HE351VE, We are for the big...

[casuprock]

HE351VE, the final frontier... Let me know what numbers this beast can put down from a built 2.3L stroker. Whenever I get around to finding out, I'll buy the closest guess a beer...

You must be logged in to view this image or video.

You must be logged in to view this image or video.

 

[casuprock]

Here's an update on this project... I've been refurbishing this turbo and collecting parts. See my article on home brew ceramic coating:

http://www.dsmtuners.com/forums/turbo-system-tech/392793-ceramic-coating-holset.html#post152468445

I elected to use a linear actuator with a potentiometer feedback to control the VGT. I am using an actuator from Firgelli Automations. This gives me direct control and 150lbf actuation force. This will be easier than using the Holset brushless motor since position feedback is more accurate and robust from the linear actuator.

My control system will be designed to seek desired boost pressure and optimal shaft speed at current pressure. HE351VE control logic. By controlling shaft speed and pressure ratio I should be able to 1) Prevent turbo from blowing hot air 2) Run turbo at efficient conditions across the boost range, up to the target peak boost.

(For reference from Holset website):

You must be logged in to view this image or video.

 

[bastarddsm]

My control system will be designed to seek desired boost pressure and optimal shaft speed at current pressure. . By controlling shaft speed and pressure ratio I should be able to. Prevent turbo from blowing hot air 2) Run turbo at efficient conditions across the boost range, up to the target peak boost.

Man, I don't know if you have a clue how turbo's work.

1. Try as you might, but a turbo has little to no influence on the amount of air the engine can concume at a given boost and RPM. The cams/displacement/intake do that.

2. Shat speed has little effect on the air flow of the unit. Shaft speed determins pressure ratio.

3. a variable geometry housing controls the exhaust wheel map.

 

[4G63Crossfire]

I almost used one of these turbos on my project. I believe you can control the VGT actuation based solely on a map sensor output controlling a programmable linear actuator.

 

[casuprock]

I almost used one of these turbos on my project. I believe you can control the VGT actuation based solely on a map sensor output controlling a programmable linear actuator.

I'm building a controller for this project right now... I'm using pressure and shaft speed as the control variables. I'll have IO on the board by breaking out an I2C bus for interfacing to external sensor systems if I find I need them.

I hope after I get this controller built, that more guys will take an interest in this turbo. It's an inexpensive turbo because nobody has tackled the control system to make it shine. The only aftermarket control system I am aware of is made by Fleece performance (for diesel engines).

I might start selling the controllers after I get it debugged and tested for a while.

My control system will be designed to seek desired boost pressure and optimal shaft speed at current pressure. . By controlling shaft speed and pressure ratio I should be able to. Prevent turbo from blowing hot air 2) Run turbo at efficient conditions across the boost range, up to the target peak boost.

Man, I don't know if you have a clue how turbo's work.

1. Try as you might, but a turbo has little to no influence on the amount of air the engine can concume at a given boost and RPM. The cams/displacement/intake do that.

2. Shat speed has little effect on the air flow of the unit. Shaft speed determins pressure ratio.

3. a variable geometry housing controls the exhaust wheel map.

Hey man, try and spell check while telling me I'm dumb.

Shaft speed has everything to do with compressor efficiency. It is also an important control variable for keeping the unit within it's design limits and to maximize pressure ratio while producing minimal waste heat. While the compressor geometry does not change, the exhaust vanes control the shaft speed and thus directly affect efficiency. Turbine shaft speed is a critical design variable. Pressure ratio + shaft speed + compressor map allows you to know the most efficient operating condition (and if the exhaust vanes should be incremented or decremented to seek a better point).

The variable reluctance shaft speed sensor on this turbo only exists because it was used in the stock application as part of a control system. A manufacturer doesn't install a $50 sensor in a device unless it is being utilized.

Aftermarket kits are available to monitor shaft speed as part of understanding how efficiently the turbo is operating. TurboByGarrett.com - Accessories

As for TB and intake mani, I've already got that covered with a Q45 TB and JMF mani acquired for the project. The car has 272 cams on 2.3L... She can breathe. I can give it more cam, but think the 272s are good since it's a stroker and I want to keep low RPM torque.

 

[bastarddsm]

You don't directly control the operating point of the turbo with the shaft speed.

Look at a typical compressor map, the wheel speed is shown as basically horizontal lines across the map. The efficiency islands are vertical in nature. So tell me how wheel speed or pressure ratio is a larger determination of the efficiency... Get a ####in clue dude. You should spend less time worrying about keeping your turbo at 78% than 77%, and more time worrying about how your going to keep a dsm transmission together at 50psi on a 2.3l.

By the way, I am by no means knocking the VGT stuff. I run a LLY Dmax turbo on my car. I control the vgt with a 1g dsm ecu. Your just going about it in the wrong way.

Once you leave the oem/emissions arena, and start concentrating on performance, throw the engineering texts away. None of that matters anymore.

 

[casuprock]

You don't directly control the operating point of the turbo with the shaft speed.

