Topband: Upon further examination of powering 26.5V relays with 12V

[Tom Rauch]

I'll go along with Thomas AC7A on this one. I measured 
several relays in an actual circuit. While in practice it is 
better than the SPICE model indicated, these circuits do NOT 
provide a sufficently long enough pulse to effectively 
"double operating voltage". The over-voltage time is far too 
short with only 22uF. The capacitor value has to go way up 
to a few hundred uF to be effective on the RJ1 vacuum relays 
I tried.

Remember a closing transfer speed increase in a relay 
operating at full rated voltage is NOT the same as doubling 
effective supply voltage on pull in.

Adding the circuit does speed up **pull in**  on all the 
common T/R relays I tried, but not nearly as well as apply a 
constant current source that is sourced from a voltage a few 
times higher than relay rated voltage.

A typical open frame 15 amp contact relay increased speed 
almost 15%, which could be enough to make the difference 
suggested.

The problem is the relay speed-up system greatly slows the 
release time, almost doubling it. This is because all 
capacitor charging current comes through the relay. Contrary 
to what a few Internet articles say, loading a relay does 
**not** speed the release time up. See 
http://www.ifwtech.co.uk/g3sek/in-prac/0204.pdf

It does the opposite! The fastest release times come when 
the relay is simply open circuited. Even a back-pulse 
cancelling diode slows it down considerably, as does adding 
this circuit. The combination of a back pulse diode and this 
circuit is even worse than either alone!!

Also the circuit has a reset time. With 25 uF nominal 
capacitance and a typical relay it was 10mS to recharge to 
80% of full voltage. That means your boost is only at 80% of 
supply minus the drops in the transistor and two diodes with 
a 10ms reset time. If you have a 12V supply a directly 
connected relay would operate at 12 volts, but with this 
circuit the steady state operating voltage is about 1 volt 
less. The boost circuit peak voltage is about 1.5 to 2 volts 
less than 12 volts, so you add around 11 volts to 9.5 volts 
for a total of about 20.5  volts from a 12 V supply but only 
for a tiny fraction of a millisecond before voltage starts 
ramping down significantly. So it isn't as rosy as the 
articles all say.

It is probably an OK patch for some specific situations, 
like a specific relay in a slow VOX mode with sufficient 
reset time, but it really isn't a good idea overall. I would 
not depend on it to use a relay at half voltage without some 
major changes. It also is unacceptable for QSK or a fast 
"double trip" with normal VOX.

I'd just add an external 12V boost  supply, like a small 
wall wart, to the station 12V supply. Then you are done with 
it for ten bucks, and you don't have to cut gear up. For a 
slow amp relay in limited cases I'd use a current limited 
external boost supply. Then it always resets no matter how 
fast the VOX double trips and you don't slow release time 
down.

73 Tom