MAP sensor question

[DarrylC]

Not upset at all, I was just surprised that he did what he has already complained about Jet doing this and then pleading ignorance.

As I have stated many times earlier I have him on a high pedestal and he has been very helpful.

As can be worked out from post 87 he was not onto it until I PMed him the info. Maybe he has been down this path before and it did not click until I PMed him and so did not realise I sent him the info. If so then I misinterpreted that he had used the info I sent him.

The post he put up was interesting but not the same IIRC. Maybe some went on to do this, Rob82 looks like he did it with the 3 bar t-MAP which is perfect for this and the 4 bar is not too bad.

What this gives is tuning up to about 410kpa. As I have emailed toads, boost is manageable from about 10psi to 45psi and all load, AFRs , ... are what the tuner expects. Low Boost, DFCO (although bugsy is making progress) and false MAP readings on the scanner are the issues but we have everything under control in normal driving and when boost goes from 22+ to over 32psi. At least you can vary the boost above 255KpA and not have to fudge the tune in the hope that the injectors, spark, ... will be OK.

I was hoping that someone would work it out and was happy to write it up and explain how and why the negative numbers are not an issue (ie how to work around them and still keep the tune as good as the factory SD). I went down this path and found the slightly negative numbers with bugsy's sensors and then thought about what they meant and guess what they can be worked around, in fact most of them occur where we cannot get to and the others at high map and are easily compensated for.

[IH8TOADS]

Negative offset numbers are a very big issue. You can enter a negative map at zero airmass in the software but the issue is the equation (and it being wrong from a physics perspective). Airflow = (Map - Offset)/Slope
What happens if the offset is negative..well you get a very wrong Airflow number as two negatives equal a positive. Does this actually happen.....it never got tested. But yes Darryl, you can manipulate the speed density values all you want to get the outcome to want...no rocket science there.

Do yourself a favour and ask Eric or one of his mates for Jerry Wroblewski's slope and offset recalculation spreadsheet he made for the Aussie SCT tuners many many moons ago. It will make your life much much easier.

In reality you guys confined to HPT aren't seeing everything that's there. B-series isn't too bad, but FG......believe me there is a lot missing and its when you start fudging major components like t-MAPs that issues may arise. B-series I wouldn't be concerned with but FG is a different kettle of fish and I would keep the tune simple to ensure you get a desired result.

[IH8TOADS]

Darryl I assume this this the part you are disgruntled about?

"HP Tuners is zero but I am hoping SCT can go lower"

I think anyone can pick up from my posts that I have a little bit of knowledge on this. An unwritten rule is to not share info from another person without their consent. I have probably done this before unintentionally but happy to be put in my place. I have learnt a lot from a ZF training course for example and was requested not to share that info....hence you don't see me posting on ZF parameters. I had shared injector scaling with Jet from a specific car, which he then later posted. That's dangerous as if used in an FG it would be lean as hell and could steer a novice in a very wrong direction. Don't put me in that boat because that's not where I fit.

[IH8TOADS]

Oh and by the way. I can actually edit the t-MAP slope and offset myself :-)

[DarrylC]

Do yourself a favour and ask Eric or one of his mates for Jerry Wroblewski's slope and offset recalculation spreadsheet he made for the Aussie SCT tuners many many moons ago. It will make your life much much easier.

In reality you guys confined to HPT aren't seeing everything that's there. B-series isn't too bad, but FG......believe me there is a lot missing and its when you start fudging major components like t-MAPs that issues may arise. B-series I wouldn't be concerned with but FG is a different kettle of fish and I would keep the tune simple to ensure you get a desired result.

I think item 1 may fall into the category of not intended for people outside the forum so even though I would have liked this many moons ago I did not pursue it and worked it out myself before I even knew it existed from one of your posts. Mainly from the US Patent, Paul Yaw's web page and answers from you. Item 2 is, unfortunately, what we have to live with and we need the FG tuners to put in requests to get this resolved but as you say they need to know what to request and without SCT that is tricky

Darryl I assume this this the part you are disgruntled about?

"HP Tuners is zero but I am hoping SCT can go lower"

I think anyone can pick up from my posts that I have a little bit of knowledge on this. An unwritten rule is to not share info from another person without their consent. I have probably done this before unintentionally but happy to be put in my place. I have learnt a lot from a ZF training course for example and was requested not to share that info....hence you don't see me posting on ZF parameters. I had shared injector scaling with Jet from a specific car, which he then later posted. That's dangerous as if used in an FG it would be lean as hell and could steer a novice in a very wrong direction. Don't put me in that boat because that's not where I fit.

