Cobb vs UpRev

FricFrac

I don't get it. First you're saying the hardware is essentially an ASIC then you says its a microcontroller.... and the box has a signal generator... and the processor is mearly a large IC. What is that suppose to mean? That's the most vauge statment in the entire thread. Then after saying the logic is locked you end off by saying they should expand the software. You have enough knowlegde to be dangerous and fool some casual bystanders but not enough for me to even remotely think you know what you are talking about. You might know something about programing but you are clueless about the hardware. Just saying in case someone reads that drivel and trys to make sense of it.

__________________
'09 370Z - '93 300ZX Vert - '83 280ZX - '83 280ZXT -'81 280ZXT stroker - '72 240Z - Stillen CBE, HFC, G3, Grounding Kit, Sways // Swift Spring // H&R 15mm // 5% Tint // Clear Bra //

1slow370

First off cut me some slack the only time i can get on here is after 12 hours of work at 3am which leads to me confusing myself sometimes. I'll try to be clearer sorry.

What I mean by saying that the hardware on these things is simpler than a computer is that instead of having a BIOS running the hardware and an OS performing various tasks, your ecm only has one level of system. So it can't run say Mac OsX, Linux, and Windows on the same computer like you can with a PC. There is no OS, just the code stored on the Eprom chip.

Because the system in an ecu is so simple, you can't simply "replace it with your own software."

The hardware in the box limits what you can do because often times there are separate chips in the ecu that do specific jobs that cannot be re-flashed so you actually can only change some of the code and have to live with what nissan put on the rest. Also the individual components on the boards limit what can be done. If you have a narrow band O2 sensor that is putting out extremely low voltages around a volt at most, and it sends that voltage to an analog to digital freq. gen. so that the data can be processed by the chip, you can't hook up a 5 volt wideband and expect to be able to program the chip to accept it. You'll just end up burning out the parts of the board that process the signal from the sensor.

As far as MAP and MAF sensors go I only have experience with the commonly used GM ones. MAP sensors come in two flavors analog and square wave. If Nissan uses square wave sensors, you would have to reprogram that whole area of the code on the ecu and any area that uses it, to accept a different frequency range used by a larger sensor, unless they all use the same frequency range in which case you could treat them like analog sensors. Analog sensors commonly used by GM are all 5volts and work on less voltage=less pressure, so a 1BAR map would have the table set to read vacuum in the low voltages and atmospheric at 5 volts. For a 2BAR MAP you would be squeezing 15 more PSI into the same 5 volts so when the ecu see's 5 volts It would normally think you were at atmospheric when you actually have pressure. With analog sensors you put greater values in all the tables that reference it to trick it into working. So on a VE table when it's at say -7 psi you would actually be putting in a value somewhere near 90% efficiency instead of the 50% it would normally have because the pressure is greater. But when you do this you are changing the accuracy of the ecu because on a table that has only like 15 values on it's axis you are now only using 7 to cover the pressure range that previously had all 15 slots. On the other hand, if the car uses a square wave sensor there may be another controller inside the ecu that takes that frequency and changes it into another frequency for processing so if you switch out sensors and are now using a larger range, the ecu won't be able to handle sensor frequencies beyond what the first converter can handle, and you won't be able to re-flash that chip to change it's range of frequencies so it will never be able to accept boost. All of this maters if you want to get rid of troublesome, restrictive and delicate MAF sensors and run speed density to avoid going super rich when the blow off opens.

You run into issues like that with every sensor and every part on the car, so in order for a company like uprev to write the code in the ecu to do different things than it normally does you can't just tell it to do it with software. You have to completely tear apart the whole car, figure out exactly what happens between each and every sensor and the ecu, and figure out at that point what can and can't be done. The only people with that kind of R&D time into making a reflash is NISSAN. They designed it so they know exactly what it can and can't do, and fat chance of them ever telling anyone anything. That is why certain companies focus on certain makes of car and why re-flashes cost so damn much. You have to invest a huge amount of time into figuring out even the smallest parts of the code in the ecu, so it makes sense to focus on cars that use similar codes so you save R&D time. Uprev won't be able to do most of things I stated because of the limitations of the hardware and the COLOSSAL amount of time it would take to figure it all out without having Nissan's design information. Now if someone from NISSAN could graciously violate their NDA and leak the design information of the ecu and it's code I have no doubt that Uprev would be able to do wonderous new things. Since it takes a huge amount of time, in 5 years from now there will be some wicked stuff for us, just like there is for the 03-06 350z, unless everyone with a 370 wants to blow cash like they own a GTR.

