EFI Systems

[Flieger]

What is the issue with the cable? I'm in the design stages of a midengine project and am probably going to end up with a cable shifted 914 trans and a cable throttle due to ease of routing. I'm using a Tilton floor mount pedal box and it's center seat so the Porsche linkage is completely out.

Is the cable ok for track but not rally cars or if the run is made straight enough can it be made reliable?

[Tippy]

There's a large cable for my throttle from the factory on my 87. As long as a quality piece, shouldn't be a problem.

[clutch-monkey]

Originally Posted by chris_seven

The carbon Air Box is interesting and nicely made but is it betterin air flow terms than a Rain Hat and a slightly angled runner on a manifold?

Again it doesn't quite meet the 'Retro' appearance we are trying to obtain but could be good in a Black gel coat.

i think trumpets without the manifold would pip it for peak power, but the carbon airbox apparently isn't that far behind and is simply better everywhere else in the rev range. The car was a front runner in period.

[clutch-monkey]

Originally Posted by chris_seven

I am interested in the Air Box concept and how to correlate ait flow changes to powerband shape as the modelling must be relatively complex.

airboxes of any half decent type will pretty much always increase midrange i found, it's just the really, really good ones don't lose much of your top end either
the modelling is complex, which is the reason i assume rothsport don't bother developing their own- they use the factory GT3 plenum as a model, porsche did the research, which makes sense.
the airbox above did not have GT3's around as an inspiration when it was built, but the research seems to be sound, as the results were convincing haha. My understanding is it isn't just about presenting cold air but providing a large reservoir of air post filter, helmhotz resonance, etc.


i've actually got an E36 M3, factory ITB's.. would be interesting to do a test with plenum on/off with trumpets

[robcf]

I've kicked the idea around for some time to try casting a set of throttle bodies for a 911.

The concept below consists of a tapered throttle body which can package up to a 50mm diam throttle plate.

The inlet side features at least a DCOE 45 bolt pattern (or up to 70mm between fasteners for a larger throttle plate).

Independent base idle adjustment via a screw actuating the throttle plate.

A throttle linkage (not shown) similar to the triumph units.




I fully intended that these could be used with a common plenum or individual runners.

I choose a short EV14 injector from an LS3 as they are easily (and affordably) available with all the injector characterization data.

For a direct to head design, there is a real challenge to balance the position of the injector due to the 3.2 and earlier head stud pattern. It is obvious why Porsche chose to change to the later 964/993 design.

[clutch-monkey]

Originally Posted by chris_seven

I can also understand that it is possible to tune pulses in an induction system to improve the charge capability into an engine but I find it hard to understand how this works inside a complex box installed in an engine bay.

cut open a 997 GT3 plenum, or for cost' sake, from memory the one found on an RX8, or the mazda K8 or KL series engines can be good examples

Originally Posted by chris_seven

What I don't understand is why, if airflow is restricted at peak power, the mid band will improve unless the cold air is a benefit until the box restricts flow which seems to say that the box isn't a good design.

no idea either. But have seen the same results on several different engines with ITB's. Good results with both, but power curve definitely different!
I no longer run open ITB's on a street weight car as a result.
The bonus for me was being able to use larger panel filters not socks or K&N.
i'm not an engineer though, so can only go off results on mine or friend's motors

Originally Posted by chris_seven

I guess my question is how much more performance will I gain if I replace the 'Rain Hats' we are planning to use with a 'box' and will it be worth the development cost.

Probably not worth the bother. If you had to, adapt or copy factory stuff would be your safest and most cost effective bet surely?

[chris_seven]

Originally Posted by robcf

I've kicked the idea around for some time to try casting a set of throttle bodies for a 911.

The concept below consists of a tapered throttle body which can package up to a 50mm diam throttle plate.

The inlet side features at least a DCOE 45 bolt pattern (or up to 70mm between fasteners for a larger throttle plate).

Independent base idle adjustment via a screw actuating the throttle plate.

A throttle linkage (not shown) similar to the triumph units.

Do you think that injector position is critical with a EFI style injector?

We have looked at the various ITB and their manifold arrangements and there is huge variation.

The Jenvey appears to point directly at the wall of their inlet manifold as does the Borla 3003.

If you want to replicate the injector position that Porsche uses on the early MFI and other systems then you will need 3 separate throttle body castings which is a pain.

