afr adjustment where to start

[gorskined]

i have a newly rebuilt 76 3.0l row turbo new stock Mahle P&C sc cams new k27 evergreen raptor turbo long neck IC . nos throttle body 3 way adjustable wur. ( cold hot and boost) rebuilt balanced fuel distributor 2 and 5 running 5% more fuel more than the other cylinders new injectors new twin 044 fuel pumps .. so basically everything is new built for reliability not HP basically stock with a few upgrades .
The fd and wur set up by Brian (rawly8) perfect fuel pressure curve . The car starts a little hard 5 seconds of cranking then idles high about 1100 a little black smoke for the first 5 seconds or so runs super rich when hot and cold. I can watch the single point afr gauge go for 12 to 13.5 to 14 then *** off the scale all within about 2 seconds of running. ( cold start injector is disconnect for now ).. i need to lower the idle and lean out the afr . as mentioned nos throttle body not sure if it is set correctly . what the big knob for? so where do i start i would assume there are steps to follow?

Thanks in advance
Regards Ned

[Black_Hat]

"Where to Start?"

Put a set of CIS gauges on and post your numbers , This will be your starting point.

[gorskined]

Originally Posted by Black_Hat

"Where to Start?"

Put a set of CIS gauges on and post your numbers , This will be your starting point.

off the top of my head i don't recall the hot , cold and boost pressures . but i did double check them. when i installed the wur and all three fell within the corresponding fuel temp /pressure curve chart .

[spuggy]

Originally Posted by gorskined

The car starts a little hard 5 seconds of cranking

This doesn't sound too out of the way. Every time? If it's the same on hot restart, perhaps check the fuel accumulator?

Maybe try jumping the fuel pump relays to run with the key on, let it build pressure for 10-20 seconds before start - still the same?

Mine used to (usually) fire immediately if daily driven. If I'd left it for a few weeks, it might want longer to wake up.

Originally Posted by gorskined

then idles high about 1100

Eh? Seems pretty normal to me. Is this your first CIS 930?

I forget whether it's the AAV or the AAR that bleeds extra air into the motor for initial cold running. It's the one that looks like an ally version of the air freshener you plug into a wall socket. A little column thingy...

It's powered, and should slowly close the orifice over a few minutes with the motor running. I knocked the power plug off mine once when messing with the WUR and didn't notice; cold start/idle was exactly the same as usual - but some miles down the road, no power to it resulted in hot idle of 1800 RPM, LOL...

Originally Posted by gorskined

a little black smoke for the first 5 seconds or so

"that's normal sir, they all do that"

At least, mine always did on CIS.

Originally Posted by gorskined

(I can watch the single point afr gauge go for 12 to 13.5 to 14 then *** off the scale all within about 2 seconds of running.

Yeh, I'm going to take those readings with a big ole bowl of salt.

Working temp for a narrowband sensor is 600F; any reading before it reaches that is, uh, "not reliable". Or "noise". Or "garbage".

A Bosch 4.2 or 4.9 wideband sensor doesn't provide usable readings until it reaches a working temperature of 1200F. Despite a dedicated heater, my Motec doesn't log readings from my 4.9 for a good 3-5 minutes after motor start.

Any unheated sensor variant just takes longer to get up to temperature (as it relies on the exhaust gas to do the job).

As the gauge reads down to 12:1 - and a narrowband only works between 14:1/15:1 or so - I'm going to presume that this is a wideband.

Originally Posted by gorskined

( cold start injector is disconnect for now )..

This might be causing your hard start. Or might not; my 930 when daily driven would fire up within 1-2 compression strokes with 1" of ice on the windshield and the TTS disconnected (thus no CS injector at all).

Can't help with your other questions. But my advice would be not to even think of touching anything until you've at least got usable readings from the lambda sensor - and preferably not before the motor is fully up to temperature as well (eg oil at working temperature).

You say the WUR is in spec - so I wouldn't touch that at all. Unless there was a compelling reason to do so, and not without more research.

If you do adjust the idle CO2, as a data sample size of 1, I found mine would idle perfectly with 3% CO2 - but setting it this would lean caused a lean surge in the mid-range at a constant (light) load and throttle setting (like climbing a hill at the speed limit).

Which went away entirely when set a little fatter. Really made no difference to the idle though, apart from the CO2 reading (checked on both a Gunson and a 5 gas analyzer).

