Jump to content
SAU Community

Recommended Posts

  • Replies 97
  • Created
  • Last Reply

Top Posters In This Topic

Top Posters In This Topic

Posted Images

Great work Shaun and Stao,

I'm interested in seeing how this progresses. For those who doubt the theory consider bernoulii's theorem.

First assume the exhaust gas is not getting compressed (density is constant) and flow within the system is ideal, then the flow out of the head must equal the flow out the turbo. If the area to the turbo is halved then its velocity must increase by a factor of approximately 1.4 (Bernouli's theorem, Conservation of Energy). Momentum is then consequently increased by about 40% => 40% faster spool time.

So in an ideal theoretical world - it works great but a lot of R&D is required to reduce ineffeciences.

Stao: Is this sort of setup possible in one of your super street turbos? ;)

You mention twin scroll manifold....youre shutting off 3 cylinders until 4000rpm?

Twin scroll turbo, single outlet manifold for this setup. You want to make all the gas go through the turbo faster by making the area smaller until it starts to create backpressure in which case you begin opening the valve on the turbo slowly until it eventually is open entirely flowing entirely AFAIK.

i still don't understand how this works, the wheel and housing stays the same size, in my head it's like putting a restrictor in any part of the manifold, the diametre decreases, so the velocity increases through the restrictor, but as soon as the diametre increases again, it simply slows down again. i know this "restrictor" is immediately before the turbo, but it's still the same volume being pushed through the same area around the wheel.

i can only think there is some merit to the idea of the pressure being directed onto a more efficient part of the blades.

guys, my set up is a full divided twin scroll 6 boost amnifold and a gt4094 full twin scrioll turbo, what has happened is part of the twin scroll in the exhaust housing has been removed and fitted with a flap. the flap closes the rear most port of the twin scroll housing but at 45 degrees, this does 2 things . directs all 6 exhaust gases through the front scroll of the turbo exhaust housing, and 2, because its at 45 degrees reduces heat build and helps to increase the gas flow.

you do not do any mods to the manifold what soever.

as mentioned above, i run a link g4 ecu, i am going to fit a linear electric actuator, wired through a relay and driven by the ecu output using rpm to determine the gate opening piont.

thanks again for the contributions guys. will update as and when

You mention twin scroll manifold....youre shutting off 3 cylinders until 4000rpm?

The exhaust gases from 3 of the cylinders on one side of the manifold travels straight through to one scroll of the turbo housing and the exhaust gases from the other 3 cylinders enter their own scroll of the turbo housing but are redirected to join the same scroll of the turbo housing.

This redirecting occurs within the turbo housing itself.

This way the exhaust flow from all 6 cylinders are using only one scroll of the twin scroll turbo until the point that the valve is opened.

I am guessing when the valve does open you might get a small dip in boost pressure as the exhaust flow has to start from scratch on the second scroll?

I would assume so Harey, unless you can open it slowly.

My design didnt modify anything, it's removable and between the TS manifold and TS turbo housing as I wanted to keep the TS when the flap was open. I ended up not making it as the GTX4508 turbo I designed it to go under was too big to fit in the engine bay with 3 inches added to the manifold height.

I would assume so Harey, unless you can open it slowly.

My design didnt modify anything, it's removable and between the TS manifold and TS turbo housing as I wanted to keep the TS when the flap was open. I ended up not making it as the GTX4508 turbo I designed it to go under was too big to fit in the engine bay with 3 inches added to the manifold height.

Ah ok so your design is an piece that sits in between the manifold and turbo housing? So would raise the turbo up by the amount of your part.

GTX4508 turbo lol ur a nutter :)

Interesting bit of gear. I've also made a prototype with the similar sort of swing valve, the problem I've found is the exhaust manifold pressure on the flapper valve was too strong for the actuator to engage, which it never opens. But there are ways around it.

Can just you just use a very long lever with lots of mechanical advantage and a strong solenoid?

i still don't understand how this works, the wheel and housing stays the same size, in my head it's like putting a restrictor in any part of the manifold, the diametre decreases, so the velocity increases through the restrictor, but as soon as the diametre increases again, it simply slows down again. i know this "restrictor" is immediately before the turbo, but it's still the same volume being pushed through the same area around the wheel.

i can only think there is some merit to the idea of the pressure being directed onto a more efficient part of the blades.

You are using a twin scroll turbo so the turbo doesn't stay the same size.

Wouldn't it be better to switch it on load rather than rpm?

Ultimately you need a 3D map with rpm and boost (not load).

Edited by Rolls

now i want one :(

maybe the next build.

im already schemeing ways of doing the flap internal to the manifold rather than the turbo housing.

and trying to make it so the gasses dont load it up too much to open with an actuator.

hmmm, time to start drawing.

You are using a twin scroll turbo so the turbo doesn't stay the same size.

ok so the twin scroll stays separated right up until exiting the wheel? i was under the impression it was only separated til just before the wheel.

i'd like to see this thing tested on track cars, additional moving parts with track temps makes me nervous.

i would rather see a back swept design, which is spring loaded (adjustable) which is opened by exhaust gases, once the exhaust gases build up enough pressure, the flap starts opening, once they reduce, it closes back up. the design could be much simpler, with less parts to fail.

  • 1 month later...

The only reason why you can't do that Nisskid is due to the fact that it would eliminate the purpose of having the twin scroll turbine

housing as you wouldn't be able to block the center where the gas switches from one side to the other. You would how ever be able to run it like a standard v-band housing, even though you would be utilising a modified twin scroll housing and a divided manifold.

