Honda / Acura K20a K24a Engine Forum banner

1 - 10 of 10 Posts

·
Registered
Joined
·
4 Posts
Discussion Starter #1
Hi all. I'm busy putting together a EK Civic AWD build here in South Africa. The aim of the car is to run mid-low 9's on an unprepped surface.

I've decided to go the 6870 route as many users state there is minimal spool loss between the 6466/6766 and the 6870. I need some guidance on choosing the turbine housing though.

  • Engine setup will be pretty stout: Ramey Racing spec K24A2
  • Car will use a dogbox and FuelTech FT600 with gearcut on E96 fuel.
  • 8200rpm limit
  • Only trick is we are at 5500ft elevation.
The manifold design will follow whatever housing I pick so not locked into a specific type yet. I was leaning towards a T4 divided 1.00 to help with spool but the simplicity of the V-band is also appealing.

Any input or experience will be very helpful.
Thanks
 

·
Registered
Joined
·
4 Posts
Discussion Starter #4
Thanks for the replies. I've always been a fan of divided, split-pulse setups and all my previous B-series cars have used this style with Holset turbos. I've also seen articles/tests showing that open aka undivided setups make more top-end. I guess that would also be influenced heavily by operating range.

The KA24 engine DSport tested on isn't a high-revving or particularly high-flowing engine in base form either, so I'd theorise that a combination revving to 8500+ might paint a different picture for the single-scroll setup.
 

·
Registered
Joined
·
60 Posts
The physics behind the dual scroll is almost entirely to do with removing the cylinders that are blowing down in the exhaust stroke from communicating with the cylinders that are seeing overlap. Cylinders 1 and 4 operate 360 crank angle degrees out of sync, just like cylinders 2 and 3. What happens with a single scroll for example is cylinder 1 blowing down and cylinder 3 sees increased pressure in the exhaust port during overlap when the exhaust and intake valve are both open. This reduces the amount of fresh charge that can get into cylinder 3, causes more internal EGR and increases trapped charge temperature, reducing the amount of spark advance you can put in before seeing knock. The effect of which is shown in the KA24 dyno pulls. I would think having longer duration cams with more overlap needed for higher revs would only make the problem worse. Furthermore, Even if a single scroll would net more peak power, I would be more than happy to give up 5 or 10% peak power to gain 20+% through the mid range. Peak numbers mean nothing as soon as you grab the next gear, maximizing the area under the curve for each gear is where the fastest times will come from.
 

·
Arouse the DAMPFHAMMER!
Joined
·
5,849 Posts
...I would think having longer duration cams with more overlap needed for higher revs would only make the problem worse.
Nice summary regarding the background to the exhaust path separation Scider! And yes, more valve overlap can make the knock and VE problem worse, but it is not a necessesity, just a possibility. In case when at overlap the intake pressure is higher then exhaust, it helps to flush/scavenge the combustion chamber. On a right designed turbo setup this is possible for a bigger amount of the revving band. On the 1200 hp-B-Series engine turbo setup I designed one can see that posivitve pressure drop over the flow path up to 9,600 rpm, than it began to change. Around 10,000 rpm the exhaust pressure at overlap is bigger then the intake pressure. It is just a question of design and applicated torque band. The K20C1 has also cam phasing on the exhaust side, for reason...the B-series not. But the K20C1 isn't the one carrying 1200 hp that easy the B-series can do it.
 

·
Registered
Joined
·
4 Posts
Discussion Starter #7
Nice summary regarding the background to the exhaust path separation Scider! And yes, more valve overlap can make the knock and VE problem worse, but it is not a necessesity, just a possibility. In case when at overlap the intake pressure is higher then exhaust, it helps to flush/scavenge the combustion chamber. On a right designed turbo setup this is possible for a bigger amount of the revving band. On the 1200 hp-B-Series engine turbo setup I designed one can see that posivitve pressure drop over the flow path up to 9,600 rpm, than it began to change. Around 10,000 rpm the exhaust pressure at overlap is bigger then the intake pressure. It is just a question of design and applicated torque band. The K20C1 has also cam phasing on the exhaust side, for reason...the B-series not. But the K20C1 isn't the one carrying 1200 hp that easy the B-series can do it.
What would your recommendation for this setup be? Realistically we will target up to 800whp with this setup and operate at 5500ft elevation. Purpose: Drag Racing

While I initially planned to use a 6466, the feedback from a number of users was that the spool trade-off between the 6466 and 6870 was marginal and the added hp potential made it worthwhile. In view of this I've been leaning towards a divided setup to help mid-range punch and widen the powerband as far as possible.

