Nasty electrical fault
Discussion
After the succes of this mornings post I'm seeking more words from the wise of PH.
I had an incident at the Oulton Park trackday. Loud bang, loss of all electrics.
It transpires that the cables from alternator and battery to the starter solenoid had come adrift and shorted the battery. The nut and lock washer must have vibrated loose.
I've reconnected the cables and charged the battery but no luck starting.
The ECU seems happy, read outs on the laptop diagnostics don't show any faults but it won't fire.
I'm assuming the short took out something in the ignition or injector drives.
So far examined the ignition control unit board and a quick look at the ECU.
I'll take the ECU PCB out and have a more thorough look for damage.
Any electrical gurus out there with suggestions for where to start checking?
Any quick checks with a multimeter on the injectors/coils that will show them working or not? Although I'm presuming I'd need an oscilloscope to do teh checks.
TIA
mikesr
I had an incident at the Oulton Park trackday. Loud bang, loss of all electrics.
It transpires that the cables from alternator and battery to the starter solenoid had come adrift and shorted the battery. The nut and lock washer must have vibrated loose.
I've reconnected the cables and charged the battery but no luck starting.
The ECU seems happy, read outs on the laptop diagnostics don't show any faults but it won't fire.
I'm assuming the short took out something in the ignition or injector drives.
So far examined the ignition control unit board and a quick look at the ECU.
I'll take the ECU PCB out and have a more thorough look for damage.
Any electrical gurus out there with suggestions for where to start checking?
Any quick checks with a multimeter on the injectors/coils that will show them working or not? Although I'm presuming I'd need an oscilloscope to do teh checks.
TIA
mikesr
If you short a standard car battery, it wouldn't be at all unusual to have lunched the battery.
The high current in conjunction with the internal resistance causes localised heating inside the battery.
The typical fault condition is that the plates become melted together or the electrolyte degraded, or both. This can result in a battery that appears to hold charge, but if you load it up (as with the starter) it just doesn't have any grunt.
The first thing to check is the battery. To test it you need a battery load tester, which will put a heavy load (but not a short) on the battery. If it can maintain a good voltage then it's not the battery. I doubt yours would pass that now.
Once it's eliminated or resolved, then you should start looking at other things.
The high current in conjunction with the internal resistance causes localised heating inside the battery.
The typical fault condition is that the plates become melted together or the electrolyte degraded, or both. This can result in a battery that appears to hold charge, but if you load it up (as with the starter) it just doesn't have any grunt.
The first thing to check is the battery. To test it you need a battery load tester, which will put a heavy load (but not a short) on the battery. If it can maintain a good voltage then it's not the battery. I doubt yours would pass that now.
Once it's eliminated or resolved, then you should start looking at other things.
It's a shame you had to replace the battery in vain.
You can get them tested, even if you don't have the tester, as I alluded.
Probably the next thing is the fuel pump. Is that running?
Then the injectors.
Then the sparks.
The trouble with all of this, is that it could be an input to the ECU. If that were the problem, it would be hard to find, because the fault manifests on the outputs of the ECU, but the individual sensors look OK.
You can get them tested, even if you don't have the tester, as I alluded.
Probably the next thing is the fuel pump. Is that running?
Then the injectors.
Then the sparks.
The trouble with all of this, is that it could be an input to the ECU. If that were the problem, it would be hard to find, because the fault manifests on the outputs of the ECU, but the individual sensors look OK.
Yes I did have the battery tested. Definitely damaged and needs replacing but there is some capacity in it.
I'm using a borrowed battery with the terminals the wrong way around to do some testing. (Usual problem of a motor factor with a book that shows the wrong battery ( pos/neg the wrong way around or protruding terminals!)
Definitely a spark at number 1 plug.
Fuel pump is priming
ECU diagnostics showing no faults
TPot readings change as expected when pressing the accelerator.
When cranking the diagnostics show an rpm of around 300. From which I'm presuming the crank sensor is OK. It also shows fuel pump on so the ECU seems to be getting the inputs it needs to power the pump.
So whats a safe way to check that the injectors are firing and fuel is being delivered?
A quick google found a page that suggests removing the fuel rail, wiring on the injectors and cranking the engine. Yikes!
I'm using a borrowed battery with the terminals the wrong way around to do some testing. (Usual problem of a motor factor with a book that shows the wrong battery ( pos/neg the wrong way around or protruding terminals!)
Definitely a spark at number 1 plug.
Fuel pump is priming
ECU diagnostics showing no faults
TPot readings change as expected when pressing the accelerator.
When cranking the diagnostics show an rpm of around 300. From which I'm presuming the crank sensor is OK. It also shows fuel pump on so the ECU seems to be getting the inputs it needs to power the pump.
So whats a safe way to check that the injectors are firing and fuel is being delivered?
A quick google found a page that suggests removing the fuel rail, wiring on the injectors and cranking the engine. Yikes!
mikesr said:
Yes I did have the battery tested. Definitely damaged and needs replacing but there is some capacity in it.
I'm using a borrowed battery with the terminals the wrong way around to do some testing. (Usual problem of a motor factor with a book that shows the wrong battery ( pos/neg the wrong way around or protruding terminals!)
Definitely a spark at number 1 plug.
Fuel pump is priming
ECU diagnostics showing no faults
TPot readings change as expected when pressing the accelerator.
When cranking the diagnostics show an rpm of around 300. From which I'm presuming the crank sensor is OK. It also shows fuel pump on so the ECU seems to be getting the inputs it needs to power the pump.
So whats a safe way to check that the injectors are firing and fuel is being delivered?
A quick google found a page that suggests removing the fuel rail, wiring on the injectors and cranking the engine. Yikes!
