Mark 2 Engine Fault Codes please

Technical Area Engine, Transmission, Exhaust etc
2

My 2001 1.8 has poor fuel economy and performance. It has a diagnostics socket in the engine bay but sadly no OBDII DLC socket. Using the LED/resistor method I have activated and counted the Flashing code sequence andam  getting a string of 4 error codes:  11, 13,14 and 15. I have a list of Mk1 codes but not Mk2. Does anybody have these? When I cross matched the codes to the Mk1 codes I got the following:

11    Code does not exist

13    Code does not exist

14    Barometric pressure sensor faulty

15    Oxygen sensor faulty

The car passed the MOT last week with no emissions issues. Can anyone help with a list of Mk2 codes? Does anybody recognise the code sequence and does it point to a possible fault (s)?

many thanks for your time

Peter.

3

The codes for NB are in groups of 4. 4 digit P codes. What you have would indicate a P1345 code (units digit of each number set). Don’t know off the top of my head what P1345 indicates.

4

It appears to be the cam position sensor code. This is normally not compatible with a working engine though, so you can try resetting the codes to see if it was a blip.

 

Edit: Read your post properly now…so it may well be faulty indeed. Check conections first though.

5

 1) Have you tried a reset? - disconnect the negative battery terminal and press the brake pedal for about 30 seconds? - then reconnect the battery.

  1. The output on the diagnostic  connector is a series of pulses, one is short and the other slightly longer. A longer pulse is a zero, but also a separator between ‘tens’ of short pulses. The code represents four digits, as an example the start of a code is always a longer pulse, if the actual code is four single digits then that would be followed by a short; long; short; long; short; long; short. = 1111 (number of short flashes)

Looking at your post, then a code 11 would be - long;long;long;long;long;short;long;short. (there isn’t one) I’ve checked through the pulses shown and there are no short pulse codes that match to your reading, nor can I match it if I include longer flashes.

Code 0110 is - long;long;short;long;short;long. Code 0115 is - long;long;short;long;short;long; short short short short short.

Never done the diode test, so I don’t know if one can discrimanate between a long or a short pulse, there’s no indication in the manual of the width of the two pulses unfortunately. Afraid you’ll have to decide? However, try the reset first!

6

 Just to keep the record straight, there is no ‘ten’ in digital counts, the count is from zero to nine, (that’s actually ten digits!) e.g. a nine would be nine short flashes. zero is a longer flash (as noted already) This follows standard arithmetic obviously, a ten would be long;long;long;long;long;short;long;long. If anyone is still puzzled, then that breaks down as long (start pulse) long (no thousands) long (space) long (no hundreds) long (space) short (one ‘tens’ digit) long (space) long (no ‘single’ digits) = 0010. (but 8 pulses - if you count them all!)

7

Thanks for the replies. The method  I used came from the Miata Haynes manual which involves counting the flashes (long & short) from the LED. From the posts it seems that I have completely misinterpreted the code sequence. 

Bear in mind that a long flash is around a second, and a short one about 0.25secs.

With the jumper and LED in place on the Diagnostic connector, when you turn on the ignition the following happens:

LONG then SHORT then a gap of about 2 secs with the LED off.

LONG then SHORT then SHORT then SHORT then a gap of about 2 secs with the LED off

LONG then SHORT then SHORT then SHORT then SHORT then a gap of about 2 secs with the LED off 

LONG then SHORT then SHORT then SHORT then SHORT then SHORT then a gap of about 2 secs with the LED off

I hope this helps - any more comments?

Thanks ever so much

Peter

8

 Ha ha - I’d signed off, and was about to shut down altogther, when the automated reply came in!

Strange one, the graph showing the peaks and troughs (pulse signal) doesn’t indicate a long gap between signals, in fact the gap looks the same as a long pulse width - however, applying logic the codes in order are

1; 3; 4; 5; = 1345 - - No SGC signal = Canshaft sensor does not detect SGC signaL Hope you’ve got a multi-meter?

