How long to make my ITB runners?
 

I am fitting Triumph Daytona 675 throttle bodies to a 964 engine and need some guidance on intake runner length.

In the pics below, the runners are over-length and will be trimmed to suit.
The minimum length I can handle between the flange and the ITBs is 110mm due to the bends. This gives about 240mm from the head to top of velocity stack as a minimum overall length.
The injectors fit just below the throttle plate.
The TBs are 44mm ID right through, which matches the ID of the 50mm alloy tube I am using for the runners and will soon match the head ports once they are opened out from 42mm.
Engine will also have DC21 cams, coil-on-plug ignition and Megasquirt engine management and is for street use.

So ... how long should they be???

http://forums.pelicanparts.com/uploa...1263646795.jpg

http://forums.pelicanparts.com/uploa...1263646310.jpg

http://forums.pelicanparts.com/uploa...1263646919.jpg

Bill,

You may be interested in research that Jaguar did in the 1950's on intake runner length while experimenting with their slide throttle injected XK engine. It has a similar hemi head with a 70 degree VIA. Every different pipe length (valve to atmosphere) will produce a pressure wave at a different rpm that will boost the torque at that rpm. The graph has been reprinted ad nausem in many books, I'll try to find one online. Roughly, a 32.5" pipe gives a peak around 3500 rpm, a 21.5" pipe around 4000 rpm, a 15" pipe around 4300 rpm and an 8" pipe around 5000 rpm. The long pipes also have an increasing negative effect at high rpm, which is why MB and others started to experiement with variable length pipes, ultimately leading to the rotating drum intake manifold.

What's the restriction of your engine bay?

Are you going to use an filter box of some kind?
You will need at least 1 diameter of free air above the air horns.

Do them as long as you can, but I doubt you can fit runners much longer than your minimum length.

You may have some fuel delivery, engine response issues with the fuel source (injectors) so far from their respective combustion chambers. Any way you can move them closer?

Sherwood

Back in the IMSA GTU days (2.5 liter engines) dyno runs showed that you needed a 6 inch runner change to even notice a difference.
Varioram intakes make a 24inch + change.
Make them any length you want and concentrate on cams to determine your engine characteristics.

Sherwood is correct. The injectors should be close to the ports with the bulk of the runner upstream. There is potentially some power to be gained by mounting the injectors upstream as the cooling effect of evaporization can give a denser charge, but the price is the likelihood of fuel collecting on and running down the runner walls and the low speed and drivability issues that ensue.
regards,
Phil

Porsche played around with this sort of thing back in the early days of racing 911s. They had high-butterfly, low butterfly, and slide-valve throttles. They had an injector placement up at the top of the velocity trumpets and an injector placement down at the heads where the stock MFI injectors go.

The factory seems to have mixed and matched in experimenting with these parts. From the period photos I have seen, it appears they settled on the low-injector placement though the trumpets still have the mounting provisions. I believe they found no rearl difference in ultimate horsepower and that they reported better drivability with the low injectors. The MFI fuel shutoff on the over-run was partly developed for this reason. Some cars had the injectors above the throttle, which would cause pooling fuel when coasting/engine braking and then a rough throttle response when accelerating again as the fuel dumped rich into the cylinders.

The high butterfly system was meant to allow the air to smooth out before getting to the cylinder- it was thought to flow better by having the turbulence and restriction further away. It was farther from the intake reversion, too. The low butterflys had the advatage of snappier response at low engine speeds because of the smaller volume to fill. The high butterfly system came in at high engine speeds when the flow was fast enough so that the added length did not add much time to the engine speed response.

Interesting input from all, thanks.

Magnus, this engine will be going into a 911 or my 930, so typical height restrictions apply due to a/c condenser under the tail.
I was planning to use a K&N filter same as used on other triple TB applications (E-3671, 12” long x 5-1/4” wide x 3-1/4” high) in a custom housing. The base of the housing will be raised up about 1” from top flange of TBs to clear fuel rails and to give about 2” clearance over the air horns.

