That's not as informative as I'd have liked. Nothing I didn't already know. Oh, well. Still, thanks for sharing. We here in California are going to be learning about this firsthand soon enough.

 

There has to be some tuned F-Types that need smog check soon...I hope they can help out.

 

Looks like that's exactly how it works. According to 13 CCR § 1968.2:
Just to confirm, have you verified that flashing back to stock returns the CVN back to the factory value on the F-Type? I assume it does if the CVN is computed only from the software and the calibration data, but it seems manufacturers have some flexibility in what exactly is included.

 

Call it what you want, but it's a checksum. Flashing back to stock seems to have the dealer's systems recognize it properly. My worry is the P0167F code that indicates that at one time or another, an aftermarket device touched the ECU. I see no mention of it in the referenced documents, so am trusting that someone else will find out before me. I just renewed my registration with no inspection required yet.

 

I didn't say it's not a checksum, it clearly is, but a checksum over what data exactly? I assume it covers only the executable code and calibration data, but there are other pieces of persistent storage in the ECU, which could potentially be covered by the CVN and not restored by the flashing tool. Hence the question if anyone has actually verified that the CVN is correctly restored when flashing back to stock on our vehicles.

And that's an example of another block of persistent storage on the ECU that changes when a tune is flashed. I suspect persistent trouble codes are not covered by the CVN, but not knowing all that much about ECU internals, I'm not sure.

I couldn't find any documentation on what data the BAR-OIS software requests from the ECU, but based on what I've read, it sounded like they only care about CAL ID and CVN when looking at ECU modifications at the moment. Obviously there's nothing that could stop them from looking at manufacturer-specific codes, such as P167F, but it doesn't sound like they do that currently.

 

The primary purpose of the checksum is to verify the integrity of the image being loaded, i.e. to ensure that is has not been corrupted in storage or transfer. As such, it cannot cover the memory locations where trouble codes are stored. If that were the case, the first permanent code would invalidate the checksum. Secondarily, and this is what CARB is doing, it can be used to match that against a database of known checksums. Depending on the checksum algorithm, it may be relatively simple to modify the image enough to have the checksum match but the image itself to be different. Unless using something cryptographically secure, they're useful to verify integrity but not authenticity.



I'm not sure about that code either, and that's the part that concerns me. It doesn't indicate any actual fault, just that at one point a non-factory device wrote to the flash RAM.

 

I've never compared a virgin stock vehicle CVN to a modified CVN then reverted back to stock with the same stool to see if the CVN would still match the stock CVN..etc

Files that are Virtual Reads, may or may not match CVN, most Virtual Read tools are owned by shops or the loaner tools. The user devices have been performing an actual read of the file (sometimes the Dimsport devices do a VR if key cycles become excessive to read the file).

Nonetheless, it's more possible to match a CVN if the stock file is read versus the tool using a Virtual Read(VR). For those that don't know the difference, the Virtual Read file basically uses the flash device to read your CAL ID and match a file on a remote server (no reading of the actual file is performed). Some ecu's cannot be read and/or it makes the read process 100x faster with a VR..it's done in seconds versus 10/20/30/120mins..etc. An Actual Read(AR)..is as read-out, it's the actual file residing inside the ecu...

The above would need to be tested to confirm CVN matches, I know 100% the CAL ID would match with either scenario. However, most of these tools including the hand held user devices, leave a hex signature in the file. This allows the device to know it was the last hardware to flash the module, meaning the entire file does NOT require to be written. If you have a file that has a hex address from 0x0000 to 0x9999, the first flash would require writing all addressed from 0x0000 to 0x9999. If the tool previously wrote the ecu and only requires changing a mismatch at 0x1000 to 0x1010, then those mismatched addresses would be overwritten and the rest left untouched. This process is 10x faster in the write process, reduces the chance of error and then only require a minor checksum correction. Nonetheless, this changing of only the mismatched section in the file would only be performed if the hex tool signature matches. If the signature does NOT match, the entire file is written from start/finish.

What I'm mainly getting at in the above paragraph; even if the actual file is written back by the tuning flash device, it's very possible the CVN will NOT match because of the tool hex signature. Going to the dealer might or might not wipe out the tuning device hex signature if the dealers SDD/SDD2 does not write the ecu AS NEW (the tech has a choice to flash as EXISTING MODULE or AS NEW MODULE). Often, if the SDD sees the same CAL ID, it will abort the flash requested by the tech and sometimes requires special clearance to force flash or to flash as new module(tech might need to jump through hooks to get SSD1 to allow option to flash AS NEW). However, lets say there's an update or the tech has some specials powers to force the flash of teh same stock file, we don't know if the SDD will overwrite the hex signature left by the tuning module.

Does SDD implant it's own hex signature at a different address?
Would SDD overwrite the previous hex signature if the ecu is written As EXISTING or only AS NEW MODULE?
Do the VR files match the stock CVN?
Would a ecu flashed back stock remove the tool signature and allow CVN to match?

Those are our own questions that would need to be cleared to confirm the CVN would match 100%. Unfortunately, all the vehicles I have on the lot have been modified. I would need a few vehicles here in-house that have a 100% untouched/untuned ECU to confirm all the above.

 

I'm very curious as to what the fix will ultimately be for all those with tuned ECUs. I can't imagine it will be cheap. A few of my injectors failed recently, and took my bank 1 catalytic converter with them. I'm glad that I now have a lot more inexpensive "legal" replacement options open to me than had I stayed in LA. All those CARB cartel restrictions really tick me off. Seriously, what is the difference between a CARB certified Magnaflow cat and a non-CARB certified Magnaflow cat, besides the exorbitant price?

