Questions on ECU's, exhausts and remapping
Discussion
Just trying to get my head around a few things re: The cerb ECU / how it works.
Things I know.
-Normally in combustion you aim to hit an air to fuel ratio (AFR) somewhere just the rich side of a stoichometric mix, to ensure high power, but with a slight safety margin. (because if you run lean, just the other side of stoichometric, temps go very high and you bork the whole caboodle)
-The car has a 'base' map which aims to do this, but things like air pressure / temp / fuel quality can all shift this perfect balance, this is why we have lambda sensors.
-lambda sensors sit in exhaust and measure whether the exhaust gasses are too rich / lean and gets the ECU to modify the fueling/ignition to correct....
this is where the questions come in
So, in the Cerb, the ECU from what I gather has 'closed loop' (i.e. using lambda) under 3KRPM (is this correct or is it a bit higher?). Above this it will go open loop. What is it exactly that it adjusts? Just fueling amount, or ignition too (seems to on my unhealthy trace at the moment). What is the purpose of shifting ignition time? How does this interact with AFR?... and why can't it just run closed loop to the redline? Is the control system just not able to cope?
When running in closed loop we can see the adaptives, and we shoot for between -10 to 0 across the range. This tells us that the ECU is having to make adjustments to (OK I'm guessing here....) ever so slightly lean (?) out the mixture from what it 'thinks' a 0 adjustment would need, this means that the throttle positions and butterflys/airflow etc. are tuned to just slightly have a rich burn with no correction, but the ECU pulls it back even closer to stoichometric perfect burn.
The reason for this is that when you then blast past 3KRPM and go into open loop (the ECU makes no more corrections?) it will just burn at whatever it's physically setup to do by mechanicals, and you still want it to be slightly rich. Therefore if your car is unhealthy, and your adaptives were all +30% (i.e. the ECU was detecting lean burn and adding a lot more to richen it up) you'd be running very lean at high revs = very bad and potentially damaging to the engine.
Is this correct? So the reason to set the adaptives where we do is so when we go open loop we know we are 'safe' in terms of burn? Or have I just totally missed the mark and more goes on / the lambdas can still do things etc..
So given all this. What about changing the exhaust? The reason I ask is that next week I'll be the proud recipient of a full SP system (wooohooo). The decat and freer flowing system should hopefully help the car breathe a bit better, but what effect will this have on the car without remapping?
I assume that I can just set it up as per usual, get the adaptives spot on and use the 'base' map. The map may not squeeze every last HP out the car (4.2), but so long as I get it setup properly there is no danger of running the car dangerously (too lean somehow) is there? You don't *have* to get it remapped after an exhaust change to remain safe, do you?
I'm sure I could have googled some of this, but hopefully those of you who understand/know the cerb works (er.... yes spifireV8 I'm looking at you
) will be happy to chime in, as some of this may be cerb specific. I'm really hoping my manifold repair and new exahust system will help cure an erratic idle I have (fear I have air leak into manifolds through cracks upsetting lambdas), but if this doesn't then I'll probably have to look even deeper and more understanding is only going to help
thanks in advance, Stu.
Things I know.
-Normally in combustion you aim to hit an air to fuel ratio (AFR) somewhere just the rich side of a stoichometric mix, to ensure high power, but with a slight safety margin. (because if you run lean, just the other side of stoichometric, temps go very high and you bork the whole caboodle)
-The car has a 'base' map which aims to do this, but things like air pressure / temp / fuel quality can all shift this perfect balance, this is why we have lambda sensors.
-lambda sensors sit in exhaust and measure whether the exhaust gasses are too rich / lean and gets the ECU to modify the fueling/ignition to correct....
this is where the questions come in
So, in the Cerb, the ECU from what I gather has 'closed loop' (i.e. using lambda) under 3KRPM (is this correct or is it a bit higher?). Above this it will go open loop. What is it exactly that it adjusts? Just fueling amount, or ignition too (seems to on my unhealthy trace at the moment). What is the purpose of shifting ignition time? How does this interact with AFR?... and why can't it just run closed loop to the redline? Is the control system just not able to cope?
