[Terry335]

Hi Gang, first let’s avoid the normal bashing and unprofessional talk. I want this to be a purely technical thread for those of us that care about knowing what’s going on with our cars.

Does anyone know exactly what signals the xede and procede are intercepting? I can't quite figure it out, and I'm too lazy to dig up a 335 wiring diagram.

Here is what I know:

Both devices intercept 4 wires, and both have a boost solenoid bypass.

I’ve assumed the 4 wires are o2 sensor a, o2 sensor b, map, and cps. The problem with this theory is then why would you bypass one of the boost solenoids? Maybe one of those wires is actually to run the remaining solenoid? Then what about the o2 sensors? If not, can you just bypass this solenoid on a stock car?

[e.n335]

Originally Posted by Terry335

Hi Gang, first let’s avoid the normal bashing and unprofessional talk. I want this to be a purely technical thread for those of us that care about knowing what’s going on with our cars.

Does anyone know exactly what signals the xede and procede are intercepting? I can't quite figure it out, and I'm too lazy to dig up a 335 wiring diagram.

Here is what I know:

Both devices intercept 4 wires, and both have a boost solenoid bypass.

I’ve assumed the 4 wires are o2 sensor a, o2 sensor b, map, and cps. The problem with this theory is then why would you bypass one of the boost solenoids? Maybe one of those wires is actually to run the remaining solenoid? Then what about the o2 sensors? If not, can you just bypass this solenoid on a stock car?

Hi Terry

I got my PROcede now and investigated exactly in this issue. It's not just intercepting, it's an overall tuning concept. Now I know why Shiv is that angry about AA picking up the method he developed, it must have took weeks to come up with that ( tricky ) solution.

I will not share what I'm knowing by now with anybody, if you are interested in find it out, it's possibe.

Cheers
Eugen

[RiXst3r]

I remember shiv saying this stuff back in the early days....

Originally Posted by shiv@vishnu

Well that's the secret part, right?

J/k.. let's break down things a bit.

Boost Target: The factory ECU reads this by reading off a T-MAP sensor. The Xede intercepts that signal and adjusts it as a function of RPM and Load. Now the car sees a revised boost signal and dials the boost up or down to compensate. New targets are reached based upon the massaged T-MAP sensor response curve. ECU is happy. The stock ECU still controls all its atmospheric/temp/etc,. compensations.

Timing Changes: The factor ECU figures out when to fire the spark plug based upon crank position (in degrees) and a corresponding value in the timing map (function of engine load/T-MAP and RPM). We don't want to change the map in the ECU since that will be stored and logged. So we intercept the crank position sensor signal and shift it in time (either delay it or advance it). This time shift will make spark timing shift as well. The stock ECU does not know that we are doing this. And it is happy to go about doing what it suppsed to do. Furthermore, the stock knock control system still functions and ready to do its thing when conditions take a turn for the worse (bad tank of gas, super hot conditions, etc,.)

Fueling Changes: The stock computer delivers fuel based upon two things (actually more but these two are the major variables): The raw fuel map and AFR sensor signals. The raw fuel map looks at engine load/T-MAP (we we already modify.. hint hint) and RPM to determine where on the table the engine is operating at. At this point, it tells the fuel injectors to spit out X amount of fuel. Then it reads the data from constantly working wideband o2 sensors. The signals tell the computer if they need to add or subtract more fuel to achieve the desired AFR target. This constant feedback loop goes on and on and on. We dont want to change the actual mapping in the ECU. That would be evasive and easily loggable. Instead, the XEDE intercepts the wideband signals and adjusts them as a function of T-MAP and RPM. The new "adjusted" data is sent to the ECU and it makes the necessary enleanments/enrichments. The end result is that we get the fueling we want and the stock ECU is happy, thinking that it is doing its usual job. The stock ECU still controll all its long and short term adaptive fuel trims just like it would in a stock car.

Those are the basics. There's some other stuff invovled as well. But we can't disclose everything for obvious reasons For example, how do we make it so the closed loop fuel control corrections are instantaneous? Or why do we include a 3rd boost control solenoid to control boost? Or why do we recalibrate both MAP sensors and not just the one used to control fuel? The list goes on and on. But there are reasons for everything. I wouldn't want to mess up my own personal car, now would i?

