VE tables vs. engine characteristics
Turbo Mini Forums

As set in TunerStudio (reset a few minutes ago) middle of screen says "multiply"



As viewed in MegaTune ( not a proper version as I have to use an early ini file



And in megatunix (look bottom left)

Schrödinger's cat - so which one am I ???

hum, so which are the issue, the M's or the T :/

I've just burned "multiply" using MTx and TS now recognises it even though it failed to burn it itself but said it had.....

There is something seriously wrong with TunerStudio.

Schrödinger's cat - so which one am I ???

perhaps a special bug that vanishes if you pay 30 quid....

You also have EAE enabled :(

Edited by Paul S on 22nd Aug, 2010.

Saul Bellow - "A great deal of intelligence can be invested in ignorance when the need for illusion is deep."
Stephen Hawking - "The greatest enemy of knowledge is not ignorance, it is the illusion of knowledge."

Edited by Rod S on 22nd Aug, 2010.

Schrödinger's cat - so which one am I ???

Medusa + injection = too much torque for the dyno ..https://youtu.be/qg5o0_tJxYM

Whilst you may feel better having found another reason to not use Tuner Studio, I do think that you really should not let anything talk to the ECU that you are not 100% sure about.

I would wipe TS and Megatune from your computers and your memory for all our sakes

Saul Bellow - "A great deal of intelligence can be invested in ignorance when the need for illusion is deep."
Stephen Hawking - "The greatest enemy of knowledge is not ignorance, it is the illusion of knowledge."

MTx has a very small number of users and I don't even know if there are actually people running engines with 3.0.3 and MTx. Practically everyone running 3.0.3 uses TS. So if there are inconsistencies, I would strongly suspect MTx over TS.

Jean

http://www.jbperf.com/

Well after nearly 8 hours I got to the bottom of it and it now runs as it should.

Paul, I would have happily carried on using MegaTune but last year that choice was efectively taken away from us when it could no longer accept the larger menu sizes of the improved code. Despite that, I still use it now to double check everything I load although I have to frig it with an early ini file to get it to work.

Jean, despite what you say both MegaTune and MegaTunix were telling the truth and TunerStudio wasn't. It took a good night's sleep to figure out why, and about 10 minutes this morning to prove it. It's down to the functionallity that was removed during the "upgrade" to 0.999.7 "Lite" and most likely wouldn't affect anyone who has paid to have the basic functionallity returned.

However, the fact that MAP=multiply had been screwed during that "upgrade" and not recorded wasn't actually the basic problem.

First, looking back at last years successful run, it's obvious the pulsewidth needs to be very low at idle



As you can see, the pulse widths were less than 2mS but I couldn't get below 3mS for the last week.

I tried re-burning the firmware with the earlier code, tried changing every setting I could think of (all this on the JimStim) but could not get below 3mS.

So I went back to basics, the fuel equation Paul quotes on the previous page, and loaded every single gauge that affects the equation and altered PRM, MAP (using a syringe) and recorded the VE table values and everything else in the equation.

Because I had vetoed everything to do with acceleration (Paul's prognosis of the fault) you can see the Gamma and enrichment gauges basically show 100% in all the tests,

1000RPM



2000RPM


5000RPM


5K plus RPM when it shifts to single pulse mode.

And just to prove I had reasonable values for CTS, MAT, TPS,


Ive done the maths on all those screen prints (too long to post) but in all cases, taking injector openning time into account, the pulse width is 1.5 - 2 mS too long.

So I went back to all the settings again looking for something that was adding 2mS and despite everyting "acceleration" turned off, I eventually spotted this,


Despite everything acceleration being turned off it had the mythical figure of 2mS showing on the screen....

But it's a RATE OF CHANGE of TPS graph so should have no effect (especially as acceleration is turned off) and there is no rate of change on the JimStim and minimal on my engine.

But low and behold.


As found,

and as input dropped from 100% to 0%


Pulse widths back as they should be......

Why this doesn't show on any of the gauges, why it works when acceleration is turned off, and why it is as large as 2mS is beyond me but, 8 hours later, I put the MS2 back in the car and it ran fine, AFRs around 13-14 at idle.

Schrödinger's cat - so which one am I ???

I did say that I thought that was the problem.

Edited by Paul S on 23rd Aug, 2010.

