NC/MK3.5 2.0L P0328 Knock Sensor Bank 1 High Input, Sensor Appears to be OK

Technical Area Engine, Transmission, Exhaust etc
1
  1. My model of MX-5 is: 2012 NC/MK3.5 2.0L
  2. I’m based near: Bedfordshire
  3. I’m looking for technical help or recommendations on: P0328 Knock Sensor Bank 1 High Input fault

Hi All,

Long story short - I’m getting P0328 Knock Sensor fault codes but the Knock Sensor appears to be ok with 4.87/4.88 megaohms measured across both pins (A and B); although i have spotted a bit of dodgy wiring on the ECU around pin 2X (Ground; Shield Wire) and both Knock Sensor and Front O2 Sensor shield wiring but this does appear to be OK with no breaks.

Longer story -

Purchased my NC at the start of the year and noticed, after a week or two, a slight loss of power where it would hold back a little (very minor to be honest) between 2-3K RPM with it being most noticeable in second or third (always with load; would rev freely out of gear) and pulling perfectly fine above 3-3.5K.

Spotted a P0328 fault code, cleared it to see if it would come back again and it did after a week with the same loss of power and holding back feeling.

Spoke to a local specialist (not one i’ve used) who suggested to run “a few full tanks of Premium” through it and perhaps a battery had been disconnected at some point.
Thought it was a bit suspect and odd advice but being new to MX-5’s assumed it was a ‘quirk’ and i did exactly that. Kept brimming the tank with Shell V-Power or Costco ‘Premium’ (car hasn’t ran on anything lower than 97RON since i’ve owned it and i haven’t, until recently, let it go below third of a tank) and after another week or two, the lost of power stopped and i didn’t see another fault code for months.

Until the beginning of this week when i ran the tank down to a quarter full and now it’s holding back with a big loss of power when pulling off in first (foot to floor and there’s very little) until it gets to about 2.5K RPM and it starts to get going. Although now there’s a lost of power and that holding back feeling throughout the whole rev range and generally the car doesn’t seem too happy.
Will add that there has been a few times, between switching off the engine, that it feels not perfect but certainly a lot better.

I am however getting the same P0328 fault code and clearing it doesn’t seem to help with it appearing again after a few minutes.

Just to add, at no point have i seen a warning light on the dashboard whilst i’ve owned the car.

So, as the Knock Sensor looks like a pig of a job i thought i’d do my due diligence and quickly run through the service manual and make sure the KS is faulty before shelling out getting it replaced.
From what i can tell it seems to be OK, measuring 4.87/4.88 megaohms across both pins (A and B).

However, i did spot a little bit of a bodge/dodgy wiring (more pictures here) on the ECU/PCM, and from the wiring diagrams, appears to be a “fix” (?) for the Knock Sensor and Front O2 Sensor grounded shield wiring -

2000×1500

Pin-outs and wiring is what i suspect but it’s a bit difficult to fully tell without removing the security/tamper/shock bolts, the connector cover and ECU/PCM connector.

So before i take it to Mazda and shout “Please fix it”, which i suspect they’ll change the KS, ‘Emission’ harness and a give me a massive bill to boot - has anyone come across anything like this before or have any theories?

Are there other areas i should look at that could possibly cause this issue and fault code?

I have read a failed or cracked Coil Pack, could possibly cause this problem or i am wondering if it’s fuel system related perhaps :man_shrugging:?

Apologies for the long winded post but hopefully someone here can shed light on this problem :grinning:

2

http://ncmiata-servicemanual.com/ServiceManual/manual/mazda/2008/miata/g5/books/g5w01/html/id010245806800.html

Don’t know if you have seen or followed this

3

I haven’t seen that page so thanks @mk375!
Although it’s interesting in that it states the MIL (Engine Warning light i assume) “illuminates” if there is an error/fault, which it hasn’t and it hasn’t be covered as it goes on an off with the ignition (first thought it had be covered).

Regarding step 5 and measuring each pin (A & B) of the knock sensor (harness-side) to ground, what should B+ be? I’m guessing less than 4.58V?
Will give this ago when the weather improves a bit :+1:

And out of curiosity, would Mazda be able to tell anymore with M-MDS or would it display the same P0328 fault code and not much more?

4

I’m no expert but I assumed B+ to be battery voltage which would indicate the high input short to power supply if that fault exists in the wire harness. The limit of my experience is measuring pin- outs with a multimeter.
Hope you get it sorted.

5

Thanks @mk375.

Quickly ran through the hardness-side of the KS sensor connector, getting 3.23V (to ground) on 2U (pin A on KS) and 0.01V on 2V (pin B on KS) with ignition set to ON (engine wasn’t started).

