Starting Problem, Tachometer Shows Erratic Revs, Eventually Starts.

[antzatgalaxy]

Hi all,

 

Having difficulty starting my Galaxy (2006 MkII 1.9 TDi Zetec [115]). I've read a lot of posts on this on this site, but none I've found mention the behaviour of the Tachometer.

 

When I turn the key, the glow plug indicator lights for a second (as normal) and the engine turns over as normal, except it doesn't start immediately. If I continue to crank it over it may eventually start (30-40 seconds) though it makes an effort over it, occasionally catching momentarily. However, I've found it best to just turn it off and try again immediately and it may start in the same laboured way, only sooner (20-30 seconds).

 

What is odd is that when initially starting the engine, there are no revs indicated on the Tachometer (though clearly the engine is turning), then the needle kicks into life after a few long seconds (as high as 5000 sometimes) and bounces around the bottom 1000, occasionally kicking higher, while trying to start. The Car doesn't start in the period the needle is zeroed, but this is only a few seconds.

 

If I crank over for too long, I might get the warning message "Alternator Workshop" or "Engine Workshop" or even "Stop" and once started I occasionally get the glow plug light flashing. I believe the glow plug flashing (accompanied by a single beep) is just the ECU telling me that it has recorded a fault and nothing to do with the glow plugs

 

Once started, the car runs fine.

 

I've been charging the battery over night every second night as I do a lot of short trips. The Battery is less than a year old and has no problem cranking the engine over for a long time and has not run flat yet in all the attempted starts.

 

The spare key has same effect, the door LED lights for a second then goes out (as it would do normally). So I don't think this is an immobiliser issue either..

 

Nothing useful has come out of engine diagnostic, but I got a garage to read these (£35). The result was that I got the Crank sensor (CPS) replaced (£37 + £30 Labour), but it has made absolutely no difference.

 

Looking for some pointers on where to go with this one and I'm no expert but willing to take advice and have a go, or at least point a mechanic in the right direction.

 

Please correct me if I'm wrong, but I gather that this engine will start eventually even without working glow plugs and I wonder if this is relevant? I also wonder if Relay 109 is relevant as I've read about this here and gone as far as finding it in the third layer of the fuse box (thanks to this site). Does this power the ECU? Is there a relay/fuse for the glow plugs? Also wonder if this is anything to do with supply of diesel to engine (Pump? Solenoid?).

 

I come back to the behaviour of the tachometer needle - This must be indicative of something?

 

I've ordered a Cam Sensor (£78) as my next move. Any thoughts appreciated.

[antzatgalaxy]

Just driven 180 miles non stop (not just out of fear of having to start the car again!) and no problems once going, so this is clearly a starting issue.

 

Where does the instrument cluster (Tachometer) get the engine revolutions reading from and is it the same place as the ECU. I've read elsewhere on this site that no fuel is delivered on starting unless the engine cranks at +250 revs. Is that true?

 

Is my engine fuel starved because it's not detecting engine rotation? Why does it all behave once started?

 

Thoughts appreciated. Cam Sensor arrived, so fitting on Tuesday....

[GrannyPower]

Hi. I recently had starting problems with my 2002 auy115 galaxy.

 

Eventually turned out to be that the timing belt had stripped a few teeth which had altered the timing so it wouldn't start.

One of the fault codes present at the time was : Cam sensor out of range.

If you unplug the sensor the ecu should run a default value and the car may start

At no point however did my rev counter do anything strange and my car would not run at all. It fired up very roughly with the sensor disconnected.

 

Chris.

[antzatgalaxy]

Hi Chris,

 

My Car has 73K on the clock, the cam belt was done at 60K and the 70K service was done a few months back. It's difficult to think back if this starting problem came on gradually or suddenly because any deterioration was probably not noticeable until it failed to start straight away (as once started it runs fine). There's been no appreciable drop in ambient temperature, maybe slightly more damp than usual, so can't really put weather conditions into the mix.

 

One of the errors brought up was Cam sensor out of range, but I'm not convinced changing the Cam sensor will solve this and it runs fine once started so I don't think there are any timing issues. No errors are ever recorded once the car is going. I'm going to change the Cam sensor on Tuesday and go from there, unless anyone can point me in the right direction. I'll also get the list of fault codes after I change it.