I don't know about removing the fuel rail.I'm using a borrowed battery with the terminals the wrong way around to do some testing. (Usual problem of a motor factor with a book that shows the wrong battery ( pos/neg the wrong way around or protruding terminals!)
Definitely a spark at number 1 plug.
Fuel pump is priming
ECU diagnostics showing no faults
TPot readings change as expected when pressing the accelerator.
When cranking the diagnostics show an rpm of around 300. From which I'm presuming the crank sensor is OK. It also shows fuel pump on so the ECU seems to be getting the inputs it needs to power the pump.
So whats a safe way to check that the injectors are firing and fuel is being delivered?
A quick google found a page that suggests removing the fuel rail, wiring on the injectors and cranking the engine. Yikes!
I'd say, for the time being at least, if you know the fuel is under pressure it's getting to the injectors. Unless the pressure regulation is electronic (which would seem odd) then it's not likely that the pressure regulation is a problem. Usually there's a mechanical regulator, which sends unwanted fuel back to the tank.
Perhaps you can hear the injectors "ticking"?
I reckon the best way to check the injectors, is to measure the current in the wires to the injectors. Do that with a clamp meter. You'll need to be able to split the individual wires out. Depending on the arrangement, there will be a main wire, and possibly a ground for each injector.
The ground path may go through the injector body, in which case things are easier. Put the clamp meter around either of the ground or the main wire, but not both. You should be able to measure the AC current by virtue of the magnetic field around the wire. If the ground goes through the injector body, then you'll probably be able to measure the signal just by putting the clamp meter around the injector body. Because of the solenoid inside that might work whatever the configuration.
Edited by dilbert on Saturday 7th August 18:06
I think you are in oscilloscope territory here. Its unlikely whatever bang on the battery took out the actual injectors. If have a definate spark then you have working coils. The fuel pump is simply disconnecting the return and turing your ignition on test.
More than that, your problem is that of a normal diagnostic, which would point to the processor being unaware of the problem. The injector drivers come from two different pins of the processor pin 29 and 34, and go via two different pins on the external connector. pin 38 for 1+7+3+5, and 53 for 2+4+6+8. They use differnt chips for line driving, but as you said, an oscilloscope on either of these pins should show a drop to 0v as the injector fires which should correspond to the injector times in the diagnostic.
The processor doesn't sense the times it shows, it merely calculates and sends them, so its possible they aren't actually appearing with the processor unaware.
If you've taken a plug out after cranking to see the spark, the plug hole ought to be full of unburnt fuel and a piece of kitchen paper offered down the plug hole should come back wet? Or even the plug replaced unconnected with a few turns should come back wet.
More than that, your problem is that of a normal diagnostic, which would point to the processor being unaware of the problem. The injector drivers come from two different pins of the processor pin 29 and 34, and go via two different pins on the external connector. pin 38 for 1+7+3+5, and 53 for 2+4+6+8. They use differnt chips for line driving, but as you said, an oscilloscope on either of these pins should show a drop to 0v as the injector fires which should correspond to the injector times in the diagnostic.
The processor doesn't sense the times it shows, it merely calculates and sends them, so its possible they aren't actually appearing with the processor unaware.
If you've taken a plug out after cranking to see the spark, the plug hole ought to be full of unburnt fuel and a piece of kitchen paper offered down the plug hole should come back wet? Or even the plug replaced unconnected with a few turns should come back wet.
Interestingly the Cerbera comes with its own injector testing kit 
The indicator repeater lights on the wings have the same mini timer plug that the injectors have. Pop a light out and plug it onto an injector connector and hey presto flashing indicator (albeit quite dim). Both banks have functioning injectors.
So injectors working, plugs are defo wet with fuel and defo a spark at numbers 1 and 2 plugs. But still no inclination to fire up.
Next was inlet manifolds and cam covers off to check for physical damage. Snapped crank shaft? Snapped cam shaft?
No to both the above. Turned the crank by hand using socket and ratchet and then pulled the fuel cut off valve so I could spin the engine with no fuel pressure. Everything looks OK.
The only odd thing was the kitchen roll stuffed in the inlets to stop debris getting in was blown out rather than sucked in ... ooerr
Next check is for cam timing at TDC on number 1 cylinder. But domestics got in the way (do children really need to eat every day?) so that will have to be next weekend.
Thanks to everyone on PH for the suggestions and to Joo for putting up with my phone calls today.
So am I going to be able to pull the engine out in a small domestic garage and then do a rebuild on the kitchen worktop?
The indicator repeater lights on the wings have the same mini timer plug that the injectors have. Pop a light out and plug it onto an injector connector and hey presto flashing indicator (albeit quite dim). Both banks have functioning injectors.
So injectors working, plugs are defo wet with fuel and defo a spark at numbers 1 and 2 plugs. But still no inclination to fire up.
Next was inlet manifolds and cam covers off to check for physical damage. Snapped crank shaft? Snapped cam shaft?
No to both the above. Turned the crank by hand using socket and ratchet and then pulled the fuel cut off valve so I could spin the engine with no fuel pressure. Everything looks OK.
The only odd thing was the kitchen roll stuffed in the inlets to stop debris getting in was blown out rather than sucked in ... ooerr
Next check is for cam timing at TDC on number 1 cylinder. But domestics got in the way (do children really need to eat every day?) so that will have to be next weekend.
Thanks to everyone on PH for the suggestions and to Joo for putting up with my phone calls today.
So am I going to be able to pull the engine out in a small domestic garage and then do a rebuild on the kitchen worktop?
Gassing Station | Cerbera | Top of Page | What's New | My Stuff