Good improvements!

You should just have put in a softer spring to solve that. A 1 bar spring would have been perfect for this engine. But back to the main issue before - could you now run higher boost? It seems you could with the revised plumbing but still you did not run higher boost looking at the chart?

 

They actually hit 35psi at one stage so there is no problem hitting boost now. Looks like our main problem is headlift again...Looks like we're going to have to take the head off again and see what's going on. Might change to Evans as well.
Next test day will be the event in a week. Just going to have to hope for the best.

 

Good to hear the boost issue is solved but damn really sorry to hear about headlift. Interesting as you didn't have head lift with the previous motor at higher hp with the same sealing setup.

Evans is not allowed in some racing series here and in the US etc. It's super slippery. Guess your rules doesn't say anything about that?

 

Yes, temps climbed from mid 80's to 110 in a pull. Too much fluid coming out. Nuisance! Maybe bolt stretch, maybe different block and it's giving too much. We'll check stud length.

I'll have to double check on the Evans. I didn't want to run it if I could avoid it, but perhaps not on this motor.

 

Always something

On Evans - it seems the engine will run hotter with it though. I'm sceptical.
http://www.norosion.com/evanstest.htm

 

Nice improvements!

His previous motor had a 8:1 CR and this motor is 9:1. This makes a pretty big difference in peak cylinder pressures and temps.

 

I was checking all day to see how it went. Not much off a engine wizz because i have two left hands but i follow You, Duke en Shawn and damn you guys have some tough times but you are doing good!

Keep at it and thank you for sharing your info, development and stories with us. Hope to finish my engine one day soon .

Keep at it Patrick looking forward to seeing the pictures and movies!

 

I had missed that. His new cam spec lowers the dynamic compression ratio a bit though so the net effect is probably less than it looks.

 

I just wonder if, since there was no mention of head lift at the last dyno session, if the change to the smaller hot side is contributing to head lift?

Comparing the dyno charts, it looks like with the .82 @ 6,500 rpm you might be closer to 540 HP @ 26 psi and still be able to go a bit higher rpm and boost wise.

I know you cannot directly compare dyno to dyno and different engines etc, but to the extent you can look at comparative dyno charts of the same engine, I suspect most of your improvements were from the other changes you made rather than change in hotside housing. It seems like you reach your boost point around 900 rpm earlier now which is a fantastic improvement.

However, if you look towards the end of this post (same one Pauly posted earlier):

http://blog.perrinperformance.com/ga...omparo-part-2/

you will see that changing from .82 to .63 A/R on the GTX3582R only seems to help spool by around 200 rpm. I guess for me, it would not make sense to lose the top end and have potential issues associated with the increased exhaust back pressure and everything having to work that much harder just for the sake of an improvement in spool of around 200 rpm.

Of course the 200 rpm is somewhat theoretical (it is based upon a 2.5 l engine though, albeit Subaru) and of course I know (as you have mentioned) the difficulties associated with making several changes at the same time but the internet is nonetheless a wonderful resource where we can learn from other's experiences, whatever they may be worth.

And that's exactly what I love about this type of thread and people like Patrick and Duke who really push these engines and share their results so that we can all learn how particular components and strategies apply specifically to what we have - great work guys!

 

Another thought... of course this is not an exact science and Garrett's stated HP figures are probably a little optimistic, so take it for what it's worth - here goes...

Garrett rate this turbo as being suitable for 450 - 750 HP (at the flywheel) and they offer it with three turbine housing A/R's: .63, .82 & 1.06

Looking at the turbine maps, the respective flow rates in lb/min are approximately: 23, 27 & 32

If then we extrapolate what Garrett has provided we can deduce that 32 lb/min is a somewhat appropriate turbine flow rate for 750 HP. If we then divide 750 by 32, that yields us around 23 HP for each lb/min of turbine flow.

Extrapolating again would suggest that the .63 may be appropriate for up to 23 x 23 = 529 HP whereas the .82 up to 27 x 23 = 621 HP.

If we accept those figures, it would indeed suggest that the upper limit of the .63 has been reached as 503.8 WHP most likely exceeds 529 HP at the flywheel.

Again guys... just "thinking out loud" - for what it's worth!

 

The almost identical peak hp figures reached for both turbine housings at pretty much the same boost tell me the 0.82 housing is, in this application, about as restrictive as the 0.63 housing.
This suggests to me the GTX3582R compressor is never going to flow as well as it can unless being used with the largest 1.06 housing. The GTX3582R is advertised to flow even more than a GT4088R which uses a higher-flowing hotside, and I don't think I have ever heard of anyone running satisfyingly a turbo as large as a GT40 on a 2.5 8V 951 engine.

The head lift now with the smaller housing tells me, considering the high CR and the large overlap on the cam, that the back pressure/boost ratio must now be pretty bad.

Apart from the capacity, your engine Patrick is getting very similar to Corleone's, with similar CR, big cam, except you are running basically twice the boost. His GT3582R 1.06 is still a more balanced cold side/hotside combination, for contained back pressure at peak power rpm, than a GTX3582R 0.XX

 

As you're aware the bulk of these parts that make up this engine were already built. So we have what we have. Quite likely it's meant to be setup as a higher revving motor with a shorter runner intake and more power at the top of the graph. Its quite possible that we're 'going against the flow' by doing what we have, but it's for a reason. We needed to get more bottom end response back after our brief test day. It was way too laggy and there was no way we were going to be able to get close to our time from the same event last year. So what you're seeing is a compromise. I am still trying to put some sort of 16v motor together for the end of the year but this may or may not happen. If that is an impossibility we may have to try a different cam in this 8v motor and see where that gets us. Keep the comments coming though. Some thoughtful insight. Thanks.

 

But also note that the boost pressure dropped at top end on the .82 housing.

 

I don't think this is quite right... looking at the dyno comparison overlay @ 7,000 rpm shows a gain of approx 20 HP with around 1.5 psi less boost for the .82 compared to the .63 which is what led me to think that 540 HP may be possible for the .82 @ 26 psi & 6,500 rpm compared to the 504 HP the .63 is achieving.

It's a little difficult for me to imagine a situation where a .82 housing would be as restrictive as a .63 housing... even if the 3" exhaust is becoming a restriction, which I suspect it is, the smaller .63 housing would be more sensitive to that - ie, a larger hotside could be more forgiving in this respect.

Here is an interesting and quite informative article "Turbo Exhaust Theory" written by a turbocharger development engineer for Garrett:

http://www.sau.com.au/forums/topic/3...xhaust-theory/

 

I would expect you will have to try the 82 housing again, perhaps after a track test, that curve looks a lot more driveable, but the 82 would probably give you a bit more now, though you never really know the trade off till you try.