Hi all...
I am currently using a homebrew 35 amp power supply that has a 50 amp
blocking diode in series with the B+ line going to a lead acid deep cycle
battery to feed two hf rigs and one vhf/uhf rig.

The power supply has no real current limiting and I want to insert a bit of
resistance in the B+ line to keep the current from exceeding 8 amps or so on
battery charge up. Can anyone give me an idea of what value would be
appropriate.

I am also thinking of replacing the deep cycle battery with a sealed
battery...one of