Hello I recently built this circuit that I designed. My problem is that it keeps on blowing the fuse. It works fine on DC, but on AC it blows the fuse. What's the problem? thanks!

I should also mention that I'm leeching the power from a power amplifier transformer.

Attached Images

	tempmon.jpg (7.4 KB, 563 views)
	power_102.jpg (4.3 KB, 558 views)

 

Have you tried it using a bridge rectifier in stead of a single diode ?

 

No, because I want this circuit to be simple as possible. I gave it a thought and maby it's the condernders fault. They get charged and then the polarity reverses and then they discharge very fast (could it be called a short circuit) and that blow the fuse. So would it help to put a diode at the Gnd?

 

The cap can't discharge back into the transfo because the diode will block it. So i don't think that is it

 

What is the rating on the fuse and what current does the fan draw?

 

630mA and the fan draws 60mA, when i measured the current consumption on DC it was 71mA

 

What does this mean? Do you mean that if you run the DC portion of the circuit off a lab supply, it works fine, but it blows the fuse when you power it off the transformer?

 

Exactly

 

I did some thinking and i found something that may be worth trying

A 7812 has a minimum input voltage of 13.70V (according to national datasheet).

When you apply 20V from a DC power supply. The diode will just conduct...20V on the 7812... No problem...

But if you apply 20V AC the diode will 'remove' the negative side off the AC voltage. This will cut your voltage in half! giving only 10V on the input of the 7812... And an 78XX not getting enough voltage might cause problems.

Like i said before, try a bridge rectifier, it won't complicate your circuit very much. Only 3 more diodes.

I'm not garantieing this is your problem, but it's worth a shot.

 

Yes I will try it.

Maybe this would also be solved with a bigger condenser? Then it could keep the 20V longer before it disharges completly? (just a thought, the diodes sound anyway better)

 

If you are using a fast blow fuse, that might be a problem. Ignoring everything beyond your 470uF filter capacitor, when you first turn on your circuit, the voltage across the capacitor is 0, which causes your cap to appear as a short circuit to the transformer/diode combination. A quick analysis, shows that the transient current on the secondary side is about 5 amps and decays for about 4ms until 0 (for a no-load situation). If your primary is 120Vrms, then the turns ratio is about 6:1, the primary current is about 0.83A which is going to blow your fuse. If it is a slow-blow fuse, it should be able to survive that transient.

 

No, I live in Finland and here we have 220VAC, hm it seems to fluxuate from 219,5 to 222,1 just measured (: And I have used slow fuses.

 

I dont know what would occur if the input voltage to the regulator is below the minimum, but the 20V from the transformer is RMS, the actual peak voltage is about 28.8V. By only using half-wave rectification, the ripple will have the same frequency as the line. Using your 470uF filter and the given specs, I found that the rms voltage on the output of your diode should be about 25.1V with a ripple of 4Vpp with a 100mA load. Your load is lower so your ripple will be smaller. But in any case you should not be coming close the 13.7V limit mentioned previously.

 

Haven't thought of what might be your problem yet, and this probably won't solve the fuse issue, but you should put a diode across the the motor to prevent any damage to the rest of your circuit when your motor current is reduced.

 

If you have slow blow fuses, and they are not blowing too quick, you should be able to put an ammeter in series with your transformer (be careful!) and see what the current draw is. An analog meter would be better as you could spot a sudden change that caused a failure. Also, can you remove some of the components from the right half of your circuit. Specifically, I would take out everything past the voltage regulator first followed by the voltage regulator itself.