What through my assumptions off was the fact that the BYP MR2 runs a very large undivided housing and achieves incredible response and peak power from that setup on the same 6870 turbo...
 

·
Registered
Joined
·
60 Posts
It looks like both of those turbos come with an amazing number of turbine choices as well:
6870:
- T3 1.06 A/R with a 3" 4-bolt discharge
- T4 (undivided) .68, .81, or .96 A/R
  • T4 (divided) .84, 1.00, 1.15, or 1.32 A/R
  • SS .81 A/R or .96 A/R V-band inlet/outlet turbine housing
6466:
- T3 .63 or .82 A/R with 4 bolt (2.5") discharge
- T3 .63 A/R with 5 bolt discharge (with or without wastegate hole)
- T3 .63 or .82 A/R with 3" OD (outer diameter) V-Band discharge
- V-Band inlet .82 A/R with V-Band discharge
- T4 .58, .68, .81 or .96 A/R with 3 5/8" OD (outer diameter) V-Band discharge
- T4 Divided .84, 1.0, 1.15 or 1.32 A/R with 3 5/8" OD (outer diameter) V-Band discharge

I'm no expert here because these near four digit HP levels just isn't something I've ever been into, but it looks like there's a number of examples of motors that surpass 800 whp with a 6466, this EVO for example: 886 WHP Eight-Second EVO IX Aiming for the Record

Fortunately they listed the T3 housing with a 0.82 A/R in the specs, so a smaller housing with the larger A/R, but a nice data point to have. Unfortunately I haven't been able to find any compressor maps or turbine maps from Precision which makes things difficult to evaluate further, but they're not a Garrett or Borg Warner. A smaller A/R would spool a bit faster but increase back pressure and reduce peak power. I would think a T4 housing with a smaller A/R to be somewhat comparable, I'd personally be curious about how a T4 divided 0.84 A/R housing would compare.


Altitude is going to work against you though. Using that Evo as an example, they hit 48 psi peak. If that's an absolute pressure, the pressure ratio would be ~3.3 in Illinois where the dyno is, and that would make the pressure ratio at 5,500 ft ~4.0, which could put you off the compressor map into the over speed range. This is where I really want a compressor map. A call to precision to see which compressor is more favorable for high pressure ratios might be in order, but I'd THINK it would be the larger 6870.
 

·
Registered
Joined
·
4 Posts
Discussion Starter #9
@Scider

The altitude added to my rationale for going for the bigger option as we lose compressor flow and as you state the pressure ratio goes very high. And yes you are correct, there are multiple turbine housings to choose from - see why this can get intricate ;-). Precision doesn't generate and publish compressor maps but we do know that their wheels outflow competitors equivalent size offerings in most cases (rear world HP).

I have a good understanding of turbo sizing and a fair amount of experience playing with the Holset stuff for customers. So most of the articles shared reinforce knowledge I already have - but its always good getting additional data points and input. Getting this combo right is more critical though as its a big $ build and I want to get it right off the bat...

The 4G63 seems to tolerate backpressure a lot better than Honda VTEC engines. Its common in the Evo world to run the 0.82 T3 or V-band to 750+whp. There are some guys that experiment with divided T4 setups but its not that common on that platform and only really once you move into larger frame turbos.

On our ex 2.0l B-series drag car, we swapped from a divided 0.70 T4 to a 1.00 and gained a lot of top-end with no sacrifice in spool due to the engine breathing better across the board. So the results can sometimes contradict the standard theory or expand the argument...

I appreciate your input and sharing of info !
 

·
Registered
Joined
·
60 Posts
I have a feeling the 4G63 is probably more tolerant of back pressure due to cam options, as it's always been turbocharged, all choices for the 4G63 are designed for that. K-series cam selection for forced induction is far more limited because VTEC benefits naturally aspirated motors. I did a bit of poking around on the STM 650 Camshafts in that Evo to try to see what they're doing, unfortunately it's common these days for manufacturers to not put out any information what so ever.

Cam selection for boosted motors is particularly important as highlighted by the work done by Jaydee in this thread: Jaydee's All Out Rotrex, TVS and Compound Forced...

I could be way off on this, but my assumption on these large turbo DSM motors is that they use milder cams than even the factory VTEC high speed cams, but once you get the turbo spooled up to 45 psi, it doesn't particularly matter, the air is going into the motor regardless of how well it would breath naturally aspirated.
 
1 - 10 of 10 Posts
Top