Welcome to our resource for MG Car Information.



MG parts spares and accessories are available for MG T Series (TA, MG TB, MG TC, MG TD, MG TF), Magnette, MGA, Twin cam, MGB, MGBGT, MGC, MGC GT, MG Midget, Sprite and other MG models from British car spares company LBCarCo.

MG Midget and Sprite Technical - Indicator problem - non MG

Here's a problem I need some logical input on, that neither the dealer garage nor the User's Group Forum have been able to help with. So if you like electrical conundrums, here's one for you!

Its a 2009 SEAT Ibiza. The right turn indicators work exactly as they should, including the dashboard light at correct speed. The left turn indicator is behaving erratically and it's the front indicator which seems to be the fault area. The side and rear lamps flash correctly at the right speed, every time.

The front nearside indicator sometimes works and sometimes doesn't. It will often flash for a period ranging from maybe 3 or 4 flashes at the correct speed, up to working for a couple of minutes. Then it stops. Wiggling the bulb may start it going again, or may not. Meanwhile the dashboard light flashes at high speed whether all three nearside indicator lights are working together, or the front one has gone on strike! The rear and side ones continue regularly at the correct speed despite the dashboard light going crazy.

I have tried cleaning and checking all the bulb connections and the single lighting multiplug to the headlamp unit. I have tried new bulbs. I have swapped the nearside and offside front bulbs together with their carriers over. The fault stays at the nearside.

I am not sure but I think if the front indicator is in a working mood, it may then stop if there is a load change elsewhere in the car eg opening the driver's door which turns on the interior courtesy lights. Its difficult to be sure of this as if left alone entirely the lamp will flash for a while, then stop. So the "load change" theory may be false.

A challenge! Any suggestions?

Does turning on the sidelights have any effect on it?
Dave O'Neill 2

Not tried that David. I will put it on the list to try. What are you thinking?
Pretty sure the sidelights work - it is of the sort of car that has permanent running lights and they work. And there is no "light error" warning light showing on the dash

A) It's Spanish
B) 'it is of the sort of car that has permanent running lights and they work' - factory fitted or user fitted?
Jeremy MkIII

Jeremy, it may be Spanish built, but it is a VW in disguise. Not that that means so much these days! Car is absolutely standard, ex factory. No modifications of any sort.

My understanding of electrics is a bit basic. I am thinking that there is too much resistance somewhere in the circuit feeding that front light, though it isn't the bulb or its carrier. Unlike "our" cars the flashing rate isn't controlled by a single flasher unit, but probably managed by the car's CPU module, which would explain the other indicators working ok even when this corner one isn't.

If the dashboard flashing indicator is sensing the excess resistance in the circuit and is maybe designed to flash rapidly to indicate that there is a fault. If the resistance in that circuit is such that the indicator will work until there is either a slight drop in voltage as something else switches on, or the resistance maybe changes as heat is generated. Would that make sense?

The loom connection including the indicator supply is via a single multiplug into the back of the headlamp unit. This is one of those modern all in one headlight/sidelight/indicator jobs so I presume has wiring tracks built into it hidden in the plastic, which feed to the indicator bulb carrier contacts. So it could be a faulty headlamp unit. Or some fault back up in the wiring loom feed to the multiplug. But how to test?

I was wondering whether it was an earthing issue, but I am assuming that the earth comes through the multiplug.

Just trying to get my head around the way it's wired, but with modern cars using canbus, it can be a bit of a nightmare.

It may even be a bad connection at the ECU end. The other puzzle is why the dash light flashes at the faster rate, even when the lights are working correctly.

Without a wiring diagram, it's just guesswork, I'm afraid.
Dave O'Neill 2

Dave, there is a lot of plastic in there, nothing to provide an earth other than through the multiplug and loom. At one stage this afternoon I though I had fixed it. I had fiddled with the bulb holder contacts and cleaned the multiplug connections with an electric spray cleaner. I already had the indicators on, rear and side flashing away steadily and dashboard one doing its own berserk dance. I refitted the bulb and it started to flash as it should, in time with the side one. I expected it to stop after 5 or 10 seconds but it kept going. After about 2 minutes I looked through the window at the dash one - still going nuts! Went back to the lights and it had stopped flashing. So I don't know if the dash lamp had slowed when the indicator was working and just started up fast or if it had been doing that all the while.