The RSR with high butterfly throttle bodies used the injector in the air trumpet and many modern motorcycle systems use twin injectors which seem interesting but are complex and costly.

Throttle plate position also seems to vary significantly and again it is difficult to quantify the influence of this variable without a great deal of work.

When I started this thread I hoped to be able to find some numbers that could point the way to differences due to this type of change but there don't seem to be much hard data available.

I have just asked for a quote for castings and will see just how expensive it is to make a start.

The direct to head design may be OK to machine from solid or may be 3-D print which could be interesting.

[MetalDoc]

My Jenvey's are attached to a 12mm phenolic resin spacer. I was told it is to limit heat transfer from the head. The engine is not running yet so I don't know how these ITBs perform.

[al lkosmal]

Originally Posted by MetalDoc

My Jenvey's are attached to a 12mm phenolic resin spacer. I was told it is to limit heat transfer from the head. The engine is not running yet so I don't know how these ITBs perform.

i understand the use of the insulator/spacer with carbs, but do not understand the use with EFI. Can anybody explain?

regards,
al

[J0hnny]

Originally Posted by chris_seven

I can understand how 'cold air' boxes work and I think that there is some foundation for this type of device although there is a body of published work that suggests that Otto Cycle engines are only indirectly influenced by changes in ambient temperature. This is in part due to the 'closed' nature of their operating cycle.

Gas Turbines which have an open cycle are much more significantly affected and do show the order of power gains suggested by ITG when ambient temperatures reduce by 15degC.

What I don't understand is why, if airflow is restricted at peak power, the mid band will improve unless the cold air is a benefit until the box restricts flow which seems to say that the box isn't a good design.


I can also understand that it is possible to tune pulses in an induction system to improve the charge capability into an engine but I find it hard to understand how this works inside a complex box installed in an engine bay. I can see that these boxes may act as a plenum but for them to have measurable effects implies that standard systems must be very restrictive.

The 9% power increase claimed by ITG must mean a huge increase in the torque developed by the engine and I struggle to understand this dramatic change.

I have never measured the temperature in the engine bay of a 911 but with a relatively high speed ducted fan drawing large volumes of air through the grille in may not be as bad as we think.

I guess my question is how much more performance will I gain if I replace the 'Rain Hats' we are planning to use with a 'box' and will it be worth the development cost.

I would also like to know the impact on efficiency as part of our objective is to gain the throttle response of carbs whilst trying maintain the fuel economy of a K-Jetronic which may be incompatible.

I'm way more familiar with the other side of the engine as I work with exhaust. However, it is gas exchange at the end of the day; when tuned properly an airbox can help in the cylinder filling by synchronizing the pressure waves with the valve opening events, this uses the pressure pulses, usually helping the mid-range. This does not happen with the hat filters, and they also tend to disturb the flow to the intake trumpets, so they will be restrictive in the upper end.

While the temperature in the engine bay is not as bad, thanks to the fan, you also have that same fan disturbing the air, the 911 engine bay is basically a huge airbox, having a separate airbox with a dedicated intake eliminates this interaction.

That said, in powertrain development, using 1-D models to optimize torque and BSFC is cost effective and expected, but I don't think the business case is there for the aftermarket, too much work for too little benefit, specially in an engine without cam phasing, fixed lift, fixed intake length, and fixed trumpet length.

[chris_seven]

I always believed that Air Filters smooth the flow of air around the intakes and that this can and often does help in a 'busy' under bonnet environment. The question is how much more improvement is there to be gained by using a 'box'.

What I am less certain of is the extent of any pulse effect in terms of cylinder filling.

Inside one of these boxes must also be quite busy as the pulses and pressure waves will potentially interact and cause problems but I imagine that these problems can be minimised with careful design.

There are many good SAE papers on this subject which seem to suggest that Air Box resonance occurs in a very narrow frequency window and has a small impact on pressure - around 10 millibars is a typical quoted value for the air boxes used on F1 cars which I would expect to be quite good.

The conclusions that seem to come out of the published work is that airboxes only provide horsepower benefits in narrow rpm bands depending on their specific design but that they are very useful in filling in 'holes' in a powerband that can arise due to poor general airflow.

With the ITB design we are planning to use the only practical measurements we will be able to make will be 'open' intakes with simple trumpets, rain hats with ITG foam elements to replace the K&N filter and a stock 'Weber' Airbox.

