quad TB's AND turbo

gse_turbo · Post by **gse_turbo** » Wed Oct 24, 2007 11:08 pm

ok i've spent about three hours on this simulator trying to get an even flow reading to all cylinders. what i came up with defies most logic and trust me i tried everything.

the magnus style did the opposite that i was thinking. with such high velocity all of the air fed the last cylinder the most.

when i finally got a really good looking pattern i lost the screen because i tried emailing the graph to myself. if needed i'll start again tomorrow and try to post it again.

keep in mind that this simulator is not 100% accurate but it does give some good insight. i'd like to see what some of the other members can come up with.

garrett

Audacity Racing · Post by **Audacity Racing** » Wed Oct 24, 2007 11:12 pm

If you want to draw up a couple things in CAD, I'll throw them in Cosmos or Fluent and see what they do (you can just do outside dimensions and a good description and I can take it from there)

gse_turbo · Post by **gse_turbo** » Wed Oct 24, 2007 11:27 pm

awesome! do you meen for the flange or the manifold idea?

garrett

Audacity Racing · Post by **Audacity Racing** » Wed Oct 24, 2007 11:27 pm

Either or...
I usually run iterations at night when I'm asleep so it takes a few days, but nothing too serious

Post by **PrecisionBoost** » Thu Oct 25, 2007 3:09 pm

Audacity Racing wrote:I guess that's my point in a nutshell... why do individual throttles if it kills your plenum volume? You'd be far better off with properly tuned runners with bellmouths inside the plenum. The throttle response argument is easily over-turned by the shift-point power argument. Having no plenum volume means when the BOV opens there is no reserve air except what is in the small amount between the throttle and the valve... meaning you'd get insane bursts of power when the air comes flooding back by... in all honesty, not driveable for the first few gears (plus who knows what you'd break in the process)

Something like a Magnus Motorsports style manifold with a large bore throttle (90mm from a Q45?) would deliver better power delivery... at least a more controllable power delivery. This is why you see such massive plenums on high performance turbo cars (i.e. Audi R8-R10 LeMans cars or the Bentley Speed 8 EXP)

Those intakes are yummy...... very nice.

I don't know if I made my point exactly..... the throttle response of your engine is allready going to be awesome (based on what I know about your build up) so I was trying to say that quad TB are probably not the best way to go.....and that they would in fact hurt your driveability in the first gear for sure.... and yes.... something would have to be done to counteract the shift-point pressure issues.

The small plenum reserve issue would definitly create some issues with low/mid RPM power levels..... it would defnintly give you some gains at 6000 RPM.... but you'd probably end up loosing power in the low/mid powerband.

Of course you could allways do something fun and make a variable geometry intake system.

As far as the shift-point power.... if your designing your own intake you could do all kinds of crazy stuff to increase plenum reserve

As an example.... what if you installed a reserve tank between the throttle body and valvetrain.

You could feed the tank from just after the TB and then create a low angle merge with the primary just before the cylinder head.

Yes.... I know that sounds crazy.... but some of the most unconventional designs end up with the best power gains.

I guess the intake design depends on what you want from your engine....

Do you want ultra quick throttle response?
Do you want maximum power at higher RPM?
Do you want lots of low and midrange power?
Do you want something very unique that is a one of a kind?

You can't have all of them..... you have to decide on a scale of 1-10 as to which ones are the most important.

Example....

TB response 5
Max Power 2
Low mid 9
Unique 8

If this was what you want..... then quad TB are not the way to go.

Quad TB would be more likely if your repsonse was

TB response 9
Max power 9
low mid 2
unique 4

Of course with this kind of design your car is probably going to drive like a honda..... nothing there until you hit 5K RPM

I personally think that the ultimate cool thing to do would be to adapt a nice variable geometery intake from another car.

Some cars are not just a two way geometery..... there are some that have four choices of internal geometry with transitional valves... so they can use more than one geometry at a time while in transition.

I don't know off hand of anyone that uses a true stepless variable geometery system but I wouldn't doubt that BMW or somebody like that is using one.

I allways thought it would be cool to have overlapping runner tubes on a stepper motor and you could pull the intake towards the head as you changed RPM

Of course this creates other issues with things like the throttle cable... but it's possible to overcome issues if your creative enough (say drive by wire TB for example )

You could even change the geometry of the plenum by adding a system that moves a wall internally (by motor or by air pressure)

At one point I was really interested in the idea of creating a dual throttle body system.... one small TB and one large.... to help control the input velocity of the intake air and shift air within the system during shifts.... but that's another long story.

