Kevin's Compound Sequential Twin Turbo Design

[V8KILR]

...

Edited November 25, 2013 by V8KILR (see edit history)

[V8KILR]

With the design pretty much finalized, the steps I see this setup should transition through are:

 

1. When running off boost the engine will draw airflow through both turbos with the majority coming through the #1 turbo. The EGCV is open and all wastegates are closed. The swing check valve (SCV) will be partially open due to vacuum from the engine.

 

2. As the #1 turbo transitions onto boost the airflow will switch to all going through the #1 turbo. The EGCV is still open and all wastegates are closed. The SCV is probably now closed. The #2 turbo is still just idling.

 

3. When the #1 turbo reaches 7psi boost, the EGCV will close and all wastegates are still closed. This will increase the exhaust flow through the #2 turbo which will now start to speed up. The #1 turbo spool speed will decrease in the rate of increase, but it will still continue to increase boost pressure.

 

4. When the #1 turbo reaches around 15psi, the #2 turbo should start making boost. The EGCV and all wastegates are closed. The SCV may start to open shortly depending on how quickly the #2 turbo boost pressure backs up in the #2 intercooler pipe. A small amount of flow from the #2 turbo may then pass through the SCV.

 

5. When the #1 turbo reaches 35psi, the #1 wastegate will be opened under ECU control. The EGCV and #2 wastegate are closed. The SCV will be partially open.

 

6. When the #2 turbo reaches 35psi, the #2 wastegate will be opened under ECU control. The EGCV is closed. The SCV will be fully open.

 

7. Win the race I was just having.

 

Now all I need to do is to save the estimated NZ $10k needed to implement this system.

[Pipedreams]

Well, with the setup you have come to I don't see the point in not running a sequential setup. You allready have implemented the necessary valves but made the exhaust side less effective by running them in series instead of parallell.

[V8KILR]

Well, with the setup you have come to I don't see the point in not running a sequential setup. You allready have implemented the necessary valves but made the exhaust side less effective by running them in series instead of parallell.

 

By running the two turbos in series on the exhaust side, I'll be able to extract more energy from the exhaust system versus running them in parallel. For example if one turbine can extract 60%, then two in series can extract 84% (60% + (40% x 0.6)). This means the #2 turbo will spool quicker as the turbines are extracting more energy from the exhaust system. The exhaust part of my setup is basically the same as the Boost Logic setup and they got 20psi at 2800rpm from their ~50mm #1 turbo.

 

My #1 turbo will spool very quickly as it has a proper 6 into 2 divided manifold which is utilizing the exhaust pulse energy which the BL system did not. To make it spool even quicker, I plan to re-use my existing divided manifold QSV on the #1 turbo to add another 4-500rpm of spool as well. Both these will result in much more exhaust gas at lower rpm, which will also help spool the #2 turbo quicker.

[jagman]

http://kth.diva-portal.org/smash/get/diva2:7928/FULLTEXT01

 

Page 34 - compare this to your system

Your system stalls the second turbo with check valve

Your system narrows the operating range

You still have inertia to overcome using the same energy source

[V8KILR]

http://kth.diva-portal.org/smash/get/diva2:7928/FULLTEXT01

 

Page 34 - compare this to your system

Your system stalls the second turbo with check valve

Your system narrows the operating range

You still have inertia to overcome using the same energy source

 

Interesting article. I'll read it in more detail later on. However the table 1.2 on page 40 (42 in the article) that shows the 7 different options that they considered does not have my setup described, so it's still a unique design to my knowledge and is not discussed in that article.

 

Your system stalls the second turbo with check valve

 

Not true. The swing check valve lets flow from the #2 turbo through as soon as it exceeds pressure at the other side of the check valve. This is exactly how the 2JZ-GTE setup works with the reed valve. Also my pipe merge design in post #46 should allow the flows to merge much sooner then the 2JZ-GTE setup does.

 

Your system narrows the operating range

 

What they said is: "However, the series-sequential system has a narrower flow range because the entire mass flow has to go through both compressors."

 

They are actually talking about the compression side of a compound setup and mine doesn't do that as the compression side is not compounded. This is also not true for my design on the exhaust side either as the two 60mm wastegates on my system bypass exhaust gas around the turbines to prevent that very problem.

 

You still have inertia to overcome using the same energy source

 

As my compound setup will extract more exhaust energy then a parallel or sequential setup possibly can, it will overcome the inertia at lower rpm.

Edited December 1, 2013 by V8KILR (see edit history)

[V8KILR]

To keep the setup simpler to make, lower cost and to make the fitment easier, I have decided to use a SPA cast manifold as the manifold part of my setup. It's in the country now and is just waiting custom clearance.