[flightlead404]

If its hard starting when its cold, but ok when its hot, it could be the csv. most likely culprit in my experience is the temp sensor. the injector itself is pretty ironclad

[gorskined]

Originally Posted by spuggy

This doesn't sound too out of the way. Every time? If it's the same on hot restart, perhaps check the fuel accumulator?

Maybe try jumping the fuel pump relays to run with the key on, let it build pressure for 10-20 seconds before start - still the same?

Mine used to (usually) fire immediately if daily driven. If I'd left it for a few weeks, it might want longer to wake up.



Eh? Seems pretty normal to me. Is this your first CIS 930?

I forget whether it's the AAV or the AAR that bleeds extra air into the motor for initial cold running. It's the one that looks like an ally version of the air freshener you plug into a wall socket. A little column thingy...

It's powered, and should slowly close the orifice over a few minutes with the motor running. I knocked the power plug off mine once when messing with the WUR and didn't notice; cold start/idle was exactly the same as usual - but some miles down the road, no power to it resulted in hot idle of 1800 RPM, LOL...



"that's normal sir, they all do that"

At least, mine always did on CIS.



Yeh, I'm going to take those readings with a big ole bowl of salt.

Working temp for a narrowband sensor is 600F; any reading before it reaches that is, uh, "not reliable". Or "noise". Or "garbage".

A Bosch 4.2 or 4.9 wideband sensor doesn't provide usable readings until it reaches a working temperature of 1200F. Despite a dedicated heater, my Motec doesn't log readings from my 4.9 for a good 3-5 minutes after motor start.

Any unheated sensor variant just takes longer to get up to temperature (as it relies on the exhaust gas to do the job).

As the gauge reads down to 12:1 - and a narrowband only works between 14:1/15:1 or so - I'm going to presume that this is a wideband.



This might be causing your hard start. Or might not; my 930 when daily driven would fire up within 1-2 compression strokes with 1" of ice on the windshield and the TTS disconnected (thus no CS injector at all).

Can't help with your other questions. But my advice would be not to even think of touching anything until you've at least got usable readings from the lambda sensor - and preferably not before the motor is fully up to temperature as well (eg oil at working temperature).

You say the WUR is in spec - so I wouldn't touch that at all. Unless there was a compelling reason to do so, and not without more research.

If you do adjust the idle CO2, as a data sample size of 1, I found mine would idle perfectly with 3% CO2 - but setting it this would lean caused a lean surge in the mid-range at a constant (light) load and throttle setting (like climbing a hill at the speed limit).

Which went away entirely when set a little fatter. Really made no difference to the idle though, apart from the CO2 reading (checked on both a Gunson and a 5 gas analyzer).

good information thanks .. guessing i will leave well enough alone until i get her out on the road and up to temp ..
i did jump out the fuel safety switch doesn't make a difference when cold starting.

i also tried starting with and without the cold start injector being connected no difference either ...

[spuggy]

Originally Posted by flightlead404

If its hard starting when its cold, but ok when its hot, it could be the csv. most likely culprit in my experience is the temp sensor. the injector itself is pretty ironclad

Eh, I know why people fixate on the "cold start injector" circuit. It does seem logical.

However, I think it contributes less than many think it does. On some CIS systems, it doesn't even come into play until ambient drops below "pretty darned cold" - like temperatures folks aren't choosing to use toy cars in, much less temperatures you're seeing in July in the contiguous 48.

I don't know what the 930 TTS is rated at (did look, but could never find it). I do know that my 930 motor would start immediately on the key after being left outside overnight in hard frost, with enough ice (1/2" or more) on the windshield that non-serious screen scrapers were useless. (Factory exchangers would get that steaming within 90 seconds of start on defrost settings though...)

And that with no functional CSV. Because sometime in the spring, tracing an intermittent crank issue due to a bad pin 14 connection, noticed the themotime switch was disconnected. Flag terminal just dangling.

Eventually, I stopped bothering to put it back on and never noticed any difference, regardless of weather conditions. Yes, I daily drove the car, sometimes in snow. Pretty sure the CS injector did nothing for me the entire time (14 years?) it ran CIS.

It usually started from cold just fine. Except for when the original 911 fuel accumulator got a bit tired after 250,000+ miles.

