Start tuning SR20DET

Discussions concerning the M7790 cpu

Moderator: Matt

kamilp1908
 

Posts: 34
Joined: Wed Oct 23, 2019 5:46 am

Start tuning SR20DET

Post by kamilp1908 »

Engine spec:
SR20DET JDM
GTX28
Hks Cams 256/260
Subaru 440ccm
Z32 MAF

I changed INJ (adjusted Load max and crank table, didn't change Load scales), then MAF (didn't change load scales). O2 sensor turned off. AFR sensor in tail pipe.
I updated inj latency to 770us. Changed K konst to 39100. With this setup I had AFR around 14.7 on idle.
I made some cruising runs 4/5th gear (2.5k-2.8k). I got AFR around 14.5ish. Looked nice.

But when I started to analise log I have found 2 problems (???):
1. After gear change AFR rises to 16.2 then drops to 11.5, but it goes to 14.5 (it takes about 1.5sec from rich to normal)
2. At 5th gear at 2.6k rpm I have TP 19 (AFR 14.2), but Min Load for this rpm is 25.

My questions:
1. Is it normal to AFR behave as like this after gear change?
2. Which TP value ECU uses to calc injector time?

Maybe my problems are connected with not changing load scale during MAF/INJ changes.

Log file recorded crusing from 3rd to 5th gear (about 2.5k-2.6k rpm)
Attachments
Nistune_2021-03-14_1115_40.csv
Crusing from 3rd to 5th gear
(221.96 KiB) Downloaded 94 times
Nissan14.04.2021_cruise_K_OK.bin
Current ECU bin
(32 KiB) Downloaded 108 times
Matt
Site Admin
 

Posts: 8961
Joined: Sun Jan 29, 2006 1:45 am
Location: Adelaide, Australia
Contact:

Re: Start tuning SR20DET

Post by Matt »

When you change gears, we see the MAF voltage drop (no load), so TP and injection time drop accordingly

We see about 15-16 AFR on decel, but then when you get the increase in load it seems like it takes a while for the wideband to register.

16:907 we have 14.7:1 AFR (4.37ms injection time)
17.891 we have 16.5:1 AFR (4.57 ms injection time)

Do you 'feel' a lag with acceleration at this point, where it affects drivability?

It's a bit strange how it goes lean, then slightly richer 14.96 (at 18.079) and then lean again before going rich
Attachments
2021-04-19 08_21_15-Nistune [PS13_RPS13 (SR20DET) (Version_ 3)].png
(55.11 KiB) Downloaded 2585 times
kamilp1908
 

Posts: 34
Joined: Wed Oct 23, 2019 5:46 am

Re: Start tuning SR20DET

Post by kamilp1908 »

We see about 15-16 AFR on decel, but then when you get the increase in load it seems like it takes a while for the wideband to register.
1. 15-16 AFR on decl is proper value?
2. I have LC-1 wideband is plugged in tail pipe so response can take some time. Maybe better place will be oem 02 sensor after turbo?
Do you 'feel' a lag with acceleration at this point, where it affects drivability?
It was my first drive to setup K const for cruising (with fuel cut at 3300 to avoid any boost spikes). Next time I will check response 'feel' after decel.
It's a bit strange how it goes lean, then slightly richer 14.96 (at 18.079) and then lean again before going rich
I have to analize others logs for this problem. Maybe it one time thing or some common problem with my car.
1. Does transition from lean (16-17) to rich (11.4) after gear change is ok? Maybe I need to lower throttle enrich?
PL
 

Posts: 957
Joined: Fri Feb 03, 2006 3:29 pm
Location: Adelaide, South Oz
Contact:

Re: Start tuning SR20DET

Post by PL »

Does this cause any driveability issues? If not then I wouldn't worry about it.

Could also be something to do with AFR sensor in tailpipe. Never ideal. Best in a dedicated fitting in the down pipe.

PL
kamilp1908
 

Posts: 34
Joined: Wed Oct 23, 2019 5:46 am

Re: Start tuning SR20DET

Post by kamilp1908 »

PL wrote: Mon Apr 19, 2021 9:31 pm Does this cause any driveability issues? If not then I wouldn't worry about it.

Could also be something to do with AFR sensor in tailpipe. Never ideal. Best in a dedicated fitting in the down pipe.

PL
Next logs I will make with O2 sensor in down pipe.
Thx PL.
kamilp1908
 

Posts: 34
Joined: Wed Oct 23, 2019 5:46 am

Re: Start tuning SR20DET

Post by kamilp1908 »

Matt wrote: Mon Apr 19, 2021 9:56 am It's a bit strange how it goes lean, then slightly richer 14.96 (at 18.079) and then lean again before going rich
I have looked in other logs, this strange lean-rich-lean behaviour occurs on every gear change (3->4, 4->5).
Matt
Site Admin
 

Posts: 8961
Joined: Sun Jan 29, 2006 1:45 am
Location: Adelaide, Australia
Contact:

Re: Start tuning SR20DET

Post by Matt »

The ECU does a fuel cut/recover on deceleration in certain RPMs bands (not expected RPM ranges in your log) but the readings from the wideband don't really match the injection time, which is why myself and PL are both thinking there is a delay with reads, and also questioning if this affects drivability (compared to before upgrading injectors/MAF)
kamilp1908
 

Posts: 34
Joined: Wed Oct 23, 2019 5:46 am

Re: Start tuning SR20DET

Post by kamilp1908 »

PL wrote: Mon Apr 19, 2021 9:31 pm Does this cause any driveability issues? If not then I wouldn't worry about it.

