OpenBCM V1.06 (Linux)

Packet Radio Mailbox


[C.D.G. OpenBCM]

 Login: GUEST

G8MNY  > TECH     14.02.19 08:00l 153 Lines 6861 Bytes #8 (0) @ WW
BID : 59090_GB7CIP
Subj: Simple SCR Regulated Charger
Sent: 190214/0659Z @:GB7CIP.#32.GBR.EURO #:59090 [Caterham Surrey GBR]

By G8MNY                                        (Updated May 13)
(8 Bit ASCII graphics use code page 437 or 850, Terminal Font)

This is a simple self regulating SCR controlled charger. It was made to give
27.5V at 15A for charging a train carriage's 500 AH 24V battery system. It is
not suitable for use with electronic equipment that can't operate on hummy
supplies! Or due to possible RFI used near Radio Rx equipment!

Using a dual winding transformer (1) or a single one with 2 large diodes so the
SCRs form part of a bridge (2), or (3) a single SCR arm following a bridge
(1)                    (2)                          (3)
___ ___    >AC1    ___ ___    _>AC1    ___ ___    _͵>>DC
L     )||(            L     )||(~͵<ͻ         L     )||(~       
60Trip)||(__          60Trip)||(                60Trip)||( ͵<ͻ 
_______)||(~~         _______)||(_͵<͹         _______)||(_       
N       ||(           N          ~)>AC2    N          ~͵>)ͼ
           )>AC2               thick                            
            >-ve                wire  >-ve                ͵<>-ve

The transformer must deliver an RMS voltage slightly greater than the battery
voltage plus diode & wiring losses. If voltage too low, triggering the SCRs can
be difficult. Tn that case use an additional 2V winding put around a toroidal
or leg of the transformer to feed the trigger circuit can keep the unit working
well on low mains (the train PSU had to deal with mains of less than 200V AC at

This circuit is very efficient, only the transformer heats up with the high
pulse current, & some SCR/diode losses, there is no huge power loss of a
dropper/linear type regulator!

For high current battery work fusing can just be "weak" wire links mounted in a
safe way not to cause fires etc, but anything to do with batteries needs some
fuse protection!

For recharging 12V batteries use 14.5V DC setting, for floating (infinite
charge rate) use 13.8V, with sealed/gel lead acid types use 13.5V MAX!

                 \ SCR1
AC1 ͵ >ͻ
/'\,/           / \.          thick   
or DC Ĵ>470ij>Ŀ  wires    With (3) DC feed the AC2 arm &
        D1  1W                 for high  D1 are not needed.
                 \ SCR2      currents 
AC2 ͵ >)͹
\,/'\           / \.                 o
      Ĵ>470Ĵ>Ĵ          / CHARGE
            1W               __         o FUSE
                      LOW    /_\`13.1V   o/o LOADS
                      oĴ   1W           FUSE
                             __       ____+
          SET VOLTS   NORM   \_/        
                > oĴ           BATTERY
-ve return           oBOOST \ /         
   ////   Make LOW = 13.1V, NORM = 13.8V, & BOOST = 14.6V.

The SCRs gates are actually pulse driven as they only see current for a short
time when the battery voltage is below the zener voltage, until the SCR fires
(turns on). Then the AC voltage input will be reduced to that of the battery
turning off the gate drive. This is important as SCR gate dissipation is quite
limited, even on large current SCRs. The gate Rs (470) should experimented with
to make sure there is good triggering for your SCRs, a general rule is to make
them 1/2 to 1/4 of the value that just triggers the gate OK.

In practice this circuit generally produces a sort of phase firing, as the
battery voltage ends up as a saw tooth dropping blow the zener voltage some
time in the next half cycle. But if the SCRs fire unevenly sometimes due to
inductance problems in the transformer, severe transformer heating can result
due to DC satuation of the core, resulting in a very low inductance across the
mains. So a "resetting thermal trip" is recommended on the transformer, just in
case it gets unduly hot.

The output voltage can be adjusted in 0.6V steps by the use of diodes in series
with the zener.

The Zener can be a virtual one for voltage adjustment...

                   \    POT
            +   NPN >10k
     100uF ===     e/     
VE ret 

The Zener can be 510V (12V system). The transistors is an NPN. Set the voltage
pot so that the Zener action is 1V higher than the required Battery voltage.
The transistor can be either an NPN as here, or the circuit can be inverted & a
PNP used.

This can be added to either zener circuit..

           NPN  Ŀ   Zener
              e/ +47uF      or
                 ===       variable
                   1k   Zener
 ve /\/\/\> Batt -ve

The transistor is a NPN. Make the shunt resistor (thick wire link e.g. 10-30cm
of coat hanger wire bolted in place as it will get hot!) so at the maximum
current wanted it drops 0.6V, enough to make the NPN conduct, & hence reduce
the zener voltage & thereby limit the current. The 1k & 47uF filtering is
needed so that the average current is controlled not the high peak current.

If the output is shorted this current limit may not be very effective depending
on the trigger threshold of the SCR etc. but it will reduce the otherwise very
high current!

Note: if the shunt is put on the supply side of zener, the voltage loss across
it will not affect the output voltage, but if it is put on the battery side it
will be reduced.

The Meter can be almost anything from 100uA to 1A movement for current
monitoring, provided it needs less than 0.6V drop, using the same current
limiter shunt. If a sensitive meter is used a switch can be provided to measure
either voltage or current.

AC/\/\/\>Batt -ve  
-ve            SHUNT          
             I     /'~`\ +   oI    V
          RCalĴMETER>oRCalZenerR>Batt +ve
                    \._./       V

Leaving the I RCal preset in line on voltage measurement will produce a small
error, but simplifies wiring. Calibrate the V & I meter scales with external
meter, adjust suitable presets for full scale. In the voltage case an
additional fixed series R (e.g. 1k) can stop meter accidents! Also a 10V zener
in series can give a more useful offset scale e.g. 10-15V.

Why Don't U send an interesting bul?

73 De John, G8MNY @ GB7CIP

Read previous mail | Read next mail

 17.02.2019 01:27:38lGo Back Go up