Alfa Romeo Forums banner

1 - 20 of 42 Posts

·
Registered
Joined
·
295 Posts
Discussion Starter #1
Today was my first day out with the new engine setup, though it is not near tuned, one big issue is starting to show. When I increase the throttle, the voltage from my O2 sensor will drop to as low as .4 volts. Currently I see .78 to .89 volts during cruise and most driving. Driving the car I can definitly feel the lean out until the O2 sensor trim kicks in and the A/F ratio returns to normal. Is there a way to quicken the response from either the O2 sensor or AFM? My AFM spring is as loose as it can possibly be and still have tension when closed. Also my idle CO adjustment on the AFM is completly closed.

My understanding of Greg Gordons tuning box is that it only modifies the signal voltage, and thus will not help here when the problem is time based.
 

·
1966-2013
Joined
·
13,741 Posts
spring is as loose as it can possibly be and still have tension when closed.
You're prolly way too loose on that AFM flap spring.

As it reads airflow, if it's too loose, then it gets way up in the 'get rich' signal range way too early (like running there most of the time) or for too long and can actually cause a bit of fuel starvation because there is no rich range beyond the rich range you're most likely already in. (loose comparision if it were a carburettor: there's no accelerator pump action because you're already running on it, so all you end up with when you step on it is a flat spot)

Alternatively, it could be the pumps aren't keeping up, the filter is restricting flow, or the regulator isn't working up to snuff.

I'm sure others with a bit more in depth information will be along at some point with better clarification or ideas.


Also my idle CO adjustment on the AFM is completly closed.
That's prolly not all that good a plan either, unless you went through that whole O2 sensor/multimeter voltage measure stuff and that's where it ended up. (pretty unlikely that it would require that radical a setting though)
 

·
Registered
Joined
·
295 Posts
Discussion Starter #3
I gave it those settings to make everything overly rich to see what happens, so I can tune from there, the settings for the previous configuration would not allow it to run below 1500rpm.

I would think that if it is dumping fuel when the flap opens, and then loosing pressure as a result, I would see a jump up to a higher voltage, and then a drop, all I see is a drop, and then an immediate return to .85 volts.

Though I know the system is not set correctly, it should give me a too rich condition.

I am worried how the system will perform when I get my new camshafts.

What it really needs is a throttle position switch.

I originally had the AFM spring 6 notches beyond the factory setting, and the problem was worse, now I am about 15 notches, and I still have that issue.
 

·
1966-2013
Joined
·
13,741 Posts
You know tinkering the AFM bypass screw fully closed pretty much hoses the way the engine runs at idle, right?

You say 'new engine setup'.

Does that mean you just did a rebuild or something?

If so, I'd have been looking for other issues if it didn't idle correctly after fire up and warm up.

Little things like a stuck open AAV, maladjusted idle air setting on the plenum, false air leaks, improperly set timing, stuck/blocked/poorly adjusted throttle linkage or poorly set throttle stop on the TB proper, correctly set and functional TPS, all the harness wires properly connected and correctly located, and check that the vac hoses for the regulator/altitude sensor/oil vapor/fuel vapor were all connected to thier proper points on the plenum (the 3 in the front are all different sized orfices for specific reasons) but somehow I very seriously doubt I've have gone knee deep into the AFM that prolly functioned just fine before.


I'd suggest resetting the AFM and bypass screw back to default, then look elsewhere for the real problem.
 

·
Registered
Joined
·
295 Posts
Discussion Starter #5
i ported and rebuilt the head for performance, including increased compression ratio.

I have picked through the system pretty well and everything seems to be operating fine.

The idle is not set properly, but is pretty good at the moment, it wants to run rich but the O2 sensor corrects it. The TPS is not set properly, but my issue is neither at idle or WOT, though it occurs on the way to WOT. These two should not affect the system in the way that I am seeing since my problems are from cruise to high throttle position.

All hoses are connected in thier proper locations.


This is exactly what I see on aftermarket FI engines before we set acceleration curves using the throttle position sensor as an input.
 