"Make them any length you want and concentrate on cams to determine your engine characteristics."
Henry, from what I’ve read on this, I’d have to agree. I sort of came to the conclusion that with the runner length I have to play with it wouldn’t be critical, especially for a non-racing application.

Sherwood & Phil, thanks for comments on injector placing.
I was taking the easy way out by just using the existing injector ports in the TBs. I hear what you are saying regarding response and driveability, but I’m currently driving a 930! Whatever response this engine has, it will be light years ahead of the 930. :D Can the difference in response be that much different with the injectors 100mm lower?
I can see that fuel collecting on runners could be a problem – that might be the bigger issue I think.

I would be interested in seeing some typical overall height dimensions from head to top of filter for TBs or carbs if anyone can measure for me.

Some musings,......

Short runners will be fine with your mild cams,....its only critical when high-RPM, max power output is required.

Your bent runners on those outside cylinders are a bigger concern when injector placement positions them above the changes in direction. Generally speaking, fuel-air mixtures don't like to turn corners (even gentle ones) and that affects fuel mixture. With carbureted engines, we make small corrections to ensure each cylinder has the same AFR, however EFI is a different matter altogether since each injector usually gets the same width injector pulse.

Food for thought. :)

JMHO, of course.

QUOTE=Steve@Rennsport;5131367]Some musings,......

Your bent runners on those outside cylinders are a bigger concern when injector placement positions them above the changes in direction. [/QUOTE]

Yes Steve, I have already had a rethink on this based on earlier comments.
I am now going to fit the injectors immediately above the head flange. This has the added cost advantage of allowing me to use the standard 964 fuel rails and hoses.

The TBs I am using don't easily lend themselves to being spaced wider (by 35mm). It introduces other complications with linkages and filters, so I am prepared to live with the trade-off of bent outer runners.

Conventional wisdom is the runnier/TB length should be about 9 inches
I've been looking at the slot plenum design with a single TB and I was about to do it on a T4 but then i got the 3.0 /6 to put in my 914
I may still try this but with a NA set-up:

Photo Album
Or for a bit of light reading on this:
Turbo Intake Plenum - HybridZ

Try and get a measurement from another car. This is the absolute maximum height I could get (on the right side) to fit under a whale tail, don't rememeber the height i used.

http://www.porsche.nu/obj/photo/f/ff...ed47b2c707.jpg

Looks v nice Magnus.
What TBs are they?
Can't quite work out what filters you have there either. More details/pics?

Thanks, the TBs are from a (3) Suzuki TS1000.

The air filters I made myself, didn't find any that fit good.
Made the fuelrail and manifold too.

http://www.porsche.nu/obj/photo/b/bc...e150fc07ce.jpg

http://www.porsche.nu/obj/photo/b/b8...c8a81af348.jpg

http://www.porsche.nu/obj/photo/e/e5...4f979f6fe9.jpg

http://www.porsche.nu/obj/photo/a/ac...a7c4fc5021.jpg

I've been looking at the slot plenum design with a single TB

The Audi manifold is an interesting design, but since +950 hp has been made with a single plenum sheet metal manifold on a 911, how much benefit could there be ? Maybe there is a different issue on a rally car, none of the F1 turbos used a dual plenum with a slot.

Magnus,
The lack of suitable long oval filters was a problem for me as well. I didn't really want the extra work and cost of using six smaller round ones either. That's one of the reasons why I have decided to keep the Triumph TBs at their original 83mm centres and use offset runners.
I will be making up a GRP or CF filter base and top cover in due course. (waiting for filters to arrive first)

One of my next problems is to design the throttle linkages. I was planning to adapt standard Triumph throttle cables or something similar. You also seem to have used cables. Do you have any pictures of how you made the connection to the existing throttle crank on the engine?