Look at a typical compressor map, the wheel speed is shown as basically horizontal lines across the map. The efficiency islands are vertical in nature. So tell me how wheel speed or pressure ratio is a larger determination of the efficiency... Get a ####in clue dude. You should spend less time worrying about keeping your turbo at 78% than 77%, and more time worrying about how your going to keep a dsm transmission together at 50psi on a 2.3l.

By the way, I am by no means knocking the VGT stuff. I run a LLY Dmax turbo on my car. I control the vgt with a 1g dsm ecu. Your just going about it in the wrong way.

Once you leave the oem/emissions arena, and start concentrating on performance, throw the engineering texts away. None of that matters anymore.

Correct engineering is performance, period. Who do you think designs turbos to begin with? I bet they have a lot of engineering books.

If shaft speed, pressure ratio, and the compressor map are insufficient in some way, I have expansion built into my controller to accept IO from other sensors or from the ECU itself. Instead of being a bummer, I'd like to hear your suggestions about other sensor inputs that would be better. Charge temp and MAF are two secondary variables that come to mind. What do you base your vane position on?

There is now a dedicated thread for updates on this controller development:
http://www.dsmtuners.com/forums/tur...he351ve-controller-project.html#post152504902

 

[bastarddsm]

I have a 3x4 table of duty cycles indexed by TPS and Boost. I think I could accomplish the same thing with a 1X5 indexed by boost alone. But then I would flag a bit to set the vanes to full open during closed loop.

Then again you could also look at rpmXboost, or load by boost, or timezonex gravitational constant. It doesn't really matter. In the end all you want the fastest spool with the biggest opening.

Your way over complicating this. What do you need a shaft speed sensor for? To tell your self your using the turbo way off the map?

I don't know how to convince you that you can't change (at least in a good way) the operating points of the compressor by altering the exhaust housing A/R.

Oh an efficancy this efficancy that...Who cares, the intake charge dillution due to your attempts to alter the operating point is gonna cost more power than you could ever gain from having a more efficient map.

Also, I don't run a MAF, I run SD. You might as well set in the corner and punch the clown for an audience as worry about changing VE's due to vane position. Set it rich, get the VGT curves knocked down, and sneak up on the AFR's. Or #### it and leave its rich/safe. Your only leaving 1-3% power on the table anyway.

 

[casuprock]

Obviously we differ on the control approach which seems to be the problem here. Running open loop has it's place and your approach has merit. Here are questions that came into my mind when I was considering open loop:
- If you run SD and also run your turbo open loop, how sure are you that you're running the best operating points?
- Without knowing shaft speed is there confidence that the turbo is matched to the flow rate of your car?
- Even with a tune, there are infinite stable operating points that work open loop, doesn't mean they are near the ideal.

Exhaust gas temp, pressure, and flow rate are heavily influenced by AFR. The exhaust turbine can experience very different conditions depending on them. If your preset map doesn't compensate, you might be under using the VTG. Not bashing your setup, just pointing out that stable/functional does not equal optimal.

Running closed loop is always more complicated, as you adamantly say, but not without advantages. The benefits of running closed loop are a more stable system, finer adjustment, and compensation for changing conditions.

Operating the turbo outside it's maps is part of the reason I will watch shaft speed. If you use rotational acceleration (differentiate shaft speed) you can know exactly when to start backing off the vanes so you don't overshoot the target speed and go off the maps/exceed bearing ratings. Control "overshoot" is a topic addressed in linear systems by 1. Damping>=Critical (suboptimal, but stable performance) or 2. Negative Feedback (optimal, stable performance) . I chose negative feedback.

Yes it's more complicated but that's what makes it interesting!

 

[bastarddsm]

Throw the engineering texts away or you'll never get out of the 13's.

You'd be better off spending your time devising a good cheap electronic boost controller, that varies the boost based on TPS and RPM.

That would allow a boost curve that puts the torque peak at whatever rpm you want, while keeping stock economy cams...

 

[juvi1624]

Damn, you guys put a lot more time and thought into this than we did when we this turbo on on my roomate's Nissan. We just used a wastegate actuator to operate the vanes. Probably wasn't the most "efficient" thing but the car flew with it like that.

 

[casuprock]

Damn, you guys put a lot more time and thought into this than we did when we this turbo on on my roomate's Nissan. We just used a wastegate actuator to operate the vanes. Probably wasn't the most "efficient" thing but the car flew with it like that.

I saw some photos of the waste gate actuator method somewhere. Maybe it was your friends car... Anyway, it was one of the things that inspired this project.

Throw the engineering texts away or you'll never get out of the 13's.

You'd be better off spending your time devising a good cheap electronic boost controller, that varies the boost based on TPS and RPM.

That would allow a boost curve that puts the torque peak at whatever rpm you want, while keeping stock economy cams...

Maybe I'll throw away all my books, or burn them for heat since the price of oil is going up.

 

[Kapok6]

What ever came of this?