When running in closed loop we can see the adaptives, and we shoot for between -10 to 0 across the range. This tells us that the ECU is having to make adjustments to (OK I'm guessing here....) ever so slightly lean (?) out the mixture from what it 'thinks' a 0 adjustment would need, this means that the throttle positions and butterflys/airflow etc. are tuned to just slightly have a rich burn with no correction, but the ECU pulls it back even closer to stoichometric perfect burn.
The reason for this is that when you then blast past 3KRPM and go into open loop (the ECU makes no more corrections?) it will just burn at whatever it's physically setup to do by mechanicals, and you still want it to be slightly rich. Therefore if your car is unhealthy, and your adaptives were all +30% (i.e. the ECU was detecting lean burn and adding a lot more to richen it up) you'd be running very lean at high revs = very bad and potentially damaging to the engine.
Is this correct? So the reason to set the adaptives where we do is so when we go open loop we know we are 'safe' in terms of burn? Or have I just totally missed the mark and more goes on / the lambdas can still do things etc..
So given all this. What about changing the exhaust? The reason I ask is that next week I'll be the proud recipient of a full SP system (wooohooo). The decat and freer flowing system should hopefully help the car breathe a bit better, but what effect will this have on the car without remapping?
I assume that I can just set it up as per usual, get the adaptives spot on and use the 'base' map. The map may not squeeze every last HP out the car (4.2), but so long as I get it setup properly there is no danger of running the car dangerously (too lean somehow) is there? You don't *have* to get it remapped after an exhaust change to remain safe, do you?
I'm sure I could have googled some of this, but hopefully those of you who understand/know the cerb works (er.... yes spifireV8 I'm looking at you
) will be happy to chime in, as some of this may be cerb specific. I'm really hoping my manifold repair and new exahust system will help cure an erratic idle I have (fear I have air leak into manifolds through cracks upsetting lambdas), but if this doesn't then I'll probably have to look even deeper and more understanding is only going to help thanks in advance, Stu.
Afaik normal lambda sensors aren't very accurate and can only tell if the mixture is too rich or too lean around the stoich level, they don't tell you how much too rich or lean you are, for that you need a wideband sensor. When you give it the beans the car needs to run richer than it does on cruise, so the standard sensor would just show "rich", which is no good for tuning. The ecu ignores the sensor and it's adaptives when the throttle is more than some percentage (40% or so?) or when the revs are above about 4k. Otherwise, when cruising, it is used to help achieve the right mixture for economy and emissions purposes. So if the car runs poorly on part throttle but fine otherwise, the sensors are something to check. Even with a wideband sensor, no ecus feedback from the sensor at wot that I know of, but you can fit an AFR gauge to keep an eye on your mixture up to the redline. Richer mixture is especially important with forced induction.
Someone will be along shortly to correct the above
Someone will be along shortly to correct the above
The Cerbera ECU is an early development of the MBE 941
There is lots of stuff in the 'real' 941 that isn't in the Cerbera ECU but you can read about features on the MBE Systems web site. Just don't expect hard/soft cut, boost, Traction control etc
The Cerbera ECu uses closed loop lambda below 4krpm and below throttle site 11 (of 16 sites 0 to 15) and only at steady throttle. This can be changed in teh target lambda map but only to turn it on or off - it always tagets lambda 1. In 'real' 941s the target can be set to a specific value line 0.9
The adaptation adjusts the fuelling only, adding or removing a percentage from the base fuel map
The fuel and ignition are adjusted by the baro pressure, air and coolant temp compensation maps
The ignition is also adjusted by the Red Rose Octane switch. It reportedly retards the ignition by 4 degrees when switched to 95 octane. I fitted one to my 4.2 a couple of weeks ago and it was retarding by 3.7ish degrees - although tweaking the ECU it could retard by whatever amount desired.
Setting the adaptives to somewhere close ensures that the butterfly and throttle pot are also somewhere close to the positions used when the original mapping was done so that the base maps will be delivering sensible fuel and ignition values.
Moving the Tpot when the throttle is near closed changes the throttle site that the ECU 'sees'
At WOT changing the TPot has little effect on the throttle site because the 2D throttle map (an exponential curve mapping voltage to throttle site) ensures that a small movement at near closed makes a big change in throttle site.
In effect there are lots of throttle sites crammed into the first part of the butterfly travel and they spread out as it moves to WOT.
Like any induction or exhaust change a full remap will get the best of the combination of components but shouldn't be a necessity - although I would expect the supplier to give advice re the need for a remap.