TB details
 

For those interested, here are a few more detailed pics of the Triumph Daytona 675 throttle bodies.

http://forums.pelicanparts.com/uploa...1263863897.jpg

http://forums.pelicanparts.com/uploa...1263863936.jpg

Billjam,

Consider Pipe Max software to get the tuned length right for both intake and exhaust. As for injector placement down low for smog and low RPM, up high for HP. If the ECU can do it staged is the answer. The high EFI location is good for quite a few HP if everything is correct.

PFM

"high EFI location is good for quite a few HP" ... can you quantify?

Billjam,

The numbers below are for a very tuned Cosworth 2LTR 4 valve 4 cylinder. The numbers are TQ on the dyno. The EFI is MoTec agian very tuned. The increase is in the 10% range, nothing to sneeze at. Not my motor just some info from a tuning acquaintance.


rpm low inj high inj
4500 82 108
5000 108 130
5500 145 162
6000 150 165
6500 148 163
7000 148 158
7500 145 155
8000 140 146

PFMhttp://forums.pelicanparts.com/uploa...1263883035.jpg

Sure, this is how I did it. I made a custom bracket and 2 stop screws. I would have preferred to use only one cable and some kind of mechanical linkage between the banks, but I didn't have time for that.
Use some high quality cables that can turn tight turns without binding. I used cheap bicycle stuff, and it isn't great, will change when I get the time.

http://www.porsche.nu/obj/photo/4/4c...d1a10def03.jpg

Thanks. That is food for thought. There's quite a difference at the lower end of the rev range - as you said, not to be sneezed at.

I'm not so naive to think I would see 10% difference in my situation, but hey, even 5% is worth chasing, especially at lower revs.
Damn. I'd virtually decided to go with low mounted injectors!
I may just stay with injectors in the TBs for now and I can always drop them down later if low speed running is a problem.

That arrangement is kinda what I had in mind as well.
I was also going to try and use some type of progressive action so that intial throttle movements weren't too severe.

Yes, that's why I used cables. The pulley on the TBs are eccentric making a progressive movement.

Engines on the dyno are at wide-open throttle during the pull. So, even though the rpms are low, the throttle plate is still open and the engine is not suffering the inefficiencies of pumping turbulent air.

Because the throttle plates are open, the issues of fuel pooling/condensing on the butterfly are not encountered. If you are running high fuel injectors at low rpm on the street, the throttle will be closed and your throttle response will be poor and lumpy upon initial acceleration.

The best thing is to use both low and high injectors. You can use small injectors low that can be controlled very accurate and when you need more fuel you open the high injectors.

F*** this!
I'm gonna just stick them half way up the runners! :)
Best of both worlds :rolleyes: and keep everyone happy.

That would be cool, but it adds a extra degree of complication (and cost).
Actually, the standard Triumph injectors would probably be good for the low position and then fit some bigger bosch units into the TBs.

But then as Fleiger implies, the power numbers from PFM probably can't be applied to a street engine at less than full throttle opening.
So I guess it's back to installing injector ports in the bottom of the runners. :confused:

If it's an all out race car install them high, but I'd never do it in a car that sees any street time.
Much like a cam, race or street, there is no magic combo that does it all, only a good compromise.

Billjam,

Just to be clear the motor ran a total of 8 injectors on the 4 cylinder, four low and four high. The EFI handled the staging of the injectors, at low RPM and part throttle applications the low injectors were used, the high injectors were staged on as the RPM and throttle angle increased. I would think you would want the throttles closer to the valves for best throttle response.

If you want the numbers for your motor ran through PipeMax PM me the motor info and cam specs you plan on running.

Regards,

PFM

"I would think you would want the throttles closer to the valves for best throttle response. "

When using fuel injection, fuel delivery is controlled by a pump (MFI) or an ECU (EFI) and various engine and ambient-condition sensors. Thus, the throttle valves in a fuel injected engine only control air flow. Because of this, the placement of the throttle valves are less of an issue than with carbs.

The only issue with FI throttle valve placement is in maximizing the tuning effects of pressure waves created by the length of the intake piping. In that regard, that's one reason to use slide valve throttle assemblies, to create a smoother air flow than a circular flat plate and pivot rod dangling in the middle of an intake pipe.

