2010 A3 1.6 TDI CR intermittent starting issue.....

My niece has a 2010 A3 convertible which she has owned for about a month, having purchased it privately.
It has done 94 k miles with FSH and is in excellent condition.
It is fitted with the 1.6 TDI common rail engine and it has the stop-start system fitted.
About a week ago she began having issues after starting it each morning.
The symptoms are that it will start on the first turn of the key but after a few seconds it will start to run roughly, with the engine juddering and will be totally unresponsive to the accelerator ie revs will not rise, at which point the engine will stall.
The car will do this about 5 times, after which it will run perfectly for the rest of the day.
A dash light did come upon the dash, think it was the glowplug symbol?
I scanned the car with VCDS after the first time this happened and it came up with a fault saying 'implausible signal' from the MAF sensor so I cleared the code and the dash light hasnt come back.
I scanned the car yesterday , to see if there was any code but no error code was found.
The car is playing up every morning.
Today I serviced the car (air, engine oil and fuel filters) and my niece was present when I did the final oil level check and wanted to start the engine to ensure that the filter housing wasn't leaking.
On starting the car up (the engine was still quite warm) it began to misbehave in the manner described above, and she yelled to me that it was doing exactly what it does each morning, except it had never done it before when the engine was warm/hot?
I had warned my niece that as the fuel filter had been replaced it could well splutter or cough for a bit until any air in the filter had been bled out (even though when I did the same job recently on my 2.0 TDI it ran perfectly)so I'm wondering if the problem could be related to the fuel system?
I should add that a replacement MAF is on order as I was a bit worried that it may fail and because it is so easy to change thought it would be worth doing.
So, any ideas what the problem could be?
The fuel filter that I removed looked quite clean with no obvious sign of gunk inside (I've seen some threads where the fuel filter housing can become gunged up due to the tandem pump failing, although I'm not sure if the 1.6 has such a pump?) .
I'm thinking that whatever it is, it isnt an electrical or electronic fault as I cant get VCDS to find anything wrong?
Thanks!

Thanks!
If you could provide any values to check with VCDS that'd be great!

vag-tech said:

Sure...Fire up VCDS -> Select control unit -> Engine -> measuring blocks -> scroll to find coolant temp and rail pressure.
Coolant temp is in the first few blocks, rail pressure is a bit further in (normally displayed in bar).
Report on fuel pressure at idle (should be stable[ish]) and coolant temp plausibility (does value look correct from cold?) during warmup.

Click to expand...

Thanks again!
My niece has gone up north to visit relatives this weekend and turned off the engine on the M5 due to congestion, when she went to pull away she had difficulty getting it going again.
(Whilst she has had these issues I'd advised her to turn off the stop/start system in case one of these components was causing the problem).
It's no longer doing it only from cold and definitely getting worse , she also had the glow plug warning light come on, then go out again.
As soon as she gets back tomorrow I'll be getting the VCDS out.....
Cheers.

Right, a bit of an update as this saga is continuing....but getting worse!
My niece went away for a few days last week but got about 100 miles away from home when suddenly the car lost all power and went into limp mode.
She called the AA but was somewhat surprised when an official VW/Audi breakdown service vehicle arrived (I've since learned the AA operate the service on behalf of the dealer network) and promptly diagnosed a faulty "auxiliary egr cooler pump".
He stated that the pump should always be active (ie making a noise) but that it didn't do anything until he began cleaning it's contacts, so he recommended changing the pump.
My niece explained she was away from home and visiting friends and wouldn't be home for about a week, he said it should be (fingers crossed) OK until then.....
She went on her way and (can you guess this is coming) about 30 minutes later the car went into limp mode again.
The AA were called again, and this time an AA-liveried van turned up.
This time the guy declared it as the EGR valve & cooler that was at fault, although he didn't offer any real reasons why....he dismissed what the 1st guy had said and said that it couldn't possibly be the auxiliary pump!
My niece ended up being recovered home and as soon as she got back I scanned the car with my VCDS and could find no fault.....
Today we visited our local main stealer just to get a price for the egr valve and cooler and I mentioned the problems with the car.
I should add that today it started and ran perfectly, no limp mode and no dash lights.
Very helpfully they offered to scan the car for faults , which they did!!!!
It came back as no faults (no surprise!) but they reckon it's fuel-related, they reckon that in 9/10 cases a faulty egr valve WILL throw a fault code.
The VERY helpful technician also told me that the car has a latest generation ECU which is why with my VCDS cable I can't get into measuring blocks to check the fuel settings....
This brings me to my dilemma as to what to recommend to my niece should do next?
The auxiliary egr cooler pump is about £100 to replace, will a faulty one really cause problems or was this AA diagnosis as unreliable as their EGR one?
I'm wondering if we should just find a diesel specialist and let them have a look?....