We all know your expertise here and as I said in the PM "Probably enough for someone like you to work out what I did". This might have been my bad as I saw it as you were not sure why the 3.5 bar sensor was an issue, I send you a PM describing the actual issue and asking if you new if SCT had the same constraints, you respond with what I have sent you as to why this is bad. So probably my bad as it looked to me that you did use the information from the PM publicly. As I said probably my bad.

Oh and by the way. I can actually edit the t-MAP slope and offset myself :-)

Congratulations, well done. Boost sensor?

Negative offset numbers are a very big issue. You can enter a negative map at zero airmass in the software but the issue is the equation (and it being wrong from a physics perspective). Airflow = (Map - Offset)/Slope
What happens if the offset is negative..well you get a very wrong Airflow number as two negatives equal a positive. Does this actually happen.....it never got tested. But yes Darryl, you can manipulate the speed density values all you want to get the outcome to want...no rocket science there.

Some points here and I am interested in your feedback.

With the 3 bar these do not occur so should be good. Now the 3.5 and 4 bar sensors. The 3.5 I doubt I can get this to work but thought the same of the 4 bar and was wrong. The 4 Bar is a different story entirely. Yes it does have negative numbers but they occur at interesting points and are "not that negative". So the software converts these to zero and we are working from a difference between what is required and what is calculated and when you do the maths on these points it is not a biggie (on the B series at least) and you can easily compensate. I gave bugsy the tune in this format to check how things were going and it was good. Even idled at 600rpm so that gave me some idea on the bottom level of the sensor.

As you say the issue with the negative numbers is that the tune will be slightly lean. So, worst case point, at map reading of 320KPA (94.5 inHg), 5000 rpm then (MAP -Offset) differs by 0.82 so a difference in air mass of 0.9%, even less at higher boost levels. I was concerned at low MAP levels and when I checked where the tune would go the difference was not an issue.

I don't see how you get the wrong airflow number by negative Offset values unless you don't change the MAP input. If you do change the MAP sensor you are already in the zone of wrong airflow and this recovers the situation. The physics may be wrong but not the actual calculation.

Yes it is a fudge but there are two people who can change the sensor data.

P.S. Herrod just responded on another forum that they can do the 3 bar so maybe they can change the setting or just doing what I am. Will ask them.

[IH8TOADS]

SCT has the t-MAP slope and offset and is editable, just not in all strategies.

[IH8TOADS]

I just put pen to paper and you're right Darryl, negative offset values will (should) work. Does HPT accept negative offset numbers, or was that what you were inferring in your PM...that HPT won't accept negative speed density offset values?

SCT will but I have never tested is personally. Most others have just fudged a little more to keep them positive.

Edit. I just re-read your whinge before and I get what you're saying now. Surely HPT can get the offset table changed so you can test if negative values do work. I might test a tune with SCT and suck it out with HPT and see what it looks like.

[DarrylC]

Yes that was what I was inferring and unfortunately HPT wont accept negative numbers and silently changes them to zero. I raised a request with them to see if this was a PCM issue or their software. Their response was that they thought it was a PCM issue and I asked them to check as it did not make sense to me and showed then the calcs and reasoning. I then PMed you to see if SCT could do it as support for the request. Now that you have confirmed my suspicion then I will change it to a request along the lines of Incorrect/Missing parameters. Thanks for that.

I could not see why we could not put in negative numbers at all. It should just work. But the 4 bar sensor will go mighty close with zero instead of -0.8 as the lowest value.

It would have been so much easier if HPT could do this

The idea of testing would be brilliant and a great help, thank you. It would also be interesting to see what the "compare copy over function" would do with this data as well.

P.S. Herrods have confirmed they can change the t-MAP sensor data, not sure about the boost sensor.

[IH8TOADS]

In regards to the t-MAP, SCT can't access all strategies. Boost sensor is not accessible (but I haven't looked for a while).

[bugsy]

I will be doing some more testing in the next couple of weeks.
So far I have been able to run closed loop boost up to 30psi.
On the test run it had no timing into it as it was just a test.
It did not bring up any fault codes and felt completely normal to drive as if it was running the 2.55bar map sensor.
Iam hoping that the HPT guys get the chance to finish the map sensor scaling so that we can all get to benefit from it.

[IH8TOADS]

Bugsy, i'm hearing this is a GTX3582r? I really don't think a 62.5mm compressor will hold 30psi. Run a datalog against RPM and see what it's doing. 30psi manifold pressure should show ~63 inHG on the stock t-MAP calibration.

[Yortt]

"On the test run it had no timing into it as it was just a test."

Depending on what this means this will allow the GTX43582r 62.5mm to run 30psi as no timing will increase exhaust temperature and driive the turbo harder and "no timing" will substantially reduce the ability of the engine to consume the available compressed air backing the pressure up.

