ITB Design for BMW - Technical Queries
Discussion
Hello,
I’m ready to start fabricating an inlet manifold for a BMW M52B28 TU (straight six 2.8) to accept ITBs. Never done this before so want to get some opinions.
Now the ports on this engine are a funny shape with an equivalent area of a round pipe with an internal diameter of 37mm. This seems a bit big for a piston that only manages 33hp per cylinder stock (however, I’m hoping for 42hp per cylinder which brings it a bit closer).
I have used a rule of thumb to calculate the total runner length of 390mm.
My next problem is the only affordable ITBs I’ve been able to find are RB26 Skyline ITBs. They cost me £120, and have almost identical spacing to the BMW. The main issue is their diameter is a massive 45mm.
So to make this work I plan on mounting the ITBs as far away form the port as possible (250mm) leaving me with short stacks to make up the 390mm. The 250mm offset is based on a guess rather than science.
Next problem is the Skyline ITBs come in pairs, and each pair don’t seem very balanced based on the visible light I can see round the valve. They are designed to seal shut and use an Idle control valve, but, as I won’t be using a ICV and the sealing material is all but gone from this set, I will be fine leaving them cranked slightly open provided the pairs balance. I can adjust between pairs, but not the pairs themselves. To solve this, I plan on using 10mm balancing pipes on each pair as shown below.
Final problem identified to date is the runners need to be a wiggly S shape (they also need to be bent in the Z axis not illustrated below). This means I can’t mount the injectors slightly away from the port to improve high RPM performance. I don’t want to mount them right by the ITB or before the ITB as I don’t won’t to compromise idle quality.
So below is my plan with the above compromises.
Does anybody feel that I’m missing something major here that will cause me problems.
Just for completeness, the stacks will have a sealed surround, then air filter, then stock MAF running stock (tuned) ECU.
I’m ready to start fabricating an inlet manifold for a BMW M52B28 TU (straight six 2.8) to accept ITBs. Never done this before so want to get some opinions.
Now the ports on this engine are a funny shape with an equivalent area of a round pipe with an internal diameter of 37mm. This seems a bit big for a piston that only manages 33hp per cylinder stock (however, I’m hoping for 42hp per cylinder which brings it a bit closer).
I have used a rule of thumb to calculate the total runner length of 390mm.
My next problem is the only affordable ITBs I’ve been able to find are RB26 Skyline ITBs. They cost me £120, and have almost identical spacing to the BMW. The main issue is their diameter is a massive 45mm.
So to make this work I plan on mounting the ITBs as far away form the port as possible (250mm) leaving me with short stacks to make up the 390mm. The 250mm offset is based on a guess rather than science.
Next problem is the Skyline ITBs come in pairs, and each pair don’t seem very balanced based on the visible light I can see round the valve. They are designed to seal shut and use an Idle control valve, but, as I won’t be using a ICV and the sealing material is all but gone from this set, I will be fine leaving them cranked slightly open provided the pairs balance. I can adjust between pairs, but not the pairs themselves. To solve this, I plan on using 10mm balancing pipes on each pair as shown below.
Final problem identified to date is the runners need to be a wiggly S shape (they also need to be bent in the Z axis not illustrated below). This means I can’t mount the injectors slightly away from the port to improve high RPM performance. I don’t want to mount them right by the ITB or before the ITB as I don’t won’t to compromise idle quality.
So below is my plan with the above compromises.
Does anybody feel that I’m missing something major here that will cause me problems.
Just for completeness, the stacks will have a sealed surround, then air filter, then stock MAF running stock (tuned) ECU.
If you don't want the throttle plates to seal at idle, it might help to slot them so that they are not so sensitive to throttle movements around idle. That would also make them less sensitive to imbalance between cylinders. By that I mean hacksaw a radial slot into the edge of the throttle plate.
I haven't done this on ITBs but did it on my STB setup with upstream fuelling. I angled the slot to induce swirl at low load to help disperse the fuel - I guess that may not be so important for you.
I haven't done this on ITBs but did it on my STB setup with upstream fuelling. I angled the slot to induce swirl at low load to help disperse the fuel - I guess that may not be so important for you.
GreenV8S said:
If you don't want the throttle plates to seal at idle, it might help to slot them so that they are not so sensitive to throttle movements around idle.
Interesting. Before I start hacking up my plates, I'll make sure I can get a spare set first, but this does sound like an accurate way of keeping them more even. Thanks.turbotoaster said:
for reference, the rover k series vvc head has 36mm inlet ports and jenvey supply 45mm ITBs to go with them, the throttles are about 3" from the port, this has been fine making up to around 260bhp NA from a 4 cylinder head(though at that power the ports are widened to 37-38mm
So this worries me a bit. If 39mm port / 45mm ITB is omptimal for 65hp, then my lonley 42hp per cylinder should have smaller ports and smaller ITBs to suit.I had no idea a K series could make 260 N/A. That's amazong.
helix402 said:
A 330 manifold fits too and gives more power.
