Thoughts on poor running/ mixture?

[Hokie6]

I'm having some trouble getting my 930 running right with the correct mixture, and looking for some ideas as to what's going on with it.

I've recently been through the motor, and had the heads redone along with fuel system refurb (WUR, fuel dist, new injectors, etc). Rebuilt the waste gate and shuttle/pop-off valve and everything associated with the turbo system (other than turbo which seems to be good). I'm running a Megasquirt ignition with COP. Basically I've gone through everything on the car to make sure it is working corectly.

The fuel pressures (system, control (hot/cold), boost) are all w/in spec. However, it wants to run rich off idle, and looking at the data, it doesn't seem to settle down. It's not really "hunting", but at about 1500 rpm, the O2WB readings are bouncing between about 11 and 13 AFR. Idle jumps from 14 to 16 AFR. This is in open loop w/o the O2 signal. The Lambda system seems to be working correctly as shown on the bottom graph, where it runs about 50% DC, until opening the throttle plate where the DC drops to about 22% for 3 seconds and back to 50%. Not sure the FV is flowing well, but it is running. MAP shown is manifold vacuum.

With the O2 system enabled, the mixture does not stabilize and experiences the same behavior.

My first thought is to try adjusting the WUR to get the AFR back to ~ 14. Other than that, I'm kinda out of ideas as to where to go with this.

Any suggestions would be appreciated. Thanks -

Dale

[flightlead404]

It's unclear from your post if this is essentially a CIS car with the MS running only the ignition and acting as a data collector, or if the MS is doing something more. I'm going to assume you have essentially just a standard CIS K-Jet with Lambda and maybe an adjustable WUR. Please lmk if not so.

If the control pressures are all in spec I would not start playing with the WUR. It's doing its job and messing with it will just band aid something else. Would be good to know what the CPs are and how you're collecting that. Do you have a digital fuel pressure sensor the MS can read?

It sounds like you are using a simulated narrow band O2 sensor output off a WBO2 to drive the closed loop idle for the original CIS. Is that accurate?

Idle mixture is controlled by the control pressure and the little 3mm adjuster in the fuel head. This mixture setting is good until about 1,500 rpm when the butterfly gets over a port in the throttle body. I believe the CIS control box also comes out of closed loop around that RPM. So, fooling with the WUR to set idle mixture will mess up your mixture off idle.

It sounds like a lot of work was done on the engine recently, I'd suspect something changed there or maybe was reassembled incorrectly before I'd jump to the conclusion that a FV suddenly went bad.

Can you provide more details on your installation?

[Hokie6]

flightlead404 -

You are correct - My car is an '86 running original CIS with the Lambda control, and the MS system is controlling ignition and logging data (base advance set to 8 BTDC). WUR is adjustable.

I really didn't want to mess with the control pressure, but thought it might be an educational exercise to see if the mixture could be leaned out. Fuel pressures were measured with a gauge, and this is what I show (they haven't changed much since they were originally set):

Sys pressure: 94 psi (6.5 bar) (a little low)
Control pressure - Cold: 36psi @ 78F (2.5 bar @ 25.6C) w/ Boost: 26psi (1.8 bar)
I'll need to check the hot control pressure again, as I don't seem to have it written down, but I believe it was in spec.

I'm using a WB sensor with a NB converter driving the Lambda system.

The engine had a top end rebuild (I checked the bottom end, and everything seems to be in good condition). Most of the fuel system was in rough shape, so all was either rebuilt or replaced. The car has actually been running well for about a year now, though it has been running richer than I would have liked (12 - 13 or so), but it was fairly consistent. I was focused on the Lambda system not being able to control it, but you're saying that it is not in play off idle.

It had the very-rich-on-boost issue (10 AFR or less), and I was working on installing a boost fuel control switch to fix that when I ran into this problem. I have reversed every thing I did which didn't make any difference, so coincidence it would seem.

I'm starting to test everything that I can think of that might have some effect on vacuum or mixture. I recently ran a smoke test, and no leaks found. Also checked operation of the AAR, AAV, TPS and shuttle/pop-off valve and all good. The air plate alignment seems to be ok. The CSV is next to make sure it's not leaking.

I'm starting to run out of things that I can test here. I guess next thing would be to go through the FD to see if I can find anything there. My understanding is that the Lambda system still has to be working correctly for proper mixture even at speed (due to the design of the later FD)?

An unrelated problem that it seems to experience is low boost pressure, but I'll cross that bridge later -

Thanks for the response. Any help would be appreciated as I'm tired of beating my head against the wall!

[Hokie6]

Funny story while troubleshooting my rich-running issue (wasn't too funny then though) -

I had removed the IC and air cleaner assy, and had let the car sit for several days. I finally got back to working on it, and replaced everything to do more testing. The car started, but ran very poorly and wouldn't even idle - the idle screw being ineffective at allowing the engine to idle faster. I parked the car and walked away. I got back to it several days later, and again removed the IC and air box. While looking over it trying to figure out my next step, I noticed some sticks jammed in idle screw hole.



