JohnP I have considered the option of reducing the field current and early on in the design, that was the prime candidate. However, following recent e-mails, I have now rejected the idea for the following reasons: 1. It is intrusive into the alternator. There is no access to the field windings externally so there is a risk of damaging the alternator. (It ain't broke, so don't fix it!) 2. It has been pointed out that there is a potential risk of damage to the 4QD motor controller by spikes form the alternator (I haven't quantified the risk, but the 4QD controllers are sufficiently expensive that it makes it worthwhile to remove that risk totally). So leaving the alternator permanently in circuit is not an option I will pursue. So, having sorted out the petrol problems last night, I plan to design a relay based circuit that will only allow battery charging from the alternator when Topsy's drive switch is in neutral and the power controller is in the "Off" position. I will also have to slug one of the relays to ensure that there is always load on the alternator. Step change doesn't bother me too much as it may happen in car e.g. turning the headlights on is an immediate 130W+ load. Cheers Peter On Wednesday, March 30, 2005 11:16 PM, John Pagett [SMTP:john.pagett@xxxxxxxxxx] wrote: > A thought on Topsy's charging circuit. <snip> > The only problem now is how to stop the load from the alternator > stalling the petrol engine. Since the field current is controlling the > alternator output, why not put a resistor in series? I realise that the > control loop will try even harder to push more current through the > field, but there is a limit when all 24V is being supplied. Of course, > the resistor would have to be found by trial and error, but I'd guess a > maximum value of 4 or 5 times the field resistance. > > I have to say I've never tried this idea, and the stability of the > control circuit may go awry, but it does have the advantage of being > cheap and providing something for the driver to warm his hands on! > > Can anyone on the list confirm that this will work or not? > > JohnP > MODEL ENGINEERING DISCUSSION LIST. To UNSUBSCRIBE from this list, send a blank email to, modeleng-request@xxxxxxxxxxxxx with the word "unsubscribe" in the subject line.