Question about ideal carb'd a/f

[Alex]

Well that's weird - to get it ready for the dyno I ultra copper RTV'd the slip fit joint.

Drove it today and I could swear that it seems much more responsive. Not "I think....maybe....it's more responsive" but "definately feels better" .... leaking slip fit joint couldn't make it that sluggish, could it?

 

[Randy V]

If air is/was getting into the system, it does change the pressure inside the tube and may also cause exhaust detonation of unspent fuel - both can cause a reduction in power, but to what level would be probably linear to the size of the leak..

In other words - It depends...

Glad you're happy with the result!

 

[Alex]

Well, the whole underside of my pipes are pitch black, so it wasn't exactly a teeeeny tinnny leak, lol.

It's too bad you can't compare dyno-to-dyno/day-to-day because I'd be really curious how my baserun with RTV'd slip-fit joint compared to my final run from the last tuning session.

 

[Alex]

Okay, we put it back on the dyno and this is just weird!

(scanner broken, so will describe)

Year ago dyno'd it, made 49X/47X (I forget exact numbers offhand). Air-fuel as rich on tip-in (around 11ish), then from 3000 onwards it normalized to 12.7 to 12.9

I only made 1 change in the year, and that was to seal my sidepipes - my sidepipes (after the collector) are slip fit, and arn't joined together by any clamps or anything. So I used high temp red rtv to seal them. After I did this, I noticed some downlow improvement (better responsivness)

Dyno'd it yesterday, made 49X/47X again. peak power was exactly the same, minus 1rwhp/2rwtq (so basically bang on). The curve was an identical overlap throughout the entire RPM band.

Now here's the weird thing - all of a sudden at 2500rpm my a/f has gone from 11ish to 13. As soon as they hit the throttle around 3000, the a/f jumps from 13 to 15, then drops to 12.9 around 3500rpm and stays at 12.7 to 12.9 from 3500 to 6500rpm.

Ran out of time at that point, but they suggested in the mean-time swapping the powervalve (holley 4150hp, 830cfm, mechanical secondaries); stock was 6.5pv, I swapped to a 4.5. After I did this i noticed some immediate down-low responsiveness.

Why would the motor all of a sudden lean out when you hit it when the only change prior to this was sealing a (massive) exhaust leak? They seemed convinced that with a carb, seeing it lean-out immediately after tip-in is a powervalve issue (not enough fuel initially), which makes sense, but I can't explain why it wouldn't show up before. We did multiple dyno runs and they all showed the same thing - lean a/f after tip in that normalized within 500rpm.

 

[Andrew Robertson]

all of a sudden at 2500rpm my a/f has gone from 11ish to 13. As soon as they hit the throttle around 3000, the a/f jumps from 13 to 15, then drops to 12.9 around 3500rpm and stays at 12.7 to 12.9 from 3500 to 6500rpm.

Ran out of time at that point, but they suggested in the mean-time swapping the powervalve (holley 4150hp, 830cfm, mechanical secondaries); stock was 6.5pv, I swapped to a 4.5. After I did this i noticed some immediate down-low responsiveness.

Can't shed any light on whether your exhaust leak has been a factor but am sceptical that going from a 6.5 to 4.5 PV would change much in a situation where an engine was rapidly loaded to WOT on a dyno - that's mostly the domain of the pump shot. When the engine is wide open, either power valve will see zero vacuum and be well and truly open.

Where the PV is relevant is in the real world when the throttle is partially opened for moderate acceleration - in this situation the PV opening point can make a major difference to mixture in light to moderate accel. This is driveability tuning as opposed to tuning for max hp and tq - I do this using a vacuum gauge and a WBo2 and develop a good understanding of what vacuum I'm pulling as I go up through the gears or pull out to pass in a variety of light and part throttle situations. It's important to roll into rather than snap the throttle open to minimise the impact of the accel pump on the test procedure.

In this procedure its pretty easy to see at what point the PV opens (you see a reasonably lumpy increase in richness for a very small additional throttle increment) and by reading the WBo2 you can then judge if you'd like that extra fuel applied at a lighter engine load, or deferred until the load is higher. On my 383SBC road car, with primaries open to the point that secondaries are just about to be engaged, the engine pulls zero vacuum at anything under 3500rpm. So when you're fiddling with PVs you're tuning for loads well below WOT and looking for some pretty subtle changes and improvements with throttle openings of around 20-40%.

