As elegant as having a special DC step start relay with charging circuit, etc.,
remember that electrolytic caps are notoriously variable as far as long term
capacitance tolerence goes. Design so that over the long term, your ciruit
doesn't eat resistors or rectifiers due to changing time constants with age.
On my SB220, I used a relatively cheap and simple scheme, maybe it was AG6K's
idea - I cannot remember now. I used the same 115 VDC coil relay that Heathkit
had used for RF, which now is done with real RF relays (vacuum and reed).
Connected it to the DC bias supply through a series 1.25K resistor, a composite
of parallel 8.2K, 5.6K and 2K, all power resistors. This was a trial and error
design; hence the gob of parallel resistors under the chassis. It ain't pretty.
With the 20 uF capacitor in the bias, and the series R to the relay coil, the
time constant is really provided by the bias transformer impedance, the
capacitor being charged up, and the fact that the series R limits the voltage
turnon risetime at the primary of the transformer, since the big HV iron is
also switched in at the same time - and has inrush which causes the voltage to
come up slowly through 20 Ohms x 2 . The R to the coil keeps heat down, from
being overdriven by the bias supply.
On the relay contacts, both poles of the 220 VAC power line are fed through a
20 Ohm 10 watt wirewound resistor, before it gets shorted out once the bias is
up to the level required to pull in the relay. Simple, works, has about a 0.5
second delay if I remember correctly. My lights don't flicker when I hit the AC
switch now. The HV starts around 1500 and then kicks up to full voltage.
K5PRO
--
FAQ on WWW: http://www.contesting.com/ampsfaq.html
Submissions: amps@contesting.com
Administrative requests: amps-REQUEST@contesting.com
Problems: owner-amps@contesting.com
Search: http://www.contesting.com/km9p/search.htm
|