A8 Parts Forum  

Go Back   A8 Parts Forum > A8 D3 > D3 - Engine Bay

D3 - Engine Bay Everything under the bonnet

Reply
 
Thread Tools Display Modes
  #11  
Old 5th August 2017, 02:07 PM
ainarssems ainarssems is offline
Senior Member
 
Join Date: Aug 2010
Location: Rushden, Northants
Posts: 3,799
Default

So you have not actually changed the battery yourself? In this case I would definitely try a new battery first. The new VCDS are either limited cars or very expensive professional versions. I would recommend buying second hand older unlimited version. If you would like to try Chinese copy there are plenty around but obviously you do not get support with them and while it is very unlikely that it will cause any problems with car sometimes they do not work well.

Plenty on aliexpress and they are cheaper than last time I bought one. These are newer 17.* versions.

https://www.aliexpress.com/w/wholesa....3&needQuery=n

I have not used them myself as I am on older 16.8 version

https://www.aliexpress.com/w/wholesa....8&needQuery=n
__________________
Currently 8less
2011 Q7 S Line 3.0TDI, 2016 Tesla Model S 90D

8 history:
2006 A8 Sport 4.2TDI quattro SOLD,
1997 S8, reached end of life with gearbox failure
Reply With Quote
  #12  
Old 6th August 2017, 03:57 PM
erubus erubus is offline
Senior Member
 
Join Date: Mar 2017
Location: highland Perthshire
Posts: 152
Default

Well i took the plunge and ordered a copy, not from ross-tech. Got the 17. whatever it was version with a hex can plug so will see what like it is. Hopefully it will let me see what the story is and play around with it until I can afford to get genuine unlimited version. I will play around with the knackered A6 first just to get a feel for it although without being able to run the engine I'm not sure what I will be able to do. It might let e see if the dodgy dealer i bought it from did something to stop the oil pressure warning from coming on - That was what killed it. Think i managed something like 90 miles in the caer before it blew up.

I haven't tried changing the battery myself, but the PO changed the battery a couple of months before I bought it to try and solve the same problem, plus Ive tried the battery out of the D2 and another one that lives in the boot as an emergency booster and the ones out of both Land rovers. It happens with all the batteries and theyre definitely not all knackered batteries. I'm reluctant to shell out another £100 or so for another new battery, although if I did i would probably get one of the glass mat ones. I can get a 950A one for about £100. I think it was a bosch one.
Reply With Quote
  #13  
Old 7th August 2017, 07:29 PM
erubus erubus is offline
Senior Member
 
Join Date: Mar 2017
Location: highland Perthshire
Posts: 152
Default

Another thing that points to it not being the battery. It doesnt even start with a jump now when hot. Been trying for the last fifteen minutes. Just spinning over, 154 rpm, no start. Interestingly the code reader said it was 77 degrees and the gauge said more like 87 degrees.

I'd be tempted to change the coolant temperature sensor as I know it can cause all sorts of problems ewith these engines. I've actually got one because i got it for the A6 and never fitted it. Its just that its in such a stupid stupid stupid place. Id have to depressurise and drain the AC system, take the front of the car off and remove all the drive belts etc and apparently the AC compressor doesnt have a clutch so it must be refilled before starting the engine. Seriously, who thought that it was a good idea to put a sensor in sucha ridiculous place? particularly one which is known to fail a lot.
Reply With Quote
  #14  
Old 8th August 2017, 08:37 PM
erubus erubus is offline
Senior Member
 
Join Date: Mar 2017
Location: highland Perthshire
Posts: 152
Default

Right then... I have now done a vcds scan. Got loads and loads of codes, although thats to be expected. Got loads of "no communication" as well. evidently a crappy copy!

a few codes that seemed relevant were:

17978 - Engine Start Blocked by Immobilizer
P1570 - 001 - - Intermittent
Freeze Frame:
RPM: 0 /min
Torque: 3.9 Nm
Speed: 0.0 km/h
Duty Cycle: 0.4 %
Voltage: 11.32 V
Bin. Bits: 00001111
Bin. Bits: 11100000
Torque: 148.2 Nm

18282 - Valve for Torque Support Mount
P1874 - 000 - Short to Plus
Freeze Frame:
RPM: 672 /min
Torque: 136.5 Nm
Speed: 1.1 km/h
Bin. Bits: 1
Bin. Bits: 0
Bin. Bits: 0
Temperature: 75.6°C

16785 - EGR System
P0401 - 000 - Insufficient Flow
Freeze Frame:
RPM: 1197 /min
Torque: 202.8 Nm
Speed: 44.8 km/h
Duty Cycle: 75.4 %
Mass Air / Rev.: 522.9 mg/str
Mass Air / Rev.: 364.3 mg/str
Mass Air / Rev.: 346.6 mg/str

17601 - Fuel Pressure Sensor (G247)
P1193 - 000 - Open or Short to Plus - Intermittent
Freeze Frame:
RPM: 21 /min
Torque: 19.5 Nm
Speed: 0.0 km/h

obviously the engine mount one and EGR one are irrelevant - maybe the EGR one might possibly cause a starting problem but it hasnt on any other car I have had that has had it.