I guess if I identify the earth and supply wires for the indicator in the multiplug, I could jury-rig wires direct to the bulb holder. That at least would show if I can eliminate a possible problem with the headlamp unit itself. They are about 120 each, though ebay might produce something cheaper. But I am not going to throw money at the problem without tracking the culprit down first!

Car is my daughter's. Due an MOT. :-(

Maybe an obvious comment but is there a relay which controls the flashers?
Jeremy MkIII

Or is this any help?
Jeremy MkIII

we've got a punto - not the same and similar, but with very similar problems.

I found that the earth on the multiplug was badly corroded (later on found out why but thats a different story)

My fix was to bypass the multiplug and solder a separate earth onto the tracking in the light unit and then ran that separately to a good earth - all the problems went away

Thanks Timmy. I will try that. Partly because I can understand the logic and how to do it. But based on years of "normal" flasher systems and their problems I am still struggling with the idea of the dash lights still going berserk during the period when all the external lights seem to be working ok - even if only for a few minutes.

Thanks for the manual link Jeremy. I do have one for it, though struggle to follow the wiring diagrams on these modern systems.

The reason I was thinking about earthing earlier, was that a bad earth can cause the bulb in question to find its earth through another bulb in the same unit which shares an earth, through the circuit for that bulb to another bulb on the same circuit and earthing through that. ergo: high resistance due to several more bulbs in series.
Dave O'Neill 2

Im going thur something similar on my dodge 2007 truck... the problem im having is trying to find a good replacemt switch

And i bet your problem is in the switch its self also

But here is were im at... the factory orginal switch burned out one of its 9 fontact lugs and went dead on the head lights

(Its one block thst controls all the lights runs thur the switch including turn signals)

Any... switch, no problem ... BUT NO GETTING A replacement switch where everything works right out of the box has been challanging tosay the worse.

At this point in happy with the only issue of not having any gauge and interior lights at night ... but ive gone thur 6 of the switches in the last few weeks and they all fail in one way or another

From only one indicator working to the high beams not working to the interior lights not shutting off and the warning buzzer staying off full time

So imnot sure what to do

But yeah if its in the 2000 era, most likely your turn circuit goes thur your head light switchat some point and the little washer burnt out on its connecting tab which its designed to do to avoid fires in an overloaded situation.

My toyata truck was very similar with its head light switch that had every illumination circut ran thur it
1 Paper

Interesting that the Seat has powers and seperate earths to the body
My VW 2006 has constant switched power supplies to the lighting and the control module does the switching on the earth side
Most modern cars use earth switching from the modules

Most likely being a Can-Bus system they are VERY touchy to dirty/corroded connectors and water in the connectors as this changes the resistance and sends alternative signals back to the controller
Every connector needs to be clean and dry

An example of this is I had a car in that wouldn't start/crank over and tracked it down to water in a door pillar connector-------
By the way I have no hair left--

If you take it to a dealer with a VAG tester they should be able to tell you which circuit the problem is on in ten seconds flat-if it's got a fault it should have logged a memory code
William Revit

As l said at the start, my understanding of electrics is a bit basic. As far as l can tell the lighting and the indicators are on completely different circuits, at least as far as switches are concerned.

Willy, what makes you think this car doesn't use the same earth switching as your VW ? It's a 2009 car and is basically a VW Polo in a different body. If it uses switched earth's, then the idea of adding an extra earth cable to the headlamp unit isn't going to work is it? Wouldn't this just turn everything on permanently ? Would trying a fly -lead earth damage anything? I know nothing about Can-bus systems. First time l have heard the word.

