Voltage Regulator for a Small Motorbike, 6V AC Current

[alec_t]

"Drain/Tab" on the drawing.

Missed that. Well spotted KISS. Interesting/annoying re the differing info, depending on where the datasheet purports to originate. I wouldn't call 8 Ohms "ultra-low" .

 

[sign216]

Here's a simple test circuit for the FET. A Vds measurement > 50mV would indicate the wrong FET type.
View attachment 88823
The same circuit can be used to test an SCR as used in the bulb bypass. R1 is optional for the FET test, but not for the SCR test.
Don't keep the battery connected for long, or it will die .

I'm building the test circuit in preparation for receiving the mailed FET, but I need to test for ">50mV?" I'm not sure my multimeter goes that low. A mV is 1/1,000 of a volt, right?

 

[KeepItSimpleStupid]

milli = 1e-3 so 50 mV = 50 E-3V or 0.050 Volts.

Some multimeters have a low voltage scale. Mine has a 300 mV scale, but even on a 1.999 or 2V scale, you should be able to tell.

 

[sign216]

milli = 1e-3 so 50 mV = 50 E-3V or 0.050 Volts.

Some multimeters have a low voltage scale. Mine has a 300 mV scale, but even on a 1.999 or 2V scale, you should be able to tell.

Kiss thanks. I checked mine, and it also has a 300 mV scale.
When I saw "mV," I just assumed it needed a expensive, precision instrument that I didn't have.

 

[sign216]

N.

My local electronics store is still hasn't gotten my special order in. Rather than wait, I'll order it from Newark Electronics, a mail order place that I've found reliable.

Alec, the MOSFETs you spec in your circuit (Int. Rectifier AUIRF3205) have to be ordered from the UK, with $20 shipping! This USA product (Fairchild HUF75344G3) appears to a be a match. Can you let me know if it'll work? I've attached the two datasheets.

 

[alec_t]

That HUF should do nicely. I'd get the TO-247 case style (HUF75344G3), which has a lower thermal resistance than the TO-220 alternative style. Don't forget a mounting kit to isolate it electrically from the heatsink/metalwork (though at a push, for this use, you could get away with a strip of thin PTFE tape).

 

[sign216]

That HUF should do nicely. I'd get the TO-247 case style (HUF75344G3), which has a lower thermal resistance than the TO-220 alternative style. Don't forget a mounting kit to isolate it electrically from the heatsink/metalwork (though at a push, for this use, you could get away with a strip of thin PTFE tape).

Thanks, I'll do exactly that.

 

[sign216]

I built the circuit on a bread board, but I must admit I don't know how to test it. If I hook a battery to the board's + and - strips, then all I get is the battery's reading.
How do I set up the circuit for testing? Attached is a copy of the diagram.

 

[Diver300]

You need to connect it to a larger voltage supply, with a suitable current limiter.

The current limiter is important.

I suggest you try a 12 V battery, and use a small bulb as a current limiter. You should see about the same voltage across your regulator with a small bulb and a medium bulb. Also, if you reverse the battery, the size of the voltage should be about the same.

 

[sign216]

Where is the battery, bulb connected? As long it's merely put on the + and - rails, all I'll read from the voltmeter is battery voltage (minus the bulb, etc).

 

[KeepItSimpleStupid]

+ Battery to bulb to (reg); (triangle terminal) to ground.

Measure (reg) to (triangle terminal).

==

Then switch the (reg) and (triangle terminal).

Measure (reg) to (triangle terminal)

==

I'd use a 12 V bulb initially. Then try a 6V.

 

[sign216]

Kiss,

As I understand it;

+ battery to bulb, bulb to + terminal of regulator
- battery to - side of regulator

Measure voltage across the + terminal of regulator, and the - side of regulator.


That's what I did (but no bulb) and all I got was battery voltage.

 

[Diver300]

Kiss,

As I understand it;

+ battery to bulb, bulb to + terminal of regulator
- battery to - side of regulator

Measure voltage across the + terminal of regulator, and the - side of regulator.


That's what I did (but no bulb) and all I got was battery voltage.

What voltage battery?

 

[KeepItSimpleStupid]

As was suggested earlier. A 12 V car battery.

Remember this thing is an AC clipper or should drop about 6V or -6V depending on polarity. It's a shunt regulator, so there HAS to be a load.

If you use a 12 V bulb and a 12 V battery, then if the regulator failed shorted, the bulb would be at full brightness. If it opened, it would not be lit.

If it works correctly, the12 V bulb would act as if 6V was applied to it. i.e. 6V across the bulb and 6V across the regulator.

You cant really break anything using the 12V bulb and a 12 V supply capable of powering the bulb.

That's what I did (but no bulb) and all I got was battery voltage.

I could ask the same questions, what battery voltage and the bulb voltage? If it was 6V, then the behavior could be OK, but it doesn't tell you if it's actually working.
The regulator won't drop exactly 6V. So, a if the regulator was working at 7V and the battery was 6 V the regulator would be off.

==

Tests could be performed with a 9V battery or a 12 V wall wart but you would have to use a resistor for the load and size it appropriately.

 

[sign216]

A car battery?!! That's an extreme test, isn't it.

Edit: I tested the circuit using a transformer for the house outlet, that yields 600mA at 13v.
At different extremes of the 2k trimpot, the circuit voltage varied between 11.25v and 11.10v.

Some kind of success, in that it's limiting voltage, but not nearly as much as I need for the bike's system.

Any ideas?


Note: A small 6v bulb was between the + transformer and the + terminal of the regulator. Bulb never lit, but must have been doing it's job.

 

[alec_t]

That's what I did (but no bulb) and all I got was battery voltage.

Without the bulb there was no (or very little) current-limiting, so the FETs almost certainly got fried. Either that, or their threshold Vgs really is 8V so they're hardly turning on.
Did you use the FET-test circuit I posted, before building the regulator?

Edit: Have you correctly identified the G, S and D terminals of the FET? The two FETs should be connected S to S.

 

[sign216]

A 9v transistor battery is going to fry the FETs? That's about 500 mA. If that's the case, will they be durable enough for the bike's system, which is 6 volts, 35 watts, so about 5 amps?

I didn't build the test circuit for the FETs, because I special ordered FETs that met the given specs. If their threshold is 8v, then with 13v they should have been functional, right?

 

[alec_t]

A 9v transistor battery

From the posts above it seemed you were using 12V (car battery) or 13VAC (transformer)!

If their threshold is 8v, then with 13v they should have been functional, right?

Yes, provided they hadn't been fried earlier .

 

[sign216]

I was using two AA 1.5v batteries, and then another test with a 9v radio battery. Both just gave me battery voltage.

Later, did a test with a bulb in series, with 13v DC 500mA transformer. In that test the circuit did function, reducing 13v to 11v.

 

[sign216]

Edit: Have you correctly identified the G, S and D terminals of the FET? The two FETs should be connected S to S.

I may have misunderstood your circuit diagram as to where FET terminals go. I identified the terminals w the FET data sheet.
Attached is how I have it. If I'm wrong, can you attach a corrected version?