 

Anthony

[alan_131]

The ecu follows a default map if it doesn't believe/can't see the cam sensor. It will run, based on where it thinks the cam should be, from the info it gets from the crank sensor. I seem to remember reading on here that you're right about the rotation speed, though I thought I'd seen 300rpm. Also, the injection system modifies its behaviour based on input from temp. sensors. If you search, you'll find a post on this from one of the forum gurus. If you have a laptop, free vcds-lite download and a "less than a tenner on ebay" lead will let you see/clear codes and read temperatures as well as lots of other info.

[antzatgalaxy]

Hi Alan,

thanks for the input. I'd like to know where the ECU gets the RPM from and if the tachometer get it's info from elsewhere or is an output of the ECU. I've got to test if fuel is actually being injected in the early stages yet.

 

If what you say about the ECU being capable of ignoring the Cam sensor and use a default map, then it doesn't sound like replacing the Cam sensor is going to make a difference to starting, but I'll give it a go.

 

I gather temperature sensors control whether glow plugs are on or not but I get the same starting issue with a warm or cold engine, so I'm looking elsewhere first.

 

I've yet to own my own Vag-Com, but it's interesting me....

[alan_131]

There's some information about the pd engines here http://www.myarchive.us/richc/VW_TDI_with_PumpeDuse.pdf that if you haven't come across before might be helpful. I don't think you'd regret getting the cable - I certainly don't. It gives you access to a LOT of useful information that combined with the help you can get from the forum makes life much simpler...

If you already have a tool to read fault codes then there's a certain amount of duplication, but I think you'd still find it gives you more information.

 

A while back there were links between tacho and alternator wiring in some VAG cars, but I have no idea (sorry!) if this is still so. Many others on here who know a lot more about this than I do...

[antzatgalaxy]

Thanks Alan, I'm convinced about VCDS-lite as I'm quite keen to increase my knowledge. Thanks for the bedtime reading also, looks very useful. I'll let everyone watching this know the result of the Cam sensor change tomorrow.

 

Anthony

[antzatgalaxy]

No improvement!! Well, no real improvement. I've now changed the Crank sensor and the Cam sensor. I could convince myself that there might be a slight improvement, but really that's insignificant to the actual problem, i.e it may start a little earlier after cranking for ages....

 

Having cleared all engine codes, these are the latest to be thrown up, I'm not saying that any more might be thrown up after more starts, but I suspect some of the later ones are a consequence.

 

16989 - ECU Defective intermittent

16705 - Engine RPM sensor - G28 intermittent

 

The description is the extent of the information I have.

 

What is sensor G28? I've seen in mentioned, but what is it? Is my ECU really defective, or is this an issue with a relay somewhere (109?)?

 

The rpm needle still doesn't move when I initially start the car before springing into life after a few seconds and then bouncing around. Is this stopping the car from starting?

 

I read that if the dual mass fly wheel is on the way out it can mess with the cam sensor, can someone confirm this. I mention this cause the mechanic I'm dealing with mentioned that it didn't sound fine. He mentioned a rattle as the engine is switched off as a pointer - any comments anyone?

 

Any help appreciated

[chrispb123456]

Engine speed sensor G28 is the crankshaft sensor

 

Camshaft sensor is G40

 

The LED on the door you said went out as normal what about when you are cranking the engine

 

Can you hear the momentary fuel pump squelch noise when first turning the key to the second position,

 

Am cautious to say change relay 109, if you have a flashing LED or no pump noise I would try a relay

[antzatgalaxy]

Hi Chris,

This is where I started, the fault codes pointed to Crank shaft, this was replaced, that didn't help. Cam shaft sensor then replaced, that hasn't helped either.

 

Right at the beginning instinct told me that despite diagnostics saying these sensors were at fault, I wasn't convinced.The engine ran fine once going. I'm aware that if the Crank shaft sensor fails while in motion, then the whole engine stops! This has never happened - so how come only faulting on starting.

 

Further investigation of Ross tech web site (I want to point out at this point, if I sound confident, I'm actually out of my depth here) that if code 16705 (G28 engine RPM sensor intermittent) came up after extended cranking and replacing the sensor made no difference, to go back and check the basics!

 

I'm now being lead down a different path and need advice. If - as I'm being told - the Dual Mass Fly wheel is on it's way out (I take that to mean disintegrating), then is it chucking crap everywhere? If so, would this effect the starter motor? Could this be a Starter Motor issue? Is it just that the started motor is not cranking fast enough?

 

Could I be looking at having to replace the dual mass flywheel AND the starter motor to get this car starting again? I've seen indications of this on the TDI forum - but I'm no expert.