PL
Last week I have done some testing. I forgot to take consult cable and 02 wideband sensor :/
So I made a little driving to feel the car. My "ass-dyno" felt normal. It looks like that cruising is ok.
kamilp1908
 

Posts: 34
Joined: Wed Oct 23, 2019 5:46 am

Re: Start tuning SR20DET

Post by kamilp1908 »

I had some time to make logs from cruising.
I have turned on the O2 sensor to check SFT and LFT. Not using any wideband O2 sensor. I just wanted to check my first tuning in normal driving.

My observations:
1. Engine behaves normal in my opinion, there is no problems with gear transitions on low load.
2. From my previous log (Sun Apr 04, 2021 6:24 pm) I have lowered K (from 39100 to 38100) and TPmin table to reduce possible STF on deaccel.
3. I have some problem with STF and LTF, because LTF is about -8% and STF can be even -17%.
Ex from current log:

2.38.704
SFT=-13%, TP=11, RPM=2,5k (is this due to TPmin for this revs is 18?)

3:38.032 2nd gear acceleration
SFT=-11%

4:01.500 5th gear cruise
SFT=0% (perfect situation)

5:42.8 idle
Very big InjTime, so it cause in short time a very big SFT=-13%


Do I have to reduce K more?
Or maybe I shoud reduce TPmin table to prevent from "big" FTS on very light cruise?
Or maybe I am missing something else?
Attachments
nistune_2021-08-11_1411_13.csv
(955.14 KiB) Downloaded 105 times
NissanSR20_TP_MIN_DOWN_rev4500.bin
(32 KiB) Downloaded 93 times
PL
 

Posts: 957
Joined: Fri Feb 03, 2006 3:29 pm
Location: Adelaide, South Oz
Contact:

Re: Start tuning SR20DET

Post by PL »

If it's running nicely I wouldn't touch it. I only ever look at the fuel trims under cruise conditions. They can be all over the place during transient operation.

Looks like it could do with a sniff more fuel if LTFT is significantly negative. I'd be tempted to just trim it out by increasing your latency figure if mixtures are good on boost.

PL
kamilp1908
 

Posts: 34
Joined: Wed Oct 23, 2019 5:46 am

Re: Start tuning SR20DET

Post by kamilp1908 »

PL wrote: Thu Aug 12, 2021 9:21 pm Looks like it could do with a sniff more fuel if LTFT is significantly negative. I'd be tempted to just trim it out by increasing your latency figure if mixtures are good on boost.
PL
I don't understand term "sniff more fuel". If LTFT is negative, so I should decrease fuel ex. by increase inj latency.
Do I understand it correctly? (I have subaru's 444cm inj with latency set to 770us)
Matt
Site Admin
 

Posts: 8961
Joined: Sun Jan 29, 2006 1:45 am
Location: Adelaide, Australia
Contact:

Re: Start tuning SR20DET

Post by Matt »

PL means a slight amount extra

With regards to LTFT negative, it means your mixtures are too rich and the ECU is pulling out fuel to maintain stoich

So in that case, add more to K/TIM (by a similar %) until it is nearer 0% on light cruise
PL
 

Posts: 957
Joined: Fri Feb 03, 2006 3:29 pm
Location: Adelaide, South Oz
Contact:

Re: Start tuning SR20DET

Post by PL »

Sorry yes I meant a sniff LESS fuel as it's trimming in the negative direction.

PL
kamilp1908
 

Posts: 34
Joined: Wed Oct 23, 2019 5:46 am

Re: Start tuning SR20DET

Post by kamilp1908 »

Matt wrote: Thu Aug 12, 2021 10:40 pm PL means a slight amount extra

With regards to LTFT negative, it means your mixtures are too rich and the ECU is pulling out fuel to maintain stoich

So in that case, add more to K/TIM (by a similar %) until it is nearer 0% on light cruise
From other topic:
Matt wrote: Mon Aug 16, 2021 11:06 am Get the vehicle at cruise (around 2000rpm) and check the mixtures here. If they are too rich then pull back TIM/K until it is stoich (for your base mixture settings)
I have some problem with those statements. I have little to rich on cruise (LTFT are about -8%). Matt suggested me to increase K, but in other thread to decrease K in case to rich AFR. Which is correct?
PL
 

Posts: 957
Joined: Fri Feb 03, 2006 3:29 pm
Location: Adelaide, South Oz
Contact:

Re: Start tuning SR20DET

Post by PL »

Increasing K, TIM and/or Injector Latency will increase fuel.

PL
Post Reply