Cheers,
shiv

Originally Posted by shiv@vishnu

The solenoid verions will provide the most flexability as we can adjust duty cycle to acheive the desired boost while allowing the factory solenoids to be the drive by the factory ECU within the duty cycle range assoicated with a stock car. Downside is that you see it when you open the hood. It's not totally obvious. But if you look, you'll see it.

The solenoid-less version works the same way but it effectively offers a fixed bleed (non adjustable). The bleed fitting we use (which basically emulates a solenoid) can be sized perfectly to acheive the desired results. The only potential problem is what happens when the conditions change (say an increase in altitude for instance). In such a case, the factory boost control system makes up for the altitude change by running its boost solenoids at a higher duty cycle. If those duty cycles operate outside of the "normal range" expected by the factory ECU, a diagnostic code is tripped.

Originally Posted by shiv@vishnu

Spent about 40 hours this week playing around with boost control. Things we've have done/learned so far:

1) The factory boost control solenoids aren't standard PWM solenoids. They are continuously variable valves. Our first attempted to drive them directly with our computer's closed loop boost control logic was an utter failure. All the pre-spool logic that is employed by the factory ECU is there for a reason and trying duplicate it with an external boost control using factory hardware is hopeless.

Next we added our own boost control solenoids (two of them) and relieved the factory boost control solenoids of all duty. Using revised closed loop boost control logic, we actually got amazing results. Instantaneous boost resonse with no spiking, oscillation, creep or any other undesired boost boost artifacts. Next, we tried control boost with just one solenoid. Under steady state conditions it worked just fine. But under normal/dynamic driving conditions, we saw boost spikes, slow spool and all other sorts of wackiness. Turns out that there are two boost control solenoids in the stock car for a reason: To adjust the vacuum pressure in the wastegate actuator lines quickly. If the pressure in the lines doesn't change quickly enough, the wastegates dont respond quickly enough (hence, boost spikes and/or slow spool depending on what the vacuum condition was at that specific time).

Ok... back to having to use two solenoids it is. But wait... isn't it kind of silly to use 2 add on solenoids when there is a perfectly function pair already there? Ok.. so back to hooking up the factory solenoids and figuring out a way to design a system that allows the turbos to generate anywhere from 5 to 20psi of boost (our goal) depending on electronics (mapping). And more importantly, it has to be able to do this without the factory ECU driving the solenoids with any more or less duty cycle than it would in a completely stock car, running stock boost. We wouldn't want there to be any wierd logged data in the ECU, now would we?

I'm happy to announce that it looks like we figured out how to do it. And it doesn't involve adding on any solenoids. And it still retains the factory ECU's low boost-till-the-engine is warm feature. And it can still run anywhere from 5psi to 20psi. And it still doesn't over or under-work the factory solenoids while doing so. So far, we've tested in sea level conditions. This weekend, I'll go up to 4000' and test it again. But so far so good.

"So what about all the current kits of the road? How is this different/better?" you ask

The approach we've employed with the early adopter beta folks works just fine. But the range of available boost is only beteen stock (7-8psi) and 11psi. And boost doesn't seem to be as rock solid as with our new approach. Of course, early adopters will easily be able to retrofit this changes once we put the setup through its paces.

It's been a productive week and I just wanted to share it with you!

Originally Posted by shiv@vishnu

It's a bit different than the method we used in the beta cars (yours included). Your setup works just fine but in a narrower range of boost performance. Your setup is limited to 10-11psi of boost. And boost pressure in the 2000-3000rpm range tends to oscilliate a few times until it stablizes. Much like the stock car. But with the extra boost, it becomes a little bit more discernible. This new approach offers better-than-stock boost stability and just-as-good boost response at higher-than-stock boost levels. That's a pretty tough performance target to achieve since usually its a zero-sum game in many respects. Usually, you end up sacrificing stability for response or visa versa. So i'm geeked up over here. See... this is me -->

-shiv

[turbocoins]

I'm waiting for the Terry tuner, v. 1.0.

[RiXst3r]

Originally Posted by turbocoins

I'm waiting for the Terry tuner, v. 1.0.

[LOK11atXEDE]

Originally Posted by Terry335

Hi Gang, first let’s avoid the normal bashing and unprofessional talk. I want this to be a purely technical thread for those of us that care about knowing what’s going on with our cars.

Does anyone know exactly what signals the xede and procede are intercepting? I can't quite figure it out, and I'm too lazy to dig up a 335 wiring diagram.

Here is what I know:

Both devices intercept 4 wires, and both have a boost solenoid bypass.