[IH8TOADS]

Your wastegate would be at 100%...and therefore essentially running open loop. The only closed loop feedback there would be would be to shut off any air to the wastegate!!

I know I probably come across as critical and negative, but im trying to be factual.

[bugsy]

I will get a data log for you guys when i put it on the dyno.
My AEM failsafe recorded a boost reading of 30.8psi.
The timing was set at 4? so maybe that pushed the boost up a little.

Within the next couple of weeks its going back onto the dyno so i will get more data and post it up for you to look at.

Darryl will have to let you know the table differences from the 2.55 bar data on the screen to what it will equal with the 4bar in it.
Toads
Feed back be it positive or negitve is always helpful.
I never take it personally.
Will get back to this with some results.

[DarrylC]

30psi manifold pressure should show ~63 inHG on the stock t-MAP calibration.

IIRC, 30psi should show up as a map reading of about 189Kpa once translated so about 56inHG.

Darryl will have to let you know the table differences from the 2.55 bar data on the screen to what it will equal with the 4bar in it.

OK here goes.

Scaling for original t-MAP sensor was MAP (inHg) = V * 15.9809 - 0.486377
Scaling for the new 4 Bar t-MAP was MAP (inHg) = V * 25.0679 + 2.23104

Now from the US patent the engineering facts are that at each point in the table the Airmass calculation is:

Airmass = (MAP - MAP@Zero)/"MAP per Airmass"

So we just need to convert what the old MAP@Zero says to the new value that the changed map sensor would read. So lets pick a nasty one as a starting point.

At 5000 RPM and 10 degress of cam angle we have a MAP@Zero of 1.7 and a MAP per Airmass of 17300.

so a reading of 1.7 inHG on the new map sensor would be:

1.7 = V * 25.0679 + 2.23104

So V = -0.021184

putting this back into the original equation would give:

MAP (inHG) = -0.021184 * 15.9809 - 0.486377
so the new MAP@Zero would be -0.824917

So we redo all the calcs and try and put the new table into HP tuners and it silently converts the negative values to 0.0. A bit confronting I thought so I logged a request with then to get the table altered to accept negative values. This morning Matt has responded that he has done it for my calibration. Anyway to go forward we still need to convert the MAP per Airmass values. This is a real simple calculation as we just scale the whole table by the ratios of the ratio of the old MAP slope/new MAP slope so new value would be:

17300 * 15.9809/25.0679 = 11028.8

All done!

I will leave it up as an exercise for the reader to do all the calculations.

Extra points for anyone who noticed!

The trick is not the calculation, it was going through all the tables in the tune to see if it would work (and hope that HP Tuners had all the ones I was interested in) and then analysing what it meant for the negative values. SFA as it tuned out

The 3.5 bar sensor is not so good because the negative numbers are not insignificant. The 3 bar does not have this issue at all and the 4 bar as stated is quite good. When HPT fixes the tables they will all be good and even better when Eric can fix the tune to allow scaling like the SCT boys are just getting.

I have heard of people scaling injector slopes to get fuelling up past the peg point, staggers belief!

P.S. had quite a few queries from the SCT boys

[bugsy]

As requested Darryl has explained it.
Thank you

[IH8TOADS]

Oh, I thought you had a 3.5 bar bugsy!

[bugsy]

Toads
I did have both the 3.5 and 4 bar.
Decided to go the 4 bar as it has a better range.
I am actually going back to the 3 bar after i fit my larger turbo as i will not be exceeding 30psi.
The 4 bar was all about testing to see if we could get it to run closed loop boost and that has worked fine.

You said that my gtx3582r would not push 30psi.

Running in open loop i have to set the wastegate duty cycle to 0.2 to keep the boost down.
Otherwise it will go past what i can read on my boost gauge.
Confusing I know but I can only go of what the boost gauge reads and how far i have to have the WG open to control it.

[IH8TOADS]

That's my stock engine BA XR6T. Just has cams, valve springs and GTX3582r on a 6BOOST manifold. You can see 30psi and I have done it also, however you won't be testing your closed loop system at 30psi with a 62mm turbo.

Have you modified any components of the PI system, proportional gain multiplier etc? Once you add a heavy actuator and change turbo's etc you don't want to be running the stock values. I have quite a few high speed runs in 4th and 5th playing with the factory PI system and it isn't a set and forget device if you want stable boost control.

[DarrylC]

Matt from HP Tuners has put the Map@Zero fix in for my strategy as well as bugsy's. So it looks like we will be able to use any sensor

He is also going to have a look at the scaling for the MAP sensors but we might have to wait until the non beta version of the software before it can be done generally.