Post by schercheeroo on Jan 26, 2023 22:51:35 GMT
I was looking though some old files and found some notes I made from when I looked through a friends megasquirt calibration that had been done professionally.
I'm going to share them here;
1) Boost limit and axis scaling
The MS2 comes with an MPX4250 2.5 Bar MAP Sensor. This will read a maximum of ~ 21.5 PSI of boost pressure.
The engine was calibrated to run at 20 PSI. This does not exceed the rating of the sensor, but it is at the edge of its stated operational range.
The sensor may be able to read higher than 21.5 PSI (I don't know this, I haven't tested it) and the manufacturer doesn't guarantee accuracy outside of this range, but I would prefer a safety margin of at least 5 psi to the maximum stated sensor range.
I feel this safety margin is necessary as changes in air temperature, barometric pressure and environmental factors can effect boost. So a 20 PSI calibration could be a 22 PSI calibration on a cold dry day winter day.
AFR and VE VE tables had the maximum break point set to 240 kPa (20.3 PSI of boost)
This is probably fine in normal running, but as mentioned above boost is not always constant or consistent. So if the engine does make a bit more boost on a slow transient in 6th gear, or because its 3°C outside and the intercooler is rejecting a load of heat,
there could be a situation where the engine is operating outside of its calibration space. The result being the ECU can not control the engine accurately.
But in reality, is the VE going to change much from 240 to 254 kPa (~2 PSI over boost)? I don't know, its probably not going to change much so it might be okay, but why take the chance? I don't imagine its too hard to set up the map axis and stick to the hardware limits.
The AFR table going out of range is less of a concern as the ecu will use the last value known value, and the AFR table was set up with sensible numbers. But again, a little forethought would give you a nice safety margin.
2)Boost control
The ECU was calibrated to run open loop boost control. That is the EBC duty cycle is determined by engine speed and TPS angle.
This control systems works well, and when set up correctly will improve drivability, improve fuel economy and reduce turbo wear.
However, this calibration appeared to have been set up with little care or attention to detail.
The map had common figures in each RPM column from 100% TPS to 0% TPS. Also the boost solenoid
control frequency was set to the incorrect value.
To be fair, this was probably all fine and it did appear to be working well. However, I was lucky enough to have a passenger ride on track in this vehicle and light turbo surge was clearly audible at part load.
I can't comment if it would be possible to fully stall the compressor on track (I suspect not in this case). It it was going to happen, it'd be more likely driving up a steep hill. In any case, there's no reason to calibrate upto the surge line and a minimum surge margin should be maintained.
The compressor maps for these turbos are available from the manufacturer so it wouldn't be hard to build some safety into the boost maps.
3) Acceleration enrichment
This was the reason my friend asked me to take a look at their calibration.
They were concerned about poor throttle response describing jerking in traffic, lack of power on fast throttle movements and in general poor drivability and an unpleasant experience.
I was able to apply the same settings as my vehicle (as the cars are running the same engine, ECU and injectors) and with a few more tweaks and about 30 minutes of on road testing the drivability was much improved.
In addition to the basic AE we also turned on the EAE settings which should help to improve throttle transients.
The lack of a proper acceleration enrichment calibration is concerning. I think this shows poor attention to detail, especially for something so fundamental to how the vehicle feels.
AE is first on my list of importance because if its not properly calibrated the rest of the tune doesn't matter as the car will be a pig to drive at anything other than full throttle.
4) Closed loop fuel correction
Closed loop fuelling correction was switched off. This was brought to my attention by the fact the engine, at normal operating temperature, was idling with an AFR of 12.5:1.
There are arguments for having the engine run in open loop at all times, engine protection from a faulty wideband springs to mind. But you can be intelligent with the closed loop setting and keep things safe.
So I set up some common sense parameters for closed loop fuelling and enabled over run fuel cut.
5)In summary
The map was a basic set up of VE and ignition tables which resulted in a safe to drive setup. Totally fine for fast road or track work.
Fuel economy and low speed drivability was poor as these were never setup to work correctly.
Axis scaling was poor and the engine was configured to operate at the edge of its MAP sensor capability giving no safety margin.
Boost control was set in a simplistic way which caused the compressor to operate at or near its surge limit at steady state.
The acceleration enrichment was not calibrated correctly which made the car unpleasant to drive as a daily driver.
Closed loop fuel correction and over run fuel cut where switched off, which is fine for a dedicated track vehicle but the features may be desirable on a daily driver or fast road car that sees some track time.
The VE table and ignition map were evidently calibrated on the dyno, but as closed loop fuelling correction was switched off there is some doubt about the VE table's accuracy.
The ignition map was almost certainly fine or at the least calibrated in a safe way. I can't make any comment about MBT or engine knock.
Other features not calibrated included idle rpm ignition advance, EAE, air temperature ignition compensation, closed loop idle (although I believe the IAC valve was deleted in this application), and environmental fuelling corrections.
I feel I've drawn a very negative picture, which is actually quite unfair to the calibrator.
In conversation, my friend said the mapping session was less than a full days work (only I few hours I believe) and only cost a few hundred pounds.
For that amount of time and money how much calibration work should you realistically expect to be delivered?
A well calibrated VE, ignition map, some basic but well set up AE and an intelligently configured TPS based boost table should be more than good enough for a safe and fun car that drives well.
Are other features like closed loop idle, closed loop fuelling, closed loop boost control, EAE and environmental corrections (hot/cold start etc) expected to be included as part of a mapping session?
