[Amps] Sort Start

[Will Matney]

See below;

*********** REPLY SEPARATOR  ***********

On 8/12/05 at 4:33 PM Ian White G/GM3SEK wrote:

>Larry Carman wrote:
>>The Henry RF generator using a 4 KW power supply at 4000 VDC and a filter
>>choke
>>only use a .5 second soft start. The 1/2 second delay seems to be plenty
>>even
>>with 30 mf of caps and no filter choke.  I've been running this
>arrangement
>>or years. Too much soft start in certain instances will cause the bypass
>>relays
>>to energize too slow and cause relay clattering amongst other problems.
>
>Agreed - most step-starts take way too long.
>
>When the step-start is about right, the first and second current steps 
>will be about equal. You can tell this, near enough, by comparing the 
>loudness of the two thumps from the transformer.
>
>The relative size of the two current steps depends on both the delay 
>time and the value of the series resistor.
>
>If you don't have any startup current limiting at all, the current surge 
>is big, but the filter caps are fully charged within the first 4-5 mains 
>cycles (<100ms) and then it's all over. A step-start shouldn't take much 
>more than twice that long, so a nice brisk 'ker-lick' from the switch 
>and the relay (or two relays) is about right... so that's something like 
>0.5s, or 1s tops.
>
>Too long a delay will simply overheat the series resistor, but with no 
>advantage.
>

That's the way I see it too. One takes a chance of burning one or both out, or overheating the relay contacts. Most use around a 25 watt resistor which is really too small to be in line very long. The test that was done showed the largest surge within the first 5-6 cycles, each cycle falling off sharpley to the next, after this (at the knee), it started leveling out but took as high as 25-26 cycles that I counted. This was a gradual drop though and not a really high current.


>
>-- 
>73 from Ian G/GM3SEK         'In Practice' columnist for RadCom (RSGB)
>http://www.ifwtech.co.uk/g3sek
>_______________________________________________
>Amps mailing list
>Amps at contesting.com
>http://lists.contesting.com/mailman/listinfo/amps