Fuel pressure sensor one is interesting though, as is the immobiliser one.

I got some live data with some of the measuring blocks. during a failed start RPM = 126, fuel injected 0, fuel pressure 62 Bar.

Engine temperature was stated as 75 -77 degrees but with 90 indicated on the gauge if that is relevant.

Does any of this suggest anything to anyone?

Gill
Reply With Quote
  #15  
Old 8th August 2017, 08:59 PM
erubus erubus is offline
Senior Member
 
Join Date: Mar 2017
Location: highland Perthshire
Posts: 152
Default

Just been thinking. The other day I tried starting the car with the larger plug on the fuel pressure regulator pulled out to see if it would start with no pressure data - it didnt but thats maybe where the open circuit intermittent fault is coming from. Also noticed the RPM on the immobiliser one is 0rpm, so the engine isnt even spinning. The first time i try starting it afer the battery has been pulled it never does anything and needs the key turned again. Possibly this is a code relating to that. so all the engine codes stored seem to be spurious or irrelevant.... maybe.
Reply With Quote
  #16  
Old 9th August 2017, 10:00 AM
erubus erubus is offline
Senior Member
 
Join Date: Mar 2017
Location: highland Perthshire
Posts: 152
Default

Had the car plugged in again this morning when i started it up. Needed a jump because I had drained the battery yesterday playing about with it trying to start it etc. With the jump leads on it was turning over about 180 ish rpm, and about 90 bar fuel pressure before it started. I had a graph up but thngs moved a little too fast to see what was happening. i couldnt figure out how to save and view the graph. When running at idle it was a touch under 400bar pressure and about 700rpm. Interestingly, 126 rpm and 62 bar gives you roughly 0.5bar per revolution and 400 bar at 700 gives you about 0.5 bar per revoluton as well.

This suggests to me that the pressure is directly related to the engine speed, therefore it is the engine speed which is holding off the injection as the pressure isnt high enough. As jumping it isnt making any difference when its hot anymore, i'm still inclined to think starter motor. I dont know if im reading the data properly but it seems a logical deduction to me. does anyone have any thoughts?

gill
Reply With Quote
  #17  
Old 9th August 2017, 09:51 PM
Mechcanico Lee Mechcanico Lee is offline
Senior Member
 
Join Date: Jul 2015
Location: St Helens Haydock
Posts: 434
Default

Hello , good therory but incorrect Erubus , if I'm correct I think this is cp3 with inlet metering valve on the pump ?? The pump is capable of making max pressure ( around 1400 bar ) at idle speeds or less , if you ' dead head ' the pump it will make max pressure just cranking .... dead heading the pump is removing inlet metering valve and fitting a blank in its place that leaves the pump fully open to make max pressure , its a test you can do as a proving test for the pump , main control of high pressure is done by inlet metering valve .

On cp3 ....again from memory injection commence starts at 250 bar .....so injectors do no start to trigger till rail pressure is at 250 bar or thereabouts , but non of this happens until ecm sees crank and cam synchronisity .

Where you say you saw 90 bar of pressure and it started may not be accurate , whilst under cranking ecm obd data can be notoriously in accurate , I would be scoping rail pressure sensor voltage for more accurate reading , you could do it with a voltage meter , just scope traces show a bigger picture .

In the data blocks can you see a block that says .....synchronised...yes / no , this is a very important factor ....with out this the whole start sequence will not happen .

At idle ,stationary ,when warm tell me the rail pressure and tell me the duty cylcle of the inlet metering valve , whilst driving on the road with vcds ( better take a passenger or log it ) put foot to the floor and at max acceleration look at fuel pressure and inlet metering valve duty cycle .

When the car does it's non start thing ...if you give it a touch of easy start does it burst into life and instantly runs as normal ?
Does th car have the electric fuel pump in the tank ....can you hear it prime on ignition on

A few things to have a go at for now
I'm a diagnostic tech ...this is what I do everyday , stay focused on individual tests ....... test ....rule out .....move on to next test
Reply With Quote
  #18  
Old 12th August 2017, 11:50 AM
erubus erubus is offline
Senior Member
 
Join Date: Mar 2017
Location: highland Perthshire
Posts: 152
Default

Wow, thanks for that info. Its getting a little over my head now though. So far I have found the VCDS program very confusing and difficult to use.

Where am I likely to find the data blocks for synchronised yes/no? I presume if it says no, then I am looking at stretched timing chains?

Fuel pressure at idle (cold) is approximately 400 bar. I will check it again when its warm. How do I find the duty cycle of the inlet metering valve?

I have resisted giving it easy start - its terrible stuff but I will give it a go and see what happens.

There will be, I'm sure, a fuel pump in the tank. I will open up the hatch and have a look to confirm. As to whether I can hear it prime or not, there are various noises when you switch on the ignition one of which does sound like something pressuring up over the course of a second or so. If I look inside the tank for the pump it will likely be easier to see if its the pump thats priming or not. Is there a measuring block to check for whether the tank pump is priming properly/at the right pressure etc?

