Intermitent spark on ! & 3

[GeoffSmith]

As far as I was aware the O² sensor completely overrides the CO pot. So fiddling with the pot will make no difference.

Correct. The pot goes to an analogue input at B12 and the O² sensor to a digital input at D7. The open loop PROM only reads B12 and the closed loop PROM only reads D7.

[Chas54]

Thanks

It good to hear confirmation of my understanding.

Looking forward now to getting a new CO meter on Wednesday.

Charlie

[Simon_P]

Fredjohn thats what I understood also. My non-Lotus colleague said it wouldn't make any difference. From what Geoff says the CO pot is completely redundant.

Geoff, where does the fuel adjustment come from when the close-loop system is not in closed loop mode? Block learn value? Are we able to see that in elanscan?

[GeoffSmith]

Geoff, where does the fuel adjustment come from when the close-loop system is not in closed loop mode? Block learn value? Are we able to see that in elanscan?

That's what I would assume but I don't think anyone has learned to read the runes. Probably not an easy one to decipher either.

[RonR]

Many moons ago I found a disassembly of a 9239 SE and a 1499 S2/USA eprom. I think the guy who did it knew a lot about ECMs!
From what I've been able to decipher there's a 1-bit option flag which sets whether the ECM operates as an open or closed loop version (Option flag 4 bit 0).
The code is almost identical on both versions, but routines for open or closed loop are used depending on the option flag.
Here's a snippet of the Open loop code from a 1499 eprom:

Code: Select all

;--------------------------------------------------------------------
; CO Pot Voltage Lookup
;   --Result is used locally to set/clr CO Pot Out Of Range flag
;   --Note there are different CO Pot ECM Pin assignments for Turbo (B12) and NA (D8)
;   --Result also used to drive TBL# 17: CO Bias vs CO POT Voltage Offset vs MAP
;--------------------------------------------------------------------

; called from $E0FE Minor Loop 1

EEDE: B6 801F         LEEDE LDAA    L801F               ; 0111_0110 [OPTION FLAGS 4]
EEE1: 85 01                 BITA    #$01                ; check b0: (CO Pot Enabled / Open Loop Cal)
EEE3: 27 1B                 BEQ     LEF00               ; exit if bit == 0 (not enabled), else..

EEE5: 86 07                 LDAA    #$07                ; ch7
EEE7: BD EE7F               JSR     LEE7F               ; A2D READ channel 7:  CO Pot Volts
EEEA: B7 0113               STAA    L0113               ; save CO_POT_A2D_RAW1 (aldl 46)

EEED: 15 2B10               BCLR    $002B,BIT4          ; clear CO Pot Out Of Range flag
EEF0: B1 8B79               CMPA    L8B79               ; =$10  16d = 0.31 Volts
EEF3: 25 05                 BCS     LEEFA               ; br if < 0.31V ("Shorted" mode)
EEF5: B1 8B7A               CMPA    L8B7A               ; =$F0  240d = 4.71 Volts
EEF8: 23 06                 BLS     LEF00               ; exit if < 4.71V

; Mode is activated if grounded or tied high
EEFA: 14 2B10         LEEFA BSET    $002B,BIT4          ; set CO Pot Out Of Range (main flag)
EEFD: 14 7002               BSET    $0070,BIT1          ; set CO Pot Out Of Range (secondary flag)
EF00: 39              LEF00 RTS                         ;


So this confirms the CO pot is never read on a 1499 Eprom.
From other parts of the disassembly it appears that the CO pot reading on open loop ECUs is in the ALDL data - Elanscan has it as as "Flags 46"

Code: Select all

8C75: 0113    DC.W      $0113  ; (aldl 46) CO Pot Voltage / EGR Temp A2D #1


and the long term BLM figure on a closed loop ECM is also in the Elanscan data as "Flags 20"

Code: Select all

8C41: 00DC    DC.W      $00DC  ; (aldl 20) CLOOP_BLM {BLM} (Long Term) Counts = N  (Z0023.b1 determined Idle or Run cell)


I've checked these on some old elanscan files and they appear to be correct.

[GeoffSmith]

Interesting, especially the reference to turbo and NA. Have you got a link Ron?

[RonR]

I'm pretty sure the zip file came from the now-defunct moates.net forum, and was downloaded in March 2019.
There's no author's identity that I can find in the file, though there are some helpful notes included.
I'm not sure if there are any privacy/copyright issues here, so Geoff, you have an email!
If it's OK with the mods it could be uploaded here or in the Wiki.

[Simon_P]

you have an email!

All your Christmases at once!

That could be really useful!

[Chas54]

No sign of new CO meter yet.
Covid ping at warehouse has delayed until next Monday.

