Not sure what you mean by that . The problem ,I believe , is a combination of the Siamese port issue mentioned earlier and the lack of port area as the exhaust port is closing.

https://cimg7.ibsrv.net/gimg/www.rx8...9ddc470295.png
More changes , boost shown on blue line above (13.5psi) , power from same log below .Feels like It's better :)
https://cimg8.ibsrv.net/gimg/www.rx8...7ab7b0d62f.png

Go 400+!!!

Boost was jumping around at over 7000 but is now very stable everywhere . Pretty happy with this

Logs indicating :
@7psi - 300whp (minimum boost possible at 7500rpm)
@10psi - 350whp
@13.5psi -394whp

If only your dyno was close.............. :)

Yep , it got fixed last week . ATM I'm still running E50 which I'll stay on till my next dyno.

Cheers .... yeah the power/psi is excellent ! I know it's only virtual dyno but that has proven very accurate now that I've refined my logging .

So what did you do differently this time? If you are willing to share.

I enjoy following your progress on this thing. I will hopefully be starting on my setup shortly, this thread has been a great source of info.

This is a log at 6psi tapering to 7psi for last 1000rpm . 313whp @ 7800 . Would probably make another 20 or so if taken out to 8500.

https://cimg4.ibsrv.net/gimg/www.rx8...e0529dc90f.png

I altered the manifold . Removed the crossover tube into the siamese/wastegate pipe from front runner. then added two small diameter balance tubes from either end. Basically took some flow away from the siamese ports and made the two outer ports flow more % wise. The idea was to create the same gas velocity through all port runners .

Have you got any pics of the modified manifold?

So .......... had another crack on the dyno today . Red line is today's run , blue line is from march . Both were at 14.5psi so I could see if there is any major improvement . As you can see , spoolup is better and there is a tad more up top . I didn't feel the improvement was significant enough to push further and was feeling a little gun-shy so we stopped the session there .
For the virtual dyno non-believers ...check out the above VD (post 1727) against the real dyno.
https://cimg9.ibsrv.net/gimg/www.rx8...4a01090a1e.png

So, basically, there are full symmetry between exhaust ports and the two inlets to turbine?

I really like the performance at low pressure. 7psi should run quite reliable at stock engine(?). Do you have some "guesstimations" what the difference would be, if you used normal 0-10% ethanol gasoline?

Symmetry would be better than before for sure but maybe not perfect. At 7psi I'm sure the engine would be happy there all day every day. As far as power with pump gas goes ; marginal difference at low boost . At higher boost (over 10psi) I just wouldn't risk it so the point is moot IMO.

I just replace my flexible intake with a solid one and seeing some good results for turbine backpressure .
Back pressure at 14psi with flexi intake :
https://cimg2.ibsrv.net/gimg/www.rx8...6f77d9e739.png
Backpressure at 14psi with solid intake:
https://cimg8.ibsrv.net/gimg/www.rx8...a5a51325fb.png
Looks like a 1.5-2psi reduction at peak ...which is equivalent to going up almost 2 steps in turbine A/R

I had to relocate OMP and go 100%premix to achieve this however . This is new OMP location :
https://cimg9.ibsrv.net/gimg/www.rx8...d1d775e3d9.jpg

You can dial up your target boost to about 15.5-16psi now :)

Rather than starting another thread, I might post this question here to continue the discussion that's already in here.

I have a series 2 GT that I am going to turbo charge. With the different sump design they have, I have decided that top mounting is the easiest route without having to swap sumps, etc.

Based on what you have found I am wondering if I am on the right thought pattern when it comes to manifold design.

A few options -
- The traditional 3 into 1 with an internally gated turbo (easiest option)
- 2 into 1 to feed the turbo and the siamese port dedicated to an external wastegate, with no link back to the other runners
- 2 into 1 to feed the turbo and the siamese port dedicated to an external wastegate, with a link back to the other runners

I think from reading this thread I am lead to believe that the second option is potentially the best?

Sorry if this is Hijacking, I am happy to make a new post if you wish, but the discussion in here is good so might bring about some good ideas.

I would go with a twin scroll (T3 if you can) and run 11/2" runners to the turbo from each end , siamese to wastegate and a small diameter balance tube from each outer runner also going to wastegate .By utilising that design, keeping all the diameters down and running the twin scroll you will maximise spoolup without sacrificing top end. You could then get away with a 21/2 " downpipe and where the WG joins increase it to 3" or (even 31/2" if you want top end) for rest of exhaust.

Why that s such a good option is that it keeps wastegate flow ,which is around 40% of the total flow at peak power , away from the engine bay . This means pipe diameters can be small maximising spoolup without any sacrifice to power potential . And the space available is very limited so this is a big advantage !

If I were to start again I'd do this utilising an EFR 8374. Wouldn't go top mount for any other turbo than that one or the 7670.

Why not EFR9180? From my knowledge, if there is no space restrictions, that looked attractive...? And it looks to be working better in the low pressure area as well. What am I missing?

The 8374 is already a big turbo for a renesis motor. A 9180 is overkill.

I don't think that is a good option for the Renesis . This engine seems unable to utilise the extra flow capacity this size turbo is capable of. You might get a bit extra up top due to the reduced backpressure but the rest of the powerband will suffer moreso than on an REW .There was a topmount 9174 fitted by Turblown a few months back and that took to 6000rpm before it was fully spooled to 20psi. That's terrible spoolup. The manifold design I suggested will help, but it still wont spoolup as well as a PP exhaust engine does.

This is the reason for my assumption:

https://cimg7.ibsrv.net/gimg/www.rx8...35ca53282a.png