Cheers, Andrew Robertson, Wellington, New Zealand

 

[Alex]

Interesting. So if my current main concern is the initial leanness (that gets leaner when you hit the throttle, then normalizes) that would more-than-likely point to either the accelerator pump or maybe a clogged discharge nozzel, as opposed to anything else?

Drivability is good (although very hard to tune for in a light car, because even bad drivability can be hidden, heh), it's just that type of change int he a/f is worrisome.....

 

[Randy V]

I'm with Andrew on the PV as being the problem..

Fixing the exhaust leak (which is also an air leak) will alter the F/A ratio seen by your o2 sensor if the sensor is downstream of the leak..

I think I would look into the Accelerator pump cam and your clearance first.. At WOT you should have .030" clearance between the pump adjuster and the accelerator pump lever.

Check also that you're getting a pump shot with even the slightest movement of the throttle shafts for both primary and secondary.

Next up would be the air orifices for the main metering system. If there's any dirt in them, they will not allow air into the emulsion system and you'll go lean for a moment until the main metering system sees enough vacuum signal to draw the liquid fuel (rather than the air emulsified fuel) up to the booster venturis..

 

[Alex]

It took several tries, but I finally got enough of the a/f scanned without corruption. Will post dyno graph when I get a new scanner, but in the mean-time this is what I'm facing.

red = yesterday
blue = a year ago

obviously something's wrong, and I agree with the above after talking with several people today - seems to be something in the accelerator pump circuit

 

[Andrew Robertson]

obviously something's wrong, and I agree with the above after talking with several people today - seems to be something in the accelerator pump circuit

How long was this pull - how many seconds did it take for the engine to go from 3200 to 4700? Assuming you loaded it in a tall gear that lean period looks like it persists too long to be a pump shot issue, although the lean spike at the start of the pull might be. After checking the pump shot adjustment as suggested and making sure your secondaries opening cleanly and fully, maybe it's time to whip the carb off, disassemble, check and blow everything out, reassemble and adjust everything to spec and running another baseline.

The engine must have been close to a lean stumble 3200-3300?

Cheers, Andrew

 

[Alex]

Pull was done in 4th - not sur ethe duration, you can see the dyno video here

YouTube - 1965 Shelby Cobra Dyno 408w

Sounds healthy, no stumbling that I can tell. I think the big problem is that the car's so light (2120lbs) and makes an excellent power to weight (a hair under 500rwhp) that it makes a lot of potential issues.

I found another guy locally who is known be an excellent carb tuner and have an appointment with him. If we sort it out, I'll let ya'll know what we changed.

 

[John]

Were you able to get this sorted out?

Inquiring minds want to know...

 

[Alex]

Maybe, heading to the dyno in a few days.

I nboticed 2 things

- lean stumble from last dyno. HOWEVER, this stumble only happened if dyno'ing in 4th. If we dyno'd in 5th, my a/f was a solid 12.7 to 12.9 througout the entire run (probably because in 4th at 3000RPM it's like FUEL ... NOW!!!!! WTF??? FUEL!!!!!!!!! ; in 5th it's slower to rev so it starts the run at 2000 and by the time it hits 3000 it's like 'okay, fuel now please...oh, hello fuel')


- when starting aggressively, the more throttle I gave it, the worse the performance. if i dumped the clutch and gave 1/8 throttle, rev'd up to 6500 and shot forward. if I repeat but with 1/2 throttle, it'd rev to 2500ish, pause for a split second, then shoot up to 6500.

4150hps come with the pink cam (2nd most aggresive cam), and a 28/29 nozzel ..... was told 28/29 was kind of dinky for my setup, so I tried a 35, which made it much much better. No lag down-low,and I can't feel any stumble anymore.


But here's my question - how do you know when to stop increasing nozzel size? From what I've read, more is not better (makes sense). Should I stop at 35 since the stumble/hesitation on aggressive start is gone, or try higher like 40?

Should I change the pink cam to the 2nd position, or try the black cam, or just leave it alone if everything is feeling fine at the moment?

 

[Alex]

By the way, another question - on mild left turns, the car likes to stumble and almost die. (i usually blip the throttle as I'm turning left to stop this) ... obviously going lean from fuel slosh (holley 4150hp 830cfm), but short of raising float levels, is there any way to stop this?

ALso, what happens if your float levels are perfect as is - i.e., you rock the car gently and fuel dribbles out over the threads .... can't set them that much higher......