1.6 TDI VDO Injector failure root cause analysis

[BenT]

Thanks Engineer Bob,

I wondered about voltage spikes too when I dissected the first injector, but having seen the crack in second one this looks more mechanical rather than electrical. I don't know enough about how the ECU generates the injector trigger waveform to know though!

I don't think VW would want to know about these failures now though - after all, these engines were released in 2009, so they'll probably want to sweep it under the carpet.

The only real fix I would be happy with is replacing either the complete engine, or just the engine mismanagement system. Since the 1.6 and 2.0 CR engines are architecturally very similar, it might be possible to swap the Bosch system from a 2.0 litre engine onto a 1.6, so injectors, high pressure pump, MAF, fuel rail, sensors and high pressure lines, wiring loom and ECU (the latter would need immobiliser removal and re-mapping too). I reckon this would cost about the same as replacing 3 or 4 injectors. Plus, there'd be less worry of the engine going into limp mode at an awkward moment! I don't know if it would work, plus I've heard from somewhere on this forum that there are differences with the cabin side wiring going to the ECU. Not a job for the faint-hearted. I'd be delighted to hear from anyone who's tried this though!

[Engineer Bob]

Hi BenT
When I was talking with the owner of the garage in Devon, where the last failed injector was replaced, about what would be the best way forward, replace engine with the 2.0, change the EMS etc, he said not to bother since they are also experiencing a larger than normal failure of injectors on the 2.0 engine!
I am still pondering what is the best way forward.

[BenT]

Very useful information, thanks!

There are numerous options of course, but if there was a truly elegant, low-cost solution the manufacturers would have probably thought of it (excepting that they want us to buy new vehicles, of course).

The newer, EA288 engines look a much better design on paper and use solenoid injectors, but they're expensive and need a completely different exhaust system front end. I don't know about electrical system compatibility either.

Or, it might be possible to revert to an old PD unit - the latter BLS engines had DPFs as an option so this would get around MOT problems. These engines were known to snap conrods though!

Maybe even a petrol engine conversion would work... But then that's major conversion territory.

I suppose anything's possible with enough time and money. But for most of us, carrying a spare and saying the odd prayer is probably the best we can do.

[Engineer Bob]

I have this morning had what I believe to be yet another injector fail. This occurred within 2 miles of home so I 'limped' home at a max speed of 15mph! Same symptoms as previous failures but yet to be diagnosed. I have just ordered an OBD II reader which should be with me tomorrow. Perhaps I should consider buying shares in VW and reap the benefits of their profits from selling replacement injectors!

[Killomiter]

Bob,

You are amongst the many!

I thought of making a post to steer people away from the so called business that I ordered injectors from as they were a right shower. One went south so I ordered a replacement - that one was duff but found out by installing it and by then I had sent my original back. Spoke to them on the phone and they sent another one out but still had the washer tied down bolt issue to replace. Once fitted it took more than two weeks to sort and still has not worked properly but had sent the second duff one back (interestingly the test and code doc was identical) - My grievance was not just the total waste of time but the promised money has never materialised back from returning the duff injectors said supplier,

Their domain name is commonraildesiels.com but trade as First Specialist Diesel ltd - Specialist Diesel Ltd., Sydney Street, Accrington, BB5 6EG, United Kingdom
My advice is do not use them!
Andy

[DW2207]

I used this chap from ebay. Bought 2 from him as I wanted a spare. Reconditioned. So far so good. Would recommend

Sent from my SM-G981B using Tapatalk

[BenT]

I always wonder if they replace the piezo actuator on these... In fact, it would be handy if one could purchase these separately. Coding would be a nightmare though!

Perhaps EngineerBob's come up with a good solution to all of this - humour!

[Engineer Bob]