If anyone is planning on getting their car mapped professionally, I would find out what is and what isn't included.
I'm going to share them here;
1) Boost limit and axis scaling
The MS2 comes with an MPX4250 2.5 Bar MAP Sensor. This will read a maximum of ~ 21.5 PSI of boost pressure.
The engine was calibrated to run at 20 PSI. This does not exceed the rating of the sensor, but it is at the edge of its stated operational range.
The sensor may be able to read higher than 21.5 PSI (I don't know this, I haven't tested it) and the manufacturer doesn't guarantee accuracy outside of this range, but I would prefer a safety margin of at least 5 psi to the maximum stated sensor range.
I feel this safety margin is necessary as changes in air temperature, barometric pressure and environmental factors can effect boost. So a 20 PSI calibration could be a 22 PSI calibration on a cold dry day winter day.
AFR and VE VE tables had the maximum break point set to 240 kPa (20.3 PSI of boost)
This is probably fine in normal running, but as mentioned above boost is not always constant or consistent. So if the engine does make a bit more boost on a slow transient in 6th gear, or because its 3°C outside and the intercooler is rejecting a load of heat,
there could be a situation where the engine is operating outside of its calibration space. The result being the ECU can not control the engine accurately.
But in reality, is the VE going to change much from 240 to 254 kPa (~2 PSI over boost)? I don't know, its probably not going to change much so it might be okay, but why take the chance? I don't imagine its too hard to set up the map axis and stick to the hardware limits.
The AFR table going out of range is less of a concern as the ecu will use the last value known value, and the AFR table was set up with sensible numbers. But again, a little forethought would give you a nice safety margin.
2)Boost control
The ECU was calibrated to run open loop boost control. That is the EBC duty cycle is determined by engine speed and TPS angle.
This control systems works well, and when set up correctly will improve drivability, improve fuel economy and reduce turbo wear.
However, this calibration appeared to have been set up with little care or attention to detail.
The map had common figures in each RPM column from 100% TPS to 0% TPS. Also the boost solenoid
control frequency was set to the incorrect value.
To be fair, this was probably all fine and it did appear to be working well. However, I was lucky enough to have a passenger ride on track in this vehicle and light turbo surge was clearly audible at part load.
I can't comment if it would be possible to fully stall the compressor on track (I suspect not in this case). It it was going to happen, it'd be more likely driving up a steep hill. In any case, there's no reason to calibrate upto the surge line and a minimum surge margin should be maintained.
The compressor maps for these turbos are available from the manufacturer so it wouldn't be hard to build some safety into the boost maps.
3) Acceleration enrichment
This was the reason my friend asked me to take a look at their calibration.
They were concerned about poor throttle response describing jerking in traffic, lack of power on fast throttle movements and in general poor drivability and an unpleasant experience.
I was able to apply the same settings as my vehicle (as the cars are running the same engine, ECU and injectors) and with a few more tweaks and about 30 minutes of on road testing the drivability was much improved.
In addition to the basic AE we also turned on the EAE settings which should help to improve throttle transients.
The lack of a proper acceleration enrichment calibration is concerning. I think this shows poor attention to detail, especially for something so fundamental to how the vehicle feels.
AE is first on my list of importance because if its not properly calibrated the rest of the tune doesn't matter as the car will be a pig to drive at anything other than full throttle.
4) Closed loop fuel correction
Closed loop fuelling correction was switched off. This was brought to my attention by the fact the engine, at normal operating temperature, was idling with an AFR of 12.5:1.
There are arguments for having the engine run in open loop at all times, engine protection from a faulty wideband springs to mind. But you can be intelligent with the closed loop setting and keep things safe.
So I set up some common sense parameters for closed loop fuelling and enabled over run fuel cut.
5)In summary
The map was a basic set up of VE and ignition tables which resulted in a safe to drive setup. Totally fine for fast road or track work.
Fuel economy and low speed drivability was poor as these were never setup to work correctly.
Axis scaling was poor and the engine was configured to operate at the edge of its MAP sensor capability giving no safety margin.
Boost control was set in a simplistic way which caused the compressor to operate at or near its surge limit at steady state.
The acceleration enrichment was not calibrated correctly which made the car unpleasant to drive as a daily driver.
Closed loop fuel correction and over run fuel cut where switched off, which is fine for a dedicated track vehicle but the features may be desirable on a daily driver or fast road car that sees some track time.
The VE table and ignition map were evidently calibrated on the dyno, but as closed loop fuelling correction was switched off there is some doubt about the VE table's accuracy.
The ignition map was almost certainly fine or at the least calibrated in a safe way. I can't make any comment about MBT or engine knock.
Other features not calibrated included idle rpm ignition advance, EAE, air temperature ignition compensation, closed loop idle (although I believe the IAC valve was deleted in this application), and environmental fuelling corrections.
I feel I've drawn a very negative picture, which is actually quite unfair to the calibrator.
In conversation, my friend said the mapping session was less than a full days work (only I few hours I believe) and only cost a few hundred pounds.
For that amount of time and money how much calibration work should you realistically expect to be delivered?
A well calibrated VE, ignition map, some basic but well set up AE and an intelligently configured TPS based boost table should be more than good enough for a safe and fun car that drives well.
Are other features like closed loop idle, closed loop fuelling, closed loop boost control, EAE and environmental corrections (hot/cold start etc) expected to be included as part of a mapping session?
If anyone is planning on getting their car mapped professionally, I would find out what is and what isn't included.
le sigh