I did take it in to the main dealer garage. They haven't fixed it but said it might be a fault with the main CPU which apparently cannot be tried on a 'sale or return' basis as it has to be factory programmed to the specific car. And they are expensive!! They did say that the OBD didn't show any error codes.


Guy, missed your original post. I'm wondering if the SEAT uses local relays for each light cluster in which case the front N/S could be faulty. Do the hazards all work?
W Bretherton

I think the hazards all work but will check next time l have access to the car. Maybe this evening.

Would adding a fly-lead earth wire between the indicator bulb holder and earth likely damage anything ?

You could check for zero ohms (or close) between bulb holder ground and body first maybe. I would have thought it should be as the flasher relay is supplying +12v.
W Bretherton

I am still struggling to conceive of a situation where the indicator bulb works at correct speed and correct intensity for anything between 5 and 100+ seconds before suddenly stopping, whilst meanwhile the dash light is very rapid throughout and the other two lamps behave themselves! It all seems so contradictory!

Guy, there will be a specific flasher relay amongst the cluster of relays, presumably under the bonnet somewhere. My understanding of CAN bus control (having read a bit more) is that the flasher relay may be solid state which wouldn't click. Could be that the front flasher wire connection is loose somewhere adjacent to the relay.
W Bretherton

Will the relay be feeding an intermittent (i.e. flashing) 12v on the output wires? would this detect with a normal digital multimeter? I have an idea in the back of my mind that they don't measure pulses very well ? I would like to determine which of the wires into the multiplug that connects to the headlamp module is the flasher one.

Does it have a towbar--
I've had similar problems with the towbar wiring connector on the rear of the car full of water causing it
Guy an extra earth could be tried but I'd have an inline fuse in it just in case as some body controllers--not sure which way the vw one goes--but some have power on one wire and earth switching on the other- BUT when the earth isn't switched on some controllers then put power into the earth side as well to monitor the circuit
In that situation it will have 12v on both wires when it's not activated
All is not black and white anymore---

You could try taking the globe out completely and see if it comes up on the dash as a blown globe--not sure if it actually tells you on indicator globes like it does on stop/park
If it does show a fault like this but not when faulting with the globe fitted then it's probably in the ecu

Being a computer it is possible it's got sh&t in it's brain which sometimes you can clear by-----
Make sure you have your radio code first--
--Disconnect a battery leed
--Turn the ign. key to the run position
--Hold your foot on the brake pedal for over one minuite--has to be a full minuite plus
--Turn the key off
--Reconnect the battery
--Reenter your radio code

It just might clear it--
good luck
William Revit

Thanks Willy, and Bill. Some things to try there. As things stand l am willing to try and test anything since the VAG technician (!) could only suggest a new CPU at several hundred pounds and no guarantee that would solve it. Pretty useless advice really!

No towbar. And other than this,it's a good condition 65,000 mile car. No real evidence of damp or rust or any other corrosion apparent anywhere.

The relay will supply an intermittent 12v, yes. Given that the rear works, you'd think the relay was ok as it will be common to front and rear I now think. The faster flashing dash indicator might mean the ECU is sensing insufficient current draw e.g. by the front flasher which seems to be going high resistance/ open circuit somewhere. If you locate the relay you could try re-inserting it a few times to clean up the contacts. But the fault seems to be somewhere between the relay and the bulbholder (or the earth).
W Bretherton

Just looked at a Seat forum and it seems that the flasher relay on many models is integrated with the hazard switch. May be worth a look but probably hard to extract as with all modern plastic fittings!
W Bretherton

The hazard switch itself could be a possible cause---they don't get used much and the contacts can be---unclean ,specially if there's been some type of silicone based dash cleaner/detailer being used with enthusiasm
Try turning the hazards on/off a few times to clean the switch contacts a bit and see if you get a result