I’ve assumed the 4 wires are o2 sensor a, o2 sensor b, map, and cps. The problem with this theory is then why would you bypass one of the boost solenoids? Maybe one of those wires is actually to run the remaining solenoid? Then what about the o2 sensors? If not, can you just bypass this solenoid on a stock car?

Terry, you certainly desire all of the technical detail don't you

As both products stand at today's date (to my knowledge):
Fundamentally, your observations are quite correct. By some clever modifications of these signals (made more effective by the "smarts" in the current Direct Injection Engine Control on this engine series), Ignition Timing, Fuel Delivery and Boost pressure are carefully modified to derive the gains we all want from this engine.
Aside from that there's two different power, and ground methods used. Both should be effective although I personally have concerns about leaving a unit powered 24/7 as the PROcede is but I have not heard of any specific failure related to this. I also think that using an external ground source for the modification of sensor data is a fault waiting to happen. Sensor readings (from the ECU's perspective) are all measured relative to sensor ground as the XEDE uses. It is my suggestion that, using a power ground or chassis ground for the PROcede ground reference will cause distortions in signal data over time and differently from car to car depending on the state of the Chassis ground relative to sensor ground.
I believe the XEDE method is better but then, we make it.

As for bypassing one of the Boost Solenoids, Shiv's observations are something that I would generally agree with. I'm also certain that the oscillation described is not lost on the BMW design team either. We'll see what changes occur in other models with this engine!!

Your (technical) comments please Eugen, Shiv or others:

Cheers

[Ninjaneer]

Ah I'm early to the show, great seats!

[e.n335]

Originally Posted by LOK11atXEDE

Aside from that there's two different power, and ground methods used. Both should be effective although I personally have concerns about leaving a unit powered 24/7 as the PROcede is but I have not heard of any specific failure related to this. I also think that using an external ground source for the modification of sensor data is a fault waiting to happen. Sensor readings (from the ECU's perspective) are all measured relative to sensor ground as the XEDE uses. It is my suggestion that, using a power ground or chassis ground for the PROcede ground reference will cause distortions in signal data over time and differently from car to car depending on the state of the Chassis ground relative to sensor ground.
I believe the XEDE method is better but then, we make it.

Hi

Thank you for your attention . You'd better don't make it. Both the PROcede and the XEDE ground methods are not correct.

The ground for powering the device should be chassis ground. For the signals use sensor ground. Otherwise you will mix it up in either way. I don't know if the XEDE has different ground connectors for signal and power ground. The PROcede does. I'll correct the harness when I install the PROcede in my car, hopefully delivered end of next week.

Leaving a unit powered on 24/7 is no issue at all. If there are no heat issues ( and there are none ), powering on/off cycles are stressing electronic devices much more, but also this is not an issue at all. I'm tyring to find a way to "wake up" the PROcede when the ECU gets this signal. Would be nice, but this is more a fancy feature and has no priority. First I have to finish my PROcede wireless remote access solution. Then the V1, then the Innovate Motorsports data logging system, then ... maybe.

Cheers
Eugen

Your (technical) comments please Eugen

Not really ... I believe that Terry has some good technical potential. His missing soft-skills are scary .

Thats how currently my desk looks like :

[LOK11atXEDE]

Originally Posted by eugen.niederreiter

Hi

Thank you for your attention . You'd better don't make it. Both the PROcede and the XEDE ground methods are not correct.

And this is the exact difficulty which I described in a previous post where I can't know at any one time, what's in a PROcede (although I get a good idea from your pics) and Shiv can't know what's in a current XEDE. So we get assumptions - the mother of all ....ups. I shall do my technical best to work through them.

Originally Posted by eugen.niederreiter

The ground for powering the device should be chassis ground. For the signals use sensor ground. Otherwise you will mix it up in either way. I don't know if the XEDE has different ground connectors for signal and power ground. The PROcede does. I'll correct the harness when I install the PROcede in my car, hopefully delivered end of next week.

So, I'm sorry, what are you suggesting?

The PROcede method is wrong and you'll correct it before installing it in YOUR car. Shiv and Vishnu will probably want that information BTW.

I have to go on the Haltech "Interceptor" diagram as I don't know the PROcede one. ( the Haltech one seems to match your pics and Shiv will correct me if they've redesigned the cct board to move around the signal and power ground planes). Mind you, you already observed it's wrong from your interpretation.