[spuggy]

Originally Posted by gorskined

good information thanks .. guessing i will leave well enough alone until i get her out on the road and up to temp .

Sounds like a plan. It's CIS, it will have some foibles

You may also find that it'll start better once it has some carbon in the heads.

Originally Posted by gorskined

i did jump out the fuel safety switch doesn't make a difference when cold starting.
i also tried starting with and without the cold start injector being connected no difference either ...

Honestly, 5 second cranking for a cold start? Wouldn't keep me awake at night. Maybe collect more data points, think about it a little more if it doesn't improve.

Do be aware of time spent cranking it over; think the handbook says 10 seconds or so max, and then let the starter cool for a while (30 seconds?). Dunno, years since I looked at it And I've blatantly ignored that advice with no observed ill effects anyway (like to get gas to the injectors the time some muppet at the docks drove it around until he used up the gas it shipped with).

If I left mine undriven for a few weeks, it'd sometimes (not often or always) take until the second set of cranking to wake up. If it'd been used 2-3 days ago, it usually seemed to fire immediately, like it was eager to go, LOL. If it'd just been used the day before, it'd either fire immediately (most often) or within a few seconds.

[gsxrken]

Originally Posted by gorskined

i. I can watch the single point afr gauge go for 12 to 13.5 to 14 then *** off the scale all within about 2 seconds of running. (
Regards Ned

Congrats on the new build! I’d suggest your AFR O2 sensor may need to be recalibrated in free air. There’s no way your WUR element is heating up so fast that your fuel pressures climb like they would if you really were leaning from 12 to off the scale in 2 seconds. The stock specs for fuel pressures are good starting places but you will drift from them a bit in fine tuning your car when your AFR gauge setup becomes trustworthy.

[gorskined]

Originally Posted by gsxrken

Congrats on the new build! I’d suggest your AFR O2 sensor may need to be recalibrated in free air. There’s no way your WUR element is heating up so fast that your fuel pressures climb like they would if you really were leaning from 12 to off the scale in 2 seconds. The stock specs for fuel pressures are good starting places but you will drift from them a bit in fine tuning your car when your AFR gauge setup becomes trustworthy.

i believe the car is starting rich because of the black smoke for a few seconds plus the burning eyes think the sensor starts low and runs through the top end of the range ...

[gorskined]

[QUOTE=spuggy;11736372]Eh, I know why people fixate on the "cold start injector" circuit. It does seem logical.

However, I think it contributes less than many think it does. On some CIS systems, it doesn't even come into play until ambient drops below "pretty darned cold" - like temperatures folks aren't choosing to use toy cars in, much less temperatures you're seeing in July in the contiguous 48.

I don't know what the 930 TTS is rated at (did look, but could never find it). I do know that my 930 motor would start immediately on the key after being left outside overnight in hard frost, with enough ice (1/2" or more) on the windshield that non-serious screen scrapers were useless. (Factory exchangers would get that steaming within 90 seconds of start on defrost settings though...)

And that with no functional CSV. Because sometime in the spring, tracing an intermittent crank issue due to a bad pin 14 connection, noticed the themotime switch was disconnected. Flag terminal just dangling.

Eventually, I stopped bothering to put it back on and never noticed any difference, regardless of weather conditions. Yes, I daily drove the car, sometimes in snow. Pretty sure the CS injector did nothing for me the entire time (14 years?) it ran CIS.

It usually started from cold just fine. Except for when the original 911 fuel accumulator got a bit tired after 250,000+ miles.[/QUOT


i found the same with my 74 911s it never worked and never made a difference. This is why i disconnected it on my 76 while looking into the running rich issue so it wasn't a factor

[gorskined]

Originally Posted by spuggy

This doesn't sound too out of the way. Every time? If it's the same on hot restart, perhaps check the fuel accumulator?

Maybe try jumping the fuel pump relays to run with the key on, let it build pressure for 10-20 seconds before start - still the same?


yes still the same

Mine used to (usually) fire immediately if daily driven. If I'd left it for a few weeks, it might want longer to wake up.



Eh? Seems pretty normal to me. Is this your first CIS 930?

I forget whether it's the AAV or the AAR that bleeds extra air into the motor for initial cold running. It's the one that looks like an ally version of the air freshener you plug into a wall socket. A little column thingy...

