Need help making a variable input constant output regulator

[DevSpur]

At 4000 rpm the stator delivers 60v @ 10A ac. This converted to dc gives around 660 watts. The regulator uses a shunt to regulate the charge to the battery and at the same time manages the charge to the battery to make sure it does not overcharge, so when the battery is charged most of the energy is transferred to heat in the regulator. The regulator also only runs at around 35% efficiency. As the rpm goes up so does the ac voltage and amperage to more than 1000w this is why we are basing our design on an average of 600w. The device we are powering uses bursts of current 15v @ +-60 amps for approx 15 min at a time with intervals of 10 min. The battery cannot sustain this. We have an option to manage the gates on mosfets via microcontroller but wanted to see if there was a solution using only components. Can you guys help?

 

[Nigel Goodwin]

Fit a bigger battery - the shunt only comes in to play when the battery is fully charged - but I'm fairly dubious it can provide such power anyway, or that the regulator can dissipate the 1000W it would have to when the battery is fully charged.

 

[ronsimpson]

I do not know how hour system works.
In my tractors, when the voltage gets to 14.5 volts the regulator changes the current in the "field coil" which causes less current to be made.
It does not just short out the too much current. No 1000 watts going into heat.
My tractors will make 100 volts with no regulator and no battery. (not normal)

If your voltage drops after a little time then you are pulling more current than you can make. I don't the regulator is the problem.

 

[DevSpur]

Car and I imagine tractor alternators work differently to motorbike stators.Alternators vary current while the stators on motorbikes are fixed. All I am asking for is if their is a solution for variable voltage in of 20v to 100v and a constant voltage of 15v out not using regulators as they are inefficient. We want to conserve as much of the original current as possible.

 

[be80be]

Buck converter

 

[DevSpur]

Buck converter

Buck converter would work but we have voltage spikes and there is no buck converter that can have an output of 60amps

 

[Nigel Goodwin]

Car and I imagine tractor alternators work differently to motorbike stators.Alternators vary current while the stators on motorbikes are fixed. All I am asking for is if their is a solution for variable voltage in of 20v to 100v and a constant voltage of 15v out not using regulators as they are inefficient. We want to conserve as much of the original current as possible.

As I've already said, the existing regulator already does that - all the alternator power goes to the battery (and the load) until the battery is fully charged, only then does it start wasting the surplus as heat. Any other system you might devise would have to do the same.

As I also mentioned, your output figures seem rather excessive, how did you measure them? - I would expect any motorbike with that kind of output to use a field coil type alternator, where it can be properly controlled.

 

[Diver300]

The open - circuit voltage of an alternator is proportional to the speed and the magnetic field. At maximum revs and maximum field the output voltage will be large.

In the case of an alternator with a field coil, the field coil current is adjusted to adjust the output voltage.

Alternator power windings have considerable inductance. The result is that shorting out an alternator doesn't result in much more current than it's normal maximum output.

The impedance of that inductance is proportional to the revs, so the short-circuit current of an alternator doesn't vary much with engine revs. High revs give high open circuit voltages but correspondingly high impedances, so the current is much the same.

Alternators with permanent magnets are usually treated as constant-current supplies. On some small motorbikes and on some pushbike generators (called dynamos) the bulbs are effectively supplied at constant current, due to the open circuit voltage changing in line with the impedance.

Regulation of a permanent magnet alternator is usually done by shunting excess current. Series regulation is difficult as the voltage rises dramatically if the load is reduced at high revs.

It is possible to use a higher voltage load and get more power, but it will not work at lower revs. I did that on a 1983 Honda CG125. It came with a 25 W headlight + 5 W tail light, on 6 V, so totalling 5 A, run straight from the alternator. I changed to a 60 W 12 V headlight (and moved the tail light to the battery) so the alternator load was still 5 A. It worked fine, but was dim low revs. I really didn't mind that the headlight wasn't bright when I was stopped with the engine idling.

The OP could get more power from the alternator by running at a higher voltage, but with the revs varying, it would need some sophisticated circuitry to always get power. It would basically need a MPPT circuit that can respond very quickly, or some other way of altering the alternator output load depending on revs. It isn't an easy task for 600 W or more.