Jump to content

Altitude Correction


Recommended Posts

As I appear to have knackered my back, and getting bored, I turned my thoughts to the upcoming 10CR.In particular the effects of altitude.

 

A bit of diging shows that at the top of Stelvio (2750m) there is approx 25% less oxygen. Some of that will be reduced as temps will be lower at altitude, but even so it explains why PI cars have issues choking up. 

However, my EFI doesn't have barometric correction. I, being simple, thought if I can work out how to switch closed loop on, that would take care of the matter. But I really do not want to give it 25% authority.....

 

The flipside is that last time I didn't have barometric correction either (at least, I don't think so, different ECU though) and things were not too noticeable. Maybe I should just cross my fingers....

 

Or any other bright ideas??

Link to comment
Share on other sites

The PI cars do just about ok IF they are running at ideal or on the lean side at sea level.  However, most are on the rich side to begin with.

 

Don't about your ECU but Mega squirt ECUs without barometric correction take a zero point when ever they are switched on.  So if climbing a big mountain it was possible to to switch off and re-start and re-zero and do manual correction that way.  No ideal if you are continually going up and down big mountains (as we will be), but possibly a workaround.  Incidently, if you don't have active idle control, they don't idle at the tops of the higher passes even with barometric correction.  I usually end up increasing the idle while we're in the mountains and turning it back down again when back to flatter lands.

 

Giving the ECU 25% authority is a bit scary but possibly ok IF your target table is dead right.  I know mine isn't.........

 

Nick

Link to comment
Share on other sites

Target table is a bit homebrew but sound principles. 15 at the cruise sites, and 14.7 everywhere else up to 25% throttle.And then gradually richer with WOT at high rpm at 12. How hard can it be  :unsure:

 

Worst part is I an headscratching about getting the closed loop switched on. Instructions are just not there in the manual, largely because they have not yet released their own version of a wideband setup.....But there are clues, I just need to be brave.

Link to comment
Share on other sites

Another MS has is a very fundamental setting called "required fuel" which is used when initially setting injector sizes, number of squirts er cycle etc.  This is very much akin to the mixture adjustment screw on an SU as changing it offsets the whole table up of down and it is a quick, easy way of making everything richer or leaner.  Better than messing up your map.....

 

Trust you back is mending?  2,500 miles in a small car with hard suspension and a bad back doesn't sound like a recipe for joy......

 

Nick

Link to comment
Share on other sites

Hmm, I did initially play with injector size to get the thing running half decently (the supplied so called "good" map had very wrong injector size and indeed fuel pressure)

 

So yes, I guess increasing the injector size by 10% in the settings will lean it off nicely.

 

I was also thinking if that was done at the outset, and give autotune 10% authority, it would correct at sea level and also at altitude. Then the realisation kicked in that autotune doesn't work when the engine is cold  etc etc....Ho hum. 

 

But tweeking the map is very quick, ideally during a coffee and cake stop.

 

As to back, slowly improving, physio booked for today. Sadly I still have 2 student houses to complete the refurb on by Monday, and things are a bit tight. Pulled in some extra help, and No2 daughter is doing a fantastic job of single-handedly assembling 12 bedrooms worth of ikea furniture. So I am placing myself in a supervisory role (ie tea maker) ......  I am hopeful I will be OK in time, and am stocking up on ibuprofen just in case.

Link to comment
Share on other sites

After some crossing of the Alps with my MegaSquirt1

the problem came up to me with the MS2.

Up to that point I did not even think about baro control on the Alps

 

I had a lot of trouble and still it does not work perfect.

The fun started with MS2 version 3.3

Just from 1000 metres I noticed leaning.

The altitude compensation on startup did not work.

I was advised at the forum to install a second baro what I did.

Now there was a compensation but much too much.

 

I tried a lot with the setup but the variables to set are very small compared

to what I would need to get it correct.

So I decided to cheat with the sensor. I tell MS it has a different sensor

than fitted and try to get the data with that better.

I am still fiddeling with the data but last test can be made in the Alps.

