Twin Turbos - No tail vs Duck vs Turbo - Airflow/Downforce?
 

I'm gathering information for my winter TT project. I will use the info to help locate the intercooler(s) and configure the type of tail, if one is used, or style wing to use.

What would be the difference in these three configurations?
Does a tail really help to direct air into the engine bay or is it needed just for downforce?
Isn't the fan alone enough to pull air into the engine bay?
If a tail is only needed for downforce can you substitute a wing?
What are the affects at speeds from 50-100mph (target)?

The tail creates a high pressure area at the engine grill which does help force more air into the engine compartment. Without the tail the area around the engine grill is a low pressure area which means it's actually wanting to suck air out of the engine compartment. So with the tail you do get more air/cooling. I think I've heard a couple of people on the board report 10-15* difference in average engine temps after adding a tail. (sounds like free HP to me)

I wouldn't think that a wing would be as good for the cooling as a tail. Then if you are looking for intercoolers there's the problem of getting them to fit without the tail. Not easy to do, but has been done.

If you want to know about the aero affects of tails go here
http://forums.pelicanparts.com/showthread.php?s=&threadid=126462&perpage=40&pagen umber=1

http://forums.pelicanparts.com/porsche-911-technical-forum/111538-downforce.html

Thanks for the links!
A wing would be used only for downforce not cooling. What you are saying sounds reasonable, in a wingless car the fan is fighting low pressure for the air it needs to push into the engine.
So I need a wing. Turbo or Duck?
Intercooler placement is a matter of custom fitment. Minimal added wing weight is the focus.

Why twin turbos?

Why not?

Just kidding. I'm gonna post for questions and advice when I get closer to actually starting the work. What I want to do is come as close as possible to eleminating boost lag. From what I have read this can only be done with extremely short exhaust manifolds and twin turbos. I have no intension whatsoever of increasing the horsepower, only reducing lag to its earliest possible threshold. I expect you to be right there with me on this quest. ;)

A ducktail is pretty cool because it reduces lift almost effectively as a Carrera tail, but actually reduces drag, unlike the Carrera/Turbo/etc. wings.

Plus, bragging rights for intercooler fitment under a ducktail are substantial. ;)

Sounds like fun. I'll be going less the gt2 tail on the street.

I've been thinking about using the duck tail for some time. The idea is to make a very shallow full bay intercooler that would fit flush with a grill cut in the duck. With most of the tail cut away it should weigh very little. The GT2 unit I currently run is very functional but also very heavy.

What do the fiberglass vs metal ducks weigh?
Are metal ones even available?

It looks like the fused metal/glass ones are still made by Gary Kempton/Made By Hand, but they're about quadruple the cost of a glass one.

My guess is the weight difference would be a wash.

Here's a link to a Rennlist discussion about fitting an intercooler under a duck.

A twin turbo size intercooler that would fit under a
"Duck" would be costly to fab.
I went with a M.A. Shaw GT-1 one piece lid/wing, you
get down force and greater cooling.
http://forums.pelicanparts.com/uploa...1101144629.jpg
http://forums.pelicanparts.com/uploa...1101144728.jpg

Thanks for the links Jack, I'll take a look.
Amac - the tail I have now is the same as the one on your car. It functions perfectly but is very heavy. I'm looking to reduce weight up high in the rear end of the car. The side scoops are functional but have to be a drag hog.

I am putting together a TT 3.6 for my 930 now. Fitting the proper intercooler under the rear decklid is very challenging in high horsepower situations. You can go with a smaller tail and a smaller custom intercooler...but why go through the effort to go twin turbo and leave a bunch of horsepower on the table with an undersized intercooler? I am modifying two 996 intercoolers...and even at that, I am going to a DP style 930 wing to make sure that I have enough room for the plumbing.
I can't speak to the effectiveness of a ductail, but I know how my car squats with the turbo tail at 130+
Rick
'78 930

Mine is a unique situation. I'm NOT building a high HP motor.
Goal is 400HP tops with as much torque and as little lag possible.
The tail needs a diet. I'm looking for alternatives.

Are you planning to keep CIS?

In case you are, I would forget about TT and go for quality single turbo with good headers.

