GTR Forum banner

41 - 60 of 67 Posts

·
Registered
Joined
·
3,522 Posts
Superb, have replied to Ben. Final version with 0.01 difference in fuelling and 1 degree off the timing. Boost is perfect, just enough to max your injectors.
nice one so you cracked with the aftermarket actuators..so in theory i should run a faster time than my 10.9 lol
 

·
Registered
Joined
·
1,216 Posts
I hope so as you're now holding a little more top end boost with 13% extra wastegate duty cycle at redline, but after a gearchange it is only similar to before. Yours is a gentle massage, nip and tuck, watch your mirrors for the stock actuator boys, just as Ben should watch you in his mirrors...

At least I can control the **[email protected]$ now, but as I put on the NAGTROC thread, I'm thinking actuators may well not be much use with stock injectors.

However, you may find if you chose to upgrade the injectors that extra boost up there we could run (you're only at 33% wastegate duty cycle) may not add much more power, and you'd be maxxing the MAF sensors too. Slippery slope, suggest leave as it is.
 

·
Registered
Joined
·
3,522 Posts
I hope so as you're now holding a little more top end boost with 13% extra wastegate duty cycle at redline, but after a gearchange it is only similar to before. Yours is a gentle massage, nip and tuck, watch your mirrors for the stock actuator boys, just as Ben should watch you in his mirrors...

At least I can control the **[email protected]$ now, but as I put on the NAGTROC thread, I'm thinking actuators may well not be much use with stock injectors.

However, you may find if you chose to upgrade the injectors that extra boost up there we could run (you're only at 33% wastegate duty cycle) may not add much more power, and you'd be maxxing the MAF sensors too. Slippery slope, suggest leave as it is.
injectors for my next shop lol http://www.g-saurus.jp/02_2_r35/02_2_R35_bosch.html .. what maf sensors are the USA guys using then ?

and whats your take on these ƒUƒEƒ‹ƒXR35
 

·
Registered
Joined
·
1,216 Posts
LOL.

Ask on NAGTROC about MAF sensors/bigger tubes, start a new post, I've not seen anyone talk about them much. I suspect most are being naughty and tuning with them pegged at 5 volts all the time which is simply bad tuning IMHO, you don't have proper control of your fuelling resulting in lean condition and knock before the ECU corrects the fuelling and the timing.

Without live tuning (coming at some point) it may be time consuming to dial in uprated MAF sensors, for now I would not attempt this on the stock ECU. But I wouldn't take the grenade approach of just running the stock ones maxxed out.

You really are at the sensible limit on the stock ECU for now. Slippery slope beckons you strongly, resist my friend.
 

·
Registered
Joined
·
3,522 Posts
LOL.

Ask on NAGTROC about MAF sensors/bigger tubes, start a new post, I've not seen anyone talk about them much. I suspect most are being naughty and tuning with them pegged at 5 volts all the time which is simply bad tuning IMHO, you don't have proper control of your fuelling resulting in lean condition and knock before the ECU corrects the fuelling and the timing.

Without live tuning (coming at some point) it may be time consuming to dial in uprated MAF sensors, for now I would not attempt this on the stock ECU. But I wouldn't take the grenade approach of just running the stock ones maxxed out.

You really are at the sensible limit on the stock ECU for now. Slippery slope beckons you strongly, resist my friend.
am happy for now HONEST and if i decide to mod anymore it will be the full WORKS turbos, gearbox gears, ecu..
 

·
Registered
Joined
·
57 Posts
So here's my graph for the day, ;) this taken from a full throttle run up through the gears.
Rich - out of interest what RPM where you changing gear? Looking to generate a similiar/comparable set of graphs for the RPM based maps that Ben has very kindly produced for me.
 

·
Registered
Joined
·
669 Posts
I "attempt" to change gear on the Red light, which I think is 7000? I change on the amber 1st to 2nd at things happen so fast. ;)

The log usually shows just over 7000 in 1st and on or just under 7000 for the other changes.