Couldn't figure out how sticks could get in there, but removed the TB and proceeded to operate on it. This is what I found:



Apparently a spider cricket (nasty critters!) had jumped in there, and then got sucked in when the engine started, blocking off the idle circuit. Traumatic ending for it no doubt.

Moral of story - check everything before re-assembly to make sure nothing has decided to move in!

[flightlead404]

Originally Posted by Hokie6

flightlead404 -

You are correct - My car is an '86 running original CIS with the Lambda control, and the MS system is controlling ignition and logging data (base advance set to 8 BTDC). WUR is adjustable.

I really didn't want to mess with the control pressure, but thought it might be an educational exercise to see if the mixture could be leaned out. Fuel pressures were measured with a gauge, and this is what I show (they haven't changed much since they were originally set):

Sys pressure: 94 psi (6.5 bar) (a little low)
Control pressure - Cold: 36psi @ 78F (2.5 bar @ 25.6C) w/ Boost: 26psi (1.8 bar)
I'll need to check the hot control pressure again, as I don't seem to have it written down, but I believe it was in spec.

I'm using a WB sensor with a NB converter driving the Lambda system.

The engine had a top end rebuild (I checked the bottom end, and everything seems to be in good condition). Most of the fuel system was in rough shape, so all was either rebuilt or replaced. The car has actually been running well for about a year now, though it has been running richer than I would have liked (12 - 13 or so), but it was fairly consistent. I was focused on the Lambda system not being able to control it, but you're saying that it is not in play off idle.

It had the very-rich-on-boost issue (10 AFR or less), and I was working on installing a boost fuel control switch to fix that when I ran into this problem. I have reversed every thing I did which didn't make any difference, so coincidence it would seem.

I'm starting to test everything that I can think of that might have some effect on vacuum or mixture. I recently ran a smoke test, and no leaks found. Also checked operation of the AAR, AAV, TPS and shuttle/pop-off valve and all good. The air plate alignment seems to be ok. The CSV is next to make sure it's not leaking.

I'm starting to run out of things that I can test here. I guess next thing would be to go through the FD to see if I can find anything there. My understanding is that the Lambda system still has to be working correctly for proper mixture even at speed (due to the design of the later FD)?

An unrelated problem that it seems to experience is low boost pressure, but I'll cross that bridge later -

Thanks for the response. Any help would be appreciated as I'm tired of beating my head against the wall!

Quite honestly you sound like a candidate for a FrankenCIS system running the eWUR and Lamda

I'm using a microsquirt for that and it runs my ignition too. Works very well.

[flightlead404]

Originally Posted by Hokie6

Apparently a spider cricket (nasty critters!) had jumped in there, and then got sucked in when the engine started, blocking off the idle circuit. Traumatic ending for it no doubt.

Moral of story - check everything before re-assembly to make sure nothing has decided to move in!

Wow! That's a riot! Thanks for sharing.

[flightlead404]

Originally Posted by Hokie6

flightlead404 -

You are correct - My car is an '86 running original CIS with the Lambda control, and the MS system is controlling ignition and logging data (base advance set to 8 BTDC). WUR is adjustable.

I really didn't want to mess with the control pressure, but thought it might be an educational exercise to see if the mixture could be leaned out. Fuel pressures were measured with a gauge, and this is what I show (they haven't changed much since they were originally set):

Sys pressure: 94 psi (6.5 bar) (a little low)
Control pressure - Cold: 36psi @ 78F (2.5 bar @ 25.6C) w/ Boost: 26psi (1.8 bar)
I'll need to check the hot control pressure again, as I don't seem to have it written down, but I believe it was in spec.

I'm using a WB sensor with a NB converter driving the Lambda system.

I have an 86 also with k-jet lambda running FrankenCIS controlling both eWUR and lambda

You can indeed control mixture with control pressure, but unless you vary it continuously (what FrankenCIS does) any changes are across the board.

Originally Posted by Hokie6

The engine had a top end rebuild (I checked the bottom end, and everything seems to be in good condition). Most of the fuel system was in rough shape, so all was either rebuilt or replaced. The car has actually been running well for about a year now, though it has been running richer than I would have liked (12 - 13 or so), but it was fairly consistent. I was focused on the Lambda system not being able to control it, but you're saying that it is not in play off idle.

You're saying after all the work and reassembly the car ran fine but was just a bit rich?

on the Lambda cars, it comes out of closed loop off idle (I forget exactly where) BUT the lambda valve is still powered on and runs at a 50% duty cycle. Unplugging the lambda would have it run at a 0% duty cycle, which (again rusty memory) would lower control pressure and therefore run rich.

cars that were originally lambda and they just unplugged it generally have other band aid solutions to overcome that mixture difference and often don't run well in either idle, off idle, or both.

