About 2 years ago I had an issue with my 58 plate 1.4TDI Greenline where the bar type fuel level gauge was not indicating a level, and the outside temperature gauge was not reading. The trip computer did not indicate a range, even though the other functions for MPG, Average MPG, Distance travelled etc continued to work. Eventually I bought a 2nd hand (£12.74) console and used VAG Key to extract the pin codes from the old and new consoles. This allowed me to fit the new/original 133K vs 109K mile console and code in the immobiliser and match the keys. This gave me a fully functioning console.

 Our electrics wizard @Breezy_Pete offered to have a look at the console for possible issues, and will now do a follow up in this post with his findings.

Thanks for introducing this Keith, and thanks for sending the cluster so I could take a look. 

 

This failure was mentioned by Keith in another recent instrument cluster related thread, and my interest was engaged because I remembered once looking at the circuit dealing with outside temperature and fuel level within a Mk1 Fabia cluster, finding them to share some components.  I thought that it might be possible to repair this cluster's failed functions if I could find such a component in a failed state.  

 

I did find a failure, which has restored those functions following repair, but it wasn't a component as such.  During the process I have sketched out the circuitry associated with the two functions, by close inspection of circuit board.  The root cause of the failure remains a mystery, unfortunately, as does the purpose of the 'thing' that failed.   Mk1 Fabia cluster doesn't seem to have the same 'feature'.

 

Pics coming this evening... 😊

 

 

 

 

Edited 27 June, 20223 yr by Breezy_Pete

I didn't spot anything visually the first time I looked at this cluster circuit board, but the second time I picked it up my eye was drawn to something near the green connector plug:

 

This strangely necked-down uncoated bit of track.  Under the microscope it was clearly ruptured, and measured open-circuit, infinite resistance with a multimeter.

Checking which pin it was going to from the right-hand end, it was pin 32.  Wiring diagrams call this connection 'sender earth' and it goes to a few senders including fuel gauge, outside air temperature, coolant lack sensor, screenwash fluid level sensor etc.  

 

It seemed clear that none of these things, where fitted, could operate as designed with this ruptured track.  

 

I conferred with Keith as to whether there'd been any known external fault, like a wiring short or similar, preceding the failure, but he was not aware of any such thing.  The replacement cluster did not suffer a similar fate, so it seemed most likely to me at the time that there must be some other fault, on the cluster circuit-board that had caused this track to fail.

So I set about reverse engineering the circuit. Not too much to it thankfully, and this is what I came up with:

MCU on the left is some big chip, presumably a microcontroller.  An output (I assume) is used to turn on a pnp transistor (not necessarily all the time) which supplies a voltage to both fuel gauge and outside temperature sensors (V share) limited by a 5.6 volt zener diode, via a couple of resistors such that the gauge senders are the bottom half of two potential dividers. The voltage V share and these two mid-points of the two potential dividers are monitored by three input pins of the MCU, I suspect analogue inputs.  I didn't map out the onboard circuitry for coolant lack or washer fluid lack sensors.

 

The necked track going to/from pin 32 just goes to the board's 0V.

 

I could not see any obvious single-component failure in this arrangement (on the board) that could cause a large current to flow back into pin 32, so I'm not really any the wiser as to why the track failed at this 'necked' point, nor why it is narrowed down like that.  A hunch I have is that it is to provide a guarantee that only a pretty small current can flow through the fuel gauge sender, which is after all in a hazardous environment?  Didn't spot a similar feature on the Mk1 cluster I have in bits though.

If anyone can explain it, I'd love to hear from you.  When I google 'track fuse' there's a fair bit of discussion about them, but mostly fairly disparaging, along the lines of 'why wouldn't you just use an actual fuse?'.

I suppose it could have corroded a little, because of the lack of solder-resist coating, making it more susceptible to runaway ohmic heating than when new? Or had a slightly dodgy bit of etching so it was even narrower than intended at some point?  It appears to have folded back on itself to the left when I zoom in? 

 

 

 

It looks like it is intended to get hot because the conformal coating or whatever it is called has been masked in a square around it.

 

I have repaired several electrical appliances (well bodged by removing a questionable safety device) by shunting a failed "thermal fuse" the items have been a coffee maker, a steam mop and an electric blanket, in all cases the fuse was designed to fail if the appliance was left plugged in and heating for an extended period, in the case of the electric blanket the fuse was squeezed between two thermal resistors whose job it was to heat it to failure if left plugged in overnight.

 

From your description it was my first thought but could it perhaps be some sort of temperature compensation for if the instrument cluster gets heated from direct sunlight?

I can't recommend bypassing designed thermal/ overcurrent cutouts, but I must admit I did solder the thinnest bit of wire I had handy across the break and returned it to Keith to try, out of curiosity to see if the faults were resolved.

 

The fuel gauge and ambient temperature readout were now operational, but there was a problem with the immobiliser which prevented full success. I suspect that maybe coding in a replacement cluster may have caused this original one to be seen as 'foreign', but my knowledge of such things is inadequate to be sure.

 

 

 

 

 As a close out to this post, I would like to thank Pete, who I consider one of the main "go too" people for electrical issues on here, for the time he has put into this issue. When the cluster was received from him I re-fitted this and, as he has stated, the fuel level and outside temperature gauges now worked as expected, but the car would only start and run for approx 2/3 seconds which is generally a sign of an immobiliser issue. This was accompanied by a clicking/buzzing sound coming from the console, almost like a relay going in/out rapidly. This had not occurred when the car was functioning with the failed fuel/outside temp functions, so I can only assume that this "noise" source may have been the cause of the "fused" track that Pete repaired - as he has stated this is not ideal/recommended but was something I agreed to him doing - with the replacement bypass wire having a greater resistance to failing. As I am wary of this causing additional items/failures I will not want to try re-coding the immobiliser and keys so will carry on with my 133K mile cluster that I matched to the ECU/keys.

 Once again, thanks to Pete for his investigation, and I hope this has added/assisted in his knowledge base of these cars electrical circuits.