I am unable to update the subject of this post. I think it may be a little misleading. It should read: Spartan 3 w/LSU ADV AFR flat spots. I forgot that 14point7 has a legacy product named Spartan 3 ADV.

I have a Rev built Mini MS3 running 1.5.1 integrated with a Spartan 3 w/a LSU ADV sensor (FW:1.08) via CAN.
I am seeing flat spots in my AFR at closed loop idle with a Spartan 3 (FW:1.08) set to perf0. Even though the PW increases or decreases the AFR remains more or less flat. If I change to perf1(10ms response time) the AFR has more noise, but then the AFR follows the PW.
Anyone else seeing this behavior with their Spartan 3

 

Not exactly sure what you are looking for, but I have two widebands currently, since adding a Spartan3. The other is a analog needle gauge Glowshift wideband.
I left them both wired for the season to see how they differed.

AFR = Glowshift wired into 02 in on a MSPNPPRO
AFR2 = Spartan3 on CANEGO

The spartan is a 'smoother' looking signal. It is set up using their instruction and I'm not sure of FW, this is as received last spring (2023) or fall of 2022.

Images:
1st is under wastegate pressure boost.
2nd is after a track session idling, just before shutdown.

 

Thanks natewin.

Can you look at your PW at idle? I wondering if this was introduced with in a later firmware iterations. A similar issue was reported back with 1.05 beta. https://14point7.discoursehosting.ne...-05-beta/179/6

In my screenshot @219.196 , the PW is increasing but the AFR does not move at all, and right after that the PW decreases, again no change in AFR. The AFR holds flat @ 14.5 for about 6 seconds.

I am adding an updated graph with PW, AFR and EGO correction plotted together with better colors.

 

Here you go. Mine looks pretty 'normal' compared to yours. I set the range on the PW to be 1 and 3 so it looks larger rather than a fairly flat line.

 

You have been a huge help. Thank you!

Now I have to figure out if I have an exhaust leak or possibly a bad sensor.

 

Just a little more info about my setup.

Purchased: Spartan 3 - Spartan 3 with LSU ADV

Unless I moved something: I think the sensor is further downstream than the old glowshift LSU 4.9 sensor which is in the downpipe about 1ft from the turbo outlet, similar to stock O2 location. I think the LSU ADV is just before a high flow cat.

 

Same, the controller and sensor came directly from 14point7. I purchased it in Dec 2023. It was shipped with FW 1.08. I just down graded the FW to 1.04. Which is probably the version you are running.

I am still seeing the flat spots, but they are not as long as in my post above. They were reduced to 2-3 sec from 5-7 sec Still not right. Not sure what that tells me, but it is another data point.

Regarding the installation location, just before the cat in my FM downpipe.

 

@natewin What firmware version are you running on your MS3?

 

MSPNPPRO (plug and play MS3 from DIY Autotune)
Serial Signature: MS3 Format 0566.05P

 

I think the signature defines the format of the mainController.ini then is applied to the .msq. Or something like that...

You can get the firmware version from the title of TS window. I am running 1.5.1

 

"MS3 1.5.2 release 20211110"
ms3-release-1.5.2.zip

 

Unplug your coils if you update it. IDK if it is still necessary on this version of MS3 but not doing so can cause problems with some hardware.

 

Solved. Replaced with another CAN wideband. This was definitely not the way I wanted it to play out.

 

What make / model is your new wide band?

DNM

 

The new one is an AEM 30-0310.

 

Did anyone attempt to contact Alan at 14Point7 about this issue?

 

Yes, I did. Alan and I exchanged emails for about two weeks. He thought my logs looked fine. All of which had similar flat spots between 14-15, like the one above. Below and above the 14-15 range it looked good. I even referred him to a post on the 14point7 support site (link above). Where another customer reported a similar issue. Honestly, I gave up because he seem convinced that it was fine.

Did I take the easy road out, probably.

 

@elp any particular reason you didn't want to run with it set at Perf=1?

I have a client's car doing the same thing, only the duration of the flat spots is much more random and less duration, maybe a max of 2 sec.

 

I wonder if the new 1.6 firmware would have helped at all?
The FW 1.6 announcement states "Add built-in support for Spartan3 CAN wideband."
https://www.msextra.com/forums/viewtopic.php?t=79196

 

Maybe, but I think this is an issue with the Spartan's way of interpreting the sensor data. I have the same issue with mine using a Link ECU. Perf=1 works as expected with a lot of noise, and Perf=0 has dead spots.