Here's
Scoffs carb setup and comments:
"As for an ideal setup, really, there isn't one, all carburettor setup involves compromise somewhere or other. I run a 26mm choke with a 1mm air corrector, a 1.2mm 2nd stage with the spring shimmed by 2mm(ish), an 0.6mm 1st stage, a 1.5mm main jet and a 2mm needle valve to keep the fuel level up. This is good for 28psi on my car before AFR creeps up to mid 12's (which means starting to go too lean). On an engine that doesn't have as good a VE (volumetric efficiency) as mine this setup may be good for higher boost. The reduction of the 1st stage helps hugely, not something alot of people seem to try. I also find that the standard needle valve is too small on a lot of big-powered GTTs, but I think often the AFR leaning at higher rpm's just isn't noticed by the owner until somone with wideband points it out to them! Quick blurps on the rollers don't show a thing. Foot to the floor in 5th gear on a real road sure does though!!
It must be understood that my setup may not be all that ideal on somebody else' engine. On my car at 20psi the AFR is very rich in the low 11's! A smaller main and everything else remaining would probably show an improvment across the board on a car not running as much boost as mine. I think after 1.2/1.3mm in the 2nd stage not alot happens.
Maximum power at any time is made at around 13:1 AFR, I've never really wanted to go much leaner than that. I think if I did I might find that the far-from-smooth transitions between stages might make things run overly lean at points and cause hesitation. With the AFR at 13 to 14:1 litte gaps in the fuel curve get covered up. Even the standard carb is far from perfect. How many gtt's suffer from a hesitation just before boost comes on ? A lot it seems!
Using a Wide Band Lambda sensor and making enough power for sub 13 sec 1/4's. 28psi, but It lives at 25psi now, it wants to pink at any more boost than that with the new cam.
I use the standard fuel pressure regulator. Actually, I use 2 in parrallel! I once got a bit paranoid about it leaning off as I managed to blow a h/g because of it suddenly leaning before I could re-act (screaming up B roads, trying to keep the car in a straight line and watching the laptop at the same time is rarther tricky). It really isn't required, but I suppose should one of the (20yr old) regulators die, the other will keep the car alive. Really I just havn't bothered undoing the pipework.
The 2mm needle valve does a good enough job of keeping the bowl full. Aparantly the 2mm valve doesn't work well on some cars, causing stalls during gearchanges or after hard pulls, possibly down to the accuracy of the home-drilling causing them not to seal properly, but worked fine for me. With an adjustable fuel pressure regulator I think maybe it doesn't take alot of extra pressure to force the needle open when it should be shut. A bigger needle is going to be even more sensitive to that you'd think. If the bowl overfills then you can imagine it would have exactly the symptoms of popping and spluttering.
I tried a
2.5mm needle valve this week, Oct06, in my car. Although the jet sealed with the usual pressure to the ball bearing, it didn't seal in the car, and fuel overspilled the bowl. I imagine there was too much fuel pressure for the float to contend with at 2.5mm! However, a 2.3mm worked ok. It didn't at first as the drilling was perhaps not absoloutly perfect (stubborn swarf on the inside of the hole?), so I applied some pressure and turning motion to the ball-bearing (needle part) of the valve in such away that it cut itself in, a little like cutting a valve into a valve seat. The result was a perfect sealing needle jet. I ran the pump for 10 minutes and had no over-fill. Nor do I have any unwanted symptoms during driving. Something worth trying the next time you can't get that drilled needle to seal.
Here's
Ant@Backyard Racings comments:
I found some interesting results messing with my carb using an LM-1 widenband lambda meter.
I was using 140main jet, 1.0 a/c jet, 25mm venturi, 23psi boost manifold.
My car was heavily over-fueling, so I was reducing jet sizes and re-testing, anyway I started by dropping the main to 130, and to my supprise to made VERY little differance, however when I put the original 125 a/c jet in that made a huge differance, and leaned right off, and was much better off boost, i.e. not in the 10`s AFR anymore.
Here's
frytols comments:
I've noticed in my engine when testing with too little boost; ~17.4psi with 140 main jet (original venturi and a/c) is that AFR was starting from eg 11.5 when at ~7.3psi and going as low as 9.5 when 17.4psi.
So the AFR was going down when revs and boost were going up.
In my opinion not using AFR and EGT sensors (on RR first before road test) when setting up your car with some 'bigger' mods is big mistake and even in Poland it's basic methodology when setting up almost any engine.
To rich is maybe not so dangerous as getting your AFR 13 or 14 but it's dangerous too. When going too rich it's possible to brake oil film and destroy your pistons and liners without propper lubrication/smearing.
Here's
Supersonics comments:
Exactly what I found when I first started using a wide band lambda. I was running a 28mm venturi with 150 main, 0.9 a/c & enlarged 2nd stage enrichment, etc, etc with 20+psi. After hooking up the lambda the spec became 135 main, 1.1 a/c, etc, etc, & reading high 11 to low 12 AFRs. Leave the 1st stage enrichment alone. 2nd stage is where it all happens while on boost. Amazing results after many years of over sized main jets!
Here's
Aves comments:
I have a Mike48 carb, although I always had problems with leaning out due to the float chamber emptying too fast.
As the drilled up needle valve didn't seal, I created
this setup.
Now it works, it's a quite flat 12.2 A/F all the way through the rev range on full boost. But I have to admit it runs very rich on part throttle.
"14.7:1 is fine for idling and off boost but at WOT the AFR should be about 12.5:1 on our cars. Why is this?"
Lee-H: 12.5:1 is as lean as you'd want to go under full load/full throttle conditions. This is the mixture that should yield the most power whilst keeping det at bay.
14.7:1 is known as stoichiometric. Any less and it's considered rich, any more and it's considered lean.
The most efficient burn occurs at stoich, but when the engine is under load or full throttle then more fuel will be required to prevent the det that would occur at stoich.
Modern EFI systems try to maintain 14.7:1 under part throttle conditions using info from the map, TPS and Lamda sensors, i.e, in a closed loop. As soon as you floor it, you shift to a completely different map, efficiency kinda goes out the window and (especially on turbo'd cars) fuel is dumped in to some degree to provide a safeguard against det.
frytol: From theoretical (scientific) point of view to burn 1kg of fuel you must use 14,7kg of air. Then it'll be stechiometric mixture.
14.7 AFR is ok when idling (truth is that is ok up to 1.5k rpm).
The best AFR for turbocharged engines is 11.0 - 12.5, and the best AFR for N/A engines is 12.5 - 13.0.
Bigger values mean too lean mixture which explodes with bigger temperature (measured by EGT sensor) resulting in 'knocking' and in extreme cases burning your pistons, valves, head - generally destroying your engine.