 

It's a great place but far from perfect.

I remember years ago when a trip to the Los Angeles Smog Basin meant burning eyes and throat and air that was visibly reddish-brown all the time. Colorful sunsets, but that's about all I can say to the positive for that. It's better now, so I try to make allowances.

Still, I understand your complaint. I make minor modifications but, for example, would never go catless on a car that came so equipped. In the 1980s I could sell used cats from salvage cars but that's been illegal for quite some time now. I don't know the difference on the Magnaflows but expect certification costs, and I wouldn't be surprised if there was a surcharge on each unit as well.

 

I'm all for having catalytic converters (not necessarily CARB-certified) fitted, no doubt, but they've gone too far this time with this new ECU tune inspection thing.

 

The Drive has answers, and tuners can get some CARB approval, apparently, and it's all murky:

California's ECU Tuning Policy: Your Questions Answered Well, all the answers we could pry from CARB and CBAR, that is... https://www.thedrive.com/cars-101/41...tune-flash-faq

California Smog Testing Will Now Sniff Out (and Fail) Tuned ECUs The toughest smog check state in the nation is getting a little tougher https://www.motortrend.com/news/cali...modifications/

--

 

I know it wouldn't be a cheap way around it, but couldn't you just buy another ECU from JLR and plug it in before you went to get your emissions test? Then lend it to all your friends when they had to go through emissions?

 

Most modern ECU's typically have the VIN coded in them so the last part of your thought likely won't work. However, the first part should.

 

It's relatively easy to read the CAL ID and the CVN using a cheap OBD2 dongle, so I suspect it would be easier to collect some CVN values from stock vehicles remotely. I realize there may be several software versions circulating, so I'm not sure how comparable the values are between vehicles.

Looking at my V8 S with a VAP tune, the CAL ID is EX53-14C204-RAH and the CVN is 09E49645. Is the CVN the same in all cars with the same software version / tune, or are the tunes customized for each vehicle?

 

The tunes are customized per CAL ID, judging on your CAL ID, its for a F-Type S "RWD" 14MY or a 15MY that was built in 2014.

Since 2011 the CVN would be the same for the exact same CAL ID. In year 2010, I believe it was 50/50 if CVN would match and pre-2010 I don't think it was logged much or made of any note to confirm matching. Most of the pre-2010 CVN's included the VIN in the calculated section, the CVN would be different as the VIN on each vehicle changes for same CVN. (Post 2011 the VIN is not included into CVN calculation)

Now back to the CAL ID, the EX53-14C204-RAH was programmed into vehicle software at factory during build. For that CAL ID, JLR released 2x dealer updates: EX53-14C204-RAG and EX53-14C204-RAJ. For 2014 it kind of makes the info gathering easier being it's the birth year of F-Type, there's currently 3 different CVN for a 2014 and 2015(built 2014) F-Type S RWD. However, for a 15MY it can classify for the 14MY software and if the 15MY was actually built in 2015 it would have those 14MY 3x CAL ID's + the CAL ID's for the same model built in 2015 which is another 3-4 ID's..making 5-6 CAL ID's for a 2015 F-Type S 5.0 RWD. If we were to add in the R and SVR CAL ID's, a single F-Type 5.0 could have over +15 different CAL ID's per year in the Americas!! The 3.0L crowd is even more crazier because of the different drivetrains and power outputs (RWD, AWD, Base, S, Manual shift base, Manual Shift S, 400PS..etc)..if all the mentioned 3.0L models were in existent during the same year we could have over +30 CAL ID's for a single model for 1 year..it's insane!!

Basing this off what's been collected over the years doesn't work, how would CARB know if the data collected over the years is NOT of a tuned vehicle? If a certain model year did have +30 different CAL ID's, and a few of each tuned model versions performed a smog test, could be over +500 CAL ID versions on the CARB database collected per year/per model (if VAP was the only company tuning JLR maybe +60, but we're not alone..).

I'm assuming the CVN would be widely different on the tuned vehicle dependent on different tuning companies editing the same CAL ID..one tuner enters 0.8003 lambda while another enters 0.8002 lambda, that would change the CVN when calculated. As mentioned in my previous post, in order for this to go smoothly and not require visiting the smog Ref, would require CARB being granted access to all the official vehicle manufacture CAL ID/CVN stored on each vehicle manufactures database.

 

Thanks for the post, that's very informative!

Agreed, the only reason to spend years collecting data from smog tests would be to see if the CVN verification works and how many vehicles would fail every year. I'm not sure why they would need five years for that, but I suppose governments move slowly.

Have you looked at all into getting your tunes CARB approved? I realize it's not going to be financially feasible for low production models like the F-Type, but I'm curious exactly how expensive that would be.

 

Let's keep it civil and on-topic. No need for more F-Type forum members to lose their privileges.

Don

 

I asked my nearby dealership if it can be done, and they quoted me nearly $7000, and as previously mentioned, would be a special order, tied to my VIN. Whether you can simply plug in another ECU, say from a junk yard, and have it work without warnings, that I do not know. My hope is that VAP can figure it all out before the majority of us have to go smog.

That said, this may be the excuse I need for myself to get back to OEM cats...(relatively) high failure rate and all.

 

Thank you. There is a lot of useful information in this thread and it would be a shame to see it locked.

 

I am not picking on you specifically, just that it's the last post in this thread. And while I love and believe in all of our rights to free speech and debate, I would respectfully ask you all to refrain from discussing this stuff in our tuning thread.