This is in a kit car (e-type Jag kit) with the BMW engine mounted vertically so the big bulky plastic lump manifolds wont fit. If I'm going to have to make a custom manifold, it's going to be ITBs. Why? I've always wanted ITBs. If I can get more power on the top end in the process then even better.montyjohn said:
Hello,
I’m ready to start fabricating an inlet manifold for a BMW M52B28 TU (straight six 2.8) to accept ITBs. Never done this before so want to get some opinions.
Now the ports on this engine are a funny shape with an equivalent area of a round pipe with an internal diameter of 37mm. This seems a bit big for a piston that only manages 33hp per cylinder stock (however, I’m hoping for 42hp per cylinder which brings it a bit closer).
I have used a rule of thumb to calculate the total runner length of 390mm.
My next problem is the only affordable ITBs I’ve been able to find are RB26 Skyline ITBs. They cost me £120, and have almost identical spacing to the BMW. The main issue is their diameter is a massive 45mm.
So to make this work I plan on mounting the ITBs as far away form the port as possible (250mm) leaving me with short stacks to make up the 390mm. The 250mm offset is based on a guess rather than science.
Next problem is the Skyline ITBs come in pairs, and each pair don’t seem very balanced based on the visible light I can see round the valve. They are designed to seal shut and use an Idle control valve, but, as I won’t be using a ICV and the sealing material is all but gone from this set, I will be fine leaving them cranked slightly open provided the pairs balance. I can adjust between pairs, but not the pairs themselves. To solve this, I plan on using 10mm balancing pipes on each pair as shown below.
Final problem identified to date is the runners need to be a wiggly S shape (they also need to be bent in the Z axis not illustrated below). This means I can’t mount the injectors slightly away from the port to improve high RPM performance. I don’t want to mount them right by the ITB or before the ITB as I don’t won’t to compromise idle quality.
So below is my plan with the above compromises.
Does anybody feel that I’m missing something major here that will cause me problems.
Just for completeness, the stacks will have a sealed surround, then air filter, then stock MAF running stock (tuned) ECU.
TBH it all sounds like a litany of problems.I’m ready to start fabricating an inlet manifold for a BMW M52B28 TU (straight six 2.8) to accept ITBs. Never done this before so want to get some opinions.
Now the ports on this engine are a funny shape with an equivalent area of a round pipe with an internal diameter of 37mm. This seems a bit big for a piston that only manages 33hp per cylinder stock (however, I’m hoping for 42hp per cylinder which brings it a bit closer).
I have used a rule of thumb to calculate the total runner length of 390mm.
My next problem is the only affordable ITBs I’ve been able to find are RB26 Skyline ITBs. They cost me £120, and have almost identical spacing to the BMW. The main issue is their diameter is a massive 45mm.
So to make this work I plan on mounting the ITBs as far away form the port as possible (250mm) leaving me with short stacks to make up the 390mm. The 250mm offset is based on a guess rather than science.
Next problem is the Skyline ITBs come in pairs, and each pair don’t seem very balanced based on the visible light I can see round the valve. They are designed to seal shut and use an Idle control valve, but, as I won’t be using a ICV and the sealing material is all but gone from this set, I will be fine leaving them cranked slightly open provided the pairs balance. I can adjust between pairs, but not the pairs themselves. To solve this, I plan on using 10mm balancing pipes on each pair as shown below.
Final problem identified to date is the runners need to be a wiggly S shape (they also need to be bent in the Z axis not illustrated below). This means I can’t mount the injectors slightly away from the port to improve high RPM performance. I don’t want to mount them right by the ITB or before the ITB as I don’t won’t to compromise idle quality.
So below is my plan with the above compromises.
Does anybody feel that I’m missing something major here that will cause me problems.
Just for completeness, the stacks will have a sealed surround, then air filter, then stock MAF running stock (tuned) ECU.
Firstly, why ITBs? What are you hoping to achieve that you can't achieve with a single correctly sized TB?
45mm is a little large for 42hp/cyl. This lands you with two problems; low load response due to a disproportionately large change in opening are per degree of shaft rotation, and lack of high load/rpm resolution as at say 75% throttle position, the TBs are flowing in excess of the engine will ever demand. This can be mitigated by using DBW.
Distance from head to TBs. The greatest advantage of TBs is how close you can mount them to the head. This reduces the amount of air which is static and needs to be accelerated when the throttle is cracked open.
Sounds like you Skyline ITBs were £120 for a reason - they are worn out. You will get away with the balance issue by joining them together, but look at other ITBs (S14/S38/S54) for ideas on how is should be done. Individual balancing screws and plates that are almost completely shut. (S14 are open only 0.1mm)
Injector distance from the back of the vave. Finding the ideal position would take a lot of dyno time and experimentation with injector spray patterns. Having them too far away will screw with transient response as the pooling of fuel in the port would require significant accel enrichment to overcome.
