Continue to Site

Welcome to our site!

Electro Tech is an online community (with over 170,000 members) who enjoy talking about and building electronic circuits, projects and gadgets. To participate you need to register. Registration is free. Click here to register now.

  • Welcome to our site! Electro Tech is an online community (with over 170,000 members) who enjoy talking about and building electronic circuits, projects and gadgets. To participate you need to register. Registration is free. Click here to register now.

Permanent Magnet DC Regulator

Status
Not open for further replies.

icedeocampo

New Member
Hello!

I'm currently designing a regulator circuit for my motorbike. I've been doing work on shunt regulators, but it seems to be producing lots of heat, and I can't figure out how to make it work.

Here's the scenario:

1. My motorbike (Yamaha Mio 110cc) has a permanent magnet that spins around a stationary stator. 2 wires that comes out from the stator, a yellow wire and a white wire, I guess this is not a 3 phase system.

The yellow wire seems to be a low current battery charging source, while the white wire seems to carry a high current.

What I do to rectify the AC voltage (white wire + chasis body ground) is I simply put a diode on the white wire (half wave). I was trying to use a full wave bridge rectifier but the bridge diodes seems to overheat when i connect the Negative side of the bridge to ground.

during Idle my bike produces around 20vAC and when in high rev it produces around 50vAC. When I try to use a shunt regulator on the high current (white) wire, everything seems to be to hot, so I guess the shunt regulator would not be a good way to control voltage.

I need to control that 20v-50v AC and get 12v-14v out of it. I hope you can help me out.
 
These systems are exceedingly crude, there doesn't seem to be any sensible way of improving them without rewinding the alternator coils? - as they don't have any extra capacity to play with.

As for a shunt regulator, it's supposed to get hot - you need big transistors on substantial heatsinks.
 
hmm.. the problem with the yellow wire is that it has a very low current while the white one has a very high current.

would you be able to give me a sample schematic of shunt regulator with a negative common ground which i can use in this bike of mine.

would it be possible to use a PNP MJ15016?
 
Last edited:
Why use a shunt regulator?

A shunt regulator needs series resistance and shunts the "excess current" to ground.

It is therefore much less efficient than a series regulator.
 
If it's how bikes I'm familiar with are, it works like this!.

The thin wire feeds a rectifier, and charges the battery - this then feeds the indicators, brake light, and side lights.

The thick wire feeds directly to the headlight, no rectifier, no regulator, the alternator winding itself is designed to be self limiting - and not provide enough power to blow the headlight bulb.

The drawbacks are the tiny battery used, and the low charging capability, plus the poor brightness of the headlight, which also dims as you slow down.

What are you trying to do?.
 
use the white (thick) wire to power HID unit's balast. I want to use that thick wire which is suppose to power the the bulbs as a rectified and regulated DC voltage (or as a better battery charger)
 
For a start you will need a MUCH larger battery, the ones provided don't work well enough for the indicators and brake light!.

What are the current requirements of the HID?, and what was the original bulb?. For that matter is the original battery 12V or only 6V?.
 
the original battery is 12v.

the original bulb was rated at 25w for the head light and 20w for the tail light , I have since replaced the tail light with LEDs. I guess have around 40-50w of power to play with for the HID.

At the moment I tapped the HID directly to the battery... draining it a little each day.. and in around 4 days... I'll be needing to charge the battery... I do that every 4-5 days :(

I'm thinking of regulating it to 13.8v and feeding it into the battery.
 
Last edited:
ljcox said:
Why use a shunt regulator?

A shunt regulator needs series resistance and shunts the "excess current" to ground.

It is therefore much less efficient than a series regulator.


can you guide me as to how i should go about doing this circuit pls :)

thanks everyone for the replies!
 
As you're already aware, there's no way of adjusting the alternator output, as you can with a car alternator (by varying the field current). So all you're doing is trying to stop the battery over-charging.

Back in the 60's Triumph did this by using a LARGE stud mounted zener diode, directly across the battery - mounted on a heatsink, and positioned between the fork legs, to get forced air cooling.

Such zener diodes would be expensive, so it's easier to use an amplified one, this is a simple example, the transistor needs to be a power type, able to sink the output of the alternator, and needs to be mounted on a decent heatsink where it get's good air flow.

