Your comment on the alternator half reminds me of an issue I had a few years ago (but only vaguely remember), hopefully someone here will be able to explain this better than I as it's a right pain when it happens: After jump starting a car I found a sensor had died in one of the cars - I think the one I was starting but may be wrong. I found out this was quite common, the inductance of the jump leads allows a voltage spike to be generated on starting which can take out random lumps of electronics. Does this ring any bells with anyone?


The cure is, where possible, to disconnect the jump leads after charging the battery in the jumpee, although I seem to remember thinking that if that battery was in a fit state to start that car then it would be of low enough internal impedance to stop the problem anyway. 


I do remember that it would have been cheaper to buy a new battery than what it cost me to replace that sensor!



On another note, this was all much more fun in the days when some cars were negative earth and some were positive earth. I remember sitting indoors having a cup of tea while waiting for my positive earth Ford Anglia to charge from my dad's negative earth Renault 4 (later mine). On a hill. The handbrake slipped on one of them, it rolled down the hill, the bumpers touched, and there was an awful lot of smoke...fortunately they were very thin (probably home made) jump leads otherwise we could have had two cars with exploded batteries! Ah, we had to make our own entertainment in them days...


Cheers,


Andy

Yep there is a high voltage surge at the end of the start phase as the peak current through the starter motor inductance is interrupted. The peak level is acceptable (for properly designed components and sensors) for a regular vehicle, but there can be awkward scenarios when the leads are attached, especially if the other car has it's engine running.

Thanks, Roger, for the detailed explanation of what can happen if a single cell has failed. Others have made points about having the engine of the rescuing car running during the jump start. Although this seems to be common practice, and even recommended in some quarters, my philosophy has always been not to have the engine running. An alternator is not designed to power a starter motor, and it does not need to in a single car in isolation. Also, as others have suggested, the surges of the jump start will affect the control circuitry of the rescue car in a manner in which they are not designed to handle.


Provided the battery of the rescue car is in good condition, it should start the other car without the assistance of the engine running.

I'm pretty certain the reason for leaving the engine running on the "donor" car is so that you don't end up with two cars with flat batteries.


The alternator will put some charge back into the battery between attempts at starting the other car, and even if you do run the battery on the donor car nearly flat, the engine is already running so you can go for a good drive to put some charge back into it.

Has anyone got any thoughts on bump starting - rolling down a hill and letting the clutch in - on modern cars? I have a sort of feeling it's a bad idea, but no real grounds for that.


By "modern" I'm thinking of cars with ECU / EMS etc...I'm aware that's showing my age again...


Cheers,


Andy

I am pretty sure it is not much help with an automatic gearbox, but always having had a manual car I have no experience to base that on.

If you have electronics controlling fuel injection then I would imagine that without the electrical power you won't get fuel to the engine, and without fuel to the engine to run it, you won't have electric power from the alternator. This doesn't make bump starting a bad idea, just a waste of time.

I have a twelve year old fiesta which will certainly start while rolling when it has battery power (having tried it at low speed when the engine has stalled due to inattention in queues), but of course that is with a healthy battery so I don't know if that counts.

Alasdair

I think that it will depend on whether turning the key or pressing the button does more than just energising the starter motor. Modern vehicles do many other things, opening the drivers door on my brother's Audi turns on the heater plugs and primes part of the fuel system. If the courtesy light switch fails you have problems starting the engine.

The battery would need to support the ECU and injection pump loads to have any chance of success. Alternators need battery voltage to start working wheras a dynamo would self start. Three large people got my Triumph Herald going fast enough to start with a completely flat battery (lights on all day  ).


Best regards


Roger

Usually a "flat" battery will have enough power to activate the ignition and ancillary circuits, if not enough to operate the starter motor so the bump start should work. If the battery is completely flat you would need enough steady speed for the dynamo to activate, as with Roger's Herald. I doubt if you could do this with an alternator so it is back to jump starting. As for automatic cars, I have never been in a position to try. If I were to be forced into this position, I'd be inclined to try manually selecting second or third gear, rather than drive. Has anyone tried this?