Battery voltage at the time 11.9/12.1V, so would i be right in assuming both the KS and wiring harness, certainly KS side, appears to OK?

Anyone have any other thoughts to what could be causing this?

Just out of interest, does anyone know what rough labour time is to replace the KS?
The part is cheap from MX5Parts but it looks to be a painful job to get to it with removing and refitting the entire inlet manifold.

Edit - Decided to install the MoA Torque Pro PID file and spotted that not only does the MIL display/light-up on Torque Pro (still not on the dash) but i’m also getting the occasional, happened three/four times in a 10 mile drive, degree (1.3-1.8) of Knock Retard under second and third gear 1.5K (or there abouts) RPM acceleration. Also had HightDetSwitch light up on Torque Pro within five minutes of the drive and didn’t light again when i switched the car off and started it back up.

I’m assuming this more likely down to “noise” from the KS rather than something to be concern about?

6

Can’t be much help I’m afraid but would try the continuity tests from KS plug contacts to ecu connector pins to rule the harness out.
Not sure if you would not expect some knock in 2nd or 3rd on load at 1.5k as you are asking a lot of the engine at that point.

7

Seems strange that when you brim the tank the fault goes away and now below quarter of a tank comes back. Check live readings for fuel pressure at fuel rail also these are prone to air flow meters giving same fault. Knock sensors are under manifold if 2ltr model not that bad to do

8

@Polly23 - I’ve brimmed the tank again (Shell V-Power) and it threw another P0328, cleared and have driven 60 odd miles without a single error or loss of power; it feels fine.

I have noticed that when the tank is low it makes an fairly loud “hiss” when undoing the fuel cap which would suggest it’s built up a vacuum in the tank. I would have thought the rollover/breather valve/fuel filler cap would prevent that happening - thinking aloud, is it possible for the vacuum to build, leaving the fuel pump struggling and the P0328 is a byproduct of lack of fuel to the engine?
I’ll crack the cap when the tank is low and see if that helps.

With the fuel rail pressure, i assume that’ll need an inline pressure gauge? Do you happen to know what an OK reading should be roughly?

Airflow meter is MAF? If so, that’s registering 0.65V with the ignition set to ON which appears to be OK from the MAF service manual. Similarly i can’t see any debris in the actual sensor itself.

@mk375 - I might give that a go if i can get the tamper bolts off, although do you know if Mazda sell replacements?

Much appreciated for the help so far guys :+1:

9

If your breather is not working properly, fuel starvation is a very real possibility imho.

10

Petrol tanks have been a closed system for years. On the US site they suggest that a leaking or insufficiently tightened cap can throw a code. The vacuum is normal.
Do you know if the plugs have been changed over the last few years?
11.9 - 12.1v is also low and would indicate a discharged battery. Double check battery connections are tight and clean as this may cloud the issue. Didn’t know there were anti tamper bolts to remove the ecu connectors. Check out this youtube vid for removal https://youtu.be/44EFhbzqWUk

11

@mk375 - Cheers again and i didn’t realise that about the fuel system, always thought they were open and “breathed”.

It had a full service (spark plugs as well) five/six weeks ago and good shout with the battery - will clean the terminals as it was looking a bit worse for wear the other day and will look at replacing it.

And i didn’t know about the tamper/shock bolts, which lock in the connect tamper plate, either until i tried to remove the PCM/ECU - you can see the recessed metal-cone-like bolt head (top-left) in the photo.
Was only when i Google’d that i found out that UK/EU models had it fitted to meet Thatcham category standards and why you don’t see it in other countries.

@999to5 - Thanks and i am starting to wonder (more so) if it’s fuel related.

Out of interest, does anyone know if Torque Pro/CarScanner/ForScan is able to measure fuel rail pressure? Or will i need to fit a gauge inline with the fuel line?
As i’m guessing when it happens again when i run the tank (or purpose) to low, measuring the fuel pressure will be a quick sure-fire way of determine if it fuel related or not.

12

Hi All,

Apologies for bumping this thread and the lengthy reply but, this issue and error has reared its head again.

Little update - towards the start of 2021 i took the “fuel pump” assembly out of the tank which appeared to be in immaculate condition and the tank looked spotless with no sign of rust. But i stripped the assembly down (clips, especially the fuel sock are a bit of a pain) any way and cleaned up the connections/connectors on both the fuel pump itself and on “top-hat”. Reassembled, cleared the P0328 error code and the car ran perfectly fine for majority of the year, even did a 1500 miles trip around Scotland/parts of the NC500 with no issue and no error codes.