Hows it going with this anyway? Any more results? Looking at doing this with my build in the near future.

ok so the twin scroll stays separated right up until exiting the wheel? i was under the impression it was only separated til just before the wheel.

If you did that the turbo still sees the same mixed exhaust pulses, it has to be separated the entire way otherwise it won't be of any benefit, hence the separating takes up space and the AR is reduced for the same physical size.

By the way has anyone thought of using a stepper motor, similar to those found on the intakes of some cars in reference to boost pressure? Just use the ecu to put out a Pulse width modulated signal in accordance to boost.

You don't need to take it too full boost as its only when the engine is trying to begin to spool a turbine that takes the longest, so say if desired boost level is 24psi just make it full open at 14psi.

By doing that you eliminate any of the issues with exhaust restrictions, boost spikes, etc as the system would be regulated via the wastegate. Realistically rpm wouldn't really matter as its only operating on boost pressure imo.

Also the stepper motor running of a pulse width would be able to be opened gradually rather than straight open and shut

Having just read a thread about tuning VCT turbos on a 1uz, you need to remember that when the valve is shut, it has a big effect on the Volumetric efficiency of the engine, fuelling and timing changes will be massive compared to when the valve is open

I think the guy who did it ended up using an exhaust manifold pressure sensor and having the valve reference it to boost pressure. Basically he said you can make heaps of boost down low, but if the valves(s) are shut to make the boost, then the motor wont make any power. BTW it was all ecu controlled.

Edited by Adriano

Create an account or sign in to comment

You need to be a member in order to leave a comment

Create an account

Sign up for a new account in our community. It's easy!

Register a new account

Sign in

Already have an account? Sign in here.

Sign In Now



  • Similar Content

  • Latest Posts

    • Hi, is the HKS  Tower Bar still available ? negotiable ? 🤔
    • From there, it is really just test and assemble. Plug the adapter cables from the unit into the back of the screen, then the other side to the car harness. Don't forget all the other plugs too! Run the cables behind the unit and screw it back into place (4 screws) and you should now have 3 cables to run from the top screen to the android unit. I ran them along the DS of the other AV units in the gap between their backets and the console, and used some corrugated tubing on the sharp edges of the bracket so the wires were safe. Plug the centre console and lower screen in temporarily and turn the car to ACC, the AV should fire up as normal. Hold the back button for 3 sec and Android should appear on the top screen. You need to set the input to Aux for audio (more on that later). I put the unit under the AC duct in the centre console, with the wifi antenna on top of the AC duct near the shifter, the bluetooth antenna on the AC duct under the centre console The GPS unit on top of the DS to AC duct; they all seem to work OK there are are out of the way. Neat cable routing is a pain. For the drive recorder I mounted it near the rear view mirror and run the cable in the headlining, across the a pillar and then down the inside of the a pillar seal to the DS lower dash. From there it goes across and to one USB input for the unit. The second USB input is attached to the ECUtec OBD dongle and the 3rd goes to the USB bulkhead connected I added in the centre console. This is how the centre console looks "tidied" up Note I didn't install the provided speaker, didn't use the 2.5mm IPod in line or the piggyback loom for the Ipod or change any DIP switches; they seem to only be required if you need to use the Ipod input rather than the AUX input. That's it, install done, I'll follow up with a separate post on how the unit works, but in summary it retains all factory functions and inputs (so I still use my phone to the car for calls), reverse still works like factory etc.
    • Place the new daughterboard in the case and mount it using the 3 small black rivets provided, and reconnect the 3 factory ribbon cables to the new board Then, use the 3 piggyback cables from the daughterboard into the factory board on top (there are stand offs in the case to keep them apart. and remember to reconnect the antenna and rear cover fan wires. 1 screw to hold the motherboard in place. Before closing the case, make a hole in the sticker covering a hole in the case and run the cable for the android unit into the plug there. The video forgot this step, so did I, so will you probably. Then redo the 4 screws on back, 2 each top and bottom, 3 each side and put the 2 brackets back on.....all ready to go and not that tricky really.      
    • Onto the android unit. You need to remove the top screen because there is a daughterboard to put inside the case. Each side vent pops out from clips; start at the bottom and carefully remove upwards (use a trim remover tool to avoid breaking anything). Then the lower screen and controls come out, 4 screws, a couple of clips (including 3 flimsy ones at the top) and 3 plugs on the rear. Then the upper screen, 4 screws and a bunch of plugs and she is out. From there, remove the mounting brackets (2 screws each), 4 screws on the rear, 2 screws top and bottom and 3 screws holding in the small plates on each side. When you remove the back cover (tight fit), watch out for the power cable for the fan, I removed it so I could put the back aside. The mainboard is held in by 1 screw in the middle, 1 aerial at the top and 3 ribbon cables. If you've ever done any laptop stuff the ribbon cables are OK to work with, just pop up the retainer and they slide out. If you are not familiar just grab a 12 year old from an iphone factory, they will know how it works The case should now look like this:
    • Switching the console was tricky. First there were 6 screws to remove, and also the little adapter loom and its screws had to come out. Also don't forget to remove the 2 screws holding the central locking receiver. Then there are 4 clips on either side....these were very tight in this case and needed careful persuading with a long flat screw driver....some force required but not enough to break them...this was probably the fiddliest part of the whole job. In my case I needed both the wiring loom and the central locking receiver module to swap across to the new one. That was it for the console, so "assembly is the reverse of disassembly"
×
×
  • Create New...