The main problem is that both turbochargers will have to draw air trough single CIS-cone. It can be solved, yes, but piping will be a maze. Then it all has to go back into single plenum trough double-throated intercooler. Lots' of work and $$$'s.

Second, there are few cheap small turbos that will fit the bill and that will survive w/o watercooling. You can try to adopt some KKK K24's or similar but i don't believe you'll win that much in response.

Chances are, you might end up with similar boost response after lot's of dough being spent on upgrade. Remember, twin turbos also mean twin air-paths, twin shafts spinning, twin oiling systems etc.

If I was you, I would ditch the twin-turbo ideas and go for short thermal-wrapped headers w/o heat, Garrett GT30R BB turbo and maybe some lighter tail.

Remember, turbo-lag isn't only in lower shaft intertia. It also depends on turbocharger size, efficiency, intercooler volume etc.

I have to agree with beepbeep, I think intercooler size has alot to do with lag. I have a monster intercooler and when you get on the boost and switch gears it's ready to go again.
http://forums.pelicanparts.com/uploa...1101181266.jpg
http://forums.pelicanparts.com/uploa...1101181374.jpg

There was a recent article in 'Excellence' about a 930 / 934 that incorporated a ducktail w/ intercooled / turbo engine underneath. Granted ther were mods to the ducktail, but the profile remained intact. An interesting alternative to the "tea tray & larger spoilers.

Jeremy

Ya bummed me out Beep.
I need significantly lowered lag threshold. The damned thing litterally jumps like a toad when boost hits. I have to ease the throttle up to 3000 to avoid the shock.

Here's what I've got:
Stock CIS
Euro Exhaust
Stock Muffler
0.7 bar spring
B&B Intercooler
K27-7200 Turbo
SC Cams

The motor has a TON of off-boost torque, it jut needs to be a less violent transition. The threshold is right at 3000rpm.

That red car looks just like mine!!

http://forums.pelicanparts.com/uploa...1101261803.jpg

Install425,
Wow. A dinner table sized intercooler. That looks even larger than the big Kokeln model. I like it.


RarlyL8,
Seems like you have several options here. Here's a few that pop into my head (demented slightly as I am):

1. Adjust driving style. Always keep the rpms above 3000rpm so full boost is always available. I guess this is only possible above 20mph or so since you are using a 915. A guy I used to see at track events told me that was the way to avoid turbo lag on the track.

2. Ducktail - would look really cool and be even cooler in combination with early hood and RSR bodywork. It'd be a shame to start over on the paint, though. Yours looks too nice already.

3. Twin fender Intercooler plumbing - Two small intercoolers in the rear fenders fed by two small turbos. Extra air ducted to the intercoolers from NACA ducts in the rear turbo fenders.

4. Six E-Rams - One for each cylinder. Sorry. Just kidding.....

5. Intercooler in the engine compartment - Feed it with extra air from hoses coming from NACA duct rear quarter windows.

6. Intercooler for twin turbos located where the muffler used to be. You don't need a muffler since the turbo muffles the sound anyway, right?

7. EFI, megasquirt on one end of the scale and Motec on the other.

8. Sell your motor and build a 3.6 turbo. Or maybe even just a 3.8 NA motor like Charlie Swanson built (3.8 w/ 50mm PMOs). Lots of torque and INSTANT throttle response.

9. Swap me your turbo motor for my 3.2 and put a turbo on it instead. :)

10. You could also go stiffer on the rear shocks or torsion bars to eliminate some turbo squat. If you haven't already, it may add to a more responsive feel upon acceleration.

Everything in the list will cost you more money including the driving style comment(more gas). Sounds like a lot of fun. I look forward to seeing what you do. I'll just live vicariously through your experience while I try to keep one car running and get the other one running for the first time.........

Just do what you want to do. If you enjoy it, then it's worth it. Don't skip out on something you want to try because people tell you it can't be done. When someone tells you it can't be done, it's only because no one's done it yet. I really enjoyed showing up to the track after getting my 8x16 and 9x17 wheels under stock SC flares. Many people told me it would never work, including some pretty experienced guys. With enough time, money and ingenuity anything can happen.

On second thought , you better ignore everything I just wrote because I'm sure I don't know what I'm talking about.