·
1966-2013
Joined
·
13,741 Posts
The idle is not set properly, but is pretty good at the moment, it wants to run rich but the O2 sensor corrects it.
There is a idle switch in the TPS, which, when at idle, puts the engine into open loop, meaning it's not using the O2 sensor at all for mixture correction, or rather that the ECU ignores input from the O2 sensor. (engines don't idle well at stoich as it's just a bit too lean for that purpose, and there's the whole other issue of the single wire O2 sensors cooling off to a level low enough to not function anyway after about 30-40 seconds of idling)

If it's running excessively rich at idle, it's because of that closed down bypass screw. (though the over loose AFM spring can't be helping the overall situation in the least)

This is exactly what I see on aftermarket FI engines before we set acceleration curves using the throttle position sensor as an input.
The TPS on the L-jet is not a potentiometer type like all the new stuff and the vast majority of aftermarket EFI rigs installed on older stuff, so trying to tinker the fuel curve with the stock one is actually a wasted effort.

Nor can you just swap in a potentiometer type as the ECU won't have the slightest clue how to interpret the signal and if it works at all, will just cause the ECU to revert back to it's default open loop mixture regardless of rpm range, engine load or throttle position. (more likely it'll destroy the ECU instead though)


Regardless of what you've done to the head and engine, you're still working with the stock induction system and have to remain within the working paremeters of that system. Sorry to say, but your mods just aren't radical enough that it shouldn't operate smoothly and effectively within the limits inherant in the L-jet system, and in fact should run quite well with 1 or 2 very minor adjustments.

Query: How about ignition, are you still on stock?

If so, and it's the motronic solid state digital one that sits back by the ECU, then why? (it's non-adjustable, which is not all that useful if you need to do things like actually tune a few degrees one way or the other)
 

·
Registered
Joined
·
295 Posts
Discussion Starter #7
There is a idle switch in the TPS, which, when at idle, puts the engine into open loop, meaning it's not using the O2 sensor at all for mixture correction, or rather that the ECU ignores input from the O2 sensor. (engines don't idle well at stoich as it's just a bit too lean for that purpose, and there's the whole other issue of the single wire O2 sensors cooling off to a level low enough to not function anyway after about 30-40 seconds of idling)

If it's running excessively rich at idle, it's because of that closed down bypass screw. (though the over loose AFM spring can't be helping the overall situation in the least)



The TPS on the L-jet is not a potentiometer type like all the new stuff and the vast majority of aftermarket EFI rigs installed on older stuff, so trying to tinker the fuel curve with the stock one is actually a wasted effort.

Nor can you just swap in a potentiometer type as the ECU won't have the slightest clue how to interpret the signal and if it works at all, will just cause the ECU to revert back to it's default open loop mixture regardless of rpm range, engine load or throttle position. (more likely it'll destroy the ECU instead though)


Regardless of what you've done to the head and engine, you're still working with the stock induction system and have to remain within the working paremeters of that system. Sorry to say, but your mods just aren't radical enough that it shouldn't operate smoothly and effectively within the limits inherant in the L-jet system, and in fact should run quite well with 1 or 2 very minor adjustments.

Query: How about ignition, are you still on stock?

If so, and it's the motronic solid state digital one that sits back by the ECU, then why? (it's non-adjustable, which is not all that useful if you need to do things like actually tune a few degrees one way or the other)

I am on the stock ignition for the moment, that may change soon, I was planning on keeping the stock one for now, until I get new camshafts (currently only have factory cams)

The current management system is sufficient for 90% of driving, the engine can idle fine with it, like i said the throttle position switch is not yet set, this post does not concern idle, I am not worried about idle now. I need a new O-ring for the idle adjustment as well, mine will not expand.

I know I can not swap in a TPS, if it were that easy I could tune out this problem. However I could possibly attach a potentiometer on the throttle or throttle linkage, and when the rate of change of the voltage output is within a specified range it would modify the O2 sensor or AFM output voltage. At this point, I think I would just megasquirt it. I have a spica manifold and TBs that would work great for a mid performance engine like this.

