Tuners Rejoice! Free Tuning For M4.4!


Recommended Posts

  • Replies 7.5k
  • Created
  • Last Reply

Top Posters In This Topic

Top Posters In This Topic

Popular Posts

Attention: The first 30 or so pages of this thread are outdated. Please refer to the M4.4 Wikia article where all the relevant information is currently being collated. Before asking any questions p

Why did I have to PM the guy I don't know him just discovered that this guy posted my results up with my registration number After I paid a trader to do work and map my car and I find out he is advisi

Crush it.

Posted Images

As far as I know, the only way is the an eprom programmer. I use a willem for reading the M4.4 Eprom

Hmm...I can flash firmware (with motronic suite), but I can't read? :blink: It's strange for me...

You cannot backup M44 software without removing the EEPROM. But you can download Motrionic software and find in the BIN folder you firmware or similar to it for your ECU part number

I don't know, but maybe firmware in my ECU has tuned, my car have blue injectors and 16t (instead of 15t), and max speed of my car 260+(instead of 235 for automatic cars)

It's very bad news for me...

Edited by HierOS
Link to post
Share on other sites

No problem, looks like tightmopedman already went through the same frustration i did, i am absolutely pissed, these are supposed to be genuine FTDI cables, apparently i got ripped off, atleast i can get back to tuning my car now :-D

Does this apply to the cable sold by GameAdvance, linked in the Wikia? When I tried to connect for logging with my 'new' laptop the driver said USBART as you mentioned before, and even after I copied the drivers from my old laptop, I cannot connect for logging with the VAG-COM. Pressing the yellow ;'connect' bar in TunerPro does nothing - previously it would turn pink and say 'connecting' - this is with a bin set to permanent logging, so even that shouldn't be necessary. With the cable connected, the COM port does now show the correct driver, though...

Link to post
Share on other sites

If you select 27C512, you can read the last 64kB in stead of only 32kB.

First 64kB does not contain communication routines, and the CPU can only see one 64kB bank at a time.

Therefore only the last 64kB is available through normal diagnose connection.

Link to post
Share on other sites

Guys, need a bit of your help please.

S70 with 16T Turbo, white injectors, 0.8594 injector constant (to have LTFT close to zero and LTFT_I as small as possible), by the wikia method. DO88 intercooler if it matters, all silicone hoses.

Been having knocking issues. Started with the rev5b manual T-5 software, and added slight modifications. I only altered three maps (lambda, load request and retarded ignition with 3 degrees in the 3000-4000rpm zone, where knock is appearing).

Still I have knocking problems, they start appearing after the third pull. I thought heat soaking but it's 10 degrees Celsius here.

I'm attaching the maps, maybe I've been too aggressive for the compression ratio ?

Boost is raising after 3000rpm (around 5PSI), going to 10-12PSI steady at 5500rpm.

Thank you a lot for your time !

maps.jpg

Edited by Midnight Caller
Link to post
Share on other sites

@tjwasiak

I'am wondering if you yourself really understand the whole load concept.

You can have the same load at different RPM's

Load is calculated by the ECU with the formula (mass air)/RPM * 1/(injector constant)* certain corrective constants.

For instance A load of 5 @ 2000RPM certainly isn't the same as a load of 5 @ 5000RPM.

Amongs other things: Volumetric efficienty @ 2000RPM is different from volumetric efficienty @ 5000RPM.

EGTs certainly won't be the same and therefore the tendency fo knock will be different.

@Piet

It seems you misunderstood what I wrote before - perhaps it is just because english is not our first language.

My idea to use at most 2 tables (in each switchable set of maps) still applies cleanly. Those maps should have engaged gear and engine speed as axis - one of them should output target load (so you can have different load target for standing still, 1st, 2nd, 3rd, 4th, 5th, 6th and even reverse - if switch would be wired into one of ECU inputs gear), while the second should output TCV duty cycle to achieve load as set/limited by previous one. How should the logic work? First map would just be new target load map instead of the default one. What additional functionality it would/should have? Setting target load to 0 at any engine speed and gear should be treated as engine speed limiter for that gear - for example you can set it up not to exceed 2500 RPM at reverse. That way it can be also used as "secondary" limiter for launch control when speed is 0. The other map ("gear dependant" TCV duty cycle) should work a little bit different then normal TCV DC map as its values should be set to achieve target load for each exact gear/RPM. If you press accelerator pedal "too much" so the load is being limited by the first map TCV DC should be set according to this (second) map (this DC map would be a TCV DC gear dependant limiter). For throttle opening such which will result in load below gear dependent limiter (first map) normal TCV DC pilot map should be used.

Now about the load at different engine speeds being not equal - of course you are right, but I have not said anywhere it is.