It's powered, and should slowly close the orifice over a few minutes with the motor running. I knocked the power plug off mine once when messing with the WUR and didn't notice; cold start/idle was exactly the same as usual - but some miles down the road, no power to it resulted in hot idle of 1800 RPM, LOL...


yes my fisrt 930 im a 356 and early 911 guy things with carbs the 76 is pre air pump no AAR



"that's normal sir, they all do that"

At least, mine always did on CIS.



Yeh, I'm going to take those readings with a big ole bowl of salt.

Working temp for a narrowband sensor is 600F; any reading before it reaches that is, uh, "not reliable". Or "noise". Or "garbage".

A Bosch 4.2 or 4.9 wideband sensor doesn't provide usable readings until it reaches a working temperature of 1200F. Despite a dedicated heater, my Motec doesn't log readings from my 4.9 for a good 3-5 minutes after motor start.

Any unheated sensor variant just takes longer to get up to temperature (as it relies on the exhaust gas to do the job).

As the gauge reads down to 12:1 - and a narrowband only works between 14:1/15:1 or so - I'm going to presume that this is a wideband.

yes its a wide band 4.9



This might be causing your hard start. Or might not; my 930 when daily driven would fire up within 1-2 compression strokes with 1" of ice on the windshield and the TTS disconnected (thus no CS injector at all).

Can't help with your other questions. But my advice would be not to even think of touching anything until you've at least got usable readings from the lambda sensor - and preferably not before the motor is fully up to temperature as well (eg oil at working temperature). pre lambda

You say the WUR is in spec - so I wouldn't touch that at all. Unless there was a compelling reason to do so, and not without more research.

If you do adjust the idle CO2, as a data sample size of 1, I found mine would idle perfectly with 3% CO2 - but setting it this would lean caused a lean surge in the mid-range at a constant (light) load and throttle setting (like climbing a hill at the speed limit).

Which went away entirely when set a little fatter. Really made no difference to the idle though, apart from the CO2 reading (checked on both a Gunson and a 5 gas analyzer).

good recommendation im going to leave well enough alone concentrate on getting her on the road and then when im convinced it wont self destruct take her to a local shop for the fine tuning .

[spuggy]

Oh, just saw your answers embedded in my quoted text. Let's touch on some of those:

Originally Posted by gorskined

yes my fisrt 930 im a 356 and early 911 guy things with carbs the 76 is pre air pump no AAR

Could have been removed. PET shows AAR #930.606.102.00 on diagram 107-05, part #44 for type 930/51/52/54 motors and 930.606.102.01 for 930/53 motors, 1975-1977.



Jim's CIS primer https://jimsbasementworkshop.com/CIS/pages/comp_page.html#aar says:

Provides extra starting air (bypassing air around the throttle body) with a cold engine. The valve slowly closes with heat, provided by an internal resistor, and to a lesser degree, from heat from the engine itself. For part number 0 280 140 200 (from a 1976), the cold resistance of the heater is about 17 ohms. This device was added in '76.

Failure mode: May remain partially open in cold ambient conditions due to open resistance element. Idle speed stays high after warm up. If it stays closed due to mechanical malfunction, the car may be hard to start.

See also auxiliary air valve (#33 on the same diagram):

Provides bypass air around the throttle when the engine is first started. When the intake manifold vacuum reaches 5 - 6 inches, the valve closes. The closing is independent of engine heat. Part number 0 280 160 400 pulls in at ~8" vacuum. This regulator supplies air to the intake manifold during warm and hot starts when the auxiliary air regulator is closed. This was added in '76.

Failure mode: When not closing properly, can cause high idle speed , or idle speed to oscillate.

(Jim's "added in '76" comment is likely for N/A middies).

Originally Posted by gorskined

yes its a wide band 4.9

Right. Well, before the sensor hits 1200F, any voltage read from it doesn't mean much of anything at all. According to the people that developed and manufacture the sensor.

In practical terms, my MoTeC M800 disregards the lambda sensor entirely for 3 minutes after start, doesn't even log any values from it while in C3 (warmup) state. And won't read/use it at all if there's no heater, or the wiring is broken incorrect etc.

I'm currently tinkering with my warmup compensation tables without lambda readings, like an animal. "Seems a little rich", followed by (next day) by "bah, overshot, too lean now" etc.

