tii linkage and tuning question for Bill Williams

[thehackmechanic]

I'm trying to dial in the FI in my '72tii. I have:

--Freshly adjusted valves

--Correct 2ohm black coil and .9ohm ballast resistor

--Newly rebuilt dizzy from Jeff at AdvancedDistributors, timed using the ball at 2400 to 2500 rpm with no audible pinging

--Pertronix (sorry, points people, but the ball is so much more stable to the timing light with the Pertronix than the points)

--New Jack Fahuna linkage rods

--Adjusted warmup regulator as per manual

--Synchronized pump and throttle as per manual

--Verified the warmup regulator piston is protruding about 10mm when warmed up

--Looked for vacuum leaks with propane and found none

It's running pretty damned well, but when I put a CO meter on it and adjust the idle mixture screw to 2-3% at idle as per spec, when I crack the throttle and put the revs into the low 1000s, it goes rich, then as the revs go up to the 3000 range, it goes lean. These are static measurements taken with the car in the garage. The CO meter is at least 30 years old so I don't really trust its absolute numerical measurements, but the basic swings from lean to rich seem valid. I'm ordering a wideband sensor as per Jim Gerock's epic thread.

But here's my question. I have read your well-photographed thread about the relationship between the linkage and the mixture screw in the tuna can:

http://www.bmw2002faq.com/component/option,com_forum/Itemid,57/page,viewtopic/t,368728/highlight,tii+regulator/

and your statement that the screw only affects mixture at idle in this thread:

http://www.bmw2002faq.com/component/option,com_forum/Itemid,0/page,viewtopic/t,356508/start,6/postdays,0/postorder,asc/highlight,/sid,b6ec39faa0d2344ef18a632841170ce1/

Why do you say that the screw ONLY affects mixture at idle, or predominantly affects it at idle? Clearly the two pieces under the tuna can are mechanically coupled, and I can see the effect on my CO meter that adjusting the screw has an effect on the mixture at higher RPM. Are you saying that, because of the shape of the cam lobe under the tuna can, the screw has the greatest effect at idle, and a lesser effect at higher RPM?

--Rob

[c.d.iesel]

....untill Bill arrives.

Glad your using a Co Meter no matter how old it is.

Still better than nose or butt meters.

The only valid reading is at idle and sitting still.

Your partial throttle openings sitting still you can forget about.

It's those partial throttle and full throttle readings under load

with the car driving or on a dyno that mean anything.

you should see 1.75 ....3.0% Co at idle

what ever gives the smoothest, rocking free idle at 775 ...850 rpm

about 2.0 ....3.5 % Co under load holding the throttle

open at about 1/6th off idle

about 5.0 % holding FULL OPEN throttle and pulling 4,500 rpm in 2nd or 3rd

gear - full load reading while holding for 5 ..8 seconds on the pull

there are some portable exhaust anylizers that you can place on the seat next to you with a exhaust probe secured along the body while you perform

these under load tests

use a vacuum gauge T 'd into the manifold so you can duplicate the exact same partial throttle open position, approx 1/6 to 1/4 open. The vacuum reading you see at 3,000 rpm while holding the pedal at 1/6th - 1/4 open and driving in 3rd gear is the same vacuum reading you use with each test run to assure same throttle opening condition and rpm testing. Then you have accureat reading from one adjustment to the next.

have fun

[bnam]

The screw inside the can advances or retards the the throttle butterfly relative to the fuel pump linkage. This relative position is changed across the full range of movement. However, since the change is very finegrain, it is material only at very small throttle openings (ie. most impact at idle) with diminishing impact as it opens up.

Byas

[BillWilliams]

Lets chat and I maybe wrong, entirely possible.

My reason. The little tiny screw inside the tuna can, when turned only increases or decreases the amount the butterfly opens at idle. It acts as a stop.

Therefore, once you are off idle, the adjustment on the little tiny screw is out of the equation.

Thoughts?

[thehackmechanic]

(nt)

[thehackmechanic]

I like bnam's answer.

Will play around more with the CO meter and c.d.'s advice and see how much effect it really has (the idle mixture screw at throttle openings larger than idle, not c.d.'s advice :^)

[bnam]

Lets chat and I maybe wrong, entirely possible.

My reason. The little tiny screw inside the tuna can, when turned only increases or decreases the amount the butterfly opens at idle. It acts as a stop.

Therefore, once you are off idle, the adjustment on the little tiny screw is out of the equation.

Thoughts?

Bill,

My reasoning is that the screw is not just a stop. A stop is active only when the lever/throttle comes in contact with it. And not in effect for the rest of the range of motion.

In the pic below, you'll notice that the spring keeps the throttle lever in constant contact with the cam -- throughout the range of motion and the screw changes the relative angle between the lever and the cam which then remains changed throughout the motion of the lever.

Throttle lever is actually 2 parts: The lower part is the cam follower. The screw is affixed to the cam follower. The upper part is really the throttle lever and this is what moves the throttle valve. THe long black line indicates the position of the throttle valve. The screw adjusts the angle between the throttle valve and the cam follower. And, this angle is constant.