If you are not maxing load nor MAF so ECU can exactly differentiate load across full engine speed range you are able to fine tune fueling and ignition advance (so it will be 100% optimal). Once set properly at each different load and engine speed you will not need any gear dependent fuel/ignition map as different level of load achieved at each gear will be sufficient factor for ECU to use different fueling and timing at each gear. Still all those values would be 100% optimal. That way you will get probably different EGT at each gear but still it would different only because of different load. And it would be optimal EGT value as everything is optimally tuned!

S70-R, on 25 Feb 2015 - 6:46 PM, said:

(...) If you think that the temperature inside the chamber has no effect over the ignition time and duration, ok. I believe it has its effects.

For you the temps in the chamber, the IAT and coolant temps only matter to avoid melting a piston but has nothing to do with ignition and knock. OK...

Lowering requested load generally means lowering torque and power output unless you are exceeding the limits of the engine and its components of course. That's the opposite we want to achieve.

By the way, I've just come from the dyno.

With this tune I had virtually no knock on the road in a 4th gear pull.

I went to the dyno with this tune and I achieve roughly the loads as on the road.

With the dyno set for the longer pull duration (which in 4th gear it takes about 50 seconds) I had a horrible total of 25 knock events.

With the dyno set for the shorter duration (which in 4th gear it takes about 29 seconds) I had "only" 5 knock events.

I achieved roughly the same loads in both runs except at low rpms. The biggest difference is that in the longer duration pull I had more torque at low rpm due to the turbo was able to spin earlier.

Max torque values were 468Nm (longer duration) vs 453Nm (shorter duration) -> not a huge and significant difference. Torque curves were identically except at low rpms.

How would you explain the huge difference about knock?

For me the higher temps in the longer duration run are the culprit.

About the things I've read I'll search for it and post it.

@S70-R

Please show me where I wrote that temperatures do not have effect on ignition timing?

You have to look at it once again - it would be wiser powerwise to get lower (limited by IAT for example) load without knock/ignition retard then trying to burn more mixture ineffectively (because of suboptimal, retarded ignition timing). If you do not believe me just check it on your own - I can say "been there seen that" :)

Regarding your dyno session - have you logged all the parameters? If you seek explanation just compare IAT on the road and during both dyno pulls. Also how long does such a 4th gear WOT pull takes at the road? You found knock because of missing IAT compensation. If you did not logged EGT we should not talk about it but I am quite sure that for proper tune (such a one that would lower load because of too high IAT meaning you will not see any knock) your EGTs logged during road pull and both dyno pulls would be more or less same at each engine speeds.

Turboforslund, on 25 Feb 2015 - 7:26 PM, said:

(...) It's a bit scary to have more aggressive timing in lower gears if you at some stage planning to go uphill/drive heavy caravans etc. for a long period. Lower speeds means bad cooling of the intake air and that will result in heavy knock (like the longer duration pull...)

Load by gear is an excellent way of saving money on tyres and transmission, but boring...

// Turboforslund

@Turboforslund

What you are talking about is a bad tune. Optimal tune would not give you such issues as ECU would react to changing IAT lowering target load.

All your answers shows your are mixing speed density logic with how Bosch load driven ECUs work. You should not work against your ECU. When it is trying to limit your load or TCV DC, retard timing or add some fuel it does it on purpose. Most of the time the logic is made so you will achieve more power then if you make it to run at higher load level but rich like a pig and badly retarded. This is why it is much safer to use stock ECU then aftermarket standalone unit in daily driven car. Most of SA (even those most expensive on the market) will not do anything to save your engine from being destroyed when tuned just slightly off while even 20+ years old 8 bit Bosch unit will.

EDIT:

Regard not being able to read out a M4.4 without desoldering flash - why an OEM (Volvo, or any of its dealers) or Bosch would need to read out any ECU when they have access to any stock file they would ever need? It is however possible to reflash the ECU so OEM/Bosch could introduce new software revisions.

Edited by tjwasiak
Link to post
Share on other sites

Hmm...I can flash firmware (with motronic suite), but I can't read? :blink: It's strange for me...

I don't know, but maybe firmware in my ECU has tuned, my car have blue injectors and 16t (instead of 15t), and max speed of my car 260+(instead of 235 for automatic cars)

It's very bad news for me...

You can always buy a spare ECU and start over. Dialing in the blue injectors and the raised speed limit isn't very hard. At least you should have a relatively safe starting point using the rev5b bin.

Theoretically you should be able to retrieve the ignition and VE maps from the bin. I doubt that would be useful though, as you can't get to the target load or tcv duty pilot maps.

Link to post
Share on other sites

Applies to any Knock off FTDI chip i guess, would you like to try the modified drivers? they will allow you to install it, then you can use FTPROG to put the hardware ID back that gets set to 0000, after that it will start working like normal... so annoying.

Edit, the only reason i recommend editing the FTDI hardware ID back is because this driver is considered "unsigned" and may not work with everything

Does this apply to the cable sold by GameAdvance, linked in the Wikia? When I tried to connect for logging with my 'new' laptop the driver said USBART as you mentioned before, and even after I copied the drivers from my old laptop, I cannot connect for logging with the VAG-COM. Pressing the yellow ;'connect' bar in TunerPro does nothing - previously it would turn pink and say 'connecting' - this is with a bin set to permanent logging, so even that shouldn't be necessary. With the cable connected, the COM port does now show the correct driver, though...

