTE>

As in a thicker head gasket? No. We (well, team members before my time) did that initially in testing in 04, but decided that it was not a very good solution. Stuffs your squish and the lower compression isn't great for efficiency anywhere you're not on full boost. My advice, build yourself a det can and try it with stock compression first. Start at low boost and conservative timing, be careful and you'll be surprised at how much boost you can run with stock compression.