 

 

It may limit max power a little bit, but with the way I'll be using it getting 1000whp will not be an issue. This manifold has previously made well over 900whp on a single turbo setup.

 

I'll port the wastegate hole out to 45mm (about the max it will take) but initially I will just bolt a cover plate on it as the T4 bolt stud pattern part of the manifold (after more porting) will be used to feed the #1 turbo and also the Precision 66mm wastegate, which will feed the #2 turbo. The 45mm wastegate port will only be used to feed the #2 turbo as well, if the back pressure in the manifold gets too high.

 

The Precision 66mm wastegate has also been ordered and is on its way.

 

 

 

Here a rough scale mockup of how it will all fit together using the SPA manifold and the Precision 66mm wastegate. Done with Paint, so definitely not Photoshop quality.

 

This is as per my original design (without an EGCV) which I'll try first.

 

Edit: drawing updated to match engine bay measurements.

 

Edited March 17, 2014 by V8KILR
drawing updated to match engine bay measurements. (see edit history)

[mellonman]

Definitely a more efficient way to run big twins, I thought you needed 2 waste gates I can only see 1 on the drawing?

 

Are you going to have modify the bonnet for clearance?

 

Glad that your going forward with this cant wait to see the end result great work so far

[V8KILR]

There are two wastegates. The second one is my existing one that I'm reusing which is at the bottom end of the yellow pipe, so I didn't bother drawing it in.

 

The drawing may be slightly out. The rear turbo will fit under the bonnet without any mods, but it may have to sit a little lower then the drawing shows.

Edited March 16, 2014 by V8KILR (see edit history)

[V8KILR]

Looking to have similar turbo positioning to the Boost Logic compound setup.

 

http://www.nissansilvia.com/forums/index.php?app=core&module=attach&section=attach&attach_rel_module=post&attach_id=232496

[Chris Wilson]

If you smack yourself on the head with a mallet you can give yourself a headache and some grief far more cheaply

[Nodalmighty]

You want to have a word with Mark Kitchen at Billet turbos in Brisbane he be able to match you a set of mega spooling nitride ceramic roller bearing turbos to suit your needs.

 

I'm doing a serial compound on my Passat TDI using the stock turbo (Small) and a BV43 (K3) as the big turbo to make about 300bhp. But both turbos are VNT so I have more control over how the interact. If you want to see how serious compound turbo setups with 3 or 4 stages work look at tractor pullers, they are just mental with 100+psi boost.

 

Lyndon.

Edited March 17, 2014 by Nodalmighty (see edit history)

[Chris Wilson]

Ball bearing, not roller bearing, surely? Roller bearings won't take the RPM or any end thrust, and have too much drag.

[Digsy]

Also doesn't having a compound sequential system really negate the need for ball bearing turbos? The quick spool should really be provided by a properly matched high pressure turbo. I agree it will help, but it seems a bit "belt and braces".

[Nodalmighty]

Ball bearing, not roller bearing, surely? Roller bearings won't take the RPM or any end thrust, and have too much drag.

 

Of course but Roller was the buzz word back when Garret started the R craze.

 

Nitride ceramic self centering ball bearing race x 2.

 

Lyndon.

[Nodalmighty]

Also doesn't having a compound sequential system really negate the need for ball bearing turbos? The quick spool should really be provided by a properly matched high pressure turbo. I agree it will help, but it seems a bit "belt and braces".

 

I don't know , would it ?

 

Lyndon.

[Digsy]

Well they are both mechanisms to improve low end torque, but the gains from a compound sequential system should be head and shoulders above what you would get from a ball bearing turbo.

[V8KILR]

Ball bearing turbos only gain around 200rpm of spool over plain bearing turbos. Where ball bearings are of most benefit is for high thrust loadings and improved off boost to on boost response at higher rpm.

 

This setup of mine should spool only just a few hundred rpm slower then factory turbos as I'll also be running a QSV on the #1 turbo. This will be about 1500rpm better spool then what I have at the moment. It is anticipated that max power will be around three times what a factory setup is capable of.

[V8KILR]

If you smack yourself on the head with a mallet you can give yourself a headache and some grief far more cheaply

 

I anticipate really bad headaches ......................................... from the high boost and huge g forces it will launch with.

 

 

You want to have a word with Mark Kitchen at Billet turbos in Brisbane he be able to match you a set of mega spooling nitride ceramic roller bearing turbos to suit your needs.

 

I'm doing a serial compound on my Passat TDI using the stock turbo (Small) and a BV43 (K3) as the big turbo to make about 300bhp. But both turbos are VNT so I have more control over how the interact. If you want to see how serious compound turbo setups with 3 or 4 stages work look at tractor pullers, they are just mental with 100+psi boost.