Edited by Stevo11811
Link to post
Share on other sites

@Piet

Those maps should have engaged gear and engine speed as axis - one of them should output target load (so you can have different load target for standing still, 1st, 2nd, 3rd, 4th, 5th, 6th and even reverse - if switch would be wired into one of ECU inputs gear), while the second should output TCV duty cycle to achieve load as set/limited by previous one.

How should the logic work? First map would just be new target load map instead of the default one. What additional functionality it would/should have? Setting target load to 0 at any engine speed and gear should be treated as engine speed limiter for that gear - for example you can set it up not to exceed 2500 RPM at reverse. That way it can be also used as "secondary" limiter for launch control when speed is 0. The other map ("gear dependant" TCV duty cycle) should work a little bit different then normal TCV DC map as its values should be set to achieve target load for each exact gear/RPM. If you press accelerator pedal "too much" so the load is being limited by the first map TCV DC should be set according to this (second) map (this DC map would be a TCV DC gear dependant limiter). For throttle opening such which will result in load below gear dependent limiter (first map) normal TCV DC pilot map should be used.

That seems way more complicated than just using the full tcv pilot and target load maps for each gear you want to control seperately, as it is done now.

Gear and RPM as axis wouldn't be very practical either, you wouldn't have any throttle control.

Unless if you want to use it as a load limiter, which would be a much more primitive method imo.

Launch control has already been achieved by Simply Volvo some time ago via spark cut I believe (not sure about that one). Not that launch control is particularly useful on a fwd vehicle imo, unless you enjoy smoking up tires, for which you don't really need launch control anyway. But that's besides the point.

You seem worried about a lack of eeprom space. Have you taken a look at the disassembly yet? 8051 assembler is pretty efficiënt as it is and there's tons of space.

Link to post
Share on other sites

That seems way more complicated than just using the full tcv pilot and target load maps for each gear you want to control seperately, as it is done now.

Gear and RPM as axis wouldn't be very practical either, you wouldn't have any throttle control.

Unless if you want to use it as a load limiter, which would be a much more primitive method imo.

Launch control has already been achieved by Simply Volvo some time ago via spark cut I believe (not sure about that one). Not that launch control is particularly useful on a fwd vehicle imo, unless you enjoy smoking up tires, for which you don't really need launch control anyway. But that's besides the point.

You seem worried about a lack of eeprom space. Have you taken a look at the disassembly yet? 8051 assembler is pretty efficiënt as it is and there's tons of space.

Yes I mean to use it as a limiter. Is it primitive? It may seem it is but I disagree. You do not need any throttle control because up to wastegate pressure you are not able to control boost anyway (unless you move to vacuum controlled actuator connected with its own vacuum pump). When boost is higher you can control it using accelerator pedal up to limit set in that map for current gear. The funniest part would be that applying more throttle should (at least when boost stabilize) give you less boost then just the opening angle at which you get that load. Why? Because wider the throttle opened the pumping loses decreases... If you have normal TCV duty cycle map properly tuned (for your setup and desired load request per throttle opening) you just do not need to alter it for different gears as you can hardly change the way engine respond to throttle angle. That is why Bosch made a huge jump to DBW and torque demand model. For ME7 and newer ECUs my solution would be bad, inefficient and primitive, while yours with full torque demand (aka driver's wish) maps for each gear proper.

I am aware that you already have working launch control but with my approach many different things could be achieved using one logic which just have to be more efficient then having many simpler functions. While I am not able (or it would be really hard for me) to program in 8051 assembler I am doing my best to understand it as it is the only way I can hack ECU my car uses :)

BTW. Have you really tested how M4.4 reacts when fed with different IAT data? For example my car would even not start with IAT sensor disconnected - ECU would try to make it so rich you can hear detonations inside intake manifold. Once I had corrupted IAT sensor which made my car run lambda in the range of 0,7-0,75 when idling and cruising (it was really hard to make it going), my ECU is a little bit older then M4.4 (M2.8.1 was used from 1995 model year) so it is really strange that both behave so differently. I can also confirm I have IAT correction for ignition timing...

Link to post
Share on other sites

Just thinking about young and ambitious people and how much I have seen of them. Appearing and disappearing. Sigh.

@tjwasiak: Have you ever thought of a more respectful and thoughtful approach?

Edited by razorx
  • Upvote 1
Link to post
Share on other sites

Join the conversation

You can post now and register later. If you have an account, sign in now to post with your account.

Guest
Reply to this topic...

×   Pasted as rich text.   Paste as plain text instead

  Only 75 emoji are allowed.

×   Your link has been automatically embedded.   Display as a link instead

×   Your previous content has been restored.   Clear editor

×   You cannot paste images directly. Upload or insert images from URL.