The XEDE has seperate Power Grounds (two of them) on Black connector pins 5 and 8 and they can thermally and electrically sink about 8 Amps. These are not required in the 335 application. I would add them if we needed any more that say 120mA to power the units internals. (they are still in the harness but not used on this application.)
The "Primary Ground" connection that the XEDE uses is "Signal Ground" because (as you rightly elude to), signal distortion is likely in time or under electrical duress, when using a power ground circuit, like the PROcede has, for 0-5V analogue signal modification.

Originally Posted by eugen.niederreiter

Leaving a unit powered 24/7 is no issue at all. If there are no heat issues ( and there are none ), powering on/off cycles are stressing electronic devices much more, but also this is no issue at all. I'm tyring to find a way to "wake up" the PROcede when the ECU gets this singal.

Well, I'm sure with a bit of smart thinking (or you could copy an XEDE harness ), you'll figure it out. My proffesional opinion is that I would not design a system for use in a car that has to remain powered up 24/7.
Your engineering opinion versus mine I guess. I have done a lot of work with Automotive Electronic Fuel injection though in the last 21 years in this field. Just kidding

Originally Posted by eugen.niederreiter

Not really ... I believe that Terry has some good technical potential. His missing soft-skills are scary .

Thats how my desk currently looks like :

While we're on that picture. Is the PROcede and especially the harness actually RoHS compliant? I guess that doesn't matter in the US anyway but where you are....................? That could be important!

Thanks Eugen,
It is nice to see some proper technical discussion on such important matters. Your dedication and enthusiastic posts are a credit to you.

Cheers

[Carnage]

. I'm impressed that the thread has actually stayed on task. But I'm sure it will go sideways soon.

[Terry335]

Originally Posted by eugen.niederreiter

Hi Terry

I got my PROcede now and investigated exactly in this issue. It's not just intercepting, it's an overall tuning concept. Now I know why Shiv is that angry about AA picking up the method he developed, it must have took weeks to come up with that ( tricky ) solution.

I will not share what I'm knowing by now with anybody, if you are interested in find it out, it's possibe.

Cheers
Eugen

I'm working on the soft skills.

So with all this talk of boost solenoids, it sounds like the xede/procede must driving the remaining solenoid. If that's the case, each device would need an extra pair of wires. So confusing!

On the topic of the method of tuning, do I have the sensors involved right in my first post?

[Terry335]

Originally Posted by turbocoins

I'm waiting for the Terry tuner, v. 1.0.

Heh I watched a tuner break apart a factory tune and develop a reflash, and let me tell you, its a lot of work! I'll stick with business software.

[WilyB]

Originally Posted by LOK11atXEDE

While we're on that picture. Is the PROcede and especially the harness actually RoHS compliant? I guess that doesn't matter in the US anyway but where you are....................? That could be important!

Being half a planet away, and given the small size of beautiful Switzerland, it may not have occurred to you, LOK11atXEDE, that the Helvetians are NOT part of the EU and that RoHS therefore does not apply to Eugen.

The RoHS Directive stands for "the restriction of the use of certain hazardous substances in electrical and electronic equipment". This Directive bans the placing on the EU market of new electrical and electronic equipment containing more than agreed levels of lead, cadmium, mercury, hexavalent chromium, polybrominated biphenyl (PBB) and polybrominated diphenyl ether (PBDE) flame retardants.

[WilyB]

SCOOP: The next 335 upgrade from Switzerland!

[e.n335]

Originally Posted by LOK11atXEDE

Well, I'm sure with a bit of smart thinking (or you could copy an XEDE harness), you'll figure it out

Hi

Thanks for reply. So you say that the AA/XEde wakes up together with the ECU when the IBS sends the command to check the sensors ? In that case, my respects !

I always said that the AA/XEde is a good product too. I also explained, why the PROcede fits more to my preferences. The main subjects are upgrade policies & possibilities, customer experience, load & stress testing and ongoing further development. I also like the great support from Shiv I ever got when requested for it.

This does not imply that the PROcede is perfect, but it fits much more to my needs. I am sure that this discussion will have consequences thus we are talking about facts and like the AA/XEde also the PROcede has potential for improvements. Don't forget we are on Hardware/Firmware Release 1.0 considering the 335i.