I hope it will be an interesting comparison.

[robcf]

The interesting thing about the Jenvey design is that they have a product which is very universal, it can be applied across various engines with success. Alone, however, it is not optimized.

The injector position is a compromise with respect to packaging and injector angle relative to gas flow. The idea is to maximise mixing of fuel with the air intake charge and minimise wall wetting. I choose to also minimise the distance from the injector to the intake port and place it after the throttle plate to increase the system response.

In my experience packaging dictated a number of design attributes, these are not necessarily optimized but a balance of engineering requirements.

I also prefer a design which implements a common airbox and resonator plenum.

BTW do you know what the nominal distance is between the intake ports (cross car)?

[chris_seven]

Originally Posted by robcf

BTW do you know what the nominal distance is between the intake ports (cross car)?

485mm between port centres.

[Tom Elder]

Hi, Chris

Great discussion here and have been following along. I have an interest in this subject but no practical experience, that would add anything to the discussion. To teach myself Solidworks, I did an air cleaner assembly as little design project. More for style than performance but it did lead to a little bit more in depth look at the possible theoretical performance improvement.

Pendulum Outlaw, Build, Adventures and Misadventures Link to the air cleaner page

During the summer I picked up a SuperFlow SF600 flow bench. I am deep into house renovations, right now, but I will throw this offer out there, if anybody wants to design and build the test fixtures, and gather the parts for a test, I would be happy to put them on the flow bench and test them. I have cleaned the bench up as it was in storage for a number of years. Computer seems to work but I have not run a calibration test yet.

Low priority for me and as a first off test learning project, I plan to have a 3d printed bell mouth made to compare to the exiting velocity stack.

[robcf]

Originally Posted by chris_seven

485mm between port centres.

thank you!!!

[J0hnny]

Originally Posted by chris_seven

I always believed that Air Filters smooth the flow of air around the intakes and that this can and often does help in a 'busy' under bonnet environment. The question is how much more improvement is there to be gained by using a 'box'.

What I am less certain of is the extent of any pulse effect in terms of cylinder filling.

Inside one of these boxes must also be quite busy as the pulses and pressure waves will potentially interact and cause problems but I imagine that these problems can be minimised with careful design.

There are many good SAE papers on this subject which seem to suggest that Air Box resonance occurs in a very narrow frequency window and has a small impact on pressure - around 10 millibars is a typical quoted value for the air boxes used on F1 cars which I would expect to be quite good.

The conclusions that seem to come out of the published work is that airboxes only provide horsepower benefits in narrow rpm bands depending on their specific design but that they are very useful in filling in 'holes' in a powerband that can arise due to poor general airflow.

With the ITB design we are planning to use the only practical measurements we will be able to make will be 'open' intakes with simple trumpets, rain hats with ITG foam elements to replace the K&N filter and a stock 'Weber' Airbox.

I hope it will be an interesting comparison.

It is very narrow, and very engine specific, the flexibility has been greatly benefited with adjustable intakes, but is not an easy thing to do with an old engine.

However, it would be interesting to measure the effect of using a varioram intake setup in the same engine.

[Raceboy]

I am in the process of replacing the throttle cable on 911 altogether in favour of E46 M3 elctronic throttle actuator and will test if two actuators work on the same output to get rid of linkage too.
Synchronizing between banks can be done with adjusting the actuator arm, no problems there.

VEMS ECU that I use can easily control the electronic throttle..

[Raceboy]

It operates exactly as OEM drive-by-wire system with dual pedal sensors (PPS1/PPS2) and dual throttle sensors, one for throttles (TPS1) and one for the motor position (TPS2).

I can create multiple profiles (progressive response etc) to improve driveability in traffic etc. Used the same system on E46 M3 with electronic throttles+double-VANOS.

In fact there are nothing too bad making DBW, just source some ETC pedal that suits best to 911 (almost any of them works, 99% of them have PPS1/PPS2) and attach throttle actuator motor. I will just use the dbw control on regular ITB's.

[billewizz]

Originally Posted by chris_seven

485mm between port centres.

Chris another outfit in the uk supplying bodies is A T power injector angle is pretty much the same as factory and they have a shaftless design to the butterfly.

[robcf]

Hows this project going Chris?