14V Voltage Regulator 18A Max

[Renegad87]

Hello,

I'm new to the forum.

I am working on a charging system for a motorcycle. It currently puts out 14.6 to 14.8 volts to the battery and the current changes as needed to charge the battery. At the most it will provide 18A or so the manual says. My problem is I need to limit it to 14.4V max, so I was shooting for 14V.

I'm very familiar with voltage regulators such as the 78XX or the 317 and these work great however they won't handle the amperage. Does anyone know of a circuit I could build that would accept the input voltage of 14.6V and provide an output of 14V all the while sustaining 18A which will change randomly?

As always I appreciate the help.
-Nick

 

[Nigel Goodwin]

The battery already does what you want, what's wrong with that?.

The 18A will be the MAXIMUM from the alternator, only at high revs.

Assuming your problem is overcharging the battery?, what kind of controller does it currently have? - I would presume a 12V system would have a regulator?.

 

[MikeMl]

Is this the spinning permanent-magnet rotor with a three-phase stator type of alternator?

Some of these are unregulated, and just have a three-phase six-rectifier stack, others have a transistorized shunt regulator. Usually the assembly is potted in back goo, so not servicable.

 

[Renegad87]

Thanks for the replies.

The problem is the battery being used has an absolute maximum charging voltage of 14.4V. No if, and, or buts about it according to the manufacturer. You are correct in assuming the 18A is a maximum at high RPM which is unlikely to occur for long duration and only if the battery is practically dead. If I can assume a lower current rating I am not sure how to determine it, I feel like there are too many variables to confidently say it will only have to handle a lower current.

The current charging system is an alternator with a full wave rectifier that has an integrated voltage regulator. It is factory set for 14.6 to 14.8V and is not adjustable. It controls the voltage by varying the excitation voltage going to the field winding of the alternator. I have access to both the positive and negative wires coming from the field winding. This year they control the negative side of the field, in previous years they controlled the positive.

 

[MikeMl]

If you can get to both sides of the field, then just use a conventional automotive alternator controller/regulator. There was a long running thread on here about that. I'll go find it.

ps: Here it is. Read it from beginning to end, and come back if you have any more questions.

 

[Nigel Goodwin]

Thanks for the replies.

The problem is the battery being used has an absolute maximum charging voltage of 14.4V. No if, and, or buts about it according to the manufacturer. You are correct in assuming the 18A is a maximum at high RPM which is unlikely to occur for long duration and only if the battery is practically dead. If I can assume a lower current rating I am not sure how to determine it, I feel like there are too many variables to confidently say it will only have to handle a lower current.

I would suggest you're worrying needlessly, lead acid batteries limit their own voltage - but why not use a proper motorbike battery if you think this one isn't suitable?.

 

[MikeMl]

How about this? leave the existing charging system alone, and add-on a "automatically turn on the headlight/running lights when the battery voltage exceeds 14.4V" circuit?

Can be done with about six cheap parts..

 

[alec_t]

Or you could lose a current-dependent ~0.4V - 0.7V in a fat Schottky diode such as this.

 

[MikeMl]

Or build a secondary regulator using a high-current PFET. Since it only has to drop a maximum of ~0.4V, it will not be dissipating much power? Again, could be done with few, cheap parts.

 

[Renegad87]

I was trying to avoid this because I didn't want to get into the battery technology debate, but the battery is not a lead-acid battery. It is a LifePo4 battery and was chosen for it's small size and high output. This motorcycle is a custom build and the battery location does not allow for a full sized lead-acid or AGM. This battery technology is extremely voltage sensitive and countless batteries have been killed due to improper charging voltages. Hense why I am at this point.

The "auto-on" light idea was tried and did not have an affect on this voltage. As lights were turned on the voltage regulator adjusted itself to maintain the 14.6 - 14.8V. It is working as intended by the OEM for the original battery technology.

The schottky diode and Power Mosfet are ideas I'm interested in. The forum that "MikeMl" linked to does bring up a good point. If I place anything after the existing voltage regulator it will have to be able to sustain the high current that may occur, however if I replace the existing voltage regulator with something else it would only need to handle the current going through the field coil which should be significantly less.

 

[alec_t]

they control the negative side of the field

Can you disconnect the reg from the controlled (negative) side of the field winding? If so, a home-brew reg could give a closely-regulated output. Now that we know your battery is a LiPo one, the Schottky probably wouldn't give a sufficiently accurate output.

 

[Renegad87]

Yes the regulator can be disconnected from the winding. It is integrated into the rectifier but has it's own external connections. One of my ideas was to just use the rectifier then make my own regulator to control the field coil voltage. From the circuit diagram in the manual it appears I could just ignore the integrated regulator and it won't have any ill effects on the rectifier.

 

[MikeMl]

Here is a prototype for a secondary regulator:

I start with the battery at 13.5V (discharged). There are some bike loads R7. When the bike starts, the existing charging system puts out an initial 18A, but the voltage at V(in) is dragged down by the loads and battery. The big PMOS M1 is fully turned on, so based on its Rds, V(out) follows V(in). The 18A available from the alternator divides, about 13A goes into the battery I(R6), and the rest into R7.

The battery voltage is increasing. When it reaches 14.4V, the input to the TL431 exceeds 2.495V, and it turns on, pulling down on the base of Q1. That begins turning off M1 at about 13s in the simulation run (actual times are arbitrary). Note that D1 is being used a ~2v Zener.

As M1 turns off, V(in) rises to its max available 14.8V. Meanwhile, the current into the battery tapers to ~zero while the load current through R7 is maintained constant.

This is just the prototype, not shown is key switching to prevent battery discharge while the bike is parked, and an adjustable sensing network.

 

[Renegad87]

Thanks for the circuit I'll get it in multisim to see if it will do what I want.

I appreciate all of the responses.

 

[shokjok]

I suggest checking your local library for books from the Amateur Radio Relay League ( www.arrl.org) for a high current power supply that uses an LM723 voltage regulator feeding four 2N3055 transistors; one I found in the 2006 edition that I intend to build with a solar panel option. You could also salvage an older computer's power supply for parts for your application.

 

[ronv]

Here is a prototype for a secondary regulator:
View attachment 92698
I start with the battery at 13.5V (discharged). There are some bike loads R7. When the bike starts, the existing charging system puts out an initial 18A, but the voltage at V(in) is dragged down by the loads and battery. The big PMOS M1 is fully turned on, so based on its Rds, V(out) follows V(in). The 18A available from the alternator divides, about 13A goes into the battery I(R6), and the rest into R7.

The battery voltage is increasing. When it reaches 14.4V, the input to the TL431 exceeds 2.495V, and it turns on, pulling down on the base of Q1. That begins turning off M1 at about 13s in the simulation run (actual times are arbitrary). Note that D1 is being used a ~2v Zener.

As M1 turns off, V(in) rises to its max available 14.8V. Meanwhile, the current into the battery tapers to ~zero while the load current through R7 is maintained constant.

This is just the prototype, not shown is key switching to prevent battery discharge while the bike is parked, and an adjustable sensing network.

Mike,
Since he can get to the field coil, why not just do a low side regulator?