In hindsight for testing it would be better to put it in Auto and let the car do the changing, that would then give very consistant results between runs. I think this changes slightly earlier? Will try this on my next run.


Rich
 

·
Premium Member
Joined
·
5,031 Posts
I think for the lower gears at least, Auto is supposed to be optimum for keeping the revs in the peak torque > peak power range.

For higher gears, you may want to switch to Manual to hold the revs longer.
 

·
Registered
Joined
·
669 Posts
Question for Thistle?

Following on from various logging sessions and the VPower Tesco test I have been staring at graphs trying to work out what is going on? One thing that strikes me is how similar they are.

Could you explain, in simple terms, for some of the key parameters which are the varibles that are read out from a map and which ones are derived dependant on feedback from sensors in the car?

So for example the WG duty if I overlay one graph over another is virtually identical, so I assume that this is an input looked up v RPM?

Wheras Ignition advance although very similar seems to joggle around a bit. Quite separate from the joggling I think you can see when the ignition is retarded after a knock event, perhaps you could confirm?

Boost also looks incredibly similar between runs, but has some variance so I assume it is an output rather than a looked up value? But what determines the boost at any given point? I still don't understand why it tails off before the WG duty gets to 99%. Is it that it is rarely limited by the wastegate?


I hope the questions make sense? I guess what I am asking in a roundabout way is how does it work? A brief explanation of how the ECU looks at tables, data from sensors and adjusts the outputs?

Finally what is this Theoretical Pulse we are logging?


Many Thanks


Rich
 

·
Registered
Joined
·
108 Posts
Question for Thistle?

Following on from various logging sessions and the VPower Tesco test I have been staring at graphs trying to work out what is going on? One thing that strikes me is how similar they are.

Could you explain, in simple terms, for some of the key parameters which are the varibles that are read out from a map and which ones are derived dependant on feedback from sensors in the car?

So for example the WG duty if I overlay one graph over another is virtually identical, so I assume that this is an input looked up v RPM?

Wheras Ignition advance although very similar seems to joggle around a bit. Quite separate from the joggling I think you can see when the ignition is retarded after a knock event, perhaps you could confirm?

Boost also looks incredibly similar between runs, but has some variance so I assume it is an output rather than a looked up value? But what determines the boost at any given point? I still don't understand why it tails off before the WG duty gets to 99%. Is it that it is rarely limited by the wastegate?


I hope the questions make sense? I guess what I am asking in a roundabout way is how does it work? A brief explanation of how the ECU looks at tables, data from sensors and adjusts the outputs?

Finally what is this Theoretical Pulse we are logging?


Many Thanks


Rich
I can answer most of your questions Rich,

The ECU has a very dynamic Timing control system that will add or remove timing based on gear, coolant temp, boost, RPM, knock, and a few other items. This is why you'll see slight variations in the timing cure during different runs. In the tuning software that we supply to our tuners there are base ignition maps that give the ECU a starting point for timing and then the ECU will dynamically advance or retard the timing depending on the different variables. The two variables that will affect your timing the most are temperature and knock. The hotter your motor is the more detonation you'll see requiring you motor to run less timing. Same with Knock the more Knock your ECU sees the less timing it will run. This is done to save your motor!

As far as the boost and WGD, were are still working through the logic of how that system works but as you said the boost curve and the WGD seem to be a very non dynamic system. There are a few other tables that deal with barometric compensations and temperature compensations but so far the cars seem to run the boost that we tell them to run.

The theoretical Pulse Width (TPW ) from everything that I have read and been able to tell is a theoretical perfect pulse width for the given moment. I think the easiest way to describe the behavior that I have see is, there is a calculated amount of fuel that the car needs to run for a given load, that value is used to find out the correct timing to run to burn all the fuel, valve timing, and a few other parameters that the ECU needs to figure out how it is going to run in order to burn all the fuel. If you log TPW and Injector Pulse Width (IPW) you'll see that they are very very close to the same. The only time they vary is when you've been on the gas and you lift your foot off of the gas. The IPW will drop more then the TPW.