It is also, absolutely, important to calibrate the lambda sensor for accurate readings, they do change over time - I also had one foul from fuel contamination. But if you got this as a kit (eg both the gauge and the sensor together), sensor controller/gauge was probably calibrated when you got it. IIRC, you can get the calibration number for the lambda sensor with a DMM too; I did it both ways when I fitted the new sensor, and the numbers were very very close (like within 0.002 V of each other).

Also, even when the sensor is up to temp, be aware that a rich misfire (which may not even be audible) will read lean, not rich, on lambda. The clue is when you keep fattening up that cell in the map, and the sensor reads leaner and leaner for each change...

Lambdas are a useful tool - but they really don't react instantly and are most accurate for readings in steady-state conditions, although they do give a vague idea of what happens when you accelerate etc.

Originally Posted by gorskined

good recommendation im going to leave well enough alone concentrate on getting her on the road and then when im convinced it wont self destruct take her to a local shop for the fine tuning .

"fine tune" in the same sentence with "CIS". LOL...

I'm sure some folks can do it. I decided I didn't have the patience/skills to keep fiddling and was just grateful it started/ran - and lived with it balking on anything other than very slow/gentle throttle tip-in during the first few miles of warmup. I also don't miss 6-12 mpg stoplight-to-stoplight in town one little bit...

HTH - good luck!

[Dr J]

Hello Ned. Jumping in here to say that I am really suspicious of your AFR sensor readings. Are you sure the sensor is properly connected? My AFR sensor (also the bosch) doesn’t begin to read right for about 10 sec or so. For example. If I power the sensor only, with the car cold and off, after around 10 sec, the afr reading starts to change and goes off the scale in about 3-4 more sec.

Also, while the car may run as high as 17 afr, it will be clear to you that it is not happy.

Cold starts at 12 afr is not unusual. Car needs to be rich on first startup. That is the warm up part of the wur- low control pressure makes for rich start and pressure increases to lean out as engine warms.

My advice is that you double check the sensor.

[spuggy]

Well, TIL (Today I Learned). LOL.

Originally Posted by spuggy

before the sensor hits 1200F, any voltage read from it doesn't mean much of anything at all. According to the people that developed and manufacture the sensor.

This is 100% true. However, the Bosch spec document Y 258 E00 015e gives

3.2 Time to activity (light-off time)
Guide value for the time to activity after switching
on the sensor heater (“light-off time”)

as "≤ 10 s". Although heater current during startup is a complicated topic, it seems (Bosch spec again):

4.2 The heater power must always be switched on power controlled (e.g. duty
cycled heater power), starting with a maximum ramp-up duty cycle as shown
in the diagram in section 1.6. This is necessary to reduce thermal stress
of the sensor element during cold starts due to high peak power in the
first seconds.
4.3 The sensor ceramic element is heated up quickly after heater start.
After heating up the ceramic all occurrence of condensation water, which
could damage the hot ceramic, must be ruled out.
To allow early heating of the sensor to reach a fast sensor activity, the
sensor installation location design must be selected in a way to minimize
exhaust-side stressing of the sensor with condensation water.
If this is not possible by design measures, the start of the sensor heater
must be delayed until demonstrably no more condensation water appears.

Originally Posted by spuggy

In practical terms, my MoTeC M800 disregards the lambda sensor entirely for 3 minutes after start,

OK - Motec's "normal" sensor heater mode doesn't even begin to heat the sensor until engine temps cross a threshold to reduce the likelihood of condensation hitting the ceramic tip of the sensor and destroying it via thermal shock. In "fast" sensor heater mode, it'll heat it as soon as the motor is running.

Not all aftermarket kits have any "motor running" indication - but heating the sensor before the motor is running is apparently a big no-no.

If using a stand-alone controller without an engine speed input, never let your controller heat the sensor prior to starting the engine. One way to guarantee this is to power the controller off of its own relay which is not turned on until after the engine is started.

From https://www.nzefi.com/bosch-lsu-wide-band-airfuel-ratio-lambda-sensors-fail-often-aftermarket-performance-applications/

Or, from the Bosch spec

Electrical heating of the sensor
The sensor heater may never be connected directly to battery voltage. It
must always be controlled by the LSU control unit or the vehicle ECU. Heat-
ing the sensor before the engine is started is not allowed.

Originally Posted by spuggy

I'm currently tinkering with my warmup compensation tables without lambda readings, like an animal.

Well, at least I know how to address that now..