Hopefully this year I get the problems wiped out with a final test.

 

As you are running Alpha/n you might not have the problem I had

with the table using MAP/density.

At high altitude the upper table is no longer in use because the total air pressure is lower.

So the are above 85 is no longer touched and full throttle take the data from lets say 80

instead of 100 as it should be. It does not touch the 100 at this altitude.

 

So my engine leaned and when staying at higher altitude I increased the total fuel amount

but when crossing the Alps I extended the EGO control to 25% I got away with it.

Link to comment
Share on other sites

Mine is MS2 on version 3 hardware with a second identical map sensor.  It runs what is now a pretty old version of the original B & G code which I've stuck with because it works and is dead stable.  I just enabled it and left the defaults - in 2009.  Been to the Pyrenees and all over the Alps (5 times) without any issues.  It just works.  Dunno if it maintains exactly but it's near enough and we've passed a number of sick PIs on hills.......

 

Not much excuse for an ECU not to have Baro-correction these days IMO.

 

Nick

Link to comment
Share on other sites

Interesting thread!

I plan also to travel through the alps, but I don't know how altitude correction will react, as 1600ft is as much I can find in Luxembourg.

So I tried to anticipate by reading as much as possible in the forums.

I've to say that most threads are quite frustrating, as I have the impression that :

  1. very few people seem to know, what's really going on in terms of altitude compensation and BaroCorr
  2. many people are confusing / contradicting etc.
  3. i might just be too dumb to understand anyway

But since I managed to map the Triumph in ITB mode, I still have a little hope that, one day, I might  hunt it up the the alps, without the need to role backwards home again.

 

So, yesterday I  read in the German MS forum a note from last year, that was making sense to me.

 

 

What strikes the author in most "baro correction" discussions is  that :

  1. different  algorithms  tend to be compared
  2. impact of "multiply map" aren't taken into consideration
  3. people think that MS SW include a specific baro correction algo. This is not the case, at least not for the later FWs. The baro correction curve is nothing but a simple multiplier. 100% means no correction , 105% means 5% enrichment. etc.

 

As an example: For a reduced atm. pressure:

  • SD Algo --> Table will be strongly leaned out, as the load = f(Map)
  • Baro%-, ITB- and AN- modes won't do that. For Baro% mode and ITB mode the load will hardly change apart from some minor non linear effects.  For AN the load remains exactly the same.

Effect of "Multiply Map":

 

Ex.:

  • Baro%- or ITB- or AN- algo
  • Base: 100% load at 100kPa
  • Instead of the 100KPa Baro we have 90kPa ambient pressure
  • For simplification the MAP signal at WOT will change from 100kPa to 90 kPa

With "Multiply Map" switched "on":

  • The effectively injected fuel (PW - DeadTime) is proportional to the MAP-value.
  • MAP goes down as ambient pressure goes down, inducing thus already a lean out
  • So the engine will run 10% leaner due to "Multiply map"
  • This might or will probablly be too lean.
  • So we set for instance the BarCorr  at 103% for 90kPa Baro
  • The injected fuel will be: 0,9 * 1,03 = 92,7%
  • This means, that the mixture leans out by 7,3%, although  the BaroCorr curve actually made the mixture richer.

Now with "multiply map" switched off:

  • The effectively injected fuel (PW - DT) is NOT proportional to the MAP signal.  Only the VE-table defines the VE value (prior all the other corrections of course...)
  • So the engine is not leaned out via "multiply map". If we set the BaroCorr curve for  instance to 92,7% at 90kPA Baro (ambient pressure)
  • The injected fuel will be: 1*0,927=92,7%

The mixture is impacted in both cases in the same manner, but:

  • in one case with a enriching BaroCorr
  • and with a leaning BaroCorr in the other case 

 

If the SD-algo is used instead, then one needs to "enrichen" even more  via the BaroCorrcurve, in order to compensate for a mixture that would lean out even more.

 

Food for thought, lapidary stuff, or simply bollocks?

 

Cheers

Patrick

Link to comment
Share on other sites

Hi Patrick, I am still working on the basics:

How to get the correct baro information into the MS2.