 

Any pointers still appreciated

 

Anthony

[antzatgalaxy]

Meant to say, there are definitely noises then I turn the key, I keep hearing them described as a squelch, but mine does sound more like a small electrical pump for a few seconds. All noises sound normal to me (as in before the problem).

 

I have no flashing door error LED codes coming up.

 

I have ordered a 109 - because I am clutching at straws and it's cheaper than a dual mass fly wheel and starter motor! - but there is no indication this has failed (no repeat clicks, no lack of glow plug dashboard light, no error codes in this direction).

 

Someone mentioned (somewhere else) trying a bump (tow) start to test if it is the Starter motor. This sounds mad, might try a hill start, but not fancying no power on the way down if it doesn't start. Plenty of hills round here!

[chrispb123456]

Do remember another member having problems with a crank sensor, his would only start with it disconnected

 

Can you confirm Door LED is not flashing when cranking and also can you hear the fuel pump not to be confused by the run on pump in the engine compartment.

 

Faulting DMF usually results in vibration and rattling, cant see a connection with G28 G40 or rev counter.

 

Have you had a look at the engine ECU wiring plug

[antzatgalaxy]

The Door LED does not flash any error codes, just the acknowledgement of the pats chip confirm (i.e. comes on and goes out after about a second).

 

I'm not sure I can hear a fuel pump, but then I've never listened for it. It's is probably the run on pump I'm hearing. But I've not noticed an absence of normal sounds, it just doesn't start properly. I've got the battery out on charge at the moment so will listen more closely when back in.

 

While battery was out I've been hunting for other problems and discovered a slightly heat damaged main fuse feed box next to the battery. Damage isn't too bad (not as terminal as those reported here), slight heat damage to insulation on end of alternator cable, little bit of over heated plastic and slightly coloured patina to fuse. Checked tightness of connection and decided to undo, clean up and tighten the only rusted connection on the alternator end of the 150A fuse. On tightening, the bolt end (supposedly held by box) twisted and snapped the fuse! So just replacing fuse now! Going to have to order new box and Alternator lead in near future. Not my day so far...

 

[seatkid]

Its been reported many times before that a burnt fuse box causes the dash to go haywire....

 

And starting problems......

Edited October 4, 2012 by seatkid

[chrispb123456]

May be coincidence but worth cleaning up and soldering both those large cables

[SilverBeast]

Definitley wants repairing/replacing. Feel the temperature of the lead after the it's been running for a while - careful you don't burn your hand! link Edited October 4, 2012 by SilverBeast

[antzatgalaxy]

I'm going to replace the fuse box now I've found it to be in this condition. However I must stress, that it really isn't that badly damaged. Only the Alternator cable connection to the fuse showed any signs of deterioration. I'd describe it as rusted. I've cleaned this up for now (and replaced the fuse I broke). The inner copper of the cable itself is not heat tempered, but I'll replace this anyway. All the other connections were fine (very low resistance). I've felt the cable several times and it doesn't get hot, but then I didn't feel it before so have no indication that this is an improvement.

 

I'm clueless to understand how the Alternator connection would have any effect on things, except in not charging the battery, which it was still is doing. However... After putting it all back together, it started a lot worse than the average bad start. Then, on thinking I'd record a video of the Tachometer needle, in front of the camera it started first time! Hasn't done that since the problem began. It was however a random occurrence. I live in hope of more random occurances....

 

I've convinced myself that the engine does turn over slightly faster the second attempt I try and start it and when it usually starts.

 

I've also managed to hear the rear fuel pump, though I had to hang out the door and it did sound more like a the windscreen washer pump than a "squelch". I have heard it before and it sound normal to me.

 

I couldn't get the ECU connections off, didn't want to force anything and couldn't figure it out. There seems to be a plastic bar holding them in that is bolted at one end behind, which I can't get to. Any hints?

 

So for now, I'm replacing the Main Fuse Feed Box, Alternator Wire and Relay 109. Not convinced that any of these will make a difference, but live in hope.

 

Any views on this now being a starter motor problem caused by a dual mass fly wheel disintegration?

[chrispb123456]

Hi there

 

Thinking about your car last night, when your cranking your car and gets to the point where it starts does it start instantly or does it try to catch intermittently eventually starting perhaps a slight misfire then clear and run properly as though there was air in the fuel system.

 

A couple of things to try before buying more parts;

 

1. remove one of the fuel hoses connected directly to the fuel filter, turn on ignition, fuel should be ejected momentarily from the filter or the pipe depending which pipe removed just to confirm intank pump is working.