 

Lyndon.

 

My setup is not compounded on the compressor side (as the compressor side is sequential), so it will only be running up to about 35 psi boost. It uses the same exhaust setup as a compound setup does, which is where the compound part of the thread title comes from.

[Noz]

Would a larger exhaust pipe be better. To help reduce the EGT, even though it would compress again into the exhaust housing. Isn't a problem with large sequential systems usually the EGT into a single small housing turbo?

[V8KILR]

Would a larger exhaust pipe be better. To help reduce the EGT, even though it would compress again into the exhaust housing. Isn't a problem with large sequential systems usually the EGT into a single small housing turbo?

 

I presume you mean between the two turbos. I did consider using a 3.5" exhaust pipe there, but as you say it does get re-compressed to go through the #2 turbo anyway, so its probably not an issue.

 

The 66mm Precision wastegate will bypass all the exhaust gas that is not required to keep the #1 turbo spooled up, so EGT shouldn't be much different then it would on a single turbo setup? If I get carried away, I may log exhaust temperature or pressure in the SPA manifold just to see if there is an issue there.

 

I don't know of any large sequential setups apart from the couple of 2JZ ones done by Sound Performance in the USA. Are there other ones as well?

[Noz]

I presume you mean between the two turbos. I did consider using a 3.5" exhaust pipe there, but as you say it does get re-compressed to go through the #2 turbo anyway, so its probably not an issue.

 

The 66mm Precision wastegate will bypass all the exhaust gas that is not required to keep the #1 turbo spooled up, so EGT shouldn't be much different then it would on a single turbo setup? If I get carried away, I may log exhaust temperature or pressure in the SPA manifold just to see if there is an issue there.

 

I don't know of any large sequential setups apart from the couple of 2JZ ones done by Sound Performance in the USA. Are there other ones as well?

 

My thinking with the EGT on single turbo compared to twin, is would you allow 100% of that exhaust gas to flow through one of the small turbine housing until given pressure changes. As long as it would be ok then should be fine. But I always thought the main problem was temp.

 

With a larger exhaust pipe you would reduce the heat, even if you increased the temperature with the re-compression at the turbine entry, it would definitely be a lower temperature. Even if only by a small amount.

 

I know this sounds stupid, but I would of thought some system to reduce the exhaust gas into the first turbo would of help made this system function exactly as planned.

 

Your best bet is to just design it and see how you get on. I wouldn't heat wrap or protect anything though I'd definitely want to remove those temperatures. You may be right, without a decent heatsink on the downpipe (which would still only give varying results) the heat may not even be a problem. I'm hoping this problem is in my head, and it works amazingly and I'm just talking nonsense. lol. Datalog the hell out of it, the learning process is going to be huge!

 

Loving just the thought process on this though dude! You could be creating one of the best supra dyno curves of all time

[V8KILR]

The big 66mm wastegate in the pic is what limits the flow through the first turbo. If the first turbo starts choking before it reaches full boost, to fix this you just open the wastegate slightly (say 5%) while still allowing enough flow to keep boost increasing. By doing this if needed, it will limit EGT and back pressure. The stock system of course chokes the engine way more on the first turbo then my setup ever will, so would have even more back pressure and higher EGT.

 

If it works as designed, it will be an awesome dyno curve similar to what a 6L 2JZ engine (if such a beast existed) would make. This is because once the first turbo is at 15-20psi, it will spool the second turbo just like it is a 6L engine.

[V8KILR]

Finally got the SPA manifold.

 

 

 

 

 

Overall I'm pleased with the casting. There are 3 places where it would need tidying up, but I guess you get that with most cast manifolds.

 

The wastegate hole is not round and is 32mm on the short side and 34mm on the long side. I'll port it out to an oval hole of around 50 x 36mm, but I won't be using it initially anyway.

 

The T4 flange hole is not rectangular and is 75mm at the wide part and 68mm at the narrow part. The height is 52mm. I'll be porting this out to an oval shape around 85 x 65mm for extra flow to the #1 turbo and the 66mm wastegate.

[V8KILR]

I spent an hour or two yesterday port matching the ports to my spare exhaust gaskets on the SPA manifold. Also ported out the T4 flange to my undivided T4 gasket as well so that's all ready.

 

I also have bought just about all the parts I need including the Turbosmart wastegates and MasterPower turbos I will be using.

 

Currently trying to find a local workshop that will fabricate the exhaust pipes and compressor pipes as per my latest design. It's been simplified a bit which should make it a lot easier to tune.