Zenith AFR Testing on a 3.0l SC
 

I am tuning my Zenith TIN using a Innovate Motorsports wide-band AFR system. The engine is a 3.0l SC with SSI headers and an old 2:1 muffler of ~1975 vintage. The engine is stock, with 122,000+ miles.

I run two types of AFR tests
1) steady state cruising
2) 2nd Gear, Uphill, WOT run from 2000 rpm to 6300 or 6500 rpm.

My goal is to get the perfect (12.7 13.1) AFR from ~3000rpm to redline under WOT, for maximum power and a reasonable cruise AFR of 13-14, with good transition.

My "code" is Idle jet/main jet/air correction jet sizes. Current I have 55 idles, 165 mains, and 175 air correction jets: 55/165/175.

Here are a few runs I did, and some 4th order polynomial fitted curves to see if there are any specific trends:
http://forums.pelicanparts.com/uploa...1287971163.jpg

After thinking about the data plotted here, and after some airflow calculations, I've concluded the shape of the AFR rise from low speed to high speed makes sense for a 3.0 with Zeniths and 34mm venturis. Based on the simple scale of 3.0l/2.2l the 27.5mm vents in the 2.2 should really go to ~31-32mm for the 3.0.

By jumping to 34mm throat, the pressure signal that pulls the fuel through the main jet (controlled by the relative sizes of the main and secondary venturis) was weakened as RPM increases. Hence a leaning out trend. I created a simple compressible flow model, considering peak flow, RPM, venturi sizes, and found that the pressure signal (vacuum) in a 3.0l engine with a 31mm vent would be about the same as a 2.2 with the 27.5mm vent. Jump to a 34mm vent, and the vacuum at high rpm drop substantially. Less vacuum, less pull of fuel, leaner mixture.

Ideally thus curve should be flat. Changing the air correction jet from 185 to 175 may have helped, But there is a long way to go. My next tweak will likely be drop down to 165 air correction jets. But I'm not confident this will have any effect down low. Note how the 57/168/185 and 57/168/175 curves are about the same. Ignore the kick up at the beginning, it's being driven by the initial lean during throttle opening.

In order to get the curves to tip, from rich to lean AFR, to flat AFR, I think the emulsion tubes need to be tweaked. Based on what Weber does, I'll look at adding a few extra holes near the top of the tube, and see if that leans out the bottom end as envisioned. Look to see those plots in November...

keep up the great work Mike! I increased my air jets to ~168 and this helped a bunch after the dissy was recurved. Would that help the AF curve (a recurved dissy)? I just got a wide band LM2.

I was just going to email you and ask how the carb project was going. Looks like it's in full swing...

James,

I ran the 168s and they worked fine, but for some reason the 165s combined with the 175 air jets seemed to give a stronger top end rush. My son and I both noticed this during the hill-climb. I have a NOS 78 distributor with 25-26 deg advance, set at 8-9 BTDC at 900rpm. It is not as aggressive in the mid-range as a recurved distributor. I'm not sure how much an effect on AFR that has. If I get unburned fuel in the exhaust, that will lower the AFR, not raise it as some might expect. So it is possible.

The best part of posting information on the BBS is all can see and critique, plus you get to commit yourself to a position and then see what people think. I'm doing it to myself, today. :)

When I said the air correction jet change didn't seem to work, I was using the 57/168/185 to 57/168/175 data. No real change. And when I ran the 57/165/175 I expected a curve between the 60/160/185 and the 57/168/175. What I got was different and I questioned the data.
http://forums.pelicanparts.com/uploa...1288004997.jpg

But after posting I started thinking about what I expected and what I got. Perhaps the 168 mains were so large that the decease in air correction jets couldn't have as much effect, but dropping to 165 mains, the air correction jet were able to influence the curve. Hence the smaller air jets enriched the high end like they were supposed to, but only when the mains were small enough?:confused:

So now I really want to try the 165 air correction jets, and test both the 165 and 160 mains to see what happens. Rain and evening commitments this week will delay this testing till Saturday:(

While my expectations are now, that the top end will flatten out more, my guess is the bottom will still be too rich. Hence, emulsion tubes....

pcar9119, vents are working fine. I have to say, I ploped them in with no fuss, spent about 3 hours dialing in the carbs, and they ran fine from the get go. I am now enjoying my hobby, by trying to get even more out of them. Fun stuff!

When I was working on those vents, I tried various waist heighths in my car to see if that would make a different in low end response. I did not find better response, but what I did find was a less progrressive roll on power with Zeniths Design. I think we discussed this on the phone that day. The proper geomtry used on my vents rather than going to a mangled copy of Zeniths vents was defintily a far better idea, in my own humble opinion. :)

My only concern with the Zenith's I have now is getting crappy gas! the carbs run fine, pull up to the rev limiter (7k), start great (3-6 pumps), but with the new mandate in WA for 10% ethanol, I get some popping at idle and, that rich exhaust gas smell. I can run 100ll avgas in it and that helps and makes it smell much better!