 

2. Fuse 14 powers relay 109 so check for 12 volts at the fuse, relay 109 has 5 terminals but only 3 are used, they are, Terminal 30 supply from fuse 14, Terminal 85 earth switched from ECU, Terminal 87 switched supply to ECU and also supply to fuse 34 so if fuse 34 is live with ignition on and continues to be live while cranking then 109 is working.

Would also be good to check supply and ground wires at ECU, can see this is awkward as it's location behind battery and coolant bottle and the supply and ground wires are at the bottom of the plug.

 

Thinking also about the rev counter does it settle down and work properly once engine has been running a while.

[antzatgalaxy]

Hi Chris,

I've taken a few videos of the tachometer while attempting starting and will try and post one that is an "average" start once I work out how to reduce it to a reasonable size. Sometimes the needle kicks more than others etc.

As soon as the engine starts, the tachometer behaves as normal, steady as a rock,engine idle speed is fine, and engine response is normal (as is everything once engine is going).

 

On turning the engine over, the engine can eventually fire a few strokes and then go back to turning over or then start. Or it can start first opportunity it catches after a period of turning over. Generally I've found if it tries to catch (fires the odd stroke) and doesn't, I get a better result if i turn the ignition off and immediately turn back on. This second start usually has a much higher success rate and starts earlier. But whose to say I'm not inventing rituals and superstitions now! It always eventually starts (so far) and I've not managed to flatten the battery yet, but I have taken it out and charged it a few times now.

 

I will check the fuel supply at the filter and let you know, I need pointers like this Thanks.

 

I have a relay 109 on the way (fairly cheap from VW), mainly because I couldn't track the info you've just given me and didn't really fancy bridging it in the fuse box, so thank you for that, I'll check power at fuses.

 

I'd like to check the connections at ECU, but haven't figured out how to get the plugs off.

 

Also, alternator wire and interior fuse supply wire in main fuse feed box are fine with normal use. But last night I turned EVERYTHING on and drove for 10 minutes and both were then very HOT near the feed box. In idle with everything on, only the interior fuse supply wire got hot first near the feed box. This has got to be contact resistance at the fuse connection so I'll give contact surface a clean here too as I only cleaned up the alternator cable before. Pick up new fuse box next week, but not going to wait till then in case this has nothing to do with it.

 

VCDC-lite cable on way too, so hopefully be able to see actual output from engine temperature sensors etc.

 

Thanks for sticking with me.

[SilverBeast]

This has got to be contact resistance at the fuse connection

 

Not always the case. The crimps onto the cable are known to be very poor so the resistance happens there rather than the contact to the fuse. The conducted heat then damages the fuse box and can also get hot enough to oxidise the copper wire braids in the cables. It is best to solder the crimped contact to the cable. In the end I replaced the fuse box and alternator cable on mine (top one) with items from ebay and soldered the crimps on the first two crimps. See in link from post 17 above. If you replace the fuse box without doing anything about the crimps, I believe it will be only a matter of time before the problem recurrs. I suggest you do your 10min drive with everything on test after you have replaced the fuse box as this should be a good indicator and I wouldn't expect it to do any damage to the fuse box in that time. Even if you replace the alternator cable with a new one I would recommend soldering the crimps.

[chrispb123456]

Agree 100% with SilverBeast the high resistance is between the cable and crimp not the crimp and fusebox.

 

Am still confident this is a power supply problem, whether it's the 109 relay (easy fix) ignition switch, or wiring fault.

Taking the cover off the 109 and looking at the circuit board it's usually pretty evident where the problem is, there's some pics somewhere on here showing the bad solder connections (dry joints)

 

Soldering those large crimps is a must even if you replace the alternator cable solder it to save any further problems, done both cables on mine five or six years ago, caught just in time before melt down got away without replacing anything just cleaned up soldered and a dab of petroleum jelly job done.

 

You will need a blow lamp with a small nozzle to do this as a soldering iron will not produce enough heat quick enough.

 

If want to show video on here it's best to upload it to YOU TUBE then copy and paste the link

Edited October 6, 2012 by chrispb123456

[seatkid]

Yup, the burning is the crimp. The cable will be oxidised in the crimp, so you need to replace the cable or at least change the crimp and cut the cable back to fresh metal.

 

This burnt cable can lead to big voltage fluctuations and this can cause the ECU/dash/sensors to misbehave.

[Lewnics]

Did we ever get to bottom of this as its currently problem that I have

[BrianH]

The crimp referred to above is a known common failure point, you need to look at that first.