Have fun.

You might want to look at your exhaust configuration and go to a setup that minimizes the lag on your K27. You may also want to investigate a different turbo with a modified impeller. I have an old style B&B setup with a k27and 1 bar spring. The boost starts to kick in gradually with load as early as 2000 rpm and is in full boost by 3,500/4,000 rpm.

I have to agree with the previous twin turbo comment from beep beep. It's alot of money and fabrication. If you look at what Stephen Kaspar (Imagine Auto) is doing with some of the single turbo setups, he is building some very powerful single setups with minimal lag. The lag is really a carefully balanced use of exhaust, electronic boost control, effecient intercooler , and well prepared turbo housing.
Rick
'78 930

Rarly8 Twin turbo 3.6 with your tail....

http://forums.pelicanparts.com/uploa...1101315981.jpg

Boost is full on at 3000rpm. Not sure at what point it begins to build. It is interesting to watch the boost gauge move up to 0.1 bar whan I climb a mountain at a sustained speed of 2500rpm. No throttle input is needed, the turbo adds the extra power. Cool.

What I would like is to have full boost at 2500rpm. Is this possible given what I have to work with? Wouldn't ANY header for a single turbo have runners that are too long? Twins are the only option for the absolute minimum runner length.
The shock that the car receives when boost hits is violent. I figure the earlier it comes in the better I can control the "hit".

Ducktail with my intercooler is definitely doable. Ducktail with a custom full bay intercooler is also doable. The unit just has to be more shallow than all the monsters we usually see. This is not a track car so that's not a problem.

No body mods will be done. I've got way too much time sweat and money in the paint now.

I know Mike the Mechanic Amac and Natchamp have twins. What is your boost threshold? How much does EFI play into this?

I just pulled a Carrera tail off my car and went to a duck. The stock tail weighed 41lbs, the FG duck weighs 11. I'm sure that staying with a FG tail, whatever shape you need to fit the intercooler will be much lighter that the stock steel stuff.

It's a multifaceted problem.

First of all, it's hard to make turbo-charged motor act as N/A one. There are certain tricks (ALS and such) but it's WRC/dragracing stuff that is hard to use on street.

Second, perception of "lag" can be divided into two components:
1. Shaft intertia = how long time does it take to spin the turbo when you floor it at, say 4500 RPM's.
2. Boost treshold = when does it start to produce positive manifold pressure if you keep your accellerator floored from 1000 RPM.

My understanding of your problem is that your turbocharger makes transition from no boost to lot's of boost in short rev range (around 3000 RPM), right?

Well, that's something that's connected to actual turbochargers compressor-map.

-- snipped text based on misscalculation, se below for correct calculations --

Also, there is lot to be won by stripping out heat and wrapping headers with thermal blanket. Using headers for heat is directly bad for turbo...turbo want's its exhaust gases as hot as possible. It lives on gas flow, and gases expand when hot -> more flow.

If your K7200 kicks hard and has steep boost buildup then I guess that left surge line on it's compressor map is quite steep. Unfortunately, there are very few compressor maps for bigger K-27 units so there is lot's of costly mumbo-jumbo and shops that sell "customized" things without any documentation. That's why I like Garrett. It's all there and there are no smoke screens.


Personal observation:

I have driven cars with small turbochargers giving lightning-quick boost from 1700 RPM and upwards and i didn't like it. Boost was ther in like, 0.2 seconds. Spoolup was so quick you couldn't notice car was turbocharged but you could still tell there was a small lag. It felt like driving big N/A engine with sticking throttle :D

I guess I'm just old-school and like decent turbo-kick :-)

Turbo map? I thought that was a dragonfly's wing.

Just kidding. Great info. Maybe I should just be happy with what I have. The motor spins relentlessly to 7000rpm. Real easy to over rev in 1st and 2nd. I don't like to rely on the rev limit switch to react quicker than I can lift.
I can manually eleminate "lag" by not stabbing the throttle off boost. The motor will follow a quick foot feed from idle to 3000rpm and then a stomp. It just happens so quickly that I have trouble getting it right every time. Violent boost bursts break parts spin tires and can cause loss of control. Maybe I should look to the transaxle for help. A very tight posi might help things out.
I have less than 400 HP. How on earth do high output turbo folks deal with these issues? And how can Juan NOT spin his tires? That's nuts!

