ELD WB input - open loop - random AFR's

s300 and SManager software questions & answers
Post Reply
Shuttle
Posts: 8
Joined: Fri Sep 04, 2020 9:32 am

ELD WB input - open loop - random AFR's

Post by Shuttle »

Hello,

Hopefully someone with more understanding than I will be able to help please.
I'm banging my head against the wall trying to figure this out as to me it makes no sense (to me!).
  • OBD0 chassis with D16A6 originally
  • Converted to P28 and swapped to D16Z6
  • 500cc RC's low impedance, resistor box fitted
  • Basemap from Hondata library stock P28
  • ELD disabled, knock, baro, dead times set
  • open loop but stock O2 fitted
  • D10 ELD selected for WB input, no offset,
  • no ELD fitted to the car, the ELD pin is only wired to the WB.
  • Innovate LM1 connected to ELD input.
  • R136/R138 removed
Log attached
shuttle_ELD_problem.s3d
(1021.83 KiB) Downloaded 121 times
Calibration attached
shuttle_2.7.2.0-base.skl
(46.73 KiB) Downloaded 123 times
and video - inline

What I see happening is that with the aux from the LM1 to the ELD disconnected the LM1 is showing on it's screen AFR's around 20-24 at idle
But idle is smooth-ish

When I plug the cable in (at 1min 15sec) immediately the idle starts bouncing around and AFR's now go to 14-17 and the display tracks the hondata AF reading very closely.
But it wont idle.
I can hold the throttle higher and it wont hold RPM.
Soon as I unplug the connection from LM1 to ECU, (at 1min 40sec) once again it idles but AFR's go to 20+
Video:: http://www.hondacrx.co.uk/AFR-or-idle-Pick-one.mp4


I tried selecting a different input for wideband (at 2min 5sec). D12, or B6, the problem persists, AFRs' track the LM1 being plugged in or not.
The videos show whats going on with the LM1 readings when they are not being logged, and the changing of inputs having no effect (realtime update was ticked!)
Video:: http://www.hondacrx.co.uk/WB-input-swap.mp4


The calibration I've up loaded onto the ECU for this test is me starting again from nothing.
I did manage to part tune the fuel this way on a previous calibration, and it would be stable
but as soon as I remove the WB connection it goes to hell.
If I try to run in closed loop with the narrow band it goes to hell too.

But, I did get the AFR's hovering around 14.4-15.5 consistently, with no load across the RPM range.


I have tested this ECU (call it A) and this S300 (also A) in a different car which is tuned by someone else.
I put the calibration from the ECU & S300 in that car (B) on to this ECU (A) and it ran the car fine.
For clarity, that's 2x ECU's and 2x S300's
I've not tried the other way around because I dont want to risk anything to the working car!! although I expect the problem would remain.

I have tried a different wiring harness (from the other car) and still same problem.
I have been over the wiring looking for any differences as not all USDM/EURO chassis pins are the same function!! and no problem there either.

Please help.
I want to get this tuned a little bit NA, and pass emissions, before fitting the turbo later.
If it's not running correctly now, I didn't want to add to problems with boost just yet.

Thanks
Dave
Shuttle
Posts: 8
Joined: Fri Sep 04, 2020 9:32 am

Re: ELD WB input - open loop - random AFR's

Post by Shuttle »

More testing today.

I have a crude test harness for bench testing ECU's out of the car.
I hooked up a pot to the ELD pin and tested it between 5V and 0V and the readings in SManager track what a DVM says is coming out of the pot.

So brain wave, back to the car, hooked up a pot to the ELD pin instead of the WB and simulated the AFR voltage from 0V to 5V on the ELD pin.
The simulated AFR's seem to track what the actual WB is reading - the WB is not connected to the ECU at all.

I fear I'm being thick...
But...

If I'm in open loop, why is the ELD input having an effect on the fueling ?
The ECU is reading the ELD to get AFR and then adjusting the fueling.
In open loop.
I thought that wasn't supposed to happen? Else how do I turn off AFR compensation ?

What am I missing ?! A simple setting to turn off tracking AFR?!

The help file itself says::

Code: Select all

Open loop (closed loop disabled)

This setting disabled closed loop. Disable closed loop while tuning. Otherwise the ECU will adjust the air/fuel ratio resulting in an inconsistent air/fuel ratio. 
But it's not working ??

bench-afr.jpg
bench-afr.jpg (134.18 KiB) Viewed 4038 times
green circle shows ELD being controlled to 0V by my potentiometer.
red circle shows AFR being read into the lambda table, it's trending downwards but it started high as ELD was set to 4.98V before testing started and it's taken about 3000 samples for this reading shown.