I know that sometimes the linked decat on a 4.2 that TCS sold did need a map tweak. Some 4.2 cars had a flat spot. Fitting the tweaked chip sorted it out.
There is lots of stuff in the 'real' 941 that isn't in the Cerbera ECU but you can read about features on the MBE Systems web site. Just don't expect hard/soft cut, boost, Traction control etc
The Cerbera ECu uses closed loop lambda below 4krpm and below throttle site 11 (of 16 sites 0 to 15) and only at steady throttle. This can be changed in teh target lambda map but only to turn it on or off - it always tagets lambda 1. In 'real' 941s the target can be set to a specific value line 0.9
The adaptation adjusts the fuelling only, adding or removing a percentage from the base fuel map
The fuel and ignition are adjusted by the baro pressure, air and coolant temp compensation maps
The ignition is also adjusted by the Red Rose Octane switch. It reportedly retards the ignition by 4 degrees when switched to 95 octane. I fitted one to my 4.2 a couple of weeks ago and it was retarding by 3.7ish degrees - although tweaking the ECU it could retard by whatever amount desired.
Setting the adaptives to somewhere close ensures that the butterfly and throttle pot are also somewhere close to the positions used when the original mapping was done so that the base maps will be delivering sensible fuel and ignition values.
Moving the Tpot when the throttle is near closed changes the throttle site that the ECU 'sees'
At WOT changing the TPot has little effect on the throttle site because the 2D throttle map (an exponential curve mapping voltage to throttle site) ensures that a small movement at near closed makes a big change in throttle site.
In effect there are lots of throttle sites crammed into the first part of the butterfly travel and they spread out as it moves to WOT.
Like any induction or exhaust change a full remap will get the best of the combination of components but shouldn't be a necessity - although I would expect the supplier to give advice re the need for a remap.
I know that sometimes the linked decat on a 4.2 that TCS sold did need a map tweak. Some 4.2 cars had a flat spot. Fitting the tweaked chip sorted it out.
Edited by mikesr on Sunday 27th September 15:02
mikesr,
Cheers, top post, although perhaps somewhat bad news for me as my thoughts on the cause of my problems are probably wrong ......
I have two problems really:
-Poor ignition trace (generally flat, but 'spikes' away from where it should be). Makes idle a bit rubbish/will stall so I lifted to 1100rpm or so.
-Lumpy adaptive maps.
I had seen in the diagnostic software that lumpy adaptive maps can be caused by manifold cracks (I now know have have one in each manifolds!), so I hoped that if I whipped these out & repaired then it would sort both problems. I had put it down to air getting into the exhaust and confusing the lamdas & therefore messing about with the ignition/fueling etc... but if the ignition is NOT controlled by anything the lambda does, only fueling duration is (from what you said), my problem of spiky ignition will NOT be cured by fixing manifolds, although maybe the adaptives may settle?
So I guess if I do manage to fix the manifolds, I'll still have to try and work out the cause of the ignition spikes, and for that I should look to baropressure, air, and coolant temp data inputs and make sure they are all inline to where they should be?
Silly car
[The other benefit of getting the SP exhaust was to decat, as mine are probably getting quite tired now, and probably harming exhaust gas flow. I had hoped sorting the entire exhaust system would sort the car, but now you have me worried!]
Cheers, top post, although perhaps somewhat bad news for me as my thoughts on the cause of my problems are probably wrong ......
I have two problems really:
-Poor ignition trace (generally flat, but 'spikes' away from where it should be). Makes idle a bit rubbish/will stall so I lifted to 1100rpm or so.
-Lumpy adaptive maps.
I had seen in the diagnostic software that lumpy adaptive maps can be caused by manifold cracks (I now know have have one in each manifolds!), so I hoped that if I whipped these out & repaired then it would sort both problems. I had put it down to air getting into the exhaust and confusing the lamdas & therefore messing about with the ignition/fueling etc... but if the ignition is NOT controlled by anything the lambda does, only fueling duration is (from what you said), my problem of spiky ignition will NOT be cured by fixing manifolds, although maybe the adaptives may settle?
So I guess if I do manage to fix the manifolds, I'll still have to try and work out the cause of the ignition spikes, and for that I should look to baropressure, air, and coolant temp data inputs and make sure they are all inline to where they should be?