Mike,

I'd send my idle air correction jet kit to you if you send your address; info@PerformanceOriented.com. I think you need to chase the lower end of your AFR curve before you chase emulsion tube modifications. The kit allows for returning to OEM jetting if you don't like what you get.

12 to 13.5 AFR @ 1k-6k would be perfect if that is obtainable.

We have E10 all the time. I have an occasional pop. Got a little worse with the 55 idle. 57 was nicer, I'll probably go back to the 57s.

After a few pops here and there today, I went back to the 57 idle jets... No AFR data to back it up, but driveability suffered a little with the 55s.

Got a set of spare emulsion tubes from neilca. Thanks. There are four sets of four holes. The sets go from the top to bottom, with .65mm, 1.0mm, .65mm and .50mm diameter. I'm going to clean them, since they have a little build up in the smallest holes at the bottom.

I'm still a little leery of the last data set, so that will be the first run this weekend, i.e. re-baseline the 57/165/175. If it matches the previous runs, then I think the next test is to install the 165 Air Correction jets, 57/165/165, followed by going back to the 160 mains, 57/160/165.

At that point I hope to know how powerful the air correction jet are. No one seems to recommend going below 165 air correction jets so that would point to the emulsion tubes.

Each idle jet change will adjust idle speed which will require a new lean-best idle mixture screw and idle air screw adjustment, perhaps the pops were from just substitution without retuning? I find that even with five of six cylinders dialed in that there is a noticeable change in idle performance when the last cylinder is tweeked, and I'm not talking about much of a tweek.

I have six Gunson Colortunes in my test engine so I am able to adjust idle mixtures quickly and I also use mercury sticks to monitor all six intake manifold vacuums simultaneously. With the Colortune I can easily determine mixture parity to 1/8 turn of the mixture screws and there is a difference to be realized in smoothness when all are set the same.

Idle air screws help balance the inaccuracies of idle air flow, a little change of throttle shaft position will result in unequal air flow responses to the three barrels on that throttle shaft. So, as laborious as it is the retuning is important to assure your data acquisition is working from a unified baseline for each test.

How much power are we talking here. All this precision, what is the true outcome your looking for. Is it just perfection? Isn't the numbers going to change from day to day? Weather, fuel, humidity. Get it dialed in for a nice driving experience and reliabilty, and drive the hell out of it. Lol, great write-up though!

Paul,

You correctly point out each idle jet change takes about a hour. Change jets, reset the needles, check synchronization & balance, set idle, check AFR, redo as needed.

Pcar9119,

Installing the Zeniths and getting a good running engine was pretty easy. And overall they work great. I'm being a little more intense about the tuning side, mainly because it's a challenge. Plus, with the AFR gauge, I think tuning is easier than in the past, and we can do more.

I'd like better fuel mileage while cruising around and be able to spend 20-30 minutes at a time running 4500-6500 rpm WOT without melting down the engine. So I'm looking for a combination of things that will make the conversion better.

As for Power? Who knows, but if I can run ~13.1 AFR, no one will argue that I could do better. Once I'm done, maybe I'll get a dyno run to see what's possible on a tired, old SC. :)

Good answer, LOL. Your tuning devotion is awesome. I wasn't trying to make light of your efforts. I think it's really inspiring the effort your putting in. I will steel your data when your done and aply it to my car, lol. Just call me GM, right? Keep up the awesome work.

More Data - But less info?
 

Over the last weekend I did a little more testing, but came away with more questions than answers. I ran the #55 idles for only a day or two because there was a slight stumble/hesitation. I reinstalled the #57 and the car ran great. (57/165/175) So I thought #55 were just a little too small. I redid the AFR runs with the 57/165/175 jets, and got about the same results as before:
http://forums.pelicanparts.com/uploa...1288739375.jpg

But, really what I learned is the AFR reading really bounce around a lot. The accuracy of the reading is affected by a lot of things, and perhaps the engine is the bigger variable. I did do a complete heater and free air calibration before this run, so the AFR system is as good as it is going to get....

My Lesson 1: AFR gauge and logging can/should be used for trending. But run-to-run variation is fairly large. Use the average, and don't sweat the extremes.

So far the 57/165/175 runs the best, has great tip-in, minimal popping, but is too rich most of the time. More to follow.....

And So it Continues, less, is still less...
 

After running the 57/165/175 jets, i was pretty sure leaner main jets would really be a good step. On 10/31/10, I went back to the original #160 mains. Here's the latest summary. See the light Blue Line:
http://forums.pelicanparts.com/uploa...1288740237.jpg

Note the 57/160/175 is better than the baseline 60/160/185, in both a leaner low end and a richer top end. So the jetting changes did help. But the magnitude is smaller than expected.

Guess what? Throttle tip-in response suffered! :confused: Just like the 55/165/175! On tip-in, the AFR soars to 16+ and on one run it hit 22! And not just on snap throttles, but gentle accelerations from ~3000 rpm see the AFR rise from 12.5 to 15.5-16+, and the engine hesitate for a 1/4 second. Both the 55/165/175 and 57/160/175 drive about the same. A little hole at 2500-3000 rpm. :(

The Next Step - Start Over!
 