I'll give you a little crash course on can-bus if you wish---
The relays, as Bill suggests are solid state and are usually inside the ecu/controller for whichever system is being used
Most cars now have multiple ecu/controllers
--These are called all sorts of names like pcm ecu controller module etc but basically they are all electronic controllers and usually the output from them are from solid state relays or more comonly known as transistors and are mounted on heatsinks to help them stay cool
These can only carry a certain amount of current before overheating and that is why the trend is for the ecu to do all it's switching on the earth side of things to reduce load on the relays-
Also there are several modules(or ecu's--controllers) in the car
A VW for example has a module inside the steering wheel for steering wheel controls which feeds directly to another controller under the column which is then connected to all the other cotrollers then via the twisted wire system known as can bus----getting a bit sidetracked here but trying to explain the basics of it all
Now all these modules are connected together
There are heaps of them----by two wires known as the can-bus or twisted wire system via serial data--you can't get in there with a multi meter or anything and measure communication between modules, it's all data based--
There will be the fuse box which will include a module for basics like electric windows door locks etc there is a module in the inst. cluster, one for the engine/trans
If it's dsg there will be one for that
There's a rear body module for rear lighting
a heating module and it goes on and on but they all have the same two wires going through them terminating at the diagnostic connector where the whole circuit can be tested for resistance--This would only have to be tested if the whole system had shut down, which is rare---don't worry about that
If the car is a dsg transmission or auto and the figures change correctly in the inst. cluster when the gearlever is moved all is well as usually these two are one each end of the system
This has got nothing to do with your fault but I hope it basically explains how the system works
I know it doesn't really as it took me ages to get my head around it but it's a little taster to get you interested in it all
So basically all the signals travel from one module to another as data and then the receiving module operates as directed to give the output required
Example would be
Turn on the rear wiper
The signal goes through the steering column module ,converts to data, travels via the twisted wire (bus) to the rear body module where the module converts to an output to operate the wiper
WHAT could possibly go wrong
William Revit

Wow ! Willy you have gone to a LOT of trouble with that explanation! Thanks. A whole new world to me, never envisaged, though it does make sense of the advertised claims of the capabilities of the electrics in many modern cars. I couldn't say I understand all of that at all, but now have a concept at least of how it is likely to be organised.

Input devices (switches, knobs, buttons, sensors) may well be operating on 12v systems, though possibly on the earth sides. Their output is fed through a controller that converts to data which can then be transmitted around the car and picked up at other locations where it is converted back to an output to activate the function required (motor,heater, light etc). This may be by activating relays to control sub-circuits.

As the data can be manipulated and integrated with other data, it explains how, in my case, the dash turn indicator can flash furiously at a different rate to the rear turn signal light which is behaving normally. The dashboard light isn't just a confirmation when the turn signal is on, as in our cars, but doubles as a fault indication when there is something wrong (rapid flashing).

Several years ago the same daughter had a Citroen which had a fault with the turn lights (none of them worked). The cure was to fit a new hazard lights switch although the hazard lights seemed to work perfectly. Very counter-intuitive but it cured the problem. But this was cured from internet advice, not by diagnosis, which I hate as it doesn't tell you anything about the "why" of the fault.

It doesn't get me much nearer the problem although as the rest of the turn lights work and it isn't the bulb or bulb holder itself, it points to a fault in the 12v sub circuit for that lamp only. Dirty contacts/ faulty relay/ poor earth seem likely possibilities, but I will investigate hazard switch as well. ;-)