Easy-he just breaks his axles!

How on earth do high output turbo folks deal with these issues? And how can Juan NOT spin his tires? That's nuts!

Well my findings...................

I recently made some more changes and we set the low boost to 22 pnds at that level even with the drag radials it will spin them so we start to work on that and at 17 psi on the low side she performs great and no tire spin, we now have the high boost set up to 30 psi.

we will run depending street /track/ either 17-22 and the high boost is at 30 psi period, we are running 1 st and 2 nd gear on low boost regardless of street/track we will on 3 er gear with a remote control .switch to the high boost ,the car will run fine with that set up, we will increase the high boost to 40 psi and the low will stay the same, our power band is all mid range to top range which helps a lot with the traction a problem everyone seems to have.

we are 27-0 on the underground world and thats with cars that run low 10 sec at the track, the closest race that we had so far we won by 4 cars, back to square one traction and the ability to deliver power from mid range to top.

And yes we are breaking axles out of the hole, and yes we have found the solution, and yes we will continue our quest is just matter of time and $.

Good Luck with your twins, but you can make a LOT of power going single, and without a proper intercooler tt will do very little for you.

Of course thats just my 0.02:)

40psi boost? And I thought I had issues.
Apparently that locked differential you have must work miracles. This may be the area I need to explore. I'm not familiar with the lock up ratios offered by Quafe and others but soon will be.
Again, I'm not after big HP numbers but increased driveability on/off boost.
So, given all the input by you fellas, is it reasonable to assume that I cannot achieve a significant decrease in boost lag by going twin turbos vs short wrapped headers and a single turbo?
Is it also reasonable to assume that no turbo exists that would pull full boost 1.0bar from 2000-7000rpm and support 400hp?

Duck tail weighs 11pounds? Where the heck to I order? What do these things typically cost? I'm checking Pelican as soon as I post this note.

I'm not sure if Pelican sells the ducks. I got mine from The Racers Group, $379. They have a CF version for $100 more.

Rennspeed is likely the least expensive.

-- snipped text based on misscalculation --

[edited, see my post below]

(well, not much more than mumbo-jumbo KKK unit with Porsche-tax added) http://www.pelicanparts.com/support/smileys/clap.gif

Bollocks!

I somehow entered 6-strokes in my whiz-bang Excel spreadsheet so numbers are all screwed!

I'll get back with new numbers later this evening!

http://www.pelicanparts.com/support/smileys/wat5.gif

Here we go again:


3300cc 930 engine will gulp 34,8lbs/min of air at 5500RPM boosting 0.7 bar and will produce 302HP with 90% VE, 29% efficency, 40deg. C intake air (I used this to "calibrate" my Excel sheet as this is typical for 930 engine)

If we look extreme case and calculate for 6000 RPM and 1 bar of boost, same engine will use 43,0lbs/min and produce 381hp, corrected for 42 deg C higher intake temp and slighly less VE of 89%.

This means that you will be far outside of GT28RS map so I beg you all not to use any of calculations I made higher up!

It will be spot-on with GT30R though.

If twin turbos are used this means that 21.5lbs/min mass flow per turbo is needed, and best matching turbo would still be GT15 (altough little more extended into less efficient region).

So answer is:

Single turbo: GT30R
Dual turbos: GT15

Calculating boost treshold gives:

It will use 14,2 lbs/min at 2000 RPM and 0.7 bar boost assuming 100% VE (which happends at low revs) and 32 deg. C intake air temp.

That's well inside GT30R surge line and my guess is that you would see boost buildup around 1500 RPM if using one of those.

Doing the same thing for GT15 gives 7.1lbs/min per turbo which is on the endge but will still work.

So final decision:


Dual GT15's will give boost slightly later than single GT30R (but still 1000 revs before your current KKK unit) and will pump air with 71% efficiency at top end. They will have shorter spool-up time at higher revs though, due to low inertia.

Single GT30R will spool up slightly earlier using identical headers as GT15 (but probably the same assuming long headers used for single turbo) and will pump air with 72% efficiency, allowing slighly more power (but it will probably get nilled by using long headers).