Inspection: step 1) Ignore that bit. “Are there other codes?”  you haven’t said so!

step 2) Is there  poor connection on canshaft sensor connector? (WD40 and ‘work the plug’ push it on and off a few times) No joy? - go to 3

Step 3) Is sensor installed normally? (is it on good!) No joy? - go to step 4

step 4) Disconnect plug at sensor, turn ign s/w on - is there battery + voltage on connector terminal A? No joy? Check for open or short circuit between main relay terminal D and sensor connector terminal A (and it doesn’t say so, but check there is voltage at terminal D of main relay!) When checking terminal D - ign s/w on; if checking the wire between terminal D and terminal A, turn the ign off (resistance or open circuit check)

steps 5,6,7 & 8 concern connection between sensor connector and PCM (ECU) - forget that, unless you want to go to a lot of trouble!

Step 9 Is it now O/K? (you won’t know till you get to step 10 - - )

sstep 10) clear DTC (code) from memory (reset the PCM, as per previous post) If you get a recurance of code, then step 11) you will have to check the wiring between sensor and PCM, then if satisfied, reset again. If all else fails - replace the camshaft sensor.

Good hunting!

 

9

Thanks Gerryn,

Yes I have a multimeter. Just off to work and I’ll try your ideas later on this afternoon. I’ll post my findings.

Thank you again for all your time & trouble

Peter

10

I thought that the LED trick only worked on mk1’s?

11

It works on all Mk2s because none of the Mk2s are OBDII compliant (don’t have a DLC). They don’t have a MIL (Check Engine) lamp either so you have to use the LED trick to get at the diags info. The DLC and hence OBDII became available on the Mk 2.5. Anyhow, following Gerryn’s advice I checked out the CPS connector and found +12V on pin A, +5V on pin B and pin C is ground. So the wiring is sound. Removed the sensor, no sign of damage or metal particles clinging to it, so re-installed it. The engine would not run at all with the cam sensor disconnected so I’m pretty sure the CPS is working correctly. I reset the codes by disconnecting the battery and pressing the brake pedal for 5 secs. Took it for a 5 minute run and reconnected the LED. Guess what? No trouble codes at all now exist. So I guess I will have to check the diagnostics say every week and see if a fault gets recorded. 

I suspect the fault codes will reappear and wonder if any of us have really understood the protocol around the LED flashes. I have a suspicion that as the MX5 was OBDII compliant in North America from 1996, the protocol is the same for a 2001 UK car, it’s just there is no DLC. Applying the logic that a long flash is a zero, and the short flashes are units, the code sequence I got originally points to an air flow sensor fault and an EGR fault (P0103 and P0405). This sounds plausible as the issues I have are related to power loss and high mpg. Does anyone know if these are common faults? 

Thanks, Peter

PS: After disconnecting the battery I now have no radio and don’t have the code for it. Great.

12

Practically the whole ECU sensor system is a common fault…on all cars. {#emotions_dlg.indif}

They can usually be tested fairly easily, but I’m not familiar with those parts on the mk2.

13

 Peter, it’s standard procedure to depress the brake pedal for 20 secs, it says so in the manual - I said 30 for safety! If it reset after 5 secs then fine, and as you say the code has cleared. (Good) When you pulled the sensor plug, then replaced it, this cleaned the pins slightly - sufficient to allow it to work at present. What you need to do is get some switch cleaner - or WD40 - and spray a small amount either into the sensor socket, or the connector plug, and ‘work it’  back and forth a few times, this helps the cleaner to clear off any oxide on the pins. If you don’t do this, then the fault will reoccur again at some point, as they say S*** happens!

Forget your original interpretation of the codes, they are irrelevant, you’ve proved that. What was entirely possible, with an ‘iffy’ connection, would be rough running, and higher petrol consumption as the PCM tried to run the car with inadequate SGS signals.  Had the connection failed completely then the engine would have stalled, but a poor connection can fool the best control system - sometimes it works, and sometimes it doesn’t. The fault would be registered (as it should be) so we did have an indication of that problem.

What I didn’t expect after looking at the pulse graphs in the manual, was the 2 sec delay between each part of the code, but now we know this happens, it will help other members to make an led diagnosis, so credit is due to you for recording the time intervals, and the pulse widtths, even if .25 secs for a bit code was a guess, it’s enough to reach a conclusion. - Well done!

14

 

Jon - considering the application (on a car) subject to wet, damp, oil and grease, and a lot of connections, the electrics and electronics on a car do remarkably well, if the wiring and connections were made to ‘Lab’ standards then they could cost as much as the car itself, making such a vehicle unmarketable. I’ve done many control systems - mainly conveyor control, and they utilise similar wiring as a car, obviously these are not normally exposed to the same environment as a car is, but failure of components is rare on these, and to the credit of the designers, failures on a car are also rare - considering the environmental difference.