vag-tech said:

The EGR valve will be continually be being checked by the ECM (by air mass) for correct operation. This will log a fault for sure if the car trips into limp due to implausible valve position/flow error, so strange that scans aren't reporting.
Really what this needed now is a good garage to properly check your EGR operation (via actuation and monitoring), fuel pressure (realtime monitoring under load [high pressure and low pressure supply]) and even check out the ancillary coolant pump (not likely to be causing fault but easy to rule out).
PS - you can definitely measure rail pressure with VCDS on a 2010 8P7.

Click to expand...

Thanks for the info, I've just found via Google that there's a known issue with failures of the CR high pressure fuel pump on cars of this era...I do hope it isn't that!
Either way I think a good diesel specialist will have to be found...
It's really odd but my VCDS cable will not read engine measuring blocks on my niece's car.
I tried twice connecting to the engine controller and everything is visible , except the measuring block which appears as a greyed-out square ....thinking the cable faulty (it's only 2 months old) I plugged it into my own BKD 2.0 2006 parked alongside and hey presto I have measuring blocks....
Very odd.
Oh, and it's a real RossTech cable bought from Gendan in May 2016.

Thanks, will do!

vag-tech said:

PPS - ancillary coolant pump signalled via ECM -> it decides when it runs

Click to expand...

Thanks , that makes some sense....

sliced said:

VCDS, for CR engines you don't use measuring blocks, you used advanced measuring values IIRC.

Click to expand...

Thanks, didn't spot your reply until now!
That gives me some hope to go and try my lead again.....
Cheers.

The car was taken to a diesel specialist today and they have said there's nothing wrong with the egr valve or cooler, HPFP or auxiliary coolant pump.
They reckon the problem is injector related and wanted to remove all 4 of them and bench test them as they suspect the solenoids on the injectors.
I advised my niece to pass on this for the time being until I was able to investigate further as surely a solenoid issue in the injector would appear as a fault with VCDS?
Reading around I've also come across lots of info that says these CR injectors aren't rebuildable too?
At a loss as to what to do next....
The car is currently behaving perfectly btw.

Thanks.
I've just tried checking the injection deviation in advanced measuring values using VCDS and thought I'd logged the data as a graph (turns out I hadn't) but the deviation values of all the cylinders seems to be jumping around.
I know the ideal deviation of all 4 cylinders together should ideally be zero but I'm finding that from one set of values to the next the same cylinder value is deviating by as much as 0.4 for one of the cylinders whilst the others are 0.2-0.3
I'm going to try and measure these values again and plot them as a graph this time, is there anything I can do to pin down which injector might be faulty?
Would an injector leak off test be worth doing, it looks simple enough?
The garage were adamant it's an injector fault.

Thought I'd post a bit of an update to this in case it helps anybody in the future....
Since early August the car behaved perfectly, covering approx 5k miles during that time.
As it was running so well my niece decided to leave well alone and just use the car.
Last week however it suddenly went into limp mode again with the engine management light lit up on the dash.
She decided to stop using the car until such time as she could bring it to me to be scanned.
I scanned it with VCDS on Sunday and this time 2 fault codes were evident , 9870 and 9867, both relating to the auxiliary coolant pump (implausible signal and open circuit).
It was decided to replace this pump and luckily Euro Car Parts locally had one in stock.
I fitted this yesterday and once I'd cleared the fault code it was taken for a drive and the fault appears to have gone.
Fingers are crossed that it really has sorted the problem!
I was interested to see that the ECP-supplied pump was made by Pierburg and was identical in every way to the Genuine item removed from the car.
The only difference was the ECP one had a small area ground-away where the VW logo and Audi rings were on the Genuine item.

Yes Chris.
The car has behaved impeccably since doing the pump.
For a car with a Common Rail engine (PD engines don't have this pump) then I'd always recommend looking at the pump before doing anything else!
Too many "specialists" all wanting to replace parts without actually guaranteeing it'll fix the problem....like I've posted above if we'd believed the garages it would have had a new EGR valve and bench-tested injectors!
It just needed a new auxiliary coolant pump, that's all!