 

What I found out is with single sensor, speed/density and initial MAP reading

the initial MAP was not accepted, although the range where it should do so

was set very wide.

 

Simply said standing in the Alps I stopped, fitted the laptop and looked for the

air pressure that dropped. Although that was displayed the MS it took the wrong

value 100 when restarting and not the reduced pressure. So no compensation

was done and engine leaned.

 

So I fitted the second baro with required setup and the values are now accepted in the MS2 but

now the calculation is pretty wrong! I can not play it with the adjustable values that all keeps in limits.

Set properly here at sea level resulted in wrong values in the Alps, but opposite to before!

Without 2nd baro it leaned, now it goes rich in the Alps!

 

So my idea was to cheat with the baro. It now tells the MS at 2000 metres that we are

only 1000 metres high and so enrichens less.

The trick is to bend the altitude curve in that way that it will fit at home and in the Alps.

Fiddeling in the Alps to get it perfect there resulted in disappointing behaviour at home.

 

What I have now is a set of data bundles to feed the 2nd baro setup in the MS that will all give

correct data at home but give different enrichments in the Alps. So I can choose the best data

at the Alps and drive down and keep the proper setup.

 

I agree, I was sitting there asking myself why do I have these problems and all the others do not?

So I went through all my settings, asking if I do something rare or special and asking what might

be the reason that with a correct declared 2nd baro the data are still wrong.

 

Happy with MS1, no reaction with initial reading on MS2 and wrong reaction with 2nd baro, I miss the

red line what connects all and what I did wrong during my improvements.

 

-Up to now I still have no answer-

Link to comment
Share on other sites

Hello all

            I run Microsquirt on my Spitfire 1500

 

I fitted a second MAP sensor same as first (Magneti Marelli PRT 03/04  105KPA) and set it at these figures that I found on the Microsquirt forums for Barometric Correction

105=101

100=100

95 =97.5

90 =94.7

85 =91.0

80 =87.5

75 =84.0

70 =80.0

65 =74.0

 

 

 

We went up to 5000 feet several times in Spain and it was Bl*****y hot and Spitty seemed ok (no black smoke and pulling ok up the hairpins etc)

 

I know my settings are on the rich side as well.

 

The only time I had trouble was when I filled up with 95 which I think is E10!!!!!!! but was ok again when back to 98 E5 (I only use 95 E5 at home)

 

Roger

 

 

post-1299-0-32655000-1504087649_thumb.jpg

post-1299-0-53614800-1504087772_thumb.jpg

Link to comment
Share on other sites

Hi Andreas, hi Roger,

 

  • what firmware are you using in your MS?  (Does Microsquirt corresponds to MS2 extra?)
  • Is "Map Multiply" switched on or off

 

Roger,

  • what algo are you using Speed Density, AlphaN, %Baro, ITB...?

 

Cheers,

Patrick

Link to comment
Share on other sites

Hi Andreas, hi Roger,

 

  • what firmware are you using in your MS?  (Does Microsquirt corresponds to MS2 extra?)
  • Is "Map Multiply" switched on or off

 

Roger,

  • what algo are you using Speed Density, AlphaN, %Baro, ITB...?

 

Cheers,

Patrick

Hello Patrick

                    I am using ITB

Link to comment
Share on other sites

Hi Roger,

 

Is "multiply Map" switched on and what Firmware are you using?

 

Patrick

Hello Patrick

                    Yes Multiply Map is on I just went with it on as it says 99% of installs need it?(and most of it is above my head!)

 

Firmware MS2/Extra 3.4.1 release 20151223

 

I have done 4000+ miles now

 

I still need to rework the throttle linkage as it is to sensitive(I need to make a cam style linkage)winter job?along with new engine?

 

I also still have the small fault that causes the tick over to go to about 1300 rpm just after full warm up is reached? but if I turn the ignition off and on it clears?