 ‘CAN’ systems (Ford are now using these, others will follow with similar ‘networked’ systems) are totally different, a lot of the electronics are now on the components, and the ECU accepts code sent from switches, sensors etc and transfers the commands to necessary components, so there is a lot more electronics on a modern car - frrankly I don’t think this is an improvement, though it is attractive to manufacturers as it saves considerably on wiring costs, there is virtually no wire other than a power cable and a signal pair (I assume) Fault finding is now down to an electronic technician, folllowed by an auto electrician, who may (or may not) be trained in servicing electronic products. As in modern domestic electronics, faulty components will be slung in the scrap bin, and if dealers don’t stock sufficient spares, then owners may have to wait for them to come from centralised parts warehouses.

With Mazda - that’s in Belgium!

 

15

Oh I totally agree. But…given the application (on a car) any engineer worth his salt would rather use mechanical means to achieve the same effect. It’s eurocrats and beardy greenies that force them to eek every last bit of efficiency out of an engine using processors, to the detriment of the owner’s pocket and convenience…

For example, my mini had a choke, operated by a 50p cable. My MX5 may well need a new ISC valve…which is just one of about 4 components that achieve the same effect…at an eyewatering cost. The difference is a few grams of CO2, to please the EUROs. As a result we’re all throwing electronic components full of exotic materials in the bin at a rate of knots instead.

16

 I have to agree - a few years ago I was at a car show - in a proverbial field somewhere; The bonnet was up on a 30’s Austin Seven, and the space round the engine plus scarity of wiring made me immediately realise how complicated our world is today. Even 70’s cars were simple in comnparison.

Take the radio and TV industry, at one time you rang someone, and a bloke came in, looked at the set, rummaged through his big parts box, pulled out a valve and replaced the duff one - he usually knew which one to replace too. Then came the wonder of electronics, and now they just come to cart the set away, which goes to a junk yard, to be shipped to Africa, where poor starving kids rummage through the pile to salavage what they can for pennies - -

Meanwhile, worldwide. millions of radio and TV engineers were out of a job. Progress? NBL!

17

Absolutely…if you quickly add up the Mazda prices for everything that now takes the place of one carburretor, you would need £thousands to buy replacements.

Granted, a carb needs more maintenance…but the point is, you COULD maintain them. My mini’s was 20 years old and was pretty temperemental, so I put it on the bench and rebuilt it with about £5 worth of parts, and about 4 hours work and had myself a new one. Or I could have bought an actual new one for about the price of just one injector. Or any mechanic in the world could sort it for me. Now, dealers take 4 figure sums for what is a basic bit of work, just because of the price of these components.

I’m not even that old, but I just can’t see that this is any kind of improvement over what mechanical components would be now if they were still being developed. Honda say that they could easily now build a carb that is considerably more efficient than injection, and more reliable, but under certain test condition it just can’t pass the EURO tests, and can’t be used.

18

OK Gerryn, will use the WD40 on the CPS contacts. What does SGS mean? The only way to really check the sensors and actuators is with a scope. So I think I’ll go out and buy one. Being able to see the actual signals being generated should eliminate any uncertainty. 

 

19

 Sorry SGS was a typo, should have been SGC, and no I don’t know what that is either! A scope will cost you, they aren’t cheap, why not buy a decent code reader if you feel one is justified, there’s plenty on Ebay. Just be sure you buy a good one. No matter what you spend, without a manual you can’t correctly diagnose a problem, you can only assume there is one, so buying any form of detector without a manual for the car is a waste of money, if you want my opinion. It’s very difficult to get a works manual from Mazda, or a dealer, as I understand it, you would need to live in the Outer Hebrides (sp?) and be able to prove there is no Mazda mechanic within three hunbdred miles before they will let you have one. Or, get lucky - as I did. Trouble is, it’s a Mk2 manual, and I wanted a Mk2.5! (you grab what you can) and no, I didn’t steal it!

20

When we were checking the codes on my mrk 2 the flashes were as , long flash -10 and short dash -single numbers so 4 longs and 2 short would be 42 which correspond with what we were checking .