The engine runs great, just not during or immediately after rapid throttle increases. I have the AFM way loose to try and get that signal earlier hoping it would cause the engine to go overly rich and then return to a better ratio via )2 sensor, but it just goes lean, despite a huge reading from the AFM.

Ive got an hour this afternoon, and I am going to set the idle better, but I cannot see this affecting my problem.
 

·
Registered
Joined
·
2,269 Posts
Hey Pillometto,

Soundz like you need to go back to basics and set a baseline. Never go nuts with your AFM and change it too much. Always record how many clicks (teeth) you moved it so you know where you are relative to factory.:eek:

Lots of guys have fooled with their AFM in search of quicker response only to ruin their idle/off idle quality. Better to run 10* static advance if you want quickness. - A few clicks loose on the AFM only Pillo!

Well, having said all that - I'll assume it's been moved from factory and you don't know how much. I guess the best way is "feel" or push on one that is factory and then set yours to similar tension (if you can find a fish scale, measure the tension). Set the TPS to factory using the factory technique.

Test all sensors to make sure there's no bad signals and also make sure cold-start valve not leaking.

If all good, set factory timing and then idle speed. Hook-up to meter and adjust Idle CO screw at idle. Re-adjust speed.

*Since you car is modded - I would recommend you tap into the Thermo-time switch circuit to energize (like an aux injector) when/if you have lean spot on top. The stock injectors should carry you to 5,000rpm.
 

·
Registered
Joined
·
295 Posts
Discussion Starter #9
With the AFM set to stock the car is undrivable, with it 10 clicks, it drives lean most of the time, with it almost completely loose and the bypass valve closed completely the car can be driven bu no rapid positive throttle inputs. I have marked the original position, but like i said undriveable. I have tested all sensors and seem to be fine.

I am going to just drive on the settings i have temporally and then trash it for a megasquirt in a spica manifold. Not having a throttle position sensor or a adjustable acceleration curve on this system really rubs me the wrong way.
 

·
1966-2013
Joined
·
13,741 Posts
Query: you still are using the stock ignition, correct?

Do you have the wide blade rotor in the dizzy or the narrow? It can make a difference AFA achiveable timing advance ranges. The narrow usually is found in points ignitions and the wide is default for Motronic. (we're talking obviously wider by about half)

The tweak of it is, the AFM is the accelerator pump, or rather it in conjunction with the TPS tells the ECU to enrich the mixture. If you've got the flap wide open in the breeze, the correct signal never goes to the ECU to enrich.

It'll run rich all the time because the ECU is confused by incoming signals and will revert to default open loop mixture regardless of the TPS (which would end up bing limited to simply a VVT signal)

'Accelerator pump' function is derived from the flap of the ECU getting into it's upper range on initial stompage of the pedal along with the TPS confirming that, yes, you're up on it. (at which time the ECU goes into enriched for power mixture until the flap settles back down to about 1/2 deflection)

Something you might want to try:

Get the engine fully up to temperature, then set the ECU back to default and leave the cover off.

As it's running rough, the flap potentiometer will occilate.

Move in 1-2 click increments at a time in *both directions* until the potentiometer just stops occilating. (try leaner/tighter/clockwise first)

*for all anyone knows, you're running way richer rather than way leaner when you're off the O2 sensor. When at idle with a warm engine, the sweeper shouldn't be more than around 1/4-1/3 from the bottom of it's range and sitting relatively static with very little fluctuation/floating around*

Close it up, leave it alone and adjust other things like timing and idle bypass.
 

·
Registered
Joined
·
295 Posts
Discussion Starter #11
I will give that a try.

I was under the understanding that the ECU would calulate fuel based on the voltage returned from the AFM and not the change in, or rate of change in the voltage from the AFM.

I do have the wide rotor in my dist.
 