As an alternative you could get far more complicated, using opamps or comparators feeding a power transistor - but something like this might be enough for your purposes?.
 

Attachments

  • amplified_zener.gif
    amplified_zener.gif
    1.3 KB · Views: 1,572
hmm... i'll be trying to do this using 2n3055

the voltage swings from 14v to 60v so I guess this transistor would be enough.

What would be the ideal resistance value of the resistor?

I was thinking... instead of using 2n3055 can i simply reverse everything and use a PNP Transistor and reverse everything so that I can put the collector heatsink as ground to the chasis?

Would I need to put a Diode if I want to Tap ve+ and ve- to the battery?

Again thanks for the super fast replies!

Here's what I have in mind (Ignore R2) would this work?

zener-replacement-gif.12596
 

Attachments

  • zener replacement.GIF
    zener replacement.GIF
    2 KB · Views: 1,781
Last edited:
i've already implemented it and it seems to be working although I'm just using it to on a 3A Transformer with 14vAC and 28vAC output.

the transistor is currently on a small piece of aluminum and it's very hot, I guess I'll be needing to get a bigger piece of aluminum and I'll try to put active cooling onto it, using a cpu fan and maybe that same heatsink as well.

would i be able to put it on the battery without shorting the battery terminals as well? won't i be needing a diode? I'm thinking of putting a diode on the emitter-collector of the transistor. hehe it's currently 10pm here, and I can't start my bike without waking up my neighbors.

and I'll be needing to go to some junk shop to get a decent heat sink, I'm looking at an old cylinder head here.

thanks for all the help, i'll post pictures as soon as I get it to work
 
icedeocampo said:
i've already implemented it and it seems to be working although I'm just using it to on a 3A Transformer with 14vAC and 28vAC output.

the transistor is currently on a small piece of aluminum and it's very hot, I guess I'll be needing to get a bigger piece of aluminum and I'll try to put active cooling onto it, using a cpu fan and maybe that same heatsink as well.

The transformer probably provides more current than the alternator does, and certainly provides it more often - the alternator output varies with engine speed. Then the regulator will only conduct when the battery is fully charged, before that the alternator output goes in the battery, and when you're not using the lights (when it goes to the lights instead) - it will have a MUCH easier life on the bike.

Don't even think 'small pieces of aluminium', use a proper finned heatsink.

would i be able to put it on the battery without shorting the battery terminals as well? won't i be needing a diode? I'm thinking of putting a diode on the emitter-collector of the transistor. hehe it's currently 10pm here, and I can't start my bike without waking up my neighbors.

No need, it won't conduct until the voltage is higher than the battery alone will provide (or at least it shouldn't!), and clamping the battery at 13.8V isn't going to do it any harm. It might be a good idea though to measure any leakage current through it when you connect it to your battery.
 
would it be better if i use a PNP Darlington Transistor? suchs as this MJ11021

i've read somewhere that a transistor in saturation produces less heat.

sorry for the n00b questions, i've only been trying electronics for around 2 weeks.
 
icedeocampo said:
would it be better if i use a PNP Darlington Transistor? suchs as this MJ11021

Depends on the base current you have through the zener - bear in mind a darlington has double the Vbe drop.

i've read somewhere that a transistor in saturation produces less heat.

Yes it does, but you don't want the transistor in saturation, it runs as an analogue device.
 
Why not use a p-channel MOSFET?
 
Hero999 said:
Why not use a p-channel MOSFET?

I don't know how they work :(, if you'd be able to come up with some diagram and those MOSFET part numbers, I'm willing to try them out =).

I'll be going to a machine shop later to drill holes onto a heatsink for the current implementation - using the PNP transistor ^_^

If you'd be able to comeup with some MOSFET implementation, I'd be more than happy to try it out
 
would it be possible to have something that is on when the voltage is below, say 14v and... and it'll turn itself off when the voltage goes beyond 14v.

It'll act like a switch... so I can maintain a 14v charging... and when it's beyond 14v It'll just turn itself off... so I won't need to dissipate heat.
 
Status
Not open for further replies.

Latest threads

New Articles From Microcontroller Tips

Back
Top