This is why standalones and piggybacks sell. The companies that make them design them to be able to support whatever components they want and the hard part for them is getting it to work with other controllers found elsewhere on the vehicle like the ABS and dash. So the farther you want to vary your setup from stock, the farther you need to modify the ecu. Is a standalone the best option for every one? Hell no, but if your going to be doing FI and want to be as safe as possible and be able to account for every little thing the engine does without just putting a bigger number in the table to half @$$ it, a stand alone is your best option right now. Sure you can get re-flash utilities to work boost applications, just not as well as a proper standalone could. They probably will both make the same peak power because you can just screw with the re-flash until the advance and the fuel are where they need to be at that one point in time, but for efficiency, reliability, and driveability it won't match the standalone unless it can do most of the things i asked for. I'm not trying to bash re-flashing it works wonders for some setups but not for all. Know that before you buy.

Edit: I think this ones a less confusing post so I might delete my last one unless I get lazy.

Edit: Oh and for the record UTEC's are total garbage that shows that not all piggybacks are created equal.

Denny McLain

Just now really getting into the capabilities and the primary fuel table may be OK for my current needs anyway.

The rpm points are

2800
3200
4000
5200
6000
6400
7000
7200

Not sure what the logic here is as it's one hell of a gap (maybe that where you tweak the MAF settings) between 4000 rpm and 5200 rpm but my car is a tad rich above 6000 rpm and there are quite a few tables above that range. At low rpm it's running 12.8 but above 6000, it's in the 12.3 to 12.5 range. Being conservative and not really knowing what to expect, I took out 1% fuel in that range and will log it to see what happens.

So far......A OK. Handy Gadget.

wstar

That is kinda weird. Looking at the rpm gaps, which are 400, 800, 1200, 800, 400, 600, 200, this is my best guess:

They expect you to tune some other engine parameters to get the big center area (1200rpms from 4000 -> 5200) about right for that whole range, and then you see the 800 and 400 gaps arranged symmetrically around that, basically meaning "the further you get away from this center, the more adjustment you'll need here". The oddball part is the very upper end, which breaks with this symmetrical hill pattern and steps out by 600 and then only 200.

__________________
7AT Track Car!
Journal thread / Car setup details

Denny McLain

My feeling also and really why I posted it. By pure luck I need to change it the most where there are the most points. Guessing it may have something to do with the ECU as I don't see the logic.

1slow370

The target AFR is only supposed to be a guideline. It tells the ecu about where you want the AFR to be so it should be set around 12.5-12.8 and leaner down low near idle midrange. You change the MAF tables to tweak how the engine runs or the VE tables which it doesn't look like you have access to. Target AFR is there to tell the ecu when to start correcting itself. You need to change the air flow tables to make the ecu run right in the first place in order to give you the afr values you want. Both tables need to be changed in order to work together but the Target AFR table shouldn't have to be changed that much maybe a few points here and there.

Basic guide to nissan tunning
http://www.plmsdevelopments.com/file...ing_basics.pdf

Denny McLain

Except for the fuel/air guideline ratio's, that is basically what Cobb is saying to fine tune. The gross tune does come from the fuel tables as I have both the stock one and the Cobb table to compare. No one seems to know (including Cobb as Calvin admits they are in a learning curve) or if someone does, are not willing to tell what is the best general AFR. Cars will vary and it's what that particular car likes best. (My Corvette liked around 13.3 to 1 - FYI, Pro Stock runs in the mid 14's - Forced induction in the 11's)

These cars appear to be different if that is the way to go as that is not what the tuners I'm used to dealing with do and certainly not what I've done in the past. (Yes, I'm an old Corvette, Chevy guy but owned a lot of imports also)

It used to be the first thing I did was either buy a larger Billet MAF or port the existing one. (More flow is better right?........ Only if it needs it Home!!) Found out later it made them harder to tune. The current trend among the people I talk to is to stay away from playing with the MAF, use stock tables as the only way to go.