·
Registered
Joined
·
6,087 Posts
I'm not sure what exactly you have done, but I'll share what I did, and hopefully it will be of some value. I had my '84 Spider head modified to a Stage V, by Sperry, many years ago. For the first several years I ran it with a mild cam upgrade. The stock Bosch set-up handled all fuel requirements without any issues. Later on, I added 10:1 compression pistons, and still ran great with the stock set-up. It was not until I installed C&B 12 mm cams that we modified the stock set-up by 1) installing a street RML ignition upgrade and 2) loosening the AFM flapper, just a tad.

Best regards,
 

·
Registered
Joined
·
85 Posts
Silly question, but have you verified fuel pressure before tinkering with the AFM flap?

If everything is in proper order (fuel delivery, sensor function and air leaks) your engine mods shouldn't require such a severe adjustment to the AFM. I understand why you *want* to megasquirt it, but I think you *should* want to sort the issues with the L-jet setup first so you aren't chasing ghosts later.
 

·
Registered
Joined
·
295 Posts
Discussion Starter #14
No, I did not check the fuel pressure, but I just don't like the L-jet to begin with and am preparing a megasquirt setup now.

Of particular annoyance to me is its cold start system with an on-off valve, and on is too much fuel for even the coldest of days that I see, with the engine on it first start. The idle air bypass on mine needs a new o ring, but even if it were good, I do not like its method of operation.
 

·
1966-2013
Joined
·
13,741 Posts
It's more costly than a megasquirt, but I gotta say I'm kinda happy jolly with the SDS system and ITB's I'm working with now.

Still farting around with dialing the mixture, but pictorial and storyline to be coming before too long on it.
 

·
Registered
Joined
·
295 Posts
Discussion Starter #16
I looked at the sds, and it was missing a lot of features I would like. Such as laptop programing, and the fuel maps seemed overly simple to me.

At work, I use Motec but it is not worth it to me to pay that much.
 

·
1966-2013
Joined
·
13,741 Posts
Lack of laptop programming: that's exactly the reason why I like the SDS :) (I don't have one, want one, or feel I should need to aquire one)

The little botton box does it all, and tucks under the seat til you're done, then you remove it. (ok, so you don't get pretty graphics and mouse-drag slope changes, but that's OK, I grew up in the era where the only thing you needed was a flat-tip screwdriver to tune, not cartoons and joysticks :D )

I don't quite get what you mean by 'overly simple' though, as for fuel alone it's adjustable every 250 rpm with values from 0 to, umm, lots, (999 mabe? I never wound it full up to check) in steps of 1 tick per click, while MAP values are from 0-255 in steps of 1, and hits 3-4x per psi of pressure/vacuum.

I guess what I'm on about is 'how complicated does it really need to be?'

Gotta love that on-the-fly adjustment knob (try that with a laptop while driving alone, at speed, with the engine under load while trying to maintain a specific rpm or manifold pressure :D Of course if you're on a dyno that's a moot point)

Still, I'm not trying to twist any arms, so please forgive the drag OT.

The one you won come pre-programmed with any kind of map, or will you be starting from scratch?

How are you fixed for sensors and such, or will you be relying on the factory stuff you've got now?

Planning on using an A/F meter to keep track of things?
 

·
Registered
Joined
·
295 Posts
Discussion Starter #19
I have to say that the knob is mighty trick for those sds systems. On our dyno engines at work we can rig up something similar in our motec computers. BTW im used to motec which is worlds beyond both of these systems.

The ecu i bought has a map for a v8 mustand now, so I will be starting from scratch. I will also be using it to control ignition via coil packs.

I intend to modify spica throttle bodies and use a MAP sensor with them.

It will have an O2 sensor, possibly wideband, and I might buy a megaview which will display the lamba on my dash, or just rig up my voltmeter in the window
 

·
Registered
Joined
·
503 Posts
I've got the stock l-jet back in but have one of the wideband O2 outputs simulating narrowband for the Ecu. I could mess around with the AFM spring and tell you guys how much you are actually affecting it with so many clicks of the spring tension. (Air/Fuel ratio)
 
1 - 20 of 42 Posts
Top