OTOH, a carburetor must meter and mix both air and fuel and deliver both through the intake tract, a more challenging task than just controlling air flow.

Sherwood

Sounds interesting. What "numbers" do you need?

Bill, MS can run dual tables for two independently controlled sets of injectors. Tuning to blend the thresholds and transitions could get interesting. You could run purely based on RPM or load. I would think that a blend of the two would be best.

I am surprised there was that much difference between injector placement. Thank you for the data. It is nice to see numbers.

Also, make sure you match spray pattern of the injectors with the location. You don't want wide angle injectors that will only wet the sidewalls and not atomize into the gas stream.

I did this exercise a while back and concluded that it was a lot of work to size the runners when you are targeting a 3rd or 4th harmonic for the rpm tuning band. The difference at these harmonics is minimal. Size them for easy fabrication, fit under the decklid and air cleaners. As others have said, you will do better to match the diameter and taper than the overall length. To be honest, with just a DC21 cam I don't think you will see any effect with runner length other than changing the pitch of the sound at WOT.

btw, it would be easy to respace your throttle bodies. Just machine a spacer and extend the mounting bolt. Also, weld a tab extension on the butterfly actuator lever. Shouldn't take more than an hour to put them directly on top of the head.

Seems like a lot of work. Why not just buy these.
http://forums.pelicanparts.com/uploa...1264088297.jpg
http://forums.pelicanparts.com/uploa...1264088339.jpg

or these


http://forums.pelicanparts.com/uploa...1264088375.jpg

What about short runners like thesehttp://http://i895.photobucket.com/a...f/HPIM3362.jpg

Lets try this againhttp://http://s895.photobucket.com/albums/ac160/apl876/stuff/?action=view&current=HPIM3362.jpg&newest=1

I would try to space them out and get rid of the bends. You just need to weld on a larger tab on the right tb in the picture.
I would then mount them just high enough over the ports to make a nice transition from tb-size to port-size.

Yeah, I know that. I can do that fabrication no sweat, but I want to use a long oval filter and housing similar that used for Webers.
I know that there will be an adverse effect on flow for the bent runners, but I can live without those one or two HP.

Your pockets might be a bit deeper than mine Henry. ;)
If money was no object, there are quite a few bolt-on goodies that I would go for, but that is really no fun at all. Half the fun here is going where no man has gone before! How many Triumph-equipped 911s are you aware of?

Thanks for trying, but I can't get link to work.

Magnus, as I mentioned above in response to Jamie, I understand the issues, but I am going to stick with bent runners. I will be mounting the injectors as low as possible as I think this injector placement is quite important.

Every project has it's cost.
The cost this project I fear is brain damage.
It's impossible to accurately equate brain cells with a monetary value but take it from someone with very few brains cells left, sometimes spending a little money is therapeutic.

Triumph parts on a Porsche may be like riding a moped:
Fun for a minute but you really don't want your friends to see it. :)

Henry - it is going to be a bit tricky to install a decent air filter over those wonderful tall stacked slide valve intakes, isn't it?

I know real racers don't use no stinking air filters - just impedes air flow. Or, at most, they use a wide mesh screek to keep leaves and other biggish bits out. At least back in the days of real racing they did. You see this on vintage cars in museums, and I've even seen it at Vintage racing events.

But I can't imagine doing that. Too much dust at the track, made either by me or by others over whom I have no control. Even more of an issue on the street.

So I understand Billjam's hesitancy here, as air filtration seems pretty important for practical motors.

When, as, and if I get around to dealing with this on a motor of mine (instead of just purchasing a bent manifold), my first thought is to fabricate a different base for one of the standard filter types, and use the stock top and filter element. I think they will be long enough, even if the two end trumpets will be fairly close to the filter sides.

And I've seen photos (on this very forum, I think) of separate filters for each throat. Don't know the brand, or the specs/efficiency.

So how would you filter air for something for which the standard one filter per side available filters doesn't seem like it will bolt on?

Walt