[Chas54]

After long wait , new CO meter arrived.

Ive posted here to stationary data sets , 1st showing 2000 faults---- I had been checking ECM connections etc so think that's where these came from.
2nd set after faults cleared with battery disconnection.

I have around 3.5 to 4% CO.

As the voltage return at the O2 sensor reads 4.2v and the injector pulse is 1 ms, I am wondering if the O2 sensor is to sooty having run rich for some time.
Despite all warning to the contrary, I may take it out to see how dirty it looks - cleaning advice varies from wash in petrol, clean with a carb or brake spray or even soak overnight in vinegar.

I can't fathom the significance of the readings on the O2 integrators. An integrator in my previous career always measured a total flow as opposed to a rate of flow ie integration being the area under a curve.

I wondered if my O2 sensor is at 4.2v , and this is the very top of its range, does the Block Learned Memory then supersede the O2 sensor reading within the ECM somewhere ?

I do still have a couple of items to try , the fuel rail pressure ( awkward getting at connection behind the plenum) and swapping to the injector set I bought from the USA (via ebay)

I wondered if there is a filter sock of some type in the petrol tank , on the petrol return. I cant make out from the picture in the manual if its on the inlet or return but it shows something.

22 aug21 2st set after fault cleared.zip

22 aug21 1st set 3.8% CO.zip

Thanks for any advice

Charlie

[Chas54]

Washed O2 sensor in petrol and blown dry with air line. No change.
I've removed the bonnet and now have loos Ned the petrol pipe bracket on the rear of the plenum. Now I can easily get at the fuel inlet line joint to insert my "T" piece . I renewed the insert a month or two ago .

[Simon_P]

Is your O2 sensor heater working?
Which O2 sensor do you have - Part number
what injectors do you currently have? Part numbers or pictures

[Chas54]

Simon
I bought the O2 sensor from SJS in April part no. A100E6024S
The injectors came from ebay USA .. they have the following no. on them 2112 17113749 I am just about to fit these.

The old injectors are 0050 1/069116..

I hopefully have included a couple of pictures.

They O2 sensor has 12v supply and the return voltage to the ECU looks good too.

I think the heater works because I see the O2 sensor return voltage change when not idling. I rises to 4v approx. after starting.

i suppose I could remove it and check temperature with the ignition on and the lead plugged in just to confirm.

Lotus new grey old black.JPG

lotus O2.JPG

regards
Charlie

[Simon_P]

Hi Charlie,

17089116 is the correct injector number.
I'm sure SJs will have supplied the correct part ( noting that the genuine Lotus part is listed at £405)
Can you check the resistance of the heater and the input voltage to be sure that it is working?

It only takes the right amount of fuel and the right amount of air and a spark at the right time. There aren't many things that affect any of those so just a case of going through them one by one.

Integration is used for error correction, the size of the error is related to the area under the curve. I'll guess that the block learn is to correct a constant offset.

Scans are most useful from cold to the fans coming on as there are more reference points. A scan of a warm engine just indicates the state not how it got there or where it went wrong.

Reset the ECU after component change or adjustment. Similarly scan after a reset to see the errors before they have been removed.

As you mention using it could help, at full chat the engine output is 121kW thats a lot of heat or 60 toasters.

[Chas54]

Simon

The O2 sensor from SJS cost about £50.

Ive just been checking the resistance and the heater circuit is 13.1 ohms.

The other circuit within the O2 sensor appears to be open circuit though. i'm not sure if this is how it should be though.

That's the variable potentiometer circuit. Ive measured between the O2 sensor cabled earth and the O2 sensor cable that would connect to the grey cable with green band on the car plug ( goes to ECM D7 terminal ). Also measured to the O2 body earth and still appears open circuit ?

I am still pretty confident the problem is overfueling and am just hoping it might have been one of the old injectors passing.

Ill make sure when I test that I collect the data from cold.

Thanks again for your sound advice

regards
Charlie

[Chas54]

Car will barely run with the new eBay injectors. 0.65 vacuum showing on elanscan so wondering if I've created a vacuum leak disturbing the plenum.
I'm going to view the operation of the injectors with the fuel inlet rail detached and on the top of the engine.

[Simon_P]

Have you reset the ECU?

[Chas54]

Yes reset ECU.

I've got the plenum off again
. I've also tested my old injectors again with the fuel rail on too of the engine .
No.3 definitely passes so will be trying the new set again . They pass my non dribble test.
Ive damaged an injector rail clip during attempts to braze something to it to make extraction easier.
I've settled for a bent gas welding rod now that hooks round the two clip legs. Hopefully I will now be able to pull these clips of 3&4 without taking the plenum off.
Regards
Charlie