Sorry for the questions, it good to have some good solid things to check for now - although they are all sounding fairly terminal in as much as they will cost more than the car is worth to fix!

edit: with the engine cold and idling fuel pressure is around 400 bar, revving it (stationary) it rises easily to well over 1000 bar. I havent as yet checked it hot.

PS: the whining vibratey noise that I could hear and feel through the steering wheel when the engine is hot is, I'm fairly sure, the PAS pump and is unrelated to the hot start issue as far as I can tell at the moment

Last edited by erubus; 12th August 2017 at 11:57 AM.
Reply With Quote
  #19  
Old 12th August 2017, 06:27 PM
Mechcanico Lee Mechcanico Lee is offline
Senior Member
 
Join Date: Jul 2015
Location: St Helens Haydock
Posts: 434
Default

I've noticed in the 8s that lots of the data blocks are not named , if you go into the advanced measurements section does that give you a clue as to what the blocks are ?
Generally the synchronised block is in the first few blocks that come up , scroll through can you see anything that says synchronisation seen or just a plain yes or no , I use ODIS at work and that is a little obscure on the 8s .

So it does sound like in tank pump is priming from what you say and no I'm pretty sure there is no data block for priming pump pressure .....you have to use an inline guage , but at least we can hear it priming which is good .

Where you say stretched timing chains can stop the synchronisity your correct but you do normally get codes for this like .....crank / cam incoherence or crank / cam synchronisity not seen and so on .Also a poor amplitude signal from crank shaft sensor can do this if it's an a/c generation type ,or very poor cranking speed can do it also .

So at this point I would be scoping cam and crank signals but obviously you need a scope for this , but if you can see a block that shows synchronisity to be good we can sort of move on and say that's good and we are most likely not getting up to start up pressure.

Just as a tester ..... go into the rail pressure block and do a few starts ups and see what pressure it shows before it starts , thing here is that the obd data is not in real time and the start up interference at the ecm can send the vcds a bit scatty for a moment ....but let's just see what we get for the hell of it .

Where I was on about easy start ....it's more to show up a symptom, if you give it a wiff of easy start and it bangs up and runs smooth instantly it sort of shows that it is not getting rid of gobs of air , if it starts imeddatley it's more likely cranking speed or low signal coming from crank sensor the wiff of sniff just jumps it past that point .....this is not definite diagnostics but is a guide when you do not have all the diagnostic tools to hand .

Few more things to have a go at .
Reply With Quote
  #20  
Old 13th August 2017, 11:32 AM
erubus erubus is offline
Senior Member
 
Join Date: Mar 2017
Location: highland Perthshire
Posts: 152
Default

You're quite right, the data blocks are very poorly named on the A8. I have loads of blocks saying duty cycle for instance but no clue as to what they actually are. After playing fr a while on the drive I plugged in the A4 to check a parking sensor issue and all the blocks are properly named. So having guessed on the A8 that 022 (rpm, pressure,pressure, duty cycle) was the one I was looking for, I got it up on the A4 and the same measuring block was rpm, fuel high pressure specified, fuel high pressure actual and inlet valve duty cycle. So I ran the 8 until it was hot, logging the whole thing. did a few starts, revved it etc.

Interestingly, all the cold starts had the duty cycle at around 10%. Pressure was more difficult as the actual start is obviously quick but here are a few snapshots...


10.22 0 351.9 0 10.5
10.85 105 351.9 27.6 44.1
11.46 147 282.9 82.8 34.4
12.08 147 282.9 96.6 33.6
12.71 735 365.7 496.8 22.3

column 1 is time stamp, 2 is rpm, 3 press specified, 3 press actual and 4 duty cycle. Thats data for one of the cold starts. They are all pretty much similar.

The next bit of data though is where it gets interesting and to be honest i dont really know what it means. I have put the data into a graph as its a lot easier to read. The first half of the graph is the engine idling, its been running for a while and is nice and hot. Specified pressure and actual pressure are equal, they dont really diverge much at all even when you red line the engine (I cut those bits out as it made the scale unreadable on the graph).

The second bit of the graph is shutdown and restart from hot...

The pressure drops immediately to zero on shutdown. The duty cycle goes up to 40 odds, specified pressure is 300 bar or thereabouts while actual pressure is hovering around the 80 bar area while its actually cranking.

Should there be a check valve on the HP pump? I woud have suspected that the pressure should be retained in the fuel rail and should actually rise with the heat soak on shut down.

I still need to find the sychronised yes/no measuring block. I will plug it into the A4 again and see if I can find it on there and go back to the 8 and see if its in the same place.

gill
Attached Thumbnails
Click image for larger version

Name:	hot strart failure.jpg
Views:	218
Size:	115.3 KB
ID:	16722  
Reply With Quote
Reply

Thread Tools
Display Modes

Posting Rules
You may not post new threads
You may not post replies
You may not post attachments
You may not edit your posts

BB code is On
Smilies are On
[IMG] code is On
HTML code is Off

Forum Jump


All times are GMT. The time now is 07:29 AM.


Powered by vBulletin® Version 3.8.0
Copyright ©2000 - 2024, Jelsoft Enterprises Ltd.