Silly car
[The other benefit of getting the SP exhaust was to decat, as mine are probably getting quite tired now, and probably harming exhaust gas flow. I had hoped sorting the entire exhaust system would sort the car, but now you have me worried!]
The compensation maps only change it by a few degrees but you would see any change in temps if you log the inputs with the engine running
When the ignition spikes is one or other of the throttle pot readings spiking at the same time?
That could take it from 17 to 27+ degrees if the TPot signal spiked to full throttle.
Similarly it could be the rpm input spiking and again logging it would show if thats happening
When the ignition spikes is one or other of the throttle pot readings spiking at the same time?
That could take it from 17 to 27+ degrees if the TPot signal spiked to full throttle.
Similarly it could be the rpm input spiking and again logging it would show if thats happening
Cheers, mike.
TPots look to be pretty ok - I've read up about looking at the throttle deltas - I'll see if they do anythign of concern.
Andy@APM actually replaced the crank sensor to try and sort the problem, and he initially thought it had, but it returned pretty quickly. It really is a bit of a conundrum - I think my best play though is to sort the manifolds, cats+exhaust & paint the chassis over the next week. I'll see where I am after that by trying a proper logging session
Stu
TPots look to be pretty ok - I've read up about looking at the throttle deltas - I'll see if they do anythign of concern.
Andy@APM actually replaced the crank sensor to try and sort the problem, and he initially thought it had, but it returned pretty quickly. It really is a bit of a conundrum - I think my best play though is to sort the manifolds, cats+exhaust & paint the chassis over the next week. I'll see where I am after that by trying a proper logging session
Stu
ro_butler,
Cheers for the input - makes me less nervous at bosching that amount of cash!
I guess your cats must have been fine to start with. I can only find 1 receipt for one in 2007 and I think the other is a lot older, so given on what Joo has seen with cerbs coming in with collapsed cats robbing oodles of power, I'm hoping I'll be in a similar position so will really get a performance increase from this. It may be that she's perfectly healthy underneath, but it seems that it's very difficult to tell without getting it on a dyno.... which will be the next thing to do in a few months when I've got all the other stuff sorted that I Want to do
Cheers,
Stu
Cheers for the input - makes me less nervous at bosching that amount of cash!
I guess your cats must have been fine to start with. I can only find 1 receipt for one in 2007 and I think the other is a lot older, so given on what Joo has seen with cerbs coming in with collapsed cats robbing oodles of power, I'm hoping I'll be in a similar position so will really get a performance increase from this
. It may be that she's perfectly healthy underneath, but it seems that it's very difficult to tell without getting it on a dyno.... which will be the next thing to do in a few months when I've got all the other stuff sorted that I Want to do Cheers,
Stu
stuthemong said:
so given on what Joo has seen with cerbs coming in with collapsed cats robbing oodles of power, I'm hoping I'll be in a similar position so will really get a performance increase from this .
Stu
Me, too. Mine was/is about 80bhp down, first thing when I get back this trip is to replace everything from the manifolds back, rattling cats and original back box. Watch this space .... Stu
Edited by Allandwf on Monday 28th September 20:48
stuthemong said:
ro_butler,
Cheers for the input - makes me less nervous at bosching that amount of cash!
I guess your cats must have been fine to start with
Yep, my cats were/are pretty much perfect. Mine has not been near a dyno since i had the zorst fitted but allegedly was 401bhpCheers for the input - makes me less nervous at bosching that amount of cash!
I guess your cats must have been fine to start with
with sports zorst and cats when it was remapped with short induction (I have the graphs for what they're worth!).
stuthemong said:
Allan,
If you want a cheap set of sports pipes with rear mufflers let me know. Mine will be up for sale, I think they are the ACT ones
Good luck chasing those horsies too - is yours booked into/been at Joo's?
Bit too far for me to get to Joos' I've got the exhaust ready, but thanks. Don't think you'll have any problem shifting it as a few of us seem to have a similsr problem.If you want a cheap set of sports pipes with rear mufflers let me know. Mine will be up for sale, I think they are the ACT ones
Good luck chasing those horsies too - is yours booked into/been at Joo's?
Edited by stuthemong on Monday 28th September 20:51
Gassing Station | Cerbera | Top of Page | What's New | My Stuff