How to choose a thermistor to protect power supply?

[Arx]

lol.. Speak of the devil.

I just popped a fuse. Admittedly, it is way to small (2A feeding a 600VA + 68000uF), but I started it several times yesterday, after completely draining the caps with a resistor.

Variables:
Fuse Fatigue? Doubt it, but maybe.
Input signal? I ran MUCH worse through it yesterday when I was looking for rail sag.

AC phase during power up? It's like the physics gods decided to give us a demonstration.

probably would have blown the first time I tried it without the thermistor.


-Nick

 

[AndrewT]

It does occur the strongest when turned on at 0v and weakest when turned on at peak

why is this so?
I believe that at first turn on, the transformer primary is equivalent to a 1r0 wirewound resistor.
If switch on coincides with zero volts crossing then initially no current flows into the resistor.
As the voltage rises, current starts to flow and rises with the increasing voltage drive and then the core starts to magnetise. We are approaching the standard inductor with AC signal source but it takes a few cycles for this to build up. That's the delay toroids need to get through the inrush problem. 10mS is just half a cycle and 40mS gets us through two full cycles, I guess it takes at least this long to get the inductor to store and release it's energy and begin behaving as a true inductor.
I further believe that this inductor phase once started does not last very long. I think the transformer reverts to a more resistive characteristic as the conduction angle re-charging the smoothing caps becomes shorter. The mains thinks it is sending short pulses into a non linear resistor rather than the commonly held view that charging current lags voltage by 90degrees.

Please explain your/somebody else's reasoning for the 90degree phase angle between I & V

 

[AndrewT]

...... Admittedly, it is way to small (2A feeding a 600VA + 68000uF), but I started it several times yesterday, after completely draining the caps with a resistor.

Variables:
Fuse Fatigue? Doubt it, but maybe.
Input signal? I ran MUCH worse through it yesterday when I was looking for rail sag.

probably would have blown the first time I tried it without the thermistor.

Hi,
2A and 120Vac gives a maximum power input of just 240VA. Just goes to show that the average power draw does indeed sit at lower than -20db below maximum output power.
It also confirms that the smoothing caps are very good at supplying all the short term peaks that the music signal demands, even loud music.

Fuse fatigue; very likely, especially since the 2A fuse will warm up, or even get hot and sag, every time the amp is switched on.

No soft start thermistor; very definitely rupture on virtually every switch on.

 

[bowdown]

Hi CJ900RR, that circuit that you posted, will it work on 240 Vac?

Cheers

Bowdown

 

[AndrewT]

Hi Bowdown,
note that no isolating transformer is used on the control circuit.
If the relay coil goes open circuit or is removed there is little to stop the caps going overvoltage.

I would recommend a power Zener (~20V) parallel to the caps and a low value fuse (~200mA) between the main power and the 330nF.

 

[CJ900RR]

Hi CJ900RR, that circuit that you posted, will it work on 240 Vac?

Cheers

Bowdown

Yes it does. I use it on 230V and I dont think that 10V would do any differense.

But someone asked me if it runs at 110V, but im not shure. I would think it does, but I cant guarantee it.

 

[Dick West]

Yes, I asked if this circuit was designed for 120 VAC (USA) or 230 VAC (European voltage)? Apparently it is designed for Europe, not USA.

One source said he had a SPICE simulation for 120 VAc but can't find it.

Here is the link for the circuit:

http://mitglied.lycos.de/Promitheus/delay_circuit_for_toroids.htm

However, scaling certain values to work with 120 VAC can't be that difficult for someone a little more knowledgeable than me.

As you look at the circuit it seems that the values for R1 and R2 need to be changed, and perhaps R3 as well. And, the value for C1 needs to be changed.

After these parts in the circuit the rest can remain the same, except for perhaps the chain of power resistors R4 through R7.

To scale parts for 120 VAC instead of 230 VAC I assume the values for R1, 2, and 3 would be decreased by about 1/2 and the value of C1 would be doubled. Is this correct?

Then, the values of C2 and C3 could be modified to get the proper time delay for the power supply capacitance loading, and the value of the R4-R7 power resistor chain could be modified to get the big caps charged in step with the time delay factor of C2 and C3.

What do you guys think?

Thanks for helping.......

 

[sek]

Hi Peranders,

regarding the Elektor circuit:

Notice that this is an Elektor design which Promitheus of some peculiar reason has put his own copyright

Meanwhile Promitheus had updated his website. He states to be the original author.

Cheers,
Sebastian.