The only shame is that AA came up with exactly the same tuning approach as the the PROcede does. We all know who developed it. Now all the discussions are about harnesses, RoHS ( I don't want to recycle my PROcede ! ) and that kind of stuff. I am sure, such discussions will result in product improvements of BOTH products.

Cheers
Eugen

[e.n335]

Originally Posted by WilyB

SCOOP: The next 335 upgrade from Switzerland!

lol you got me ...

[LOK11atXEDE]

Originally Posted by WilyB

Being half a planet away, and given the small size of beautiful Switzerland, it may not have occurred to you, LOK11atXEDE, that the Helvetians are NOT part of the EU and that RoHS therefore does not apply to Eugen.

Happy to stand corrected on that. I did not know Switzerland was not a party to the RoHS Compliance.

Lachlan

[LOK11atXEDE]

Originally Posted by eugen.niederreiter

Hi

Thanks for reply. So you say that the AA/XEde wakes up together with the ECU when the IBS sends the command to check the sensors ? In that case, my respects !

Thank you sir. It was, in our opinion, the best method. Leaving the unit on 24/7 was not.

Originally Posted by eugen.niederreiter

I always said that the AA/XEde is a good product too. I also explained, why the PROcede fits more to my preferences. The main subjects are upgrade policies & possibilities, customer experience, load & stress testing and ongoing further development. I also like the great support from Shiv I ever got when requested for it.

This does not imply that the PROcede is perfect, but it fits much more to my needs. I am sure that this discussion will have consequences thus we are talking about facts and like the AA/XEde also the PROcede has potential for improvements. Don't forget we are on Hardware/Firmware Release 1.0 considering the 335i.

The only shame is that AA came up with exactly the same tuning approach as the the PROcede does. We all know who developed it. Now all the discussions are about harnesses, RoHS ( I don't want to recycle my PROcede ! ) and that kind of stuff. I am sure, such discussions will result in product improvements of BOTH products.

Cheers
Eugen

Good competition leads to good products and pricing. I respect your decisions for purchasing and your reasons. The choice is still up to the individual and so it should be. The XEDE can be software and firmware upgraded now and in the future. How the Vendor wants to facilitate that to ensure the best results for their own customers is up to them. AA has chosen the XEDE as the method of choice for the 335i and it works. I don't think that just because it is the same method of tuning a 335 as PROcede uses, AA's selection should be discouraged.

Back on technical,
You didn't address my comments on the PROcede/XEDE ground issue and since you have a PROcede and I don't; can you enlighten us please on what the PROcede chassis ground does right or wrong. You mentioned it in this thread and another thread earlier in the week.

Cheers

[jonm42]

I just wanted to say thank you for staying civilized and on topic.

[e.n335]

Originally Posted by LOK11atXEDE

Back on technical,
You didn't address my comments on the PROcede/XEDE ground issue and since you have a PROcede and I don't; can you enlighten us please on what the PROcede chassis ground does right or wrong. You mentioned it in this thread and another thread earlier in the week.

Cheers

I am sorry, I don't explain the issue. I'd rather prefer an explanation from vishnutuning.

Thank you for understanding.

BTW, I already have a concept now so that I can hook up the PROcede to 12V switched power. Completely different, completely new. No need to wake up the PROcede. You will see it in a couple of days. The ECU will see everything as usual. I'll start with this tomorrow . It's nice to have the PROcede already in advance of the car and a little bit of time .

This will be a pretty good first result out of this discussion. Let you surprise.

@WilyB: and no, it has nothing to do with my super secret prototype intake trumpets

Cheers
Eugen

[Terry335]

Originally Posted by eugen.niederreiter

I am sorry, I don't explain the issue. I'd rather prefer an explanation from vishnutuning.

Thank you for understanding.

BTW, I already have a concept now so that I can hook up the PROcede to 12V switched power. Completely different, completely new. No need to wake up the PROcede. You will see it in a couple of days. The ECU will see everything as usual. I'll start with this tomorrow . It's nice to have the PROcede already in advance of the car and a little bit of time .

This will be a pretty good first result out of this discussion. Let you surprise.

@WilyB: and no, it has nothing to do with my super secret prototype intake trumpets

Cheers
Eugen

Just connect a bunch of relays so the sensors are connected to the factory ECU when the procede is off.

[e.n335]

Originally Posted by Terry335

Just connect a bunch of relays so the sensors are connected to the factory ECU when the procede is off.

Only two. I am nearly finished.

Looks like this ( intermediate state: )



Cheers
Eugen