 

I have also fitted a led were the choke knob used to be so I know when its reached about 120/130 F

 

That,s just reminded me I need to lead off the first bit of the cold start enrichment it about AFR 10.0 at the moment and I can smell petrol

 

But other than that it starts after 2 complete revolutions as it supposed to and with no touching of the throttle and seems to drive off smoothly(other than the very light throttle) I need some of those racing driver shoes!

 

Roger

 

Roger

Link to comment
Share on other sites

Hi Andreas, hi Roger,

 

  • what firmware are you using in your MS?  (Does Microsquirt corresponds to MS2 extra?)
  • Is "Map Multiply" switched on or off

 

Roger,

  • what algo are you using Speed Density, AlphaN, %Baro, ITB...?

 

Cheers,

Patrick

I have MS2 3.3 Grant Slender 2.8 what is a modification for TPS smoothing

I have Speed density and no multiply

Link to comment
Share on other sites

I have MS2 3.3 Grant Slender 2.8 what is a modification for TPS smoothing

I have Speed density and no multiply

 

 

 

Hi Andreas,

 

My understanding with BaroCorr is that it does not correct the PW by default.

The Baro curves are laid out to generate no correction out of the box.

The user has to adapt this curve, since each engine requires specific Baro correction.

 

What may have happened, when you were using the single MAP sensor at startup for Baro measurement, is that the Baro value was well measured and displayed, but it had no impact as the Baro correction inf the Baro curve for any measured Baro value is 0,

Unless you changed the curve...

 

So no wonder you had  to increase the fuel in the VE table to pass the alps the first time, as SD depends only on MAP and will be for instance only be 80kPa at WOT. This will then also command only a VE for 80kPa load at WOT

It's good you have "Multiply MAP" switched off, otherwise it would have been even leaner.

 

 

For your latest setup,  with discrete Baro sensor, you tweaked the Baro curves, if I understand well.

Don't forget, that with pre MS 3.4.x FW, the Baro correction is not a muliplier, but it devides the PW, by the Baro correction value.

 

Below are quotes out of the release notes from the 2 latest FW families.

 

MS2/Extra 3.4.x firmware

Key changes and Gotchas

=======================

1. Baro

Previously the code used to divide by the barometer when calculating the fuel

pulsewidth. If using an old tune you need to enable the "old style" baro

calculation to enable the old behaviour.

 

 

MS2/Extra 3.3.x firmware

Key changes and Gotchas

=======================

2. Barometric correction

The recommended settings for baro correction in 3.3.x have changed.

100% means un-altered fuelling.

 

If you are upgrading from a previous firmware version, to change you settings

to the current method:

a. Basic/Load Settings -> barometric correction is typically all 100% with

small adjustments up or down as required.

b. Tools -> Calibrate MAP/baro

"At total vacuum" = 0

"Rate" = 0.

You must perform steps a AND b. See also the tooltips [?] on those pages.

 

 

Cheers

Patrick

Link to comment
Share on other sites

That's very interesting Patrick.  Thanks for posting that up.

 

I think something to beware of with this and with Megasquirt in general is that some of the features work in very different ways dependent on the version of firmware you have, so you have to check the details written for the actual version of firmware you are using.

 

My Vitesse MS2 ECU uses B & G firmware.  It's a bit basic by current standards, but it does nearly everything I want (no over-run fuel cut-off) and it's stable and reliable.  My PI used the same initially and then I switched to what was the latest version of Extra at the time to gain the additional table resolution, fuel cut etc.  I had all kinds of problems with it and nearly went back - then I realised I was doing stuff wrong because I assumed it worked the same way I was used to.  Some things did, but not all and once I realised this, things went better!  I never did find that firmware as stable, though I was never quite sure if this was a "noise" problem with the install.  Felt like the issues started with the Extra firmware.

 

For example, with the B & G code I'm pretty sure that the Baro correction function works without having to mess with the curves.

 

Nick

Link to comment
Share on other sites

Create an account or sign in to comment

You need to be a member in order to leave a comment

Create an account

Sign up for a new account in our community. It's easy!

Register a new account

Sign in

Already have an account? Sign in here.

Sign In Now
×
×
  • Create New...