Different strokes for different folks and what works.....works. An old dog can change his stripes. I'll read the guide as the software in it looks more like I'm used to as the Cobb software is certainly different. Certainly is a learning curve to all of this.

Thanks for the guide.

1slow370

you can always do what we used to to find out what works. Simply find someone with a technosquare tune as it makes the most power right now and rip it from thier ecu then copy and paste baby Used to do it with .bin from the mclaren special grand prix tgp turbocharged 3.1 l engine when running forced induction on our cars. I'm sure you could rip the nismo tune too.

Denny McLain

I doubt the Nismo tune is nothing more than raising the red line. I'd be very curious as to the Technosquare tune as I'm still wondering if the Cobb tune helped or not. (yes, the air/fuel changed) Or...... is it a matter of different dyno's or...... the computer somehow needs time to learn and is holding it back as the car did not immediately respond. May make another trip to Cobb and see if I can get another baseline on their dyno to compare apples to apples.

1slow370

Ok and a minor update on the Cobb vs. Uprev issue Apparently Uprev exists they just won't give it to us yet but it is in fact in use and will say no more than that for now.

Hi-TecDesigns

Reading this thread with heavy interest as I'm trying to decide which system will prove more useful. I'm an EE that does embedded hardware/firmware design, so I can attest to the fact that reverse-engineering a system is not easy, though certainly doable with enough time.

I have no specific knowledge of what any particular company is doing, but in uprev's case the use of a complete custom OS may be a bit of a misnomer. By the time all modifications are made I imagine it would appear to be completely custom, but it most likely grew out of the stock system little by little. My approach would be similar to the following:
1) Determine where maps exist (regardless of what they do), not terribly hard as they usually consist of large chunks of slowly changing values
2) Once these are known, it's a fairly simple matter of determining the code that access each maps (still may not know what each does).
3) Modify code form step 2 to branch to another location, one that includes custom code allowing for returning map data from multiple custom maps located elsewhere in memory.

This is only the beginning, mind you, but it would allow for a fairly quick modification of a stock ECU to run with (multiple) custom maps. Granted, it does not grant the programmer with any special knowledge as to what those maps actually do (that's where the experience with tuning helps to recognize data), but it shows you how easy a simplistic mod can be. Understanding what effects any changed maps will have on other portions of the system is beyond the scope of such a simplistic change, and that's where the time sink comes in.

On to our coding exercise:
4) Trace back sensors to the processor pins (not always easy due to multi-layer PCBs).
5) Pull out processor data sheets to determine what pieces of code are accessing each of those pins.
6) Start making connections between each piece of code identified in step 5) and the maps those same pieces of code access.
7) Begin to see a pattern emerging, then realize you've merely had too much Red Bull and Twinkies.
8) Put down Red Bull and Twinkies, then start praying you still see a pattern emerging.

That's the process, in a nutshell. It's not necessarily difficult, but it is slow, tedious, and often times not straightforward.

So, who'll be the first to tell me to ?

__________________
http://hi-tecdesigns.com/images/pict...on%20Small.gif
2000 S2000 Brickyard Red Metallic w/ Romanesque Crimson Kandy, custom everything else

FuszNissan

....lol

__________________
Mods: Too Many to list!
LouFuszNissan
ZDAYZ 2012

Ken in AZ

I thought UpRev patched the ROM file to the RAM thus making it capable to switch tunes on the fly? What I mean is that they take the rom data and put it in the RAM - since it is in the ram it is now edit-able?

I'm no EE but this is the way it was described to me.

brody8877

holy f-ed i was answering question on the 1st page and not realize there is 11 pages of reading i need to do.


edited - its 3am in the morning....


btw I have Cobb AP v2 on my wrx wagon, I hope they will release maps for the 370z soon.

theDreamer

That is one of the things they do, usual control in changing tune is through the cruise control buttons. Though some of us are wondering if they might try to use the digital display we have now, which would be awesome.

__________________
theDreamer's Z // Fast Intentions // Uprev // GTM // HKS // TEIN
theDreamer's Silvia // URAS // GREDDY
Houston Zs // Facebook // Twitter // Instagram