For the screw to be a stop IMO -- it should have been mounted on a fixed part and not on the cam follower. At least in the later model throttle body, that is not the case.

Byas

[BillWilliams]

I agree with what you say, your explanation is more accurate than mine.

When I referred to it as a stop, what I was trying to convey is that once the intermittent linkage adjustment is set, synching air and fuel, the little screw in the tuna can functions as an adjustment to set the butterfly position at idle and holds it at that relative position.

What I am trying to say, is even with your explanation which I agree with, once you come off idle, the position of that screw really does not affect CO to any discernible degree.

I have seen some try to get their off-idle CO correct by adjusting that little screw.

[thehackmechanic]

After playing with it last night, I have direct evidence the idle mixture screw is affecting more than idle, but clearly it's not the mechanism you want to use to control mixture at open throttle.

I actually own two exhaust gas analyzers -- a "Kar Check" which I bought in a pawn shop in Austin in 1982 and which you can drive with it in place, with the sensor strapped to the bumper and the hose shoved in the tailpipe, and a Gunison, which is new, but I don't like it -- it's twitchy and drifty as hell, and designed only to take a static idle measurement. But the Gunison probably has some degree of absolute accuracy to it, if you catch it when it isn't drifting, whereas the Kar Check I only believe in relative terms (ie, really lean versus lean versus rich versus really rich).

I set the linkage up by the book (pinning the pump and the throttle cam in the tuna can) and adjusted the CO at idle with the Gunison to between 2 and 3% as per spec. I then drove the car and watched the Kar Check analyzer.

I didn't need the analyzer to tell it's too lean -- it buffeted and hesitated at a wide range of even throttles like driving into an inconsistent headwind -- but the analyzer confirmed it. WAAAY lean. I loosened the mixture screw by 1/4 turn. Yes I realize that is a lot. I drove the car again. The hesitation was nearly gone. The analyzer showed the mixture still lean but closer to center.

I backed off another 1/4 turn, and the results were better still (car ran smoother, exhaust analyzer was more centered, less lean), but by that time, at idle, the idle lever was completely off the cam, not touching it at all, indicating something else is amiss in terms of the adjustment.

Ordered the wideband gauge so I can see accurately what's going on. Will attack as per Gerock's long post:

http://www.bmw2002faq.com/component/option,com_forum/Itemid,57/page,viewtopic/t,339832/highlight,bucking/

[BillWilliams]

Turning that screw that far will affect WOT but then your idle CO has to be way different than what it was, either way leaner or richer.

I still contend, CO at off idle is controlled by the intermediate shaft adjustment (linkage length, timing, valves and all that being a given) and idle CO is controlled by the small screw, just as the manual states.

You will get it one way or the other and we will all learn from your efforts.

[stone02]

Hi,

after all the reading and thoughts i think that you(rob)

need to eliminate all other vacuum leaks. start with the

air bypass hose under the plenum. either block it off or

ensure it is new and sealed. next you need to check the

cold start valve for no leak condition if you do not block

off the air bypass. next you need to confirm the small

vacuum line from the throttle body to the cylinder head

blow-by hose is intacted and solid. next confirm that the

blow-by hose is good with no cracks in it's length.check

the bellows from the air-cleaner to the throttle body.

also test that there is no leak along the booster hose.

plug off the vapor return line at the air-cleaner. is the

plastic cap on the bottom of the throttle body?

finally if all this leaves you lean still then you can leed

the throttle body with pump so that the entire range of the

test is richer to start out and then lean adjustable to get

the CO you need. lastly if the car has sat for a long term

spring parts in the pump may still be contaminated and

running slower when traveling along the fuel cone.

i will keep thinking of better stuff.

good luck,

stone

[thehackmechanic]

Stone, I agree!

[jgerock]

Beating a proverbial dead horse.....

The factory FI settings are only good for a baseline adjustment on the tii. I think I described this in detail in one of my many threads on trying to get the tii running correctly. Both Rob Torres and Bill Williams told me my car was running too lean even though I made every effort to have all items "spot on" regarding the synchronization and TB settings. Sounds like you may have to change the vertical pivot linkage index at the bottom of the throttle body.

More stuff..

That tiny return spring inside the tuna can is another weak area on tii's.

If you make the accelerator linkage TOO tight, then the engine may not want to drop to idle RPM making you think something else is wrong (happened to me). If you have to push the cam inside the tuna can toward the idle speed screw something is wrong.

Faulty plug wires will mask fuel injection running issues (stuttering, missing, sputtering, etc). The "lifetime" Bosch plug wires may crap out in 1-2 years.

[02fanatic]

I agree with that Jim....I had my original stock '73tii for 2 or 3 years before I realized the return spring on the throttle body was missing! Car ran fine though....if the linkage had been too tight it would have been a different story!

......If you make the accelerator linkage TOO tight, then the engine may not want to drop to idle RPM making you think something else is wrong (happened to me).