[Amps] soft start circuits

[Ian White, G3SEK]

PAUL HEWITT wrote:
>
>
>> -----Original Message-----
>> From: amps-bounces at contesting.com [mailto:amps-bounces at contesting.com]On
>> Behalf Of David Lisney
>> Sent: Thursday, January 08, 2004 9:05 AM

>>one minor bug is that the lightly loaded solid
>> state relay has once operated by itself, probably due to a mains
>> transient, and the power supply has powered up by itself, hardly
>> a major concern in a fully enclosed low voltage power supply but
>> not something you'd want to trust entirely. This glitch has only
>> occurred twice in maybe 10 years but still not acceptable for a
>> high voltage supply.
>>
>>
>The SSR manufacturers recommend MOV's at the line side terminal to avoid
>DV/DT turn-on.

Even better would be a mains filter, then the master mains switch, and 
then the MOVs on the SSR.

The filter protects the whole amp, including the transformer and 
rectifiers, from incoming mains spikes. The spikes are attenuated and 
stretched enough that they may not affect the SSRs anyway, and the MOVs 
are a good backup. Since MOVs can be 'worn out' by having to handle too 
many transients, it's best to connect them downstream of the master 
mains switch, so they only see mains voltage when the amp is actually 
switched on.

Although Paul wrote earlier that he'd never had an SSR "weld shut", I 
still feel uneasy about that, because semiconductor devices do tend to 
fail short rather than open. I haven't tried SSRs yet, but I just wonder 
about the minimum guaranteed current breaking capacity for any load, any 
phase angle?

OK, failing short-circuit is not likely to happen in a soft-start 
circuit, and wouldn't be a big problem if it did... but the mains relay 
that feeds the HV transformer is another story. That relay *must* open 
reliably for safety reasons. The bigger the current surge, the more 
important it is that the relay can stop it.

Maybe I'm being old-fashioned, but where safety shutdown is involved, I 
somehow feel more confident in a device whose contacts physically move 
apart.

(BTW, my standard shutdown test is to drop a piece of copper pipe into 
the anode compartment so it crowbars the HV. The protection circuit has 
to shut down the HV quietly, with no big bangs and without even blowing 
the mains fuse... and then do it again and again, to cover all phase 
angles of the mains cycle.)



-- 
73 from Ian G3SEK         'In Practice' columnist for RadCom (RSGB)
                            Editor, 'The VHF/UHF DX Book'
http://www.ifwtech.co.uk/g3sek