I hope this post has been informative and answered your questions. All the information above is information that Trey, Thistle and I have figured out. It might not be correct but it is how I understand the systems to work.

Joe
GT-R Hacker
Cobb Tuning
 

·
Registered
Joined
·
669 Posts
Hi Joe

Thanks for the explanations, can I check a few things?

First off it sounds like although when tuning you have control over a lot of variables the "ECU" retains contol / overrides some of what you attemp to achieve to protect the engine?

So if I understand it correctly the dynamic Timing control system that you describe takes account of a whole load of variables and chooses to compensate for / back off the timing, and you have no control over this / how much it intervenes?

Also thinking back to some earlier problems with boost cut am I right to assume the "ECU" will overide / back of the timing up to a point, then if things go to far it says enough is enough and backs things right off?

This intervention also implies that even without a new map, improving the exhaust, using a higher RON Fuel or running cooler you would make more power? Or is the ECU intervention only a one way downward trip?

So on the boost is this a parammeter that you program, or is controlled by setting the WG Duty?

I will chose to not understand TPW for the moment. ;) However in case it prompts any thoughts I have knocked out a busy graph in the same format including all the logged items.



Also included is a calculated Injector Duty Cycle taking account of the pulse width & the RPM ( I hope I have got the maths right?)

MAF seems very constant, perhaps you could comment on that?


Many Thanks


Rich
 

·
Registered
Joined
·
1,216 Posts
Injector duty cycle=Injector pulse * RPM / 1200

Mass Airflow measured in volts goes through a non-linear lookup table to convert to airflow.

Theoretical pulse width is proportional to airflow/RPM (except when it clips at 25.5). It is proportional to the injector pulse required to give a stoichiometric (lambda 1.00, AFR 14.7:1) mixture, although the units presently logged are not the same.

MAF is consistent because I'm trying to hold your car near peak power all the time during your acceleration run.

Nearly all the control variables you're looking at have open and closed loop control processes.

Open loop means you lookup mapped tables with a value you want to output.

Closed loop means you then measure the effect and correct it to a target.

Cruise control is a closed loop process using throttle position to control vehicle speed.

Throttle position is a closed and open loop process where the throttles are moved by motors to a desired position which is measured.

Valve timing is a closed and open loop process where the oil control solenoids change oil pressure to rotate the inlet cams.

Knock control is a closed and open loop process where mapped tables and calculations give an ignition timing value that should be just right (enough to make good torque and emissions but not too much to cause significant knock), but then feedback from the knock sensors after each combustion event trims the next ignition timing event to keep it right.

Boost control is a closed and open loop process where we lookup wastegate duty from a map to give our intended boost level and feedback from the turbo pressure sensors allows it to be corrected to our target (but this bit is not yet well understood). However, you'll note the recent boost control improvements have come from changing the wastegate duty lookup to be based on RPM rather than airflow as we thought this would and confirmed it did give tighter control, safely allowing us to run more top end boost.
 

·
Registered
Joined
·
1,216 Posts
A few other questions for you to check understanding ;) :

1. Why does injector pulse (and theoretical pulse width if it didn't have the clip at 25.5) look like a torque curve?

2. Why does ignition timing increase with RPM and decrease with boost?

3. Why indeed do I let your boost start tapering in each gear before I have the wastegate duty at 100%?
 

·
Registered
Joined
·
669 Posts
Hmm Master... Well on the one hand I want to say "I vill ask ze questions" on the other I feel like a small boy at the back of the class who has not been listening...

Really not sure that I know the answer to any of these, but here goes.

1. Why does injector pulse (and theoretical pulse width if it didn't have the clip at 25.5) look like a torque curve?

Well I can see that Injector pulse, Theretical Pulse (Why does it clip at 25.5?) and for than matter Boost all resemble a Torque Curve. I can only presume that this is because they are all closely related and that max Torque is at max boost / injector pulse?