Saul Bellow - "A great deal of intelligence can be invested in ignorance when the need for illusion is deep."
Stephen Hawking - "The greatest enemy of knowledge is not ignorance, it is the illusion of knowledge."

Edited by Rod S on 23rd Aug, 2010.

Schrödinger's cat - so which one am I ???

Edited by Paul S on 23rd Aug, 2010.

Saul Bellow - "A great deal of intelligence can be invested in ignorance when the need for illusion is deep."
Stephen Hawking - "The greatest enemy of knowledge is not ignorance, it is the illusion of knowledge."

Well hot restarts are fine now, even with the inlet plenum too hot to touch.

The next issue is my IAVC control loop is screwed (or maybe the valve itself) as it's not resetting to zero between starts so the restarts are with a very high idle.

Next problem to fix before I address the VE tables, which I'm sure now is the way forward.

Paul, in your case (above, previous page) did you add the lines above 100% MAP (Fuel load) manually or did some other setting in MS2 create them for you ??

Schrödinger's cat - so which one am I ???

Saul Bellow - "A great deal of intelligence can be invested in ignorance when the need for illusion is deep."
Stephen Hawking - "The greatest enemy of knowledge is not ignorance, it is the illusion of knowledge."

Rod,

I'm glad you found out that I was wrong about which software to trust but I'm sorry if I caused any additional effort on your part. And I'm really annoyed (obviously not as much as you) by the root cause of this.

I'm glad to see you're making progress. Have you tried the test mode for the IAC to make sure it's behaving as it should? It might be missing steps so you may need to play with the settings but you may also have to bypass the current limiting resistors on the bottom of the MS card.

Jean

http://www.jbperf.com/

Jean,

The software "issue" was me not thinking entirely logically at the time (hence a good night's sleep and I realised) - part of the functionallity removed from TunerStudio during the "upgrade" to "Lite" is it's ability to read what's actually loaded to the ECU so whenever you click on any settings panel it shows you what it thinks should be there (most recent msq???) not what is actually there. The old Megatune has a button to read the settings from the ECU and MegaTunix reads the ECU by default.

The root cause is strange, in as much as it appears to be triggered by tiny fluctuations in the power supply affecting the TPS signal.

I presume the default of 2mS added is because it's about right for an american V8 - the default "Reqdfuel" loaded is 15mS so an erroneous 2mS won't make much difference, but it makes a hell of a difference to a 3.4mS (actually 1.7 per injector in sequential mode) setting.

Re. the IACV, I've got a spare identical one sitting on my desk ready to test on the JimStim tomorrow, I just have to make a suitable jig to mount it in to mimic the "stop" inside the valve body otherwise the pintle just winds out the end and falls on the floor when it's trying to zero itself....

Schrödinger's cat - so which one am I ???

well done rod and paul and jean for finding a solution ...

.shame there isnt a george... then we could change rod to ringo and have a set !

Medusa + injection = too much torque for the dyno ..https://youtu.be/qg5o0_tJxYM

Schrödinger's cat - so which one am I ???

I'm going to have to get smarter! Or less lazy and read things properly.

If something is worth doing, it's worth doing half of.

It was the TPS d/dt that I factored up by 10, but you are correct, you also need to significantly reduce the added pulse width.

At least this has helped me further understand AE and how it works!

Saul Bellow - "A great deal of intelligence can be invested in ignorance when the need for illusion is deep."
Stephen Hawking - "The greatest enemy of knowledge is not ignorance, it is the illusion of knowledge."

The TPSdot threshold has to be set to a relatively high value because even a difference of a few mV in the sampled signal will result in a significant TPSdot because the delta t is very small. And even without any noise from the electrical components there will still be some sampling noise which will always result in a non-null TPSdot.

Jean

http://www.jbperf.com/

Jean,

I think I understand - even though the TPS voltage spikes are VERY small in absolute voltage, the ones I have logged are very "sharp" and regular. That's using a standard PC PSU (so switched mode) for the power supply. It will have been worse on the car having an un-regulated battery charger connected to compensate for the knackered battery.

Because they are sharp spikes that means d"Voltage"/dt (ie, TPSdot) is going to be large and very regular.

What do you think is a suitable threshold or is it just trial and error ???

Rod.

Schrödinger's cat - so which one am I ???