Leaning out with revs

[ScotcH]

Hey all,

Finally got the beast running basically ok ... changed fuel pump, lines, filters, dist cap rotor, wires, points, and cleaned and lubed up the distributor guts. It had a major miss/sputter/something while under load, could not accelerate, but that is now solved. Runs well, and pulls well (don't know what actually fixed it, but it all needed doing).

Here is the new problem ... or maybe not? The car leans out the higher I rev ... starts around 12:1 at idle, and climbs to 16:1 at the top. Cruising on the highway, around 110 kph, it's around 14:1. This is obviously very lean ... certainly in my BMWs it would melt the engine in no time. But it seems to be happy there, and runs fine?

The AFR is being measured with an LM-1, O2 bung welded in right after the Y to go to single pipe. Now it's entirely possible that the weld sucks (did it with the pipe on the car) ... could it be sucking in clean air and skewing the reading? Thoughts?

[ScotcH]

No one runs a AFR gauge? Any guesses as to why it would progressively lean? What affects that during acceleration ... the main jets? This thing has dual DCNF40 carbs ... the bores too big, and not getting enough fuel in there?

I fiddled with the knobs a bit more ... the engine is really running well now, pulls all the way to redline, and can easily get up to 130 kph, but at that point I chicken out since I don't trust the suspension and wheels/tires ... plus it shakes like a mofo (yes, need to change out some of the 30 year old components).

Base timing is set for 10º BTDC, at least according to the plastic shroud makrs, which I have no idea how accurate. Advance is working, brings the timing mark to about 12 o'clock when revving, so maybe 30-40º, which seems right. A bit of a burble off throttle ... it has headers and an ANSA, so I think that's normal. No puffing or backfires, and no pinging that I can hear/feel.

[Auspider]

Isn't 14.7:1 suppose to be the optimum AFR for an engine? From doing a prac at uni I think the results were that the engine was more efficient at higher rpm, was thirsty at idle. That was a Toyota twin can though

[ORFORD2004]

15 or 16 or higher is very lean not 14.1

[ScotcH]

Isn't 14.7:1 suppose to be the optimum AFR for an engine? From doing a prac at uni I think the results were that the engine was more efficient at higher rpm, was thirsty at idle. That was a Toyota twin can though

Yes, 14.7:1 is optimal combustion for gasoline ... but most engines I've dealt with don't like to run above 13:1 ... they just build too much heat, and detonation happens. Mind you, I usually play with 13:1 compression ratio engines (BMW race motors), so that might be what's comfusing me.

Hence the thread ... I don't know what's normal for these cars, so I'm asking

[majicwrench]

Not really sure why you are dealing with FA numbers. Not really sure if such an add-on oxygen sensor is accurate. It runs well?? GREAT.

That said, nothing wrong w 14:1. And air leak, exhaust leak etc will skew numbers lean.

[ScotcH]

It runs well?? GREAT.

That's pretty much my attitude at this point

I put in the AFR just for kicks ... I had it at the shop, and figured it might help me nail the mixture a bit better than by ear. They are very accurate, the LM-1 (and other Innovate products) are used on many race/track cars, and used by many dyno tuners where being off by .1 can blow up an engine. It uses a wide band Bosch LSU

having said all that, tuning by ear seems to work just fine, though it's a bit tedious with 4 miture screws to play with.

[majicwrench]

I remember back in the 70's when I went to a Jag class and they were showing us OXYGEN SENSORS!!!!! At the time, never thought we would see em on cars, despite what the Bosch guy said. But........

So, yes, they are accurate, but system has to be air tight upstream, or it will show lean. Any mis-fire will also show as lean, unburned O2 in exhaust. Idle w lumpy cam, wil also show lean, such poor combustion at idle. Unburned O2. Bad valve?? Unburned O2, sensor will show lean.

[AndyS]

I have done a lot of tuning of Webers on a Dyno and with an a/f meter. 16 to 1 at high rpms under load and you will eventually melt a piston. you are losing a lot of power as well. try running a few size smaller air jets-the air jets affect the high rpm mixture more, the main jets the lower and mid rpms. smaller air jets will enrich the higher rpms.

14.7 to one is ideal only for electronic fuel injected cars at ldle and cruising speeds.Under hard throttle, efi cars will run more in the 12.5 to 13.5 range. with carburated cars, you must also remember you are looking at an average A/F number, and its often the case not all cylinders are matched like on an EFI car with more controllable mixtures. always better to be safe and err on the slightly rich side so that one of the cylinders isn't running too lean.

I set up my older cars to run the 12-13 to 1 range under load.

[kilrwail]

You need a kit of main jets and air correctors, so you can experiment - say + or - two sizes for each. I suspect Andy is right - your problem is with the air correctors. You'll want smaller to richen it at higher rpms.

[sptcoupe]

I've actually been running an AFR gauge quite a lot lately on the coupe. A full house 2.0L with 44 IDFs. What you describe is normal for what you will see on an AFR gauge with carbs, which are a compromise across the rpm range. If you are accelerating from stop, it will go the 11s. If you are slowly accelerating up to speed it will be in the high 12s. Once at speed and cruising, it will be in the upper 13s or 14s.

[ScotcH]

Good info guys! Once I pull the exhaust and check the welding (I need to redo the joints as well ... some leaks later in the system) I can do some more testing. Then I need to see what's actually in the carbs as far as jets and valves and stuff, and go from there ... I need to go see that guy you mentioned who was all the 308 carb parts and/or experience

Fun times!

[chrisfiat]

your car is running well and all is good, don't go looking for problems or to muck it up with meaningless info ( yes an o2 sensor on a non electrically controlled carb is meaningless ) the carb will never give you optimal stokieometric ratios which is why we all have highly controlled fuel injection now, so we keep our emissions within check and our power up, if it is wumming well and the plugs read well, drive it

[kilrwail]

O2 sensors driving an AFR gauge and lap top software is not meaningless on a carbureted car. It's essential to choosing the correct jets and air correctors for optimal performance across the rpm range and to obtain a smooth idle and transition to open throttle. I couldn't run my Porsche with 6 carbs without it and avoid destruction caused by a lean mixture at WOT.

[AndyS]

O2 sensors driving an AFR gauge and lap top software is not meaningless on a carbureted car. It's essential to choosing the correct jets and air correctors for optimal performance across the rpm range and to obtain a smooth idle and transition to open throttle. I couldn't run my Porsche with 6 carbs without it and avoid destruction caused by a lean mixture at WOT.

Exactly, especially since you are testing it under load where you can really see what is going on. without this info, you never really know if you have a lean burn at high rpm or are just dumping in too much gas.

Seat of the pants tuning with no data is not usually optimal. I know this, because every car I set up by feel ended up being at least slightly off when I got a chance to run the car ran under load on a dyno with a gas analyzer, or on the road with a high band A/F gauge.

A friend had a Lancia Scorpion with a built up 2L that was jetted by the engine builder. I drove it, and wasn't that impressed, and sent him to a dyno place I knew about. On the Dyno, it was obvious it was running too lean and needed bigger chokes as well. He ended up picking up over 30 HP just by tuning the twin Webers! Now the thing throws you back in the seat. (made 125 at the tires afterwards)