Throttle body conversion

s0ck

s0ck

s0ck

Soldato
Joined
18 Oct 2002
Posts
5,299
OK, so we've had the 'how do I turbo my car' threads etc. How do you go about converting your NASP wagon for use with throttle bodies?
Speculation welcome but I'd prefer some facts too :)

So far, I'm thinking, inlet manifold comes off, TB's/injectors attach to head. The car I've got in mind is the Rover K series which I believe uses a MAP sensor for fuelling, so it will fuel based on manifold pressure but wait, we just removed the manifold!
Also, how is this done for cars with a MAF? Air is sucked in and metered just after the filter but now each 'trumpet' (I think that's the correct term?) so I'm not sure how that would work at all.
After market management... OK, we all know to get the most out of the engine it's always the best route but we aint talking forced induction here so I was thinking, perhaps, uprated the fuel regulator/increase rail pressure slightly to compensate. Perhaps along with injectors/fuel pump. Then, book an hour on a rolling road or someone who has a wideband lambda to make sure the fuelling is right and er, wham bam job done? :confused:
I suspect I've overlooked something major but please feel free to pick holes in my plan! This is possibly going to be my track car btw. It will most likely start life as a Rover Metro 1.4 16v. They're cheap to run/maintain, nice and light for slinging round the track and hopefully, not too hard to release some extra horses.

If you're sitting there thinking, WTF?! then the basis for all this is that the Lotus Exige used the 1.8 K series with throttle bodies and they made good power, circa 200BHP. Obviously, I don't expect anywhere near this performance from a home brew hybrid rust bucket but that's where the inspiration comes from :)
 
Throttle bodies themselves aren't that expensive, its the standalone ecu and remapping that costs a lot. Worth it on a car like an Integra, on a metro I'd say not.
 
Well, if nothing else, I might learn a bit more about how engines work :)

Just found this:

Airflow is not measured, but calculated based upon engine speed and throttle position.

I'm fairly sure they use a TPS sensor by default and seeing how all the throttles would be moving at the same rate, it might just be a case of attaching it to one of the throttles. Would need some kind of compensation for the extra air and I think this is where it might be an idea to increase the rail pressure slightly. Still hunting for more info...
 
You'll need an aftermarket ECU to run a alpha-n map.

Really need headwork and cams to make the cost justifiable, single throttle setups can perform nearly aswell with the correct headwork
 
With throttle bodies you usualy run Alpha-N load measurement, (essentialy using throttle angle to measure engine load) rather than the tradition speed-density system that uses a MAP sensor. The reason being that due to the loss of the plenum chamber, getting a steady MAP signal is more complicated. If you want to stick with speed-density I've seen it used by running small pipes to all throttles and combining them with some kind of surge tank to smooth out the pulses.

Off the top of my head you will need the following:
Throttle bodies
An appropriate manifold (unless TB's are direct to head type)
Injectors (unless the originals fit the TB's and have appropriate flow rates)
Fuel rail
Fuel pressure regulator
Mappable ECU (OMEX, DTA, Emerald, Megasquirt etc.)
Any sensors for above (i.e. you may need a throttle sensor)
Wiring loom to suit
Misc fittings for fuel pipes etc.

You would also idealy have a wideband lambda sensor if you want to do some mapping yourself.
 
Thanks for the input :)
I'm more concerned, initially, with just getting it to start and run rather than the performance side of things. With this in mind, I will try and stick with the OE ECU and speed/density, if at all possible. I would hope to use the injectors from the standard car; does the fuel rail attach directly to the injectors and is it metal cast or tubing of some sort? :confused: :D
Looks like MS can do Alpha-N so that should fit in perfectly with my track car on a budget :)

PS: I've no idea if any of this is going to work but it sounds like fun.
 
s0ck said:
Thanks for the input :)
I'm more concerned, initially, with just getting it to start and run rather than the performance side of things. With this in mind, I will try and stick with the OE ECU and speed/density, if at all possible. I would hope to use the injectors from the standard car; does the fuel rail attach directly to the injectors and is it metal cast or tubing of some sort? :confused: :D
Looks like MS can do Alpha-N so that should fit in perfectly with my track car on a budget :)

PS: I've no idea if any of this is going to work but it sounds like fun.
Click to expand...

The MEMS ECU is pretty much unuseable by itself to be honest, there are very few options for mapping it. You could use some kind of piggyback ECU but getting a proper ECU is definately the way forward. Not wishing to put you off, but this isn't a trivial project if you have little experience, half measures are unlikely to see the results you want. The Megasquirt is a very capable ECU, but it's also somewhat "experimental" rather than a well proven commercial quality design, which means there's a fair bit of information you need to absorb just to get the ECU up and running and have a chance to getting the mapping somewhere close.

Can't remember if the fuel rail on the K series is cast metal or molded plastic to be honest, but either way the tops of the injectors sit in recesses in the fuel rail, and the fuel rail is bolted down to the manifold, clamping the injectors in place.
 
Dogbreath said:
Can't remember if the fuel rail on the K series is cast metal or molded plastic to be honest, but either way the tops of the injectors sit in recesses in the fuel rail, and the fuel rail is bolted down to the manifold, clamping the injectors in place.
Click to expand...

Depends on the K series, the older ones use plastic manifolds and the newer ones are alloy. Another one of the causes of HGF, plastic manifolds warp, all the coolant leaks and the engine overheats.
 
ratface said:
With throttle bodies bolted straight to the head they require a balancer pipe infront of each of the butterflys, this pipe allows air pressure to equalise between all the chambers and means you have somewhere to connect the map sensor too and a servo if required, usually a manifold is still required even with throttle bodies fitted
Click to expand...

Virtualy no installations I have seen have had a balancer pipe, most of them use a servo tapping on just one cylinder, and the vast majority use alpha-n so no need for MAP sensor.
 
Never seen a balancer on any setup and that includes the OEM bike setups. Some bike setups use MAP as the ignition load measurement with TPS for fuel.

Ideally you need a synchronisation tool aswell to ensure each cylinder has a similar amount of air at low throttle angles otherwise it will totally wreck the drivablility of the car aswell as uneven cylinder temperatures etc.

Are you planning head work and decent spec cams? If its a 1.4 K head you really need to do some decent work, the ports will remain the restriction not the intake system and therefore should really be the first port of call IMO.
 
That looks like a Honda B series flange, maybe someone is trying runing it on speed-density, a lot did before the DIY ECU stuff had throttle body support.
 
I think it's fair to say to say this project will never make serious power :p
The 1.4k has the narrowest ports of all, lowest duration/lift, etc.
Considerations are obtaining a VVC head (same spec as VHPD) and blanking off the VVC mech which immediately kills a chunk of power but going from there with the TB idea should yield considerably more power. I think ultimately, keeping VVC and not going the TB route would give the biggest gains, or perhaps jumping to a forged T16 turbo setup, although I'm trying to avoid turbo, for the track.
Possibilities are endless but my funds are not :)
 
If you kept the VVC head + mechs on a 1.4 block wouldn't you need aftermarket management of the mechanism? AFAIK there isn't a system that can control the VVC yet (although apparently emerald are close).
 
You must log in or register to reply here.
Back
Top Bottom