Intake manifold design question

[litch]

Recently I've been thinking of a way to move the intake manifold so I can make room for a large turbo. My father suggested having the intake manifold make a wide 180 degree turn, and go over the top of the valve cover to the other side of the engine.

 

I measured it out, and it seems like there would be clearance, but I'm not too sure on the specifications of things. The runners would end up being 26" inches long or so, not to mention making a large bend, but considering there are oem manifolds that are similar, I can't imagine it would be absolutely terrible.

 

Can somebody explain to me what sort of effect a manifold like this would have, and how a bend in a runner would affect resonant waves? I think I have a basic grasp, but I'm not an engineer by any means.

[johnc]

Intake runner length (from a plenum) is highley dependent on valve timing.  A longer runner tends to damp out harmonics so we can assume a 26" runner (at least 3.5 times longer then ideal) will have very little harmonic benefit.  Its just a pressurized pipe.  There will be an increase in lag and the engine will be less responsive to throttle even on boost.  There could be a lot of surging.

[litch]

Oh, ok. I assumed that a longer runner would produce stronger waves, at a lower rpm. I can't remember where it was calculated, but something I read said it would be a first order harmonic (I can't remember what it's actually called) at around 4500 rpm.

 

So, essentially there would be no benefit in such a thing, besides making room. I would assume that using itb's would eliminate the lag problem then?

[Leon]

Longer runners drop the resonant point lower in the rpm band.

 

With 40degC IAT and 248deg valve duration, a 26" runner will hit the 1st harmonic around 11000RPM, nowhere near 4500RPM! However, the 2nd harmonic will be around 5500RPM and a 3rd harmonic close to 3750RPM.

 

I've written some detailed (probably TOO detailed) responses about intake runner tuning, which can be found in a search: http://forums.hybridz.org/topic/105783-ideal-intake-runner-taper/

[skirkland1980]

Ok so resonant tuning is sound waves bouncing toward the intake valve then...where?...and back to the intake valve. What does the sound wave bounce off of to bounce back to the valve? Seems like only certain intake designs will work for this type of tuning.

[Six_Shooter]

The resonance happens between the valve and the open plenum. The volume of air in the plenum, whether it is a defined space like many intakes or infinitely large, like ITBs with no filters, the resonance acts the same. 

 

Generally speaking, long runners promote bottom end torque, while shorter runners promote high end power. Engines with shorter runners generally tend to rev higher, or rev easier, not solely due to the intake runner though, it just helps with the rest of the components to promote that ability.

 

This is why many engines over the years have used two runner lengths that switch at certain RPM and/or plenum vacuum point.

 

You should take a look at what some drag racers of the '50's and '60's were doing. Some runner lengths were measured in feet, not inches. lol

Edited February 20, 2013 by Six_Shooter

[R3VO 3VOM]

Intake runner length (from a plenum) is highley dependent on valve timing.  A longer runner tends to damp out harmonics so we can assume a 26" runner (at least 3.5 times longer then ideal) will have very little harmonic benefit.  Its just a pressurized pipe.  There will be an increase in lag and the engine will be less responsive to throttle even on boost.  There could be a lot of surging.

Could this lag (not due to the turbo lag) be solved with ITB's since we're talking about a totally custom intake? They wouldn't be cheap, but they would reduce the intake runner length tremendously. This lag is caused by the elongated distance between the TB and the valves, so when you step on the gas and open the throttle, it will take that air a while to travel the 26" to the head. With the ITB's in place, this length for the air to travel would be reduced.

 

Referring back to the original post, why can you just run an angled spacer? Just thinking aloud here; but one like this off Ebay:

http://www.ebay.com/itm/T4-TO-T3-TURBO-FLANGE-ADAPTER-W-ANGLE-CA18DET-KA24DET-180SX-RB25DETT-R32-/200833291856?pt=Motors_Car_Truck_Parts_Accessories&hash=item2ec298d650&vxp=mtr

Granted, it is a T3 to T4, but you could cut off the T4 flange and put on a T3 flange or whatever flange turbo you are running.