2. Why does ignition timing increase with RPM and decrease with boost?

I guess it needs to increase with RPM to compensate for the lesser time that you have from when it goes bang to TDC? I guess it needs to decrease with boost because the boost gets in more charge and needs less advance? Do these two reqirements not conflict with eachother to some degree?


3. Why indeed do I let your boost start tapering in each gear before I have the wastegate duty at 100%?

Very unsure on this one. In fact I wonder if it's just an artifact of the gear change? I say this because if you look at this more conventional pull in third gear wihout the change from second you get a more sensible looking rise in boost that then slowly drops as the revs rise.



So just to confirm that I am not understanding, one more power hungry question?

Given that the drop off starts before the WG is at 99%, does this mean that you could flatten it a bit? ;)


Rich
 

·
Registered
Joined
·
1,216 Posts
I would say you are correct on all three.

Torque is closely related to the air mass that you can get into the cylinder which depends on amongst other things, turbo efficiency, head flow/cam profile & lift, engine speed. It is also related to the thermal efficiency of the combustion event which follows, which is related to ignition timing and air fuel ratio amongst others, which are limited by knock in most turbo engines. Theoretical pulse width clips at 25.5 because Nissan designed in an area for a small overboost, but we're gobbling all that up. At some point it would be nice to keep the theoretical pulse width from clipping either by increasing the 25.5 ceiling or rescaling things so we only use say 20 theoretical pulse width where we used 25.5 before.

You could ramp up to 100% wastegate duty cycle 500 RPM earlier to hold another 0.5 PSI from 5500-6000 RPM in your 3rd gear example, but in other gears and situations it might be different. It shows less taper than you might expect from studying through the gears plots. When I ramped up earlier I found I got more overboost on gearchanges or risked the boost rising in this area before falling again. The torque is falling off in this area, to maintain the same boost you'd end up with less timing and more heat, and it probably wouldn't go faster.

As it is, the ramp up of duty cycle is far more aggressive than other tuners have been using, and a lot more than needed on other platforms. I think that is because the wastegate flappers are quite big, so exhaust manifold pressure pushes them open. Usually people think of wastegates being moved by the compressor air from the turbo outlet that we reduce in pressure through the wastegate solenoid, but on the GTR after 4500 RPM we're fighting against the exhaust gas trying to open the wastegate. It works though because it gives us more airflow, more fuel use, more acceleration and we can control the knock.
 

·
Registered
Joined
·
669 Posts
Thanks Thistle and Joe for sharing your knowledge, it is much appreciated.:thumbsup:

I guess you can be one of a couple of types of user. Load the map, do a quick log and send it off, wait for the new map, or like me try and have at least a basic understanding of what is going on. In other non automotive areas where I am more in my comfort zone I often find the questions and observations from users contain something that I have missed.

So thanks to Nissan for producing a reprogrammable ECU, so much better for the level of tune a lot of us are interested in than a replacement. Thanks to Joe & the team at Cobb for coming up with a great product in the AP. Thanks to Ben and Thistle for the great support and work they have done taking the software and maps forward.

Welcome to the world of e-tuning. :D


Rich
 
Joined
·
3,666 Posts
Thistle,

Looking at the datalogging I see your running around 12 to 1 AFR presume this is the reading from the cars lambda,s? have you check this number agaisnt a wide band set up?

I feel 12 to 1 is a little lean for a big BHP car? whats you thoughts about this please?
 

·
Registered
Joined
·
1,216 Posts
Agree entirely. Thankfully it is a bit richer than it appears.

External widebands read consistently about 0.5 richer in a few independent tests, as long as there are no cats in their way. I'm ignoring post cat readings now as they read inconsistently 0.5 to 2.0 richer (not what I expected!), if you retuned to 11.5 on that external wideband you could be 13.0 in reality.
 
Joined
·
3,666 Posts
I'm ignoring post cat readings now as they read inconsistently 0.5 to 2.0 richer (not what I expected!), if you retuned to 11.5 on that external wideband you could be 13.0 in reality
This is with a Wide band Lambda I presume? reading from the tail pipe?
 
41 - 60 of 67 Posts
Top