227bhp said:
They need mounting as close to the head as possible otherwise there is less advantage of fitting them at all.
I don't think they are too big.
I've been a little worried about this. The extra volume will reduce throttle response.I don't think they are too big.
I've heard a lot of people who fit M3 ITBs to my engine end up with a turkey as the diameter is 50mm so it robs all the low down power. On the M3 they get around this, partially because it's a bigger engine, but also they fit them further away from the port (I just don't know how far, it would be handy to know).
These guys have fairly long runner lengths
http://racehead.com.au/products-page/throttle-body...
Again, not sure how long they are, but the overall length looks a bit too short for the peak power RPM of this engine.
I suspect the length was limited to fit in a stock BMW engine bay so it may not be the best example to follow.
And then there's these:
https://store.jenvey.co.uk/throttle-body-kits/bmw/...
These look very short, if I was to guess I'd say 250mm from valve to trumpet, which would make peak power at 8500 rpm (length = 2100/rpm). Hmmm, that doesn't seem right? What am I missing?
stevesingo said:
TBH it all sounds like a litany of problems.
Firstly, why ITBs? What are you hoping to achieve that you can't achieve with a single correctly sized TB?
45mm is a little large for 42hp/cyl. This lands you with two problems; low load response due to a disproportionately large change in opening are per degree of shaft rotation, and lack of high load/rpm resolution as at say 75% throttle position, the TBs are flowing in excess of the engine will ever demand. This can be mitigated by using DBW.
Distance from head to TBs. The greatest advantage of TBs is how close you can mount them to the head. This reduces the amount of air which is static and needs to be accelerated when the throttle is cracked open.
Sounds like you Skyline ITBs were £120 for a reason - they are worn out. You will get away with the balance issue by joining them together, but look at other ITBs (S14/S38/S54) for ideas on how is should be done. Individual balancing screws and plates that are almost completely shut. (S14 are open only 0.1mm)
Injector distance from the back of the vave. Finding the ideal position would take a lot of dyno time and experimentation with injector spray patterns. Having them too far away will screw with transient response as the pooling of fuel in the port would require significant accel enrichment to overcome.
Prolems? Yes, so many problems. The whole car is a kaleidoscope of problems but that is part of the fun for me (it's a kit car by the way).Firstly, why ITBs? What are you hoping to achieve that you can't achieve with a single correctly sized TB?
45mm is a little large for 42hp/cyl. This lands you with two problems; low load response due to a disproportionately large change in opening are per degree of shaft rotation, and lack of high load/rpm resolution as at say 75% throttle position, the TBs are flowing in excess of the engine will ever demand. This can be mitigated by using DBW.
Distance from head to TBs. The greatest advantage of TBs is how close you can mount them to the head. This reduces the amount of air which is static and needs to be accelerated when the throttle is cracked open.
Sounds like you Skyline ITBs were £120 for a reason - they are worn out. You will get away with the balance issue by joining them together, but look at other ITBs (S14/S38/S54) for ideas on how is should be done. Individual balancing screws and plates that are almost completely shut. (S14 are open only 0.1mm)
Injector distance from the back of the vave. Finding the ideal position would take a lot of dyno time and experimentation with injector spray patterns. Having them too far away will screw with transient response as the pooling of fuel in the port would require significant accel enrichment to overcome.
The standard BMW inlet manifolds wont fit in my kit car engine bay. The space lends itself more for someting long and skinny, not bulky like the BMW manifold. Now I could build a manifold that uses a single TB, but I've always wanted ITBs. It's as sinmple as that. If I applied logic to all aspects of my car hobby, I wouldn't be building this car
Agreed on the distance, I just want to mitigate the oversized ITBs as best as possible, however, I may be over doing it with 250mm (+ port length) and may lean to a shorter distance. I just need to make sure the valves are away from any high velocty flows so deciding this value is tricky.
The skyline ITBs are in pretty good shape, it's just the seals in the bores look like they've been cleaned away at some point. There's no visible wear on the bores. Maybe the plates are a bit worn, but these can hopefuly be replaced ones I've got a tried and tested running rig. The valves can only shut if you are using an ICV. Without one, they will need to be cranked open a bit.
On injector distance, I've almost settled on stock location (i.e. pointing staight into the inlet port). If I shorten the runners, then this will be the only location left that I can fit them (once I discount upstream on the ITBs).
Useful response, thank you. It sounds like getting this runner length downstream of the ITBs is the key variable I need to get right. I can then experiment with upstream stack lengths to fine tune later. Injectors poiting into the ports. Yes, 45mm is a bit big, but all after market ITBs I've found use 45mm, and hopefully, the slightly lengthy runner length (but not too lengthy) will mitigate this somewhat.