 

But I'm not sure the clearances with the framerail and steering and such would be. 

[Six_Shooter]

The resonance doesn't happen between the valve and the throttle plate, but the valve and the end of the runner.

 

Think of it this way, when an ITB is wide open, the length of the runner is from the end of the inlet "trumpet" to the valve, the throttle plate itself is not an influence here.

 

Placing the throttle plates as close as possible to the valves can help with throttle transitions is why people will use ITBs, but once they go wide open, they act just the same as an intake system that uses a single TB and plenum (provided that the plenum is sufficiently large enough to feed the engine with an appropriate amount of air).

[Leon]

Ok so resonant tuning is sound waves bouncing toward the intake valve then...where?...and back to the intake valve. What does the sound wave bounce off of to bounce back to the valve? Seems like only certain intake designs will work for this type of tuning.

 

Any and every 2- and 4-stroke intake design works like this. Valve opens, air starts moving, wave propogates away from valve, wave hits end of runner, opposite wave comes back and hits the intake valve just before it closes (at resonance).

Edited February 20, 2013 by Leon

[litch]

I suppose my calculations were off, not surprising. I'm not looking for ultra efficiency, just the best I can get with easily accessible tubing and parts. My plans may Change (I'm 19), so that may be exactly what I want. Is there a way you can give me a number comparison between 1st, 2nd, and 3rd wave?

 

Leon, what you wrote was good stuff, easy to understand. I'm just having trouble applying it to my goals, I suppose.

 

R3VO, the exhaust manifold is going to be custom, and I'd like to make it top mount so I can easily access the turbo.

Edited February 21, 2013 by litch

[Leon]

I suppose my calculations were off, not surprising. I'm not looking for ultra efficiency, just the best I can get with easily accessible tubing and parts. My plans may Change (I'm 19), so that may be exactly what I want. Is there a way you can give me a number comparison between 1st, 2nd, and 3rd wave?

 

Leon, what you wrote was good stuff, easy to understand. I'm just having trouble applying it to my goals, I suppose.

 

R3VO, the exhaust manifold is going to be custom, and I'd like to make it top mount so I can easily access the turbo.

 

The first harmonic is strongest with the rest following suit. Hitting the 1st harmonic is ideal as far as resonance tuning goes, but it isn't practical (think 5 foot long runners). Shoot for the lowest harmonic that is practically achievable for you. I'm not sure what you mean by "a number comparison between 1st, 2nd, and 3rd wave"?

 

To answer the original question, bends don't affect wave propogation, at least not meaningfully. Bends do increase pumping losses, but a "dry" runner will experience less losses (lower density) and by the sound of it you're using a dry runner (port injection).

[johnc]

And the length of the pipe reduces the effect of the wave through frictional losses and the inherent damping mass of the volume of air.  To take it to the extreme, Edward Teller calcualted that a hydrogen bomb greater then 100MT would not yield more then 100MT due to the damping effect of the atmosphere.  Plus. adding 26" to the length of the pressureized column of air from the comrpessor isn't a postive thing.

 

You can always lay the engine over on the passenger side like Electramotive did to get more room for stuff.

[Xnke]

From looking at a LOT of photos of race intake designs, including Bob Sharp and Electramotive, and running some math a lot of times, it looks like a 15.5" long runner is a reasonable "local maxima" for torque production..which should happen about 5300RPM. Look at runner cross-sectional area next, and size appropriately. Again, drawing cues from photos, and even more time with the pencil, a runner Cross-Sectional-Area of 1.76" is also sized for 5300RPM. Hard for me to say what the area of the Bob Sharp and Electramotive cars was, as the runners are tapered. The runner opening as it entered the plenum on the BSR car was 65mm, though...tapering down to a 40mm intake port. The taper wasn't constant though, IIRC.

 

If you're not moving the engine over, and you're not snaking the runners up over the valve cover, 15.5" or so is about as close to the shock tower as you'd want to get. You can go longer; but not by much!

 

Remember to add 4" to the length of the manifold runners, to count for the port length inside the cylinder head. So if you have a 6" long intake manifold from cylinder head face to the end of the runner, then you are actually running a 10" long column of air.

 

If you do roll the intake manifold over the valve cover, angle the ports up about 15* when they enter the head, basically match them up to the port floor. This will make for a U-turn of slightly less than 180*, and will give you even more room for your big turbo, as well as giving a little room to lift the port roof, straightening out the intake port more and *usually* yielding a little more flow. Not sure how well it would do with a big U-turn intake runner, though.

[Leon]

From looking at a LOT of photos of race intake designs, including Bob Sharp and Electramotive, and running some math a lot of times, it looks like a 15.5" long runner is a reasonable "local maxima" for torque production..which should happen about 5300RPM.

 

You can't say that without assuming some key factors. There is no "one-size-fits-all" with runner lengths, it is dependent upon the application (i.e. cam profile, air temp...).

 

Is the port really 4 inches long? I haven't measured so I don't have an answer but that seems a bit long.

[Xnke]

Yep, It's 4 inches from the manifold mating face to the back of the intake valve, measured along the port centerline.

 

I can say that, since it's a recommended length by four engine builders on three continents, plus many of the intake manifolds, including the Nissan Comp, Cannon, and the longer of the two mikuni-type carb intakes with performance velocity stacks, all of these intakes end up between 13 and 15 inches long, The Bob Sharp center-throttle design also ends up at this length, AND the somewhat-questionable formulas found on the internet and in the two textbooks I've got here predict good results with this length.

 

The "local maxima" statement means that I believe you'll find a nice torque peak as well as a wide powerband on most well-prepped L28's when fitted with a manifold 15.5" long. It may not be the "peak" or absolute maxima of the curve, but I think it'll be a reasonably large peak, if it isn't the largest peak, if you graphed power output along the vertical axis and manifold runner length along the horizontal axis.

[Leon]

Yep, It's 4 inches from the manifold mating face to the back of the intake valve, measured along the port centerline.

 

I can say that, since it's a recommended length by four engine builders on three continents, plus many of the intake manifolds, including the Nissan Comp, Cannon, and the longer of the two mikuni-type carb intakes with performance velocity stacks, all of these intakes end up between 13 and 15 inches long, The Bob Sharp center-throttle design also ends up at this length, AND the somewhat-questionable formulas found on the internet and in the two textbooks I've got here predict good results with this length.

 

The "local maxima" statement means that I believe you'll find a nice torque peak as well as a wide powerband on most well-prepped L28's when fitted with a manifold 15.5" long. It may not be the "peak" or absolute maxima of the curve, but I think it'll be a reasonably large peak, if it isn't the largest peak, if you graphed power output along the vertical axis and manifold runner length along the horizontal axis.

 

Good to know about port length, thanks for that info.

 

All I'm saying is that the tuned RPM is a function of more than just runner length. Rules of thumb can be used with varying degrees of success.

Edited February 23, 2013 by Leon

[johnc]

The custom intake that Sunbelt built for my racing engine was very specific to the cam I was running.  The intake runners from the 1.5L plenum were 7.5 inches from the tip of the velocity stack bell in the plenum to the head, with a taper from 2" to the port diameter of 1.625".  So total intake length was 11.5"

[johnc]

BTW... The OP posted the question in the context of Turbo setup.  A lot of the discussion in this thread is a about wave tuning for a NA application.  IMHO, I think the biggest issue he his faced with is the long intake path creating lag and slow throttle response.

[Leon]

BTW... The OP posted the question in the context of Turbo setup.  A lot of the discussion in this thread is a about wave tuning for a NA application.  IMHO, I think the biggest issue he his faced with is the long intake path creating lag and slow throttle response.

 

Yes, but turbo or not, the engine will benefit from wave tuning (and he did specifically ask about resonance effects). I don't think an extra 12-15" of intake length will have a perceptible effect on throttle response.