And my personal recomendation:

GT30R all the way. It's single, it's easier to install, there is less piping to fabricate, it will have lower boost treshold than twin GT15's and breathe as well at high end provided you don't boost more than 1 bar @ 6000 RPM.


Sorry if I messed things up with my earlier post, mistakes do happend. I changed "4" into "6" noth thinking it's number of strokes and not number of cylinders.

Cheers!

And here are some butterfly wings to match:

http://forums.pelicanparts.com/uploa...1101764689.jpg
http://forums.pelicanparts.com/uploa...1101764704.jpg
http://forums.pelicanparts.com/uploa...1101764719.jpg
http://forums.pelicanparts.com/uploa...1101764738.jpg
http://forums.pelicanparts.com/uploa...1101764757.jpg

Thanks for the info!

Now one question. How the heck do you read those funky graphs?
Can you explain to me how one big turbo is able to spin up quicker than two small ones? Not only are the impellers smaller but the headers are shorter.

One possible reason is they are also getting half as much of the exhaust gases each.

Wayne doesn't sell it, but here's a good turbocharger book.

http://www.amazon.com/exec/obidos/tg/detail/-/0895861356/qid=1101783280/sr=8-2/ref=sr_8_xs_ap_i2_xgl14/102-5911360-6472956?v=glance&s=books&n=507846

I like it but some people reviewed it and called it "dated".
Other people liked the Corky Bell book, but I haven't read it yet.
http://www.amazon.com/exec/obidos/tg/detail/-/0837601606/ref=pd_bxgy_text_1/102-5911360-6472956?v=glance&s=books&st=*

Corky Bell's is one of the best reads you'll have. You won't regret the $$ spent on it and you'll feel in better command of a turbo package. Great read for both the virgin and the experienced.

beepbeep

if you were to compare side by side the T series turbos vs the Gt series for example

T- 66 = GT 28 and so on, it be cool to have a side by side comparasion since you are doing such of good job:)

Easy. Vertical axle is pressure ratio. If it says 1.7 it means 0.7 bar of pressure difference = 0.7 bar of boost.

Horisontal axle is mass flow. It's how much air you move per minute. "the wings" is how good compressor is working at certain regime.

For example, if you use GT15 and try to achieve 1 bar of pressure into engine that will only accept 5 lbs/min it won't work. It will just stall like clogged vacuum-cleaner.

If your engine uses 15lbs/min at same boost, you'll achieve approx 74% efficiency, which means that 74% of energy used to spin the turbine will be compressing the air and rest will be heating it.


It's not just a matter of spinning up. It's also a matter of internal geometry, flow, shaft frictional-losses etc etc.
(I didn't take into account header heat losses as it's very hard to estimate).

Twin turbochargers will also get half of flow, so I started by calculating how much your engine would be using in certain range (0.7 bar @ 2000 RPM) and plotting it into graphs.

It's 7.1 lbs/min for GT15 and 14,2 for single GT30R.

For 1 bar @ 6000 RPM mass flow figures are 43 and 21.5 lbs/min respectivly.

GT30R will pump with 72% efficiency there and GT15 will be around 68%...For low boost, GT15 will pump with 65% efficiency and GT30R around 68%.

GT30R is single unit with just one ball bearing shaft. Dual GT15's have twin shafts with plain-bearings (=more frictional losses)

GT30R can be fitted on same place as KKK papperweight, minimizing the amount of plumbing, both lubrication-, intake- and exhaust-wise.

Dual GT15's will have shorter headers (=good), approximately same boost treshold, pump more hot air at top-end (=worse), use dual intake and exhaust plumbing (=worse), have twin plainbeared shafts that need to be spun (=worse).

To me, case is simple. Your engine size and desired power level makes twin turbocharging unnecessary and costly project, especially when using state-of-the-art single turbocharger. Even more so considering you still want to keep CIS.

It's up to you...


http://forums.pelicanparts.com/uploa...1101839272.jpg

I've read both those books and more. Good stuff.

Thanks again for all this info. Much more specific than what I've read. This all leads to an additional question:
Given all the above info, why did Porsche twin turbocharge the 993T ?