sizing throttle body
Turbo Mini Forums

how do you go about sizing a throttle body? there has to be a balance between airflow and throttle control so biggest cant be best?

Well, I believe other factors like if there is a compund lever on the TB linkage should be considered as well. I have a BMW (i think) TB at 65mm that has a brilliant compounding ratio - the last 25-35% or so of opening takes just 3mm of pull on the cable. The first 25% of opening takes something like 4 times this amount.

This 70mm size would be considered 'oversize' on a 1300 turbo with a standard linkage, but I feel it'll work really well.

One good question - how big is your intercooler pipe? If it's 50mm ID, an 80mm TB could be overkill LOL

lol,

ditto what dave says about the linkage,

On the 7 port i have a 56-57mm chavalier tb, to get it in i had to lose all the progressive linkage,

the throttle is a bit abrupt, town driving is all done under about 12% of throttle opening and normal crusing under about 30%

for the scooby lump i'd say go for somthing about 65mm

The one I got for the Mini project (at 1360cc) is a 54mm Rover one.

I would have liked the compounding linkage but it was a case of "take what you can get that will fit". This one has the next best thing, a varying radius cable pull - hard to describe but it means the initial cable pull is at a very large radius from the spindle but gets closer and closer as the spindle rotates.

Schrödinger's cat - so which one am I ???

Im currently building an engine using 2 single Jenvey TBs. I sized mine using flow area.
I did:

(pi x Inlet Valve Diameter x valve lift) x 2 (2 inlets but 1 port) x 0.6 for flow losses = ans

sqrRt (ans / pi) x 2 = Rough diameter for Butterfly

Alex

Are you using the Jenvey TBs for a turbo setup? Which ECU are you planning to use? And are those the single or dual injector TBs?

Jean

http://www.jbperf.com/

I hope to use it for turbo eventually, but as its a uni project and i ahvnt got much time its going to be NA for now. Ill be using a VEMS unit. Im planning to have one injector per runner, mounted on the runners themselves. So only two.

Alex

You are aware that the VEMS doesn't have an option to do port injection on the Mini engine, right? You may be able to make the engine run with it but it will be a compromise at best.

There's a reason (actually many of them) why there's been so much discussion and work on developing a Megasquirt version to handle the specifics of the siamese-port engine.

Jean

http://www.jbperf.com/

oh yeh of course lol, ive done plenty of reading on the subject for a while now. If i was going to port injection then i wouldnt be using VEMS. I havnt just thought im going to inject the 5 port and gone ahead thinking its gna work with the siamese ports.
The point of the project is to prove that an EFI system is much more efficient than a carbed system across engine range, and the reason for using 2 throttle bodies is to try and eliminate any irregular pulse waves in the inlet manifold.

Alex

Jean's point is that as you have decribed it above "two throttle bodies" and "one injector per runner", that is essentially port injection.....

The fact you have two throttle bodies (most of us experimenting with this are using a single throttle body and a plenum/airbox before the runners) doesn't alter the fact that you say you are injecting into the runners..... individual runners = ports.
EDIT - unless you are intending to make the runners really, really long and put the injectors as far away from the head as possible, when it might start to behave like a wet manifold.....

Rod.

Edited by Rod S on 21st Feb, 2009.

Schrödinger's cat - so which one am I ???

hmmm i see your point now.
My thinking went along the lines of, if i could make inject the fuel so it was sufficiently mixed in with the air before it enters the port then it should have a similar effect to using a carb, where teh mixture is prepared before hand.
I am using a custom manifold, where i am placing the injectors as far up the runner as possible facing into the airstream, and angle down at the bottom of the runner where it will bounce off of the bottom, similar to as if we were injecting into the back of the inlet valve.
Will this not act as a wet manifold?

Alex

The problem is that the injector fires over a percentage of the engine cycle. At low load this may be less than 25%, or 180 degrees of the cycle.

With the siamese inlet port, unless the injection point is precisely timed, then all the fuel will go into the inner cylinders and nothing in the outers.

I know because I've tried.

VEMS is good for normal injection but it cannot handle the requirements of a siamese inlet port.

Saul Bellow - "A great deal of intelligence can be invested in ignorance when the need for illusion is deep."
Stephen Hawking - "The greatest enemy of knowledge is not ignorance, it is the illusion of knowledge."

Don't forget that even if you try to make it more akin to wet manifold than port injection by putting the injectors as far away as possible and pointing in strange directions, the siamese "problem" exists with carb(s) or single point injection anyway.

It's inherent in the design (or rather the valve cycle) of a siamese port that the inners will get a richer mixture than the outers.

That's why Leyland/Rover changed from SPI (an electronic carb with wet manifold) to MPI with the injectors really close to the ports and some very clever injector timing to try and compensate for the inherant mixture differences.

Jean's new code for MS is intended to better this.

Your description now sounds like twin electronic carbs !!! Don't take that the wrong way - it's an interesting twist and you may do better than twin carbs, but will not overcome the fundamental problem.

Or prove us wrong by setting up twin widebands like this



Rod

Schrödinger's cat - so which one am I ???

Ok, i understand what your saying, and im not taking it the wrong way lol, i understand that u were probly just warning me incase i thought that i could simply port inject and map as normal without taking into account charge robbing.

My plan is to use the TBs as electronic carbs and do expect to have charge robbing occur. I have mentioned in my writeups for my project that the only way to try and eradicate port robbing is to use perfectly time sequential injection, build a new crank to change the firing order or to actually physically separate the ports.

Alex

That's good.....

If you get all that into your write-up and have a lecturer who understands it, you should do well.

Obviously I wouldn't go for the new crank..... the seperation of the ports is well accepted with the hybrid 7 port head, a few "A" series 8 port (expensive) heads and all those (many on this forum) who are using 8 port bike heads instead (some even with 16v).

However, perfectly timed sequential injection on the standard 5 port head is the way forward for a few of us....

If your project funds will run to it though, I would seriously suggest the twin widebands (as per my photo) to see just what you can achieve with twin "electronic carbs".

Rod.

Schrödinger's cat - so which one am I ???