Nistune Forums

I ran into something interesting last night while tuning a KA-T with a 1995 ecu. The car had a stock SR T25 turbo on it, and made boost at very low throttle angles. From tuning the last 95 ecu with a VE table, I learned that you cannot change the flag to open loop in the fuel tables. So if the flags no longer work, my question is, what toggles the open / closed loop operation. I noticed that I could set the cell values in the air fuel table, around 50 TP and steady state rpm, lets say to 11 to 1, and at part throttle the ecu would stay at 14.5 to one, this is with the O2 sensor disabled. If I continued to give it throttle the air fuel would all of a sudden drop close to the commanded air fuel of 11 to 1, and if I slowly backed out of the throttle, I could see that the air fuel would remain at the 11 to 1 ratio in TP load areas that were 14.5 to one when rolling into the same load value. This leads me to believe there may be a throttle position based power enrichment threshold and hysteresis. The second thing I noticed was that at idle and light loads, I saw no effect to the air fuel ratio, through values in the VE table by making very large changes. In a previous post by Matt, it sounds like this table does affect the tuning, at least on some ecus , I wonder if it is only referenced in open loop or during power enrichment. These Nissan ecu's are still a bit new to me, but I know that US domestic manufactures started adding logic to delay power enrichment for emission reasons in ecus around the time OBDII came out, it seems likely that Nissan would have some sort of strategy to delay power enrichment as well. Maybe someone can share a disassembly of these VE table type ecu's and we can step through the fuel routines to find out whats going on .

- Chris

http://www.dcperformance.com

I ran into the same problem with a european model S14a (82F06 ECU)
Below ~ 3000RPM and up to ~ 30-40 tp it refused to get out of closed loop, whatever I tried.
for a car with stock injectors/afm/turbo this is probably ok, but not if you had to rescale TP because of different injectors/AFM and a fast spooling turbo.

There's probably a couple of variables or some code to ignore the fuelmap (flags) below a certain TP/RPM.

I'm surprised that no one else has commented on this besides Eric. The more I've had to deal with these later ecu's the more I understand why many tuners prefer to convert over to the earlier S13 ecus . If this open loop / closed loop threshold calibration function could be figured out and defined in Nistune, this would make tuning these controllers a lot more practical (and safe) for boosted applications.

- Chris

http://www.dcperformance.com

Agreed its something that needs further investigation. I had nothing to add at this time since the O2 sensor code is quite complicated and we dont know enough about it

Honestly it is not something I have had to deal with yet when tuning my ER34 which also has VE tables. But the S14A that Eric mentions doesnt even have these VE tables. The parameters would have to be defined somewhere so its a matter of looking through the ECU code to trace through the conditions to enable/disable closed loop

I've added this to my investigation list

Eric wrote:I ran into the same problem with a european model S14a (82F06 ECU)
Below ~ 3000RPM and up to ~ 30-40 tp it refused to get out of closed loop, whatever I tried.
for a car with stock injectors/afm/turbo this is probably ok, but not if you had to rescale TP because of different injectors/AFM and a fast spooling turbo.

There's probably a couple of variables or some code to ignore the fuelmap (flags) below a certain TP/RPM.

eric,
you'll have to patch in the new address files which can be found in a thread in the type 3/4 section
but can you try the tps enrichment trigger by zero'ing off the table to see if this fixes the issue?
i've got a suspection the trigger is also some how linked into the o2 sensing aswell but have only tried it on my car as of yet

Bump,

I have a feeling my ECU has this issue too. It's a VE ECU and has the same symptoms described.

This is describing a problem i had with r32 HCR32 ecu on rb20det with o2 flagged on as normal it would seem to hold o2 values outside flagged o2 cells, it was mentioned to me at the time it was unusual and to me seemed i was on my own with this issue.
Others must have experience with working around this issue, I disabled o2 feedback as a last resort.
Would like to find out more when possible as well.

Are you guys unplugging the o2 from the harness as well or just removing it via nistune?

I leave o2 sensor connected and set the o2 feedback temp high so its out of range.

Could you define what is "VE" ECU !?
Could you point out this "VE" table ?

I posted this few time ago, you should read it:
viewtopic.php?f=6&t=2344&p=17069&hilit= ... oef#p17069

VE tables are only in later model ECUs which have Fuel Map (RPM vs TP) and Volumetric Efficiency map (RPM vs TPS)

Coefficients from both maps are added together (dependent on TPS position and values referenced in AlphaN Increase fuel table)

Nistune has called these AF Alpha coefficients STFT and LTFT in the latest version to be consistent with other tuning packages. They always affect the current injection time

My issue is similar in that if my primary o2 is connected, the car will not only hold on to closed loop, but I get a nasty lean spike to 16-17.0afr and once the revs go past 3200 or so it clears up and holds onto closed loop if im accelerating slowly, or map values if WOT. If I accell briskly from a stop its a HUGE hesitation that makes the car buck, as if im holding the brakes at the same time.

If I disconnect the primary o2 and run open loop I get no lean spikes AT ALL, regardless of throttle position. This is with 3 different sensors ( 2 oem denso (1 wire and 3 wire) and 1 bosch 3 wire) so I know it has nothing to do with the sensor.

Weird.

I posted this few time ago, you should read it:
viewtopic.php?f=6&t=2344&p=17069&hilit= ... oef#p17069

Been going through the Z32 code over the last few days. I've put together a symbol table based from your feedback and what I have been working through previously

The 8 bit Z32 ECUs use those async/sync throttle enrich tables (some of these I need to play with on the bench to understand them better). This is similar to how early SR20 ECUs do their enrichment

The 16 bit ones appear to use the VE table instead but I am still going through both sets of code. This is similar to ER34 RB25, B14 SR20DE, S14 KA24DE and anything else later year

Closing out older issues.

These problems noted are to do with the shifting of TP (by moving K constant) when resizing injectors/MAF

Keeping TP in the normal factory load ranges and using the added Total Injection Multiplier instead for fueling adjustment resolves this issue. Still to locate O2 feedback parameters (separate issue) for adjustment anyway

Found O2 feedback parameters use load scaling in a RAM map. Adjusting the 'Closed loop TP scale' when adjusting K constant will stop the ECU 'holding' onto closed loop

Refer updated Nistune Mapping document (Page 19) and 1.2.97 with updates for SR20/K11