Had some spare time today, so I played with my VW
Result was---
Turned the indicator on--flashed normally at approx 1 flash per second
Removed the globe and holder from the head light and turned on the indicator again, it flashed normally on the dash for about four flashes then sped up to at least twice as fast----rear and side flashers still flashing at normal speed but dash light and audible click at the faster speed
Also a yellow warning light in the tacho
Refitted the globe and turned it on again
Flashed quickly for about four or five flashes and then returned to normal and the warning light extinguished
Sounds familiar--??
From this I reckon the quick flashrate and sound are a warning that there is a fault
I'm thinking your problem is between the globe and the headlight plug somewhere,
You have tried the holder from the other side so it's not that, so it must be either the connection where the holder fits in or the connection between there and the headlight wiring plug---does the globe holder click over into position properly, when I was refitting mine it was easy to get it jammed in there but not clicked around into position
Maybe if you turn the indicator on and have a good push and pull around while it's going you might be able to find the bad connection
Don't forget though that it is going to flash four or five times before it changes status so you will have to push-wait- wriggle- wait etc
I'm a bit suspect that someone might have got the bulb holder jammed(like I did)and maybe bent the connections where the holder fits in
Good Luck
Also I tried to trick my hazard switch by pushing it in really slowly and jamming it sideways and all sorts but couldn't get it to fault there--I'm sure you've got a headlight problem
William Revit


I'm a bit confused by this...

"These can only carry a certain amount of current before overheating and that is why the trend is for the ecu to do all it's switching on the earth side of things to reduce load on the relays- "

Surely the current flow is the same on the earth side as it is on the supply side?
Dave O'Neill 2

Thanks Willy,
You have replicated similar conditions to mine, though not identical.

The binnacle light has continued to flash at the high rate when indicating left, over a period of some months. Even when all three lamps (globes) have been flashing correctly on that side it has still continued at its faster rate. I suspect that even when the front (faulty) lamp can be persuaded to flash correctly, there is still a higher than expected resistance that it continues to detect.

I haven't been able to get access to the car since starting this thread,so as yet haven't tried any of the suggestions. Prior to that I had played around with the contacts on the bulb holder but not attempted to bend or otherwise alter the corresponding metal tags in the headlamp unit.

The bulb holder clicks properly into place although I agree that it is easy to get it apparently installed when in fact it isn't positioned correctly. So I am alert to that. The other things I have tried is "wiggle testing" the various wires and contacts and also I cleaned the multiplug with a spray can of electrical contact cleaner.

My suspicion (as mentioned several posts ago!) is a fault with the headlamp unit - either the inbuilt metal tracks or the contacts where it connects to bulb holder and to multiplug. Hence the thought of bypassing at least the earth connection with a fly lead direct from bulb holder to earth. If I could determine which was the live feed wire at the multiplug (there are about a dozen cables to it) then I could wire it to by-pass the headlamp unit completely - at least as a test.

Willy, Dave
I think two different things are being mixed up possibly. The job of an ECU, ultimately, is to switch a device on or off e.g. a lamp, horn, motor etc. My understanding is that it usually does this by powering an external relay on or off. The ECU output which powers the relay often grounds the relay by means of an internal transistorised switch which goes to low resistance I.e. 0 ohms. This is easier for ECU design than powering the relay with +12v. The relay then switches on the appropriate device by switching the battery supply +ve. The powering of the relay and the powering of the device are two separate circuits. I hope I'm not teaching my grandmother to suck eggs here.....
W Bretherton

Dave - yes but switching on the earth side stops voltage spikes etc being fed back into the ecu during switching operation specially for things like injectors etc
Bill- yes-no probably both---it would depend on the location of the 'relays' -internal or external

Guy--one last little thing
While you are testing--
Some ecu's go into failure mode when there is a fault--I'm not sure if this one does or not but if it does what will happen is--
If you remove the bulb the controller will sense a problem and turn that circuit off-It will monitor the circuit every few seconds and if the circuit returns to normal (refit the globe) it will notice that and turn the circuit back on
This can be a real pain when testing with a meter or test light when there is a fault present as the controller won;t be giving anything out
I have proved with the indicator that it self corrects after refitting but with say a taillight globe if you remove one of those while it's turned on and then refit it, it won't reset until you turn the lights off and turn them back on-----My vw is like this and so are the Ford Focus/Mondeo things not sure about your car
I'm off to the island for a few days , I'm looking forward to see what you find
William Revit

This thread was discussed between 17/06/2017 and 22/06/2017

MG Midget and Sprite Technical index

This thread is from the archive. The Live MG Midget and Sprite Technical BBS is active now.