My point is that there are a billion different designs.... some crazy complex ones.... some simple ones.... what you have to figure out is what your end goal is before you start.

gse_turbo · Post by **gse_turbo** » Thu Oct 25, 2007 4:28 pm

good point,

i guess my biggest interest is that all four cylinders get the same volume and velocity at the same time once the throttle is open.

my next concern would be that when the throttle is closed that air is cut off equally from all four at the same time.

finally i would say that i would want a manifold that didn't create turbulence or scavenging between closed and wide open throttle.

all and all just a really efficient manifold, power and response will come from efficiency.

garrett

Post by **PrecisionBoost** » Thu Oct 25, 2007 6:19 pm

I was thinking that a quad throttle body design with a large intake runner would be interesting if a combination of a hot wire mass air flow sensors and drive by wire TB's were used.

Basicly you could create a feeback loop to ensure that the airflow is exactly identical under all conditions.

The hot wire MAF sensors don't cause a pile of turbulance so they should work well in a situation like this.

What happens is that you send the quad drive by wire throttle bodies a target throttle position, and as they move to this new position they use the MAF to make small changes to the position so that each intake runner has the exact same flow levels.

It's not all that different than fuel control systems..... you have the Long term fuel trim to decide where the long term movement of the duty cycle is going to be...... then you have the short term trim to try and keep the A/F at the stoich point (14.7)

In this case the TPS/TB would be like the LTFT and the feedback loop would act like the STFT.

One of these days I'm going to have to make myself a flow bench..... basicly I just need a number of MAF and MAP sensors, and a decent air pressure source.

I'd really like to fool around with a system that can vary the runner tube diameter, runner tube length and geometry of the intake plenum.

I also want to do some work on a variable geometry exhaust system that uses eddy currents to get rid of reversion into the other pipes..... but I need to set up the flow bench so that it can create individul pressure pulses into each exhaust runner.

Audacity Racing · Post by **Audacity Racing** » Thu Oct 25, 2007 6:51 pm

I was thinking about this while meeting a client today...

We're so concerned with pulse separation and such in the exhaust stream, why not in the intake tract? Why don't you see a single throttle plate with then sweeping runners of equal distance but separated... like a turbo manifold in reverse...

This way you maintain smooth flow, have large enough volume to accommodate plenum pressure change, and have a single inlet point... I think it's not only do-able but worth investigating.

Also, Chris, BMW does have a stepless design... first in the industry to my knowledge. It's found in the 745i along with Valvetronic and Vanos technologies...

gse_turbo · Post by **gse_turbo** » Thu Oct 25, 2007 6:54 pm

so then with this system would you need individual TB's with separate drive by wire system? i was thinking one cable throttle linkage connecting all four of the throttle plates mechanically.

with some of the ideas put out here we'd have to make a custom quad TB plate. most of the parts are available like throttle plates in different sizes, we could use the TPS and what not from the factory unit but what provisions would have to built into the throttle plate (i.e. passage for air flow metering and so on)? could idle air be accurately controlled by the throttle being cracked on slightly?

garrett

Audacity Racing · Post by **Audacity Racing** » Thu Oct 25, 2007 6:57 pm

Here is a similar idea found on an FSAE car... I think it could be optimized significantly more and made to have cleaner air splitting than that one (which is of course NA)

It would be EASY to make that multi-plane with length added into the center restrictor

Audacity Racing · Post by **Audacity Racing** » Thu Oct 25, 2007 6:59 pm

Chris's idea is precisely the point I was trying to make initially. You'd need this massive array of sensors and throttle control valves to able to tune the system, but in reality a tuned intake manifold will deliver air to all cylinders equally (see my previous two posts).

The only upside to the electro-mechanical system I can honestly see is that you're no longer limited by engine bay constraints or having to do math for shaping of the plenum...

gse_turbo · Post by **gse_turbo** » Thu Oct 25, 2007 7:29 pm

these were done on the wind tunal simulator that was posted on the first page. it's not exactly a CAD program so the measurments are only as good as my eye. these by no means are all the manifold styles i 'tested' nor are these as accurate as a flow bench but good enough to get ideas out.

garrett

Audacity Racing · Post by **Audacity Racing** » Thu Oct 25, 2007 7:32 pm

I'm glad I could pass that along...

Audacity Racing · Post by **Audacity Racing** » Thu Oct 25, 2007 7:33 pm

BTW, I'm sure you noticed this too... placing the bellmouths above the floor of the manifold makes no difference as to the flow... It's easier and cheaper to weld them flt into the floor, haha

gse_turbo · Post by **gse_turbo** » Thu Oct 25, 2007 8:03 pm

thanks for the tool. it's a good substitute for not having a flow bench.

you're very right, the same could be said for several of the manifold styles that are supposed to add big HP. that's why yesterday i said some of the tests were a bit counter intuitive or didn't work as some manufacturers claim.

garrett