Once again I've tried this on two ECU's a P28-A02 and a P30-G00 with J12, R136, R138 removed.
Erasing the ECU and uploading a new base calibration before testing.
Shuttle
Posts: 8
Joined: Fri Sep 04, 2020 9:32 am

Re: ELD WB input - open loop - random AFR's

Post by Shuttle »

Any suggestion please ??

This seems to me like:

1) I'm doing something stupid - what ?
2) A bug in the S300 V1 hardware/software!

if ELD is not enabled for WB input, it's voltage still affects the AFR in open loop, no matter if you set WB input to EGR, or anything else.
That's entirely against how it's supposed to work, no ?
Shuttle
Posts: 8
Joined: Fri Sep 04, 2020 9:32 am

Re: ELD WB input - open loop - random AFR's

Post by Shuttle »

I really didn't want to mess with my other car but needs must.

It's also a S300 V1.
It was tuned before I bought it.

The ECU has R136/R138 still fitted.
It runs fine in the Chassis it is in. With boost.
it's set to open loop in the map.
WB is not connected up to it yet - thankfully!


Put this into the shuttle and tested the ELD.
No WB connected it actually runs reasonably B16 vs D16, both RC550's).
Let the car warm up.
Plug in the WB and AFR's go crazy.

So as far as I can see, this is a problem with open loop in S300 V1
ie: it doesn't work. ELD always affects the AFR's regardless.


Similar posts allude to this:

viewtopic.php?t=8259
"Using ELD input as wideband is not a good idea. S300 disable ELD error only, but ELD correction is still working! You can tune maps with wideband at ELD connected, but when you detach wideband from ELD input - AFR will be poor."

viewtopic.php?t=19823
"One huge thing I'm noticing is that ELD is still receiving voltage. I had open loop and I selected D14 pin (currently D14 is disconnected from O2 sensor)
How would ELD be receiving voltage without any wires attached to D10?
"


This appears to be a known issue.
It makes open loop unusable as recommended, so one would think it would get fixed?
Is S300 V1essentially now abandoned for support ?
Does V3 have the same bug ?
DaX
Posts: 654
Joined: Tue Mar 18, 2003 3:48 pm

Re: ELD WB input - open loop - random AFR's

Post by DaX »

Here is my setup:

S300V3 in a P28
R136 and R138 removed
Wideband (PLX Devices) wired into ELD input (D10) on ECU
Open Loop

I do not have any issues with my AFR or the ECU acting like it's in closed loop when it's in open loop like you are describing.
Shuttle
Posts: 8
Joined: Fri Sep 04, 2020 9:32 am

Re: ELD WB input - open loop - random AFR's

Post by Shuttle »

Thanks!!

watch your AFR's to get a baseline
pull the WB connection to the ECU so the ELD pin floats **
watch your WB readings change

If that doesn't happen, can you please share:
  • your calibration,
  • version of SManger
  • firmware on your S300,

I note you said you have a V3, I have a V1 I think this is the issue...

** the resistors are pull up/down to create a potential divider and give constant voltage when not overridden by the stock ELD circuit. Removing them and not connecting a WB the ELD pin floats and picks up electrical noise expressed as a random voltage.
DaX
Posts: 654
Joined: Tue Mar 18, 2003 3:48 pm

Re: ELD WB input - open loop - random AFR's

Post by DaX »

I don't have an easy way to disconnect the wideband from the ECU, so I'm unable to perform that test - sorry.
My calibration is attached.
SManager 2.7.5.0
I don't know how to verify firmware on the S300.
Yes, mine is a V3.
Attachments
2020-08-20.skl
(43.49 KiB) Downloaded 124 times
Shuttle
Posts: 8
Joined: Fri Sep 04, 2020 9:32 am

Re: ELD WB input - open loop - random AFR's

Post by Shuttle »

Thanks, I took a quick look but I will check it out page by page later to compare.

Saw that you have flex fuel input configured to ELD as well as your wideband.
That would mean they conflict with each other !?

From the online menu, information, will tell you the firmware version.
V3 is waaay ahead of V1.
DaX
Posts: 654
Joined: Tue Mar 18, 2003 3:48 pm

Re: ELD WB input - open loop - random AFR's

Post by DaX »

Yes, I am running flex fuel, but I don't know where you're seeing that I'm using D10 as the input for that, because I'm not. It is configured as AN0 for flex fuel input.
Post Reply