Today, with modern high performance engine management, the ONLY reason (IMO) to fit port throttles is to allow one to fit aggressive cams. And if you fit those throttle way away from the intake valve, you can't run such an aggressive cam, and defeat the entire point!
Do everything you can to get the plates as close as possible to the valves!
What you could do is to internally "sleeve" the throttles down a bit, even with a taper section to facilitate a blend to the port, then fit smaller throttle plate. Just get some ally tubular sleeves turned up, shink fit them into the bores (if you are worried, you can also epoxy them into place, but that's probably unnecessary), then rebore the centres and cross drill for the throttle plate spindle.
Do everything you can to get the plates as close as possible to the valves!
What you could do is to internally "sleeve" the throttles down a bit, even with a taper section to facilitate a blend to the port, then fit smaller throttle plate. Just get some ally tubular sleeves turned up, shink fit them into the bores (if you are worried, you can also epoxy them into place, but that's probably unnecessary), then rebore the centres and cross drill for the throttle plate spindle.
And rather than mess around with idle valves and such, just go DBW and control the idle directly with plate position. Use a BMW M3 throttle actuator or a small brushed DC gear motor with crank and position sensor (sensor directly on the spindle to avoid any lash). I use a 20Watt MAXON DC motor for this task on my ITB'd 4 pot.
(i also run 4 high speed map sensors and regenerate a single load signal for the 4 runners by coherently measuring the intake pressure at IVC (ie sync'd to crank position)
(i also run 4 high speed map sensors and regenerate a single load signal for the 4 runners by coherently measuring the intake pressure at IVC (ie sync'd to crank position)
Edited by anonymous-user on Sunday 28th July 12:54
Finaly, with the heavy throttle assy mounted direct to the head, you can fabircate a really lightweight, and high thermal impedance, low pressure pre throttle intake from carbonfibre or glassfibre, whcih because it is simple, can be easily hacked around and tuned for length, trumpet geometry and taper angle etc
Sounds like a lot of work, here’s an off the shelf solution:
https://dbilas-shop.com/en/intake-tract/throttle-b...
https://dbilas-shop.com/en/intake-tract/throttle-b...
Oh, and the stock MAF cal won't work great, because if you change in the intake volume and filling the "manifold model" will be wrong, and your transient fuelling will be all over the place!
(the MAF cal itself (v/kgm3) may or may not change, depnding on how different the pipe work is that it is mounted in)
And, don't worry about trying to get the injectors away from the head, pointless on anything but a 100% engine (ie F1 etc), just ruun high fuel pressure through a smaller set of injectors. You can normally run up to around 5 to 6 bar through std road injectors with standard drivers in the ecu, and up to 10 bar with a peak and hold current controlled driver
(the MAF cal itself (v/kgm3) may or may not change, depnding on how different the pipe work is that it is mounted in)
And, don't worry about trying to get the injectors away from the head, pointless on anything but a 100% engine (ie F1 etc), just ruun high fuel pressure through a smaller set of injectors. You can normally run up to around 5 to 6 bar through std road injectors with standard drivers in the ecu, and up to 10 bar with a peak and hold current controlled driver
helix402 said:
Sounds like a lot of work, here’s an off the shelf solution:
https://dbilas-shop.com/en/intake-tract/throttle-b...
yuk, what a mess, pretty clearly designed by someone who has no idea how air actually flows! (The AFR distribution is going to be horrible with that ^^^ system......)https://dbilas-shop.com/en/intake-tract/throttle-b...
Max_Torque said:
And rather than mess around with idle valves and such, just go DBW and control the idle directly with plate position. Use a BMW M3 throttle actuator or a small brushed DC gear motor with crank and position sensor (sensor directly on the spindle to avoid any lash). I use a 20Watt MAXON DC motor for this task on my ITB'd 4 pot.
(i also run 4 high speed map sensors and regenerate a single load signal for the 4 runners by coherently measuring the intake pressure at IVC (ie sync'd to crank position)
Sounds like you know what you’re talking about re engine design. I know a fair bit about standardish BMW engines, rather than custom solutions. The M52tu has an electric throttle motor and a cable. (i also run 4 high speed map sensors and regenerate a single load signal for the 4 runners by coherently measuring the intake pressure at IVC (ie sync'd to crank position)
Edited by Max_Torque on Sunday 28th July 12:54
I think the choice of an engine that only fits the car with a set of custom throttle bodies is giving the op a lot of work.
The M52tu is a great engine, I used to have one which I fitted an M54 B30 inlet manifold and DISA valve too. Perhaps a better engine for the project would be that fits off the shelf.
Gassing Station | Engines & Drivetrain | Top of Page | What's New | My Stuff