Either way, a few weeks ago the issue popped up again - throws a P0328 (“Knock Sensor 1 Circuit High Input Bank 1 or Single Sensor”) and you can see, using FORScan/Torque Pro etc, the ECU/PCM pulling/adding 3 to 8 degrees worth of timing (Knock Retard).
And as before, no MIL/warning light is shown on the dashboard yet the MIL PID shows as being ‘on’.

Again, i’ve tested the resistance through the KS and it’s the same 4.87 megaohms (within spec). I’ve even jury rigged a brand new KS (and grounded it as if it was connected to the block) in place and it stills shows the error and pulls timing and that’s after driving it down the street and back.

The biggest difference now is that the error code(s) rarely clear, cleared once in about twenty or so attempts, and it’s now logging under “PCM” as well (shows two P0328 errors) - which seems to suggest, from reading around, that the error must not return within a certain amount of distance or time before it’ll clear (or possibly it needs Mazda to clear it?).

It’s been in with a local Mazda dealership for an hours worth of diagnostics (~£150) and they couldn’t find an issue with the car. Only thought they offered was to go through the fuel system but they wanted, at minimum, £800/900 to take the fuel tank out and were addiment that is the only way to access the pump even though i explained there’s a top access panel on the floor behind the roll-hoops/PRHT control box.
Spoken to another local Mazda dealership and they basically said they would keep going over the car, if you agreed, at £120+VAT per hour until they find the fault and if they can’t, then they would forward it to Mazda Europe for information - which doesn’t particularly fill me with confidence other than a big bill.

Have also spoken to four MX-5 specialists (all seemed like really sound guys) around the country and they’ve never seen anything like it and they’re all completely baffled with the issue although, they all seemed to agree that it’s unlikely the KS (failure is rare on the NC allegedly) but couldn’t quite understand why P0328 would show otherwise.

Suggestions were that is it fuel system but couldn’t understand what; it’s a fault with the PCM, which again is rare, but it might be worth ripping it out (those security bolts look a right pain) and sending it to a ECU place to get it checked over; and/or, to throw the car on a dyno and see what it’s doing.
Unfortunately, none of these places are local (2-4hrs each way) and whilst it’ll probably be fine (famous last words), I’m a bit hesitate doing a long drive and causing further problems, more so if it’s possibly running lean.

Just to add, i’ve thrown a new Yuasa YBX5053 battery in as the old/original Panasonic was looking ropey - issue still remains.

Because of the last “fix”, i’m thinking it’s fuel related/starvation so perhaps i should go through that pump assembly again and possibly replace the pump whilst i’m there but i’m not entirely sure what i did last time to “fix” the problem.
I know Moss Europe sells a pump which will probably be fine but, as i’m sure it’s aftermarket/pattern-part, does anyone know where i can get a new OEM part?
I believe it’s a Denso pump but i can’t seem to find a product code for it (mine had nothing stamped on the outer casing) and the Mazda item (#LFG1-13-350 / #13-350A) doesn’t return many UK results - both local dealerships will only supply the entire assembly for £300(odd)+VAT.

And is there a way i can measure fuel pressure or flow, either through OBDII or jury rigging something to the fuel rail/line?
Just thinking it would be a sure way of seeing if it is fuel related.

@Polly23 did mention “air flow meters” throwing the same error code; would this be the MAF?

Otherwise, has anyone got any additional thoughts on what could be causing this or where/what i could look at next?

The only alternative is to get it into Mazda, or chance it/trailer it to one of the specialists, and suck it up on the cost front :expressionless:

Again, much appreciated for the info, has helped massively so far :slight_smile:

Edit - Just to add, does anyone know if this would be an MOT failure? As said, no MIL/engine warning light is continuously displayed and it functions as normal (comes on and goes out when you start the car).

13

My car is a UK model 2014 NC 1.8 and I have the “check engine” warning light coming on along with the above error code when starting from cold. The car is very, very flat in the 1500-2000 rpm range and mpg is poor, perhaps 10-20% lower than normal. Resetting the error code when cold does not work - it just comes back when engine is restarted, When car is thoroughly warmed up it is possible to clear the code and it stays off until the next cold start. In this warm/code cleared state the car has plenty of go at low rpm (expected from Mazda torque curves in technical guide) and petrol consumption returns to being OK.

With the torque app. I can monitor the ignition advance and in the above ‘flat’ state there is almost no advance at all. Mazda tech guide says that KS retard is applied under conditions of high load so if KS were faulty I guess the MCU will apply max retard to be safe. Attempts to monitor the knock retard PID give no data so I guess the 1,8 L car does not use this particular PID.

My guess is that the KS is intermittent, being faulty when cold and OK when hot.

Any other ideas?