On a dull, rainy, Saturday morning what better to do than disassemble a failed injector from my caddy.
Following on from the excellent failure analysis undertaken by BenT, I removed the piezo module from a failed injector that I had persuaded the garage (they were very reluctant) to give me in November 2021 following its replacement.
Sticking to the same process used by BenT and using his pictures as a reference, I took the injector apart and removed the piezo module. The injector was marked 03L130277S with date code RJ05 (5th Oct 2018?). The component connected in parallel with the electrodes of the piezo stack had colour bands of brown, orange, black, black, red. After disconnection, the resistance of the component was measured and found to be 130k ohms. The colour banding of the component appears to be unconventional since the colours would normally indicate 1, 3, 0, with 0 multiplier, 2% tolerance, which should be 130 ohms.
Having carefully removed the plastic holder the stack fell apart into two pieces near the top. The surfaces of the piezo material at the break had the same discolouration that BenT had observed, if the discolouration is viewed obliquely it has a metallic sheen. The rest of the stack then easily broke into a further six pieces, none of the other pieces showing any discolouration. There does appear to be an outer surface coating to the stack which made it very difficult to separate the stack further.
After all of this still no conclusions as to the exact method of failure can be deduced, my gut feeling is that the failure is related to heat stress and/or high voltage breakdown.
I am having my seventh injector replaced this week and have asked the garage to replace another one so that all of my injectors will have been replaced since November 2021, which, touch wood, should not fail for at least another 2 to 3 years. The saga continues!!!!

[BenT]

Nice work Engineer Bob!

I've been wanting to take one of the "S" ones apart for a while to see what the differences are (this would provide a clue as the to the failure cause, perhaps). Did you notice anything different between your injector and the one I've photographed? Also, was the failed wafer near the top? Sorry for all of the questions!

It's hard to believe that you've had so many fail, maybe best not to think about what else you could have spent the cash on...

When mine failed the week before last, I noticed that the piezo actuators were surprisingly hot. Perhaps we need to fashion little cooling jackets for them! Although, if a multi-million (billion?) pound company like Siemens haven't been able to fix it there's not much chance for any of us!

[Engineer Bob]

In answer to your questions, BenT, the only physical difference I noted was that the white insulator ferrules covering the electrode posts (your picture 4) were not present. There was, however, a similar moulding which is now integral to the black connector cap. Other than that everything looked identical.

When I removed the silicon sealant from the stack it easily split into two pieces where the 'damage' had occurred. The piezo stack was virtually impossible to break apart further without causing breakage to the material, I only managed to split it into a further five pieces, but there was no further evidence of damage. The damage was located approx 5.5mm from the top of the stack. I could not measure any meaningful resistance with the stack removed from the casing due to breakages in the minute wire conductors.

I did take pictures during the disassembly process but found that I could not include them with my post. How did you attach yours to your failure analysis report?

Regarding your comment re cooling sleeves, I think that the inner perforated sleeve of the piezo module is possibly acting as a small heatsink. However with the silicon sealant acting as an insulator I don't know how efficient the heatsink is. I still think that the failure is more electrically related, the mark on the faces of the piezo cells is reminiscent of an electrical flashover. Again how do you check the electrical feed to the injectors?

Geronimo (my Caddy) has been collected this morning for fitment of two replacement injectors. The guy who did the recovery (LV=BR) said that in his experience this was a common reason for Caddy's being recovered! The guy who runs the garage who are doing the replacement for me said to me this morning that when he ordered the replacement parts from VW yesterday morning (delivered to them yesterday evening) there was plenty of stock available, more than he would have normally expected.

[BenT]

Thanks, that's interesting! Adds weight to it being an electrical problem I suppose. Perhaps a lower resistance would make it easier for the piezo stack to dump its charge between firing events, reducing the likelihood of arcing. I actually tried heating one of the wafers up to orange heat with a blow lamp and holding it there for a few minutes - the whole thing survived seemingly unchanged, excepting that the silicone burned off very quickly. Whatever causes these to fail must generate intense heat for a very brief moment, for the electrode and piezoelectric material to fuse while leaving the silicone undamaged.

I actually made my original post on the TDIclub.com website and uploaded all of the photos there. For this forum, I just copied links to the TDIclub forum photos. I don't know how to do it otherwise!

Hope you get your van back safe and sound! I may paint mine green, stick some googly eyes on the front and call it "touchwood"...

[Engineer Bob]

BenT,
'Torchwood' may be more appropriate!

[Ricky302]

Number 4 injector on my 2011 caddy went, bought a reconditioned one of ebay, the old injector was stuck and I didn't have the tools to remove it, replaced just the electrical part of the injector, working perfectly 15K miles and ongoing., I see the electrical parts can be bought separately on aliexpress.

[BenT]

I'd wondered if that would work, thanks for posting! Did you have any calibration issues, i.e. rough running? I saw some actuators on aliexpress the other day for about £47, but wasn't sure if they were genuine. Perhaps worth a shot! No point in replacing the whole unit for the sake of one (relatively inexpensive) faulty part.

[Ricky302]

The running wasn't rougher but the diesel 'knock' was definitely louder, after about 500 miles it quietened down to normal.