Voltage Regulator: Design, Testing & Replacing

[Adder_Noir]

Hi,

I started a project a few months ago to replace the voltage regulator on my car's alternator but I had to stop due to lack of knowledge and testing ideas.

I've since learned a lot more about electronics and have decided to pick it up again. Straight away I've made a design change after reading some of the sections in The Art of Electronics by Horowitz & Hill. Hard going but a good book!

Here is my suggested circuit. The left side shows feedback from the stator. This goes through a low pass filter with the 3db point at around 1.75Hz and a charge time to full of 0.5s (time constant of 0.1s). This then overcomes the zener if too high and opens the transistor collector/emitter path and dumps current to Earth through a low resistor R3.

Normally the rotor coils would be being charged at the same current the existing regulator consumes at idle speed, which is settable by changing the value of R4.

I propose to make R1 and R2 very high maybe 100k plus to prevent circuit loading and to keep the signal good. I've arrived at some theoretical values for all components involved I just have one more thing to do before acquiring a second hand alternator and testing this out. So my question is can anyone see any major problems or design flaws with it before I go to testing stage?

http://img486.imageshack.us/img486/3997/circuitca2.gif"

I'd appreciate any feedback offered thanks

*Edit*
One thing I have thought of is that when the transistor is fully on I should arrange for the parallel resistance of R4 and R3 to be the same as just R4 in series.

 

[waht]

What is the voltage of the input signal, it is AC?

The circuit would work, by it is generality not practiced to put a zener diode in series with the base. The values for R1 and R2 in 100K range are too high, and most likely there wouldn't be enough current to switch the transistor on. I recommend do in the 10K range. Also because the transistor is switching an inductive load, you need put a regular diode in reverse to the rotor coil, otherwise by switching the coil on and off high voltages will develop and kill the transistor instantly.

If you provide more info, maybe I can help a bit more.

Hope that helps.

 

[Averagesupernova]

Usually alternators have the field (rotor) with one side attached to B+. The other side is tied to the collector of a transistor which has a grounded emitter. Doing it the way you describe will waste considerable amount of power in R4.

 

[Adder_Noir]

What is the voltage of the input signal, it is AC?

No it's a battery fed DC supply.

The circuit would work, by it is generality not practiced to put a zener diode in series with the base. The values for R1 and R2 in 100K range are too high, and most likely there wouldn't be enough current to switch the transistor on. I recommend do in the 10K range.

Deal, thanks for pointing that out.

Also because the transistor is switching an inductive load, you need put a regular diode in reverse to the rotor coil, otherwise by switching the coil on and off high voltages will develop and kill the transistor instantly.

Yep forgot that one!

Hope that helps.

It did thanks mate.

Usually alternators have the field (rotor) with one side attached to B+. The other side is tied to the collector of a transistor which has a grounded emitter. Doing it the way you describe will waste considerable amount of power in R4.

I see thanks for mentioning that pal. The rotor coil has a resistance of 5 ohms roughly I'll just leave it at that as the charging resistance. Thanks mate

By the way men is my 100k 1uF 0.1 time constant low frequency filter component choice values ok?

 

[Adder_Noir]

I also noticed today when disconnecting the excitation wire but turning on the ignition without turning over the engine there was a loud whirring sound. This stops when re-making the connection with the exciter wire.

Is this sound the sound of the battery trying to drive the alternator as if it were a motor? There was 0.15 amps running through the exciter wire even when the engine was not turning over. I assume there has to be some torque applied to the rotor to make it counteract the attempts of the battery to turn it?

 

[MountainDew]

I also noticed today when disconnecting the excitation wire but turning on the ignition without turning over the engine there was a loud whirring sound. This stops when re-making the connection with the exciter wire.

Is this sound the sound of the battery trying to drive the alternator as if it were a motor? There was 0.15 amps running through the exciter wire even when the engine was not turning over. I assume there has to be some torque applied to the rotor to make it counteract the attempts of the battery to turn it?

Please don't take this the wrong way cause i am not in any way saying you don't understand the workings of it, i just wanted to put my 2 cents in if that's ok.

Ground the housing on the alternator to the negative terminal, take the FAT wire on the alternator, hook it to the + terminal of the battery "this is the terminal for charging your battery, or running lights. turn the alternator, nothing will happen, now excite the alternator by touching the smaller wire to the +12 terminal. the alternator will start to hum if not spinning "coil viberations humming", if spinning, torque will be greater or harder to turn.

The alternator Produces AC, there is a rectifier converting the ac to dc, after the ac is converted to dc it's then feed to a voltage regulator of say maybee 12 volts? the amount of amps the alternator produces should be stickered to the side of the alternator, basic oem car alternators is fairly low in amps, fleabay has high amp alternators, the biggest i saw there was 200 amps where the one on my geo is only 55 amps. i also think that whatever voltage the alternator produces, that is not in used, is feed back into the field coil.

how does the car stay running when you pull the positive terminal off? when you get the alternator generating, if generation was enough, it can continue to excite the wire causing the system to keep generating power, as long as there is horse power turning the alternator there is no need for a battery, allthough, great loads can blow your rectifier or voltage regulator, they are not made to be ran without a battery even though its possable to do so.

Usually there is a hole on the back side of the alternator, it is there to short the circuit, you simply stick id say a nail in it, and it will generate the max the generator can produce determaned by the battery and speed torq etc,, on the alternator. i don't recommend this, and if you do do it, make sure you know what you're doing.

 

[light_bulb]

the circuit won't work as shown, it's best to leave the alternator unregulated and have a tap off the b+ and put a simple choke and bypass cap on the line, car electronics are made to handle the voltage swings.

 

[Adder_Noir]

Ah thanks for the replies chaps

MountainDew thanks for the detailed reply. I now think that the whirring sound was the current running into the hydro-electric steering motor instead of through what is probably a low resistance path to ground through the voltage regulator.

Lightbulb I appreciate the idea (it's a damn good one) but I'd prefer the challenge of replicating something Vauxhall's made.

Does anyone know if it's possible to get a transistor which opens rather than closes when current is run into the base? I'll post up my new circuit design in a bit if someone can answer the question.

 

[light_bulb]

a reverse biased mosfet or you could use two transistors.

 

[Adder_Noir]

Sorry for being so rude as to not reply but I've since done some more testing on my car and I've found something I didn't expect. I'll be in touch again when more progress occurs.

 

[NoTime]

R4 sets the minimum field coil current.
Partially because it limits inductive kickback from the field coil and also allows the alternator to produce some power if the transistor fails open.

R3 sets the maximum field current. Helps preserve the alternator if the transistor shorts of if the load is too great.

Cf Rf also help limit inductive kickback from the field coil by preventing the field drive current from changing rapidly.

You need another transistor to invert the control voltage signal.
You should still have a kickbavk diode on the field coil.

 

[Adder_Noir]

R4 sets the minimum field coil current.
Partially because it limits inductive kickback from the field coil and also allows the alternator to produce some power if the transistor fails open.

R3 sets the maximum field current. Helps preserve the alternator if the transistor shorts of if the load is too great.

Cf Rf also help limit inductive kickback from the field coil by preventing the field drive current from changing rapidly.

You need another transistor to invert the control voltage signal.
You should still have a kickbavk diode on the field coil.

Thankyou NoTime again for your help and insight