Paul Abbott (Performance Oriented) and I have been trading e-mails the past couple of weeks. It's been a good, offline discussion, and after this past weekend of frustration, what he said made real sense. He mentioned, and I had read other places and posts, that getting the transition circuit right was critical. And he made an interesting observation: perhaps I'm running on the transition circuit at idle.

For a 2.2L engine with Zeniths, all is well. Add in 0.8l of extra displacement, it is very likely to get enough air to run at idle, the butterflies are cracked open much more than a 2.2l would have them. That is probably enough to uncover the 1st transition port.

If this is true, then I am under-flowing the Idle Mixture Screws and making it up with the 1st port flow. Leaving me LESS fuel for transition. Maybe that is why, when I tried to reasonably lean out the mid-range (from two different directions!) I ran into the lurking issue: mid-rpm hesitation due to fuel starvation.

Two Options to increase idle air flow with closed butterflies:
1) open up the Air Bleed Screws (easy)
2) Drill a small (1mm) hole in the butterfly (Winter Project)

OK, it's dark, and 38F. Not going to happen tonight. ;) Saturday I'll get out the synchrometer, open up the Air Bleeds and reset the Needles and idle speed.

In the mean time, the Emulsion Tube Mods (thanks neilca for sending a spare set for our shared use) will be held off until I can get the Transition better. Paul, sent me a nice kit for changing idle air bleed jets. That is also a carb disassembly and rebuild task, which will wait.

I'm also running out of weather: snow flurries on Saturday....:(

Opened Air Bleed Screws helped transition!
 

Good news. I opened up the air bleed screws as much as I could and still synchronize the Zeniths. I was able to set idle by closing the butterflies as low as 700 rpm, but raised it to 900-950 rpm. I think that means the throttle plates were not completely closed at 900 rpm, which is good. I backed the idle mixture screws out to about 2 full turns, and set idle AFR at 13+. The engine idles smoothly without any pops. BTW, I kept the 57/160/175 combination that exhibited a little tip-in hesitation.

I went for a test drive in town and on the highway. The tip-in hesitation is gone, and the AFR spikes are much smaller (they are expected, but no longer 16-17+) The instant I roll the throttle on, the car starts to accelerate. The only odd thing is sometimes the idle would drop to 700 rpm. It would idle just fine there, but then pick up with a quick blip of the throttle. Cruise AFR went up a little, which is nice.

Since the air bleed screws are out so far, I'm a little concerned they will back out due to vibration. The little black plastic thread locker is 40 years old! I'll put a drop of silicon RTV to hold them for the rest of this season (3 weeks maybe...)

Per Paul's Abbott's suggestion, I think drilling the butterflies with a small bleed air hole would get these carbs to where they need to be for a 3.0 sized engine. This will be an over the winter project. I'll figure out what the area of the air bleed port is, and select a drill size to match, or be some fraction of it.

Today, I'll try the smaller air correction jets (57/160/165), and see how that affects the top end. Then, maybe I'll start to play with the Emulsion Tubes.

Air Correction Results Look Good
 

I ran the 57/160/165 configuration today, and overall it seemed pretty good. Still a little lean vs. Ideal on top. Reducing the Air Correction Jet from 175 to 165 did have an effect. Combined with the modifications to the idle (using the air bleeds) I was quite happy to see cruise AFR rise to 12.5-13, which should help mileage.

There were AFR spikes to 16+ on snaps. I'm coming to the conclusion this is normal, as long as it recovers quickly. I did not notice any hesitation.

Here is the latest comparison Plot. The orange line is the latest run.
http://forums.pelicanparts.com/uploa...1289179573.jpg

You can see the air correction jet rolled the AFR over a little on the high end, as expected. I also overlayed the Power vs. AFR curve everyone references. It's better to ere on the rich side. So right now I'm pretty darn close. Perhaps #162 or #163 jets would be ideal.

Altitude and Temperature Correction for Ideal AFR at LRP
 

Take a look at this plot:
http://forums.pelicanparts.com/uploa...1289180085.jpg

It's busy, but, I calculated the effect of altitude, air temperature, and fuel temperature on AFR. As fuel gets colder, it also gets denser. Not as fast as air, but enough to have an opposing effect on AFR (see how the orange line's slope is opposite that of the dark blue "air effect only" line).

Lime Rock Park is at ~550ft AGL. It is going to be 35-55F on Friday. I have made a curve for AFR @ LRP, with 12.9AFR ideal at 50F. If an engine was jetted perfectly at LRP at 50F, then AFR would vary in the green hatched box, as air temp and the barometer changed throughout the day.

I also made a curve for that same engine, for Windsor, CT, at today's barometric reading. Note the engine should be 0.3 points leaner. I plotted a star for the composite average of my measurements: 46F and ~13.4 AFR. Based on the plotted star I am 0.1 points higher than the curve.

So this is where I go out on a limb. If I leave the jetting as is, I should be at 13.0 AFR WOT at LRP this Friday. Now I just have to steal a glance at the gauge on the main straight. I have a couple of days to decide If I will keep the 57/160/165, or swap back to my favorite 57/165/175.

Any thoughts?