Originally Posted by Hokie6

It had the very-rich-on-boost issue (10 AFR or less), and I was working on installing a boost fuel control switch to fix that when I ran into this problem. I have reversed every thing I did which didn't make any difference, so coincidence it would seem.

I'm starting to test everything that I can think of that might have some effect on vacuum or mixture. I recently ran a smoke test, and no leaks found. Also checked operation of the AAR, AAV, TPS and shuttle/pop-off valve and all good. The air plate alignment seems to be ok. The CSV is next to make sure it's not leaking.

I'm starting to run out of things that I can test here. I guess next thing would be to go through the FD to see if I can find anything there. My understanding is that the Lambda system still has to be working correctly for proper mixture even at speed (due to the design of the later FD)?

An unrelated problem that it seems to experience is low boost pressure, but I'll cross that bridge later -

Thanks for the response. Any help would be appreciated as I'm tired of beating my head against the wall!

The vacuum system is mostly used for ignition control, not mixture control. It does things like allow or block advance, retard at different temps. The one mixture control thing it does do is boost enrichment via the WUR.

[dos531]

Step 1, throw all CIS components in the trash

[Hokie6]

Quite honestly you sound like a candidate for a FrankenCIS system running the eWUR and Lamda

I'm using a microsquirt for that and it runs my ignition too. Works very well.

I've looked at it, and have some interest, but probably a ways down the road. I need to get it running well again first. My MS2/3 might be a bit overtaxed at the moment, but it's something I'd like to implement in the future if possible.

You're saying after all the work and reassembly the car ran fine but was just a bit rich?

Yes - It ran well, and then something happened. That's what I'm trying to figure out

on the Lambda cars, it comes out of closed loop off idle (I forget exactly where) BUT the lambda valve is still powered on and runs at a 50% duty cycle. Unplugging the lambda would have it run at a 0% duty cycle, which (again rusty memory) would lower control pressure and therefore run rich.

cars that were originally lambda and they just unplugged it generally have other band aid solutions to overcome that mixture difference and often don't run well in either idle, off idle, or both.

I don't believe this is correct. Based on everything I've seen and read, the system pretty much stays in closed loop when warm, unless accelerating or in transition areas (or under boost?) I have confirmed this with my SC using an AFR gauge. The system also runs at 50% DC with an open O2 sensor connection, unless with accel and WOT. Again, I have confirmed this with my car.

The vacuum system is mostly used for ignition control, not mixture control. It does things like allow or block advance, retard at different temps. The one mixture control thing it does do is boost enrichment via the WUR.

This is true, but I'm going through everything while I'm in there. Vacuum leaks can cause strange symptons, so making sure there are no leaks. I did find a problem with AAR, so I feel a little better about getting something fixed!

dos531 - Step 1, throw all CIS components in the trash

Nope. Another rabbit hole to go down that I'm not ready for. I've had good luck with CIS on my SC, and will stick with it for now -

[Speedy Squirrel]

Is there data available on WRO2 and the resulting switching sensor simulated voltage voltage?

[Hokie6]

Is there data available on WRO2 and the resulting switching sensor simulated voltage voltage?

Here is some sample logging that shows RPM, manifold vacuum, AFR, and converted NBO2. The resolution isn't the best as the sampling was low. Max boost was about 160kPA (which is about all I can get, though the boost gauge is pegging 1+ bar). WBO2 is bouncing between 14/15 AFR with NB switching 1/0.

[flightlead404]

Why does the AFR look normal but the narrow band output go crazy?

[Hokie6]

The NB is working correctly. When the mixture is rich (AFR< ~14.7), the output of the NB goes high (i.e 1V), and when mixture is lean (AFR> ~14.7), output goes low(0V). The NB output is constantly switching between these two states as the ECU corrects for mixture. This constant switching will average out to about 14.7 AFR on these cars. Hope that makes sense -

[flightlead404]

Originally Posted by Hokie6

The NB is working correctly. When the mixture is rich (AFR< ~14.7), the output of the NB goes high (i.e 1V), and when mixture is lean (AFR> ~14.7), output goes low(0V). The NB output is constantly switching between these two states as the ECU corrects for mixture. This constant switching will average out to about 14.7 AFR on these cars. Hope that makes sense -

ok wondered if that was it.

[gorskined]

when my wide band was going bad My afr always went to 14.7 most of the new bosch ones you buy on line are counterfeit and fail relatively quickly . i would replace with one from a known bosch product distributor. The real ones are registered when you buy them.

[Hokie6]

I've had this unit for a few years now, and am confident that it's good. It will reliably show AFRs from 10 - 17 while driving which all seems believable. I'm dealing with an issue that is causing it to be too rich. I wish it would show me 14.7 lol