Hi All, I have just had to replace the power transformer in an AWA model 573 valve radio. The 6X4 rectifier had shorted and taken out the power transformer. Is the live lead to the transformer input the best place to position the fuse to avoid future problems?. bowler

Preferably, the secondary should be fused. Under some circumstances, and depending on the transformer's design, the secondary can be severely overloaded without there being sufficient current in the primary to blow a fuse. (Aside: Any fuse in the primary needs to be of the slow-blow type to accommodate inrush current at switch on.)

Some tube circuit designs incorporate a 'fusistor' -- that is low ohms resistor in series designed to burn out should there be excessive current in the circuit.

Fuses in the primary tend to be about as useful as an RCD & circuit breaker, if the secondary faults. Got a nice photo of a transformer melt down in a Thorn where the fuse wedge got melted & I salvaged the fuse.

With 6X5 there was a tendency (AWA did it & I have done it) to put a 100 Ohm 1/2 Watt resistor in series with the plates. Early designs of them particularly in Zenith sets were a horror.

I have done the same with diodes and other rectifiers as that also has some effect on inrush current. The idea is to have the resistor "sailing with the wind" so its wattage allows it to burn on overload before the transformer gets to critical temp.

The biggest issue with a fuse is inrush. The bench PSU has a critical fuse on its reformer, however, should one forget and switch it on with B+ alive the 10μF cap will take it out.

Marc

I have a tendency to agree with Marc about protecting a transformer from a rectifier fault.
A line fuse is there for when a wire shorts out and pulls 20 amps or so on the mains line, then the CB or fuse pops.

You guys will have noticed in a lot of my projects I put a resistor(s) in the plate or diode circuits.
Its far better to have a 100 ohm resistor smoking than the HT windings.
I have had a bridge fail in one of my projects and the resistor in the diode shorted leg smoked furiously.
I am a bit slow but the smoke and smell made me wake up quicker than smelling varnish!

In normal operation 100 ohm 1w resistors only drop a couple of volts, negligible.
Recommended if there is any doubt.

Fred.

Yes I agree with Marc and Fred.

Small transformers are difficult to protect with a fuse. The best solution is a thermal fuse in intimate contact with the winding.

"Fusible resistors" are a good idea.

Pye used to use a 1/2 watt Morganite resistor with a spring around it, so designed that the heat would melt a solder joint and the spring would open the circuit. Once you fixed the fault you would re-solder the spring and it was re-armed.

They called this a "circuit sentry".

They were typically used to protect 6CM5s from a dead damper diode or horizontal oscillator failure. The resistor was in the G2 circuit where the difference between normal operation and a fault condition was about 50%.

This cheap but effective idea was carried forward into SS TVs where it avoided the holes burnt in circuit boards seen in other makes.

Nice touch.

I can suggest that another approach is to add a 1N4007 in series with each 6X4 anode, as that protects against the 6X4 arcing and short circuiting across the power transformer secondary - which may be what happened.

The other likely failure mode is the 6X4 common cathode shorting to heater, and that would require a fuse in the CT leg of the full-wave secondary. However there is a 150 ohm backbias resistor, which with the winding resistance may limit the fault current for a fuse not to be effective. Anyway, the fuse value should be the next rating up from the actual measured operating CT rms current - which would need say a 1 or 10 ohm resistor inserted in that CT leg so that a true rms meter could make a current measurement.

The other problem with fuses today is that like oils, "all fuses aint fuses!"

Fuse manufactuers put a rating on them that should comply with a failure curve.
Thus a 0.5 amp fuse should never blow at 0.5 but then fail at ??? 0.6? 1??depending on the curve.
So you have to look at the rating curves that depend on, amp level and temperature with time.
Thats when you start to see that fuses are not much chop unless the fault current is 5 or 10 times the rating!

Has anybody ever tested what is sold for fuses today?
Well I did about 10 years ago.
I can tell you that the 3AG wire fuses are not too bad complying with standard fusing curves pretty closely but any thing else was rubbish.
I had plastic mini fuses rated at 5 amps hanging in there with 25 amps through them, the case of the fuse buckling and melting, but no blow.
Those short little wire fuses and the coloured ones used in cars are pretty broad in the rating, you need to add the word "about" in front of the fuse rating.
Some of the car audio fuse or circuit breakers are just rubbish. You have devices that are supposed to handle 50 or 150 amps about the size of a postage stamp! No break distance internally no spark quenching and at 24 volt they just arc over and keep going until they melt in a cloud of smoke.

End of rant.
Fred.

Fred, mains AC fuses must comply with a standard - here we typically use IEC 60127-2 (sheets 2 & 3) which have 5 windows of performance - the manufacturer may show a bogey line but its the window limits that have to be met. I doubt people could test a fuse to any level of technical rigour as it would mean setting up constant rms current source, a zero crossing gate, and blowing at least 10 samples for statistical significance and applying all the test standard window conditions, so lots and lots of tests. End of rant, Tim.

Many thanks for the comprehensive advice. Bowler