Interesting little HK fault today ...

HK Actor, chassis only - the shop had removed it from the cabinet for me, as they always do on these to save workshop space. Reported problem was "powers up but no sound". At a first look, that did appear to be the case. However, when the board with the PSU on it was moved ever so slightly, it came on, but then stayed on until it was next turned off. Once on, no amount of provocation would make it go off again. Likewise, when it was wrong, no amount of tapping or hitting would make it work, but as soon as you turned the chassis over, on it came. I then noticed that when it was wrong, although the mains switch was lit, giving the impression that it was powered up, the fan wasn't running. This is powered from its own personal supply with a single diode and filter cap, straight from one side of the split secondary winding of the power transformer, so if there is power to the transformer's primary, then the fan pretty much has to run.

A bit more observation of what happened when it did come on, revealed that a relay clicked. This is across the inrush thermistor, and again, its coil is fed from a simple supply derived directly from the transformer secondary. It then dawned on me what was happening, and a quick tug on the thermistor proved it. One leg had sheared off inside the hole passing through the PCB, making the thermistor effectively open circuit, so no supply to the transformer primary. The switch of course lit up, as the neon is internal to it. The gap in the broken leg was obviously big enough that the vibration from tapping, was not enough to close it up, but when the board was flexed, that was enough, and the leg momentarily made again. This was enough to get the relay coil supply up, the relay contacts then bridging the thermistor, so it didn't matter any more that it had gone open circuit again. And once the relay contacts were shorting the thermistor, no amount of disturbance would make it die again, until it was actually turned off.

What a pleasant change to get a nice positive fault with reasoning that works .... :-)

Arfa

Reply to
Arfa Daily
Loading thread data ...

as

"powers

However,

powered

a

is

It

PCB,

to

flexed,

get

I've always expected to see problems with polyswitch "fuses", considering the temp they run at, but I don't think I've seen a problem one. I expected a "welded" lead to come loose or internal cracking of the material

Reply to
N_Cook

Dry joints on these thermistors are surprisingly common. I think it may be because the higher powered ones are kind of tall and weighty and have no support other than their leads. Powered speakers are a favourite for this, possibly because they get a lot more mechanical shocks in transport than your average punter would give to, say, his desk or FX rack, and also because they get a LOT of vibration in use.

I always wiggle them whenever I see them, and if the PCB comes out I will always resolder them (along with the usual suspects, e.g. zener diodes, power resistors that get hot etc).

Cheers,

Gareth.

Reply to
Gareth Magennis

"spamtrap1888"

We eventually decided not to use them for loudspeaker protection, because they cycled so rapidly that we decided consumers would complain.

** Polyswitches respond slowly and reset only after a long wait.

Even though these were the same consumers who didn't notice that their speaker's voice coil former was bottoming out against the motor structure.

** Bollocks.

.... Phil

Reply to
Phil Allison

ElectronDepot website is not affiliated with any of the manufacturers or service providers discussed here. All logos and trade names are the property of their respective owners.