Throttle Bodies - why?
Discussion
So the current trend seems to be to stick throttle bodies on absolutely everything. But what are the actual benefits of doing this? I understand how roller or slide types will increase airflow at WOT because there's no butterfly spindle in the way; and I guess that the lack of a shared plenum could prevent problems with inlet pulses from other cylinders. But apart from that, it seems to me that unless an engine has a particularly poorly designed inlet setup, the gains aren't going to be that great; and what gains there are will only really be at WOT anyway.
So given the price of a throttle body setup, is it more about looks and fashion than actual performance? Do they provide some other improvement that I'm not aware of? Or are they something which have a large effect on some engines with restrictive inlet tracts, and which people have then just copied, assuming they will give the same improvement everywhere?
So given the price of a throttle body setup, is it more about looks and fashion than actual performance? Do they provide some other improvement that I'm not aware of? Or are they something which have a large effect on some engines with restrictive inlet tracts, and which people have then just copied, assuming they will give the same improvement everywhere?
A normal plenum is usually designed to :
1) fit a tight space
2) give good driveability
3) be ecomonical to make..
4) ulitmate power is not a priority
TB's are quite no compromise and besides point 4) do not necessarily do points 1 to 3 that well..
be if not too large can also give good drivability due to less contamination from exhaust gases from previous cylinders.
1) fit a tight space
2) give good driveability
3) be ecomonical to make..
4) ulitmate power is not a priority
TB's are quite no compromise and besides point 4) do not necessarily do points 1 to 3 that well..
be if not too large can also give good drivability due to less contamination from exhaust gases from previous cylinders.
Here is something Rob Robertson did earlier.. and he knows his stuff :
Throttle bodies against plenums and cams
What I briefly said earlier with large overlap cams, this will murder an engine running a plenum, where as on throttle bodies it tends to smooth things out a bit. Take a single plenum and all 8 cylinders draw from this one opening, which is good as each cylinder can draw as much air as it wants with no restrictions as far as air flow is concerned. The BIG DOWNSIDE TO A PLENUM, on the overlap period part of the cam, this is where both exhaust and inlet valve are open at the same time, i.e. at the end of the exhaust stroke where the piston is forcing the burnt gas out the exhaust port the inlet valve opens before the exhaust is shut, instead of the unwanted rubbish going out the exhaust it is sent back up past the inlet valve and into the plenum. This is more aggravated by the fact that the next cylinder is sucking hard and will suck the waste out of the disposing cylinder into the good one, so instead of the new cylinder getting a good charge of fresh air and fuel it has 20 or 30% of nothing that cant be ignited, hence poor combustion poor idle and low power. This is only a problem at low rpm as at high rpm you have the advantage that the exhaust manifolds "should" be scavenging the fresh inlet charge into the cylinder. If you have the money to go to throttle bodies then this is where the biggest gain is going to come in the low to mid range, on the overlap period each cylinder can only contaminate its own cylinder, and part of this will be lost to the atmosphere anyway, so generally lets say at below 3000 rpm a plenum is giving a cylinder 70% of fresh charge to be ignited a set of throttle bodies will be giving 90% at the equivalent rpm all this = more ft-lb of torque, just be careful you don’t go too small on the throttle bodies or this will hurt the top end breathing, you have gone from a 72 mm plenum and as I said each cylinder will see 72 mm down to what ever size your throttle bodies is.
Rob Robinson
Throttle bodies against plenums and cams
What I briefly said earlier with large overlap cams, this will murder an engine running a plenum, where as on throttle bodies it tends to smooth things out a bit. Take a single plenum and all 8 cylinders draw from this one opening, which is good as each cylinder can draw as much air as it wants with no restrictions as far as air flow is concerned. The BIG DOWNSIDE TO A PLENUM, on the overlap period part of the cam, this is where both exhaust and inlet valve are open at the same time, i.e. at the end of the exhaust stroke where the piston is forcing the burnt gas out the exhaust port the inlet valve opens before the exhaust is shut, instead of the unwanted rubbish going out the exhaust it is sent back up past the inlet valve and into the plenum. This is more aggravated by the fact that the next cylinder is sucking hard and will suck the waste out of the disposing cylinder into the good one, so instead of the new cylinder getting a good charge of fresh air and fuel it has 20 or 30% of nothing that cant be ignited, hence poor combustion poor idle and low power. This is only a problem at low rpm as at high rpm you have the advantage that the exhaust manifolds "should" be scavenging the fresh inlet charge into the cylinder. If you have the money to go to throttle bodies then this is where the biggest gain is going to come in the low to mid range, on the overlap period each cylinder can only contaminate its own cylinder, and part of this will be lost to the atmosphere anyway, so generally lets say at below 3000 rpm a plenum is giving a cylinder 70% of fresh charge to be ignited a set of throttle bodies will be giving 90% at the equivalent rpm all this = more ft-lb of torque, just be careful you don’t go too small on the throttle bodies or this will hurt the top end breathing, you have gone from a 72 mm plenum and as I said each cylinder will see 72 mm down to what ever size your throttle bodies is.
Rob Robinson
The comments by Rob are simplified at best.
The benefits of a Port throttled lay out are as follows and one usually reaps the benefits if done from the outset as part of a total engine mod.
Individual throttle bodies allow, wilder cams with more overlap- as the volume between the throttle and inlet valve is smaller. This means there is less of a tendency for exhaust gas residual to flow back into the intake. This in turn means that one can have more overlap before idle quality and part load combustion quality is compromised.
Throttle response is much much faster as the throttled volume- That between the throttle butterflies and the inlet valves is smaller
One can speficy a massive volume for the inlet plenum with individual throttles and because this is no longer part of the throttled volume- we are not constrained on this to think about throttle response and can size it for ultimate power- how do you think the BMW M3CSL makes over 14 bar BMEP?
I've seen some part load fuel economy benefit with port throttles- due to less pumping work- as your pressure recovery during the pumping cycle is much faster.
Drawbacks include- more difficult to set up at idle - to get even airflow distribution
The presence of the butter fly itself can be a restriction and restrict power- if this wasnt taken into account and catered for from the outset
The individual throttles can dampen out part load in cylinder motion (depending on where the throttles are cited) and therefore limit combustion stability at part load- effecting things like lean misfire limit or EGR tolerance (thus limiting fuel economy at some points) and emissions for catalyst warm up
The benefits of a Port throttled lay out are as follows and one usually reaps the benefits if done from the outset as part of a total engine mod.
Individual throttle bodies allow, wilder cams with more overlap- as the volume between the throttle and inlet valve is smaller. This means there is less of a tendency for exhaust gas residual to flow back into the intake. This in turn means that one can have more overlap before idle quality and part load combustion quality is compromised.
Throttle response is much much faster as the throttled volume- That between the throttle butterflies and the inlet valves is smaller
One can speficy a massive volume for the inlet plenum with individual throttles and because this is no longer part of the throttled volume- we are not constrained on this to think about throttle response and can size it for ultimate power- how do you think the BMW M3CSL makes over 14 bar BMEP?
I've seen some part load fuel economy benefit with port throttles- due to less pumping work- as your pressure recovery during the pumping cycle is much faster.
Drawbacks include- more difficult to set up at idle - to get even airflow distribution
The presence of the butter fly itself can be a restriction and restrict power- if this wasnt taken into account and catered for from the outset
The individual throttles can dampen out part load in cylinder motion (depending on where the throttles are cited) and therefore limit combustion stability at part load- effecting things like lean misfire limit or EGR tolerance (thus limiting fuel economy at some points) and emissions for catalyst warm up
Marquis_Rex said:
The individual throttles can dampen out part load in cylinder motion (depending on where the throttles are cited) and therefore limit combustion stability at part load- effecting things like lean misfire limit or EGR tolerance (thus limiting fuel economy at some points) and emissions for catalyst warm up
I'm trying to understand this last part. Is this something to do with the physical obstruction of the throttle plate affecting the nature of the flow past the valve? Or is this some side effect of the cyclic changes in manifold depression?Black Flash, are you also on Locostbuilders? If so 
, although you'll know me as DIY Si, the other bloke thinking of the Alfa V6. I'm looking at fitting ITBs to the Alfa engine to free up the inlet and allow me to set the engine up properly to allow full power. Mainly because the Alfa inlet isn't really very good, and I think I can do better as I don't care much about induction roar or fuel economy. As said though, ITBs should give better power through out the rev range if sized correctly. This may in turn lead to better MPG and should give better drivability. Also, if set up correctly, they can sound very good too!
, although you'll know me as DIY Si, the other bloke thinking of the Alfa V6. I'm looking at fitting ITBs to the Alfa engine to free up the inlet and allow me to set the engine up properly to allow full power. Mainly because the Alfa inlet isn't really very good, and I think I can do better as I don't care much about induction roar or fuel economy. As said though, ITBs should give better power through out the rev range if sized correctly. This may in turn lead to better MPG and should give better drivability. Also, if set up correctly, they can sound very good too!Snake the Sniper said:
Black Flash, are you also on Locostbuilders? If so , although you'll know me as DIY Si, the other bloke thinking of the Alfa V6. I'm looking at fitting ITBs to the Alfa engine to free up the inlet and allow me to set the engine up properly to allow full power. Mainly because the Alfa inlet isn't really very good, and I think I can do better as I don't care much about induction roar or fuel economy. As said though, ITBs should give better power through out the rev range if sized correctly. This may in turn lead to better MPG and should give better drivability. Also, if set up correctly, they can sound very good too!
Yes, that's me! And on about a million other forums trying to get information about a myriad of things! I mainly post here and on the Dax boards though., although you'll know me as DIY Si, the other bloke thinking of the Alfa V6. I'm looking at fitting ITBs to the Alfa engine to free up the inlet and allow me to set the engine up properly to allow full power. Mainly because the Alfa inlet isn't really very good, and I think I can do better as I don't care much about induction roar or fuel economy. As said though, ITBs should give better power through out the rev range if sized correctly. This may in turn lead to better MPG and should give better drivability. Also, if set up correctly, they can sound very good too!Trying to work out how best to spend my monies on tuning...better cams and inlet work, increase capacity, stick a supercharger on it...? Good to know there's someone else in the same boat.
GreenV8S said:
Marquis_Rex said:
The individual throttles can dampen out part load in cylinder motion (depending on where the throttles are cited) and therefore limit combustion stability at part load- effecting things like lean misfire limit or EGR tolerance (thus limiting fuel economy at some points) and emissions for catalyst warm up
I'm trying to understand this last part. The Black Flash said:
GreenV8S said:
Marquis_Rex said:
The individual throttles can dampen out part load in cylinder motion (depending on where the throttles are cited) and therefore limit combustion stability at part load- effecting things like lean misfire limit or EGR tolerance (thus limiting fuel economy at some points) and emissions for catalyst warm up
I'm trying to understand this last part. I stumbled across this phenomenon while testing an engine with valve tronic.
Later on one day I was looking over some ignition maps of an S54 M3 engine- over the whole engine speed load range and comparing them to a 328i M52 Tu engine, and I found that the S54 needed masses of ignition advance at low load higher speeds- way more than the usual scatter for pent roof 4 valve engines. (Especially seeing as the S54 M3 engine has a shallower combustion chamber and alot of tumble-both good for fast burn). Then I tweaked.
This could at least partially be explained by the elasticity of air? ..
on a plenum with throttle upstream you have a large volume to draw on so on the first part of the induction stroke you will have a higher velocity of air into the chamber before the elasticity to vacuum brings the port velocity lower again. It would matter little that the rest of the inlet suck was at reduced velocity (by that i mean below the mean velocity of the port over the time interval of the whole inlet stroke) because the swirl /tumble is already set in motion in the chamber by the initial high velocity gas at the very start of the inlet stroke..
In a low volume individual throttle situation this velocity gradient happens much more rapidly therefore the net effect is more gas time at low port velocities hence less tumble/swirl. For this to occur you have to assume the vacuum in a throttle-upstream plenum is transient per inlet suck ie the vacuum is not uniform and stable per individual suck .. this allows for the port velocity gradient with time model to work (though the vacuum is uniform and stable within practical limits over an extended time interval or you'd have an unsteady idle speed).
or am I talking b
ks?
on a plenum with throttle upstream you have a large volume to draw on so on the first part of the induction stroke you will have a higher velocity of air into the chamber before the elasticity to vacuum brings the port velocity lower again. It would matter little that the rest of the inlet suck was at reduced velocity (by that i mean below the mean velocity of the port over the time interval of the whole inlet stroke) because the swirl /tumble is already set in motion in the chamber by the initial high velocity gas at the very start of the inlet stroke..
In a low volume individual throttle situation this velocity gradient happens much more rapidly therefore the net effect is more gas time at low port velocities hence less tumble/swirl. For this to occur you have to assume the vacuum in a throttle-upstream plenum is transient per inlet suck ie the vacuum is not uniform and stable per individual suck .. this allows for the port velocity gradient with time model to work (though the vacuum is uniform and stable within practical limits over an extended time interval or you'd have an unsteady idle speed).
or am I talking b
ks?Edited by trackcar on Monday 30th June 23:24
trackcar said:
This could at least partially be explained by the elasticity of air? ..
on a plenum with throttle upstream you have a large volume to draw on so on the first part of the induction stroke you will have a higher velocity of air into the chamber before the elasticity to vacuum brings the port velocity lower again. It would matter little that the rest of the inlet suck was at reduced velocity (by that i mean below the mean velocity of the port over the time interval of the whole inlet stroke) because the swirl /tumble is already set in motion in the chamber by the initial high velocity gas at the very start of the inlet stroke..
In a low volume individual throttle situation this velocity gradient happens much more rapidly therefore the net effect is more gas time at low port velocities hence less tumble/swirl. For this to occur you have to assume the vacuum in a throttle-upstream plenum is transient per inlet suck ie the vacuum is not uniform and stable per individual suck .. this allows for the port velocity gradient with time model to work (though the vacuum is uniform and stable within practical limits over an extended time interval or you'd have an unsteady idle speed).
or am I talking bks?
We used CFD or Computational Fluid Dynamics to evaluate what was happening.on a plenum with throttle upstream you have a large volume to draw on so on the first part of the induction stroke you will have a higher velocity of air into the chamber before the elasticity to vacuum brings the port velocity lower again. It would matter little that the rest of the inlet suck was at reduced velocity (by that i mean below the mean velocity of the port over the time interval of the whole inlet stroke) because the swirl /tumble is already set in motion in the chamber by the initial high velocity gas at the very start of the inlet stroke..
In a low volume individual throttle situation this velocity gradient happens much more rapidly therefore the net effect is more gas time at low port velocities hence less tumble/swirl. For this to occur you have to assume the vacuum in a throttle-upstream plenum is transient per inlet suck ie the vacuum is not uniform and stable per individual suck .. this allows for the port velocity gradient with time model to work (though the vacuum is uniform and stable within practical limits over an extended time interval or you'd have an unsteady idle speed).
or am I talking b
ks?Edited by trackcar on Monday 30th June 23:24
Its simply that tumble motion comes from a bias of flow within the port- which them imparts a momentum to the inlet charge- as perhaps more flow flows over the top of the valve for instance. With the port throttle barely open- like under part load operation- it will simply disrupt this bias of the flow- and therefore the motion generation. Its worse with Valvetronic at part loads- as the valve itself is doing the throttling and at low loads, valve lift could be 1.5mm or thereabouts- and no motion can get through. BMW got around this on their later Valvetronic 2 engines, but staggering the valve lifts to try to impart some swirl motion in liu of tumble....
Edited by Marquis_Rex on Tuesday 1st July 00:08
trackcar said:
This could at least partially be explained by the elasticity of air? ..
on a plenum with throttle upstream you have a large volume to draw on so on the first part of the induction stroke you will have a higher velocity of air into the chamber before the elasticity to vacuum brings the port velocity lower again. It would matter little that the rest of the inlet suck was at reduced velocity (by that i mean below the mean velocity of the port over the time interval of the whole inlet stroke) because the swirl /tumble is already set in motion in the chamber by the initial high velocity gas at the very start of the inlet stroke..
In a low volume individual throttle situation this velocity gradient happens much more rapidly therefore the net effect is more gas time at low port velocities hence less tumble/swirl. For this to occur you have to assume the vacuum in a throttle-upstream plenum is transient per inlet suck ie the vacuum is not uniform and stable per individual suck .. this allows for the port velocity gradient with time model to work (though the vacuum is uniform and stable within practical limits over an extended time interval or you'd have an unsteady idle speed).
or am I talking bks?
I think you will have to put these idea's of yours in the cubboard next to the hydolastic suspension bubble and the underhead over hangers on a plenum with throttle upstream you have a large volume to draw on so on the first part of the induction stroke you will have a higher velocity of air into the chamber before the elasticity to vacuum brings the port velocity lower again. It would matter little that the rest of the inlet suck was at reduced velocity (by that i mean below the mean velocity of the port over the time interval of the whole inlet stroke) because the swirl /tumble is already set in motion in the chamber by the initial high velocity gas at the very start of the inlet stroke..
In a low volume individual throttle situation this velocity gradient happens much more rapidly therefore the net effect is more gas time at low port velocities hence less tumble/swirl. For this to occur you have to assume the vacuum in a throttle-upstream plenum is transient per inlet suck ie the vacuum is not uniform and stable per individual suck .. this allows for the port velocity gradient with time model to work (though the vacuum is uniform and stable within practical limits over an extended time interval or you'd have an unsteady idle speed).
or am I talking b
ks?Edited by trackcar on Monday 30th June 23:24
trackcar said:
Bums. So it's a simple deflection issue and nothing so grand as I had hoped. I'd got an air de-elasticiser machine all drawn out in my head an' all 
Would that not be a supercharger? So it still seems to me that they're of most benefit on fairly tuned engines using cams with high overlap, and / or which will run at WOT a lot, and / or which have restrictive inlets to start with, and of much less benefit on a mildly tuned road engine ?
The Black Flash said:
So it still seems to me that they're of most benefit on fairly tuned engines using cams with high overlap, and / or which will run at WOT a lot, and / or which have restrictive inlets to start with, and of much less benefit on a mildly tuned road engine ?
I think that in the latter case, the throttle response advantage still stands. Gassing Station | Engines & Drivetrain | Top of Page | What's New | My Stuff