what size throttle body for adequate air flow?

KZjim · Post by **KZjim** » Tue Jul 05, 2005 5:17 pm

Is there a formula for figuring out throttle body size/air flow?... i'm planning a 1356cc in-line 4 cylinder Kawi.... with a gpz1100 16 valve head....

Jack · Post by **Jack** » Tue Jul 05, 2005 7:12 pm

Search for this one. There is a ton of info on the site.

Jack

KZjim · Post by **KZjim** » Wed Jul 06, 2005 4:46 pm

I have searched quite a few threads and havent found what im looking for.. i have done the calculation from the manual # of TB's/size of TB=hp, which is great.. The info i really want is how much air i need to move to support a given engine displacement. My engine will displace 1356cc's will the stock GPz 41mm TB's be enough?

KZjim · Post by **KZjim** » Thu Jul 07, 2005 8:49 am

Lance,
Thanks for the great info, thats what I was looking for, I'll start with the 41mm units and go from there....

Bow · Post by **Bow** » Fri Sep 30, 2005 6:38 am

lance wrote: However, as a basic esimate, you might consider the engine's displacement as 100% filled at it's peak rpm, and use that as a 'safe' estimate. So in your case, 1356cc is about 83 cubic inches, and suppose it revs to 10,000 rpm. That means you would be moving:

83 cid * 10,000/2 = 415,000 cubic inches of air per minute (at 100% VE).

1 cubic foot is 12Ã‚Â³ cubic inches = 1728 cubic inches/cubic foot

so in the above example you need about 240 cubic feet per minute.

OK.. I did this far and I'm stuck...

Engine: 2389 cc [145.47 ci]
Max RPM: 7200
Cubic Inches of air per minute (at 100% VE): 525,734.59
Engine CFM: 303.67

Throttle Bodies: 42mm each [1.65 in]
Throttle Body Area: 2.15 sq.in. [each]
Total Throttle Body Area: 8.59 sq.in. [2.15 * 4]

While a lot of aftermarket throttle bodies are rated in cfm, you are probably looking for a diameter. This invloves some guesswork. As a comparison, my TBI throttle body, with four 1.750" bores, was claimed to flow 900 cfm. This means that 9.62 square inches of throttle area is supporting 900 cfm. So we should be able to support 240 cfm with 240/900*9.62 = 2.57 square inches.

I got to:

303.67/900*9.62= 3.25 sq.in.

I am stuck on how to get this:

This would suggest that four throttles of 0.90" (23mm) in diameter *should* work.

Can anyone help me out?

Thanks,

Bow

ransom · Post by **ransom** » Fri Sep 30, 2005 11:52 am

If you're running four throttles and need 3.25 sq. in. then each one
needs to be 3.25/4 or ~0.81 sq. in.

area is pi times the radius squared, so 0.81 = 3.14 * r^2.

Then r^2 = 0.81/3.14, which is about 0.26 (rounding up), and r = sq
root of 0.26 which is just over 0.5". So the radius is 0.5" and the
diameter is one inch, or 25.4mm.

Of course, this is before Lance's 20% calculation for using ITBs
instead of a single TB, but this way you can run through the calcs
yourself, which is satisfying when it works out. At least, I think so.
But I'm a big geek

Hope this helps,
Jesse Ransom

Posted by email.

whittlebeast · Post by **whittlebeast** » Fri Sep 30, 2005 12:18 pm

Are you thinking that you need/want 4 throttle blades or one single throttle blade. If you are going to 4 ITB's the I would lean in the direction of the factory motorcycle setups. About 42mm. If you are going for one big hole I would look at the Honda Accord piece. Remember that small angle changes at first tip-in will make big changes in torque output. Going to a ever changing throttle arm radious cable linkage will help drivability. Plan on using MAPdot AE tuning on any setup of this type. Also plan on RPM based AE, as traditional AE will just not work.

Hope this helps

AW

Bow · Post by **Bow** » Sun Oct 02, 2005 7:30 am

Thanks guys..

Yes Jesse, I do get a big, goofy grin when I figure it out on paper... ;)

AW,

Thanks you also.

Yes, I am planning on running 4 Throttle Bodies, the Engine I am building it on already has 1 big TB... how boring... ;)

I picked up a set of 42mm 2002 GSX-R 750 throttle bodies and I wanted to make sure that they would be big enough to handle the task at hand.

Thanks again gents!

Bow · Post by **Bow** » Mon Oct 03, 2005 6:23 am

Hey all...

I was playing with Excel, trying to put all of this knowledge into an easy to use form...

I am pretty sure I messed it up somewhere in the 20% rule...

can anyone check it over?

chodjinn · Post by **chodjinn** » Mon Oct 03, 2005 9:17 am

Hi,

I've just done the calculation for my engine. I'm running a 3.5litre V8 (215 cu) and used peak rpm as 6,000. I plan on using a single throttle body when I move over to Injection. According to the calculation I need a single throttle body with 20sq.in, which results in a 5" throttle body!

I was planning on using either a single 72mm or 90mm?

Anybody provide any advice?

Post by **efahl** » Mon Oct 03, 2005 10:43 am

Hmm, I get a much smaller number than that.

215/(12**3) * 6000/2 = 373.3 CFM

Using 300 fpm as the maximum velocity, and using an 8 mm diameter throttle shaft, I calculate the following:

Flow 373.3 cfm
Velocity 300.0 fps
TB size of 3.667 in^2 = 2.161 in dia = 54.88 mm

What sort of flow velocity are you guys modelling here? Even with 200 fps, I get much smaller TB size:

Velocity 200.0 fps
TB size of 5.297 in^2 = 2.597 in dia = 65.96 mm

(I really should build an on-line calculator, as my algorithm uses a Newtonian solver to converge on bore diameter as a function of flow and velocity, which requires subtracting out the area of the throttle shaft at each diameter and then re-solving. I'd like to get boundary layer effects in, too, probably just a simple fixed-thickness model.)

Eric

jsmcortina · Post by **jsmcortina** » Mon Oct 03, 2005 10:59 am

I'm using a pair of 50mm throttles on my 3.5 V8 (because that's what I had from two Sierra pinto EFI manifolds) and while it works well, they are actually too big. That equates to the same area as one 70mm.

James

Post by **efahl** » Mon Oct 03, 2005 10:59 am

Here we go, same calcs with a fixed 1 mm boundary layer, assuming mean boundary velocity is 0.5*mean velocity through the TB...

Total Flow = 373.3 cfm
Mean Velocity = 200.0 fps
Number of bores = 1
TB area per bore = 5.580 in^2
Diameter = 2.666 in = 67.70 mm

Now go here and play with various settings that net you various velocities (velocity matter is all that matters, flow and diameter just get you there), specifically look at the last two numbers for pressure drop (see if Corky Bell is correct with his suggestion of 300 fps max):

http://www.not2fast.com/gasflow/velocity.shtml

chodjinn · Post by **chodjinn** » Tue Oct 04, 2005 12:21 am

I was going on power as a factor as well (before I found this thread). I've seen a 300bhp SD1 (same engine as mine) running twin 65mm group A throttle bodies. I used a much simpler formjla by just comparing x-sectional area:

65mm = 3.25cm Radius = 33.24cm2

x2 = 66.48cm2 for 2 65mm throttle bodies

A single 90mm gives:

4.5 x 4.5 x 3.147 = 63.73cm2

Therefore I worked out the 90mm was almost equivalent to 2x 65mm, and as I want to run a similar power level surely that should be sufficient?

Tommy · Post by **Tommy** » Tue Oct 04, 2005 3:39 am

I see that in the calcs CFM is used. How would one calc this for a supercharged engine, would the Pressure Ratio affect the # used for the calcs? Or should the uncompressed CFM be use?

Tommy

chodjinn · Post by **chodjinn** » Tue Oct 04, 2005 4:56 am

^^^ Same question here actually, I've got two turbos bolted to mine. Do you have to consider boost? I would imagine you would, as forced induction increases the air flow. If so a large single throttle body like I mentioned would probably suffice (90mm)?

cheers

PSIG · Post by **PSIG** » Wed Oct 05, 2005 1:25 pm

efahl wrote: . . . Now go here and play with various settings that net you various velocities (velocity matter is all that matters, flow and diameter just get you there), specifically look at the last two numbers for pressure drop (see if Corky Bell is correct with his suggestion of 300 fps max):

http://www.not2fast.com/gasflow/velocity.shtml

Cool calculator Eric. Is there a 'rule of thumb' range we could then derive to set reasonable operational boundaries between tip-in troubles and performance losses? Perhaps back-calc from driving examples through the calculator? It would be great to have an appended line on the injector calculator that suggests the TB size range appropriate to a vehicle with the entered values as a sort-of educated-guess package deal.

David

Post by **efahl** » Wed Oct 05, 2005 6:28 pm

David,

I was thinking along the same lines with regard to the injector sizing thing... I looked at my "turbo" calculator and thought about adding TB sizing to it, but then decided to work out a better scheme for doing the TB estimation first before I integrate it into anything else. I've got a friend who is an aerodynamicist and he's thinking about how to model a nice axisymmetric TB including a wide-open blade obstruction. I had not thought about the tip-in phase of flow, I'll mention that to him and see what if we can get a simple model that does something useful.

Eric

whittlebeast · Post by **whittlebeast** » Wed Oct 05, 2005 6:58 pm

I like to keep the intake air velosity below about 90 ft/sec or about 60 MPH. My simple minded logic is if you ever stick your hand out the window at 60 MPH that is about the time the loads start going up. Remember that drag is a squared function.

For fun I ran a few napkin calcs on a typical 1200 cc motorcycle pulling 13000 RPM with air going thru 4-42mm intakes and got about 50mph airflow in the throttles. (1200 GSXR stock)

Next I ran a 302 CI motor running 5000 RPM thru a single 80mm MAF and got 90 MPH (Mustang with a 80 mm MAF)

Last I ran that same motor pulling air thru a duel bore 62mm throttle body and got 75MPH (Mustang with a 62mm duel throttle body)

A little food for thought

AW

Post by **efahl** » Wed Oct 05, 2005 7:07 pm

whittlebeast wrote:For fun I ran a few napkin calcs and got about 45 MPH airflow in a typical 1200 cc motorcycle pulling 13000 RPM with air going thru 4-42mm intakes and got about 50mph airflow in the throttles. (1200 GSXR stock)

Andy,

I get about 110 fps for that, assuming 7 mm throttle shafts and ignoring boundary layer flow... I think there is a lot more going on wrt turbulent flow in the tb than first meets the eye, especially considering how big the Reynolds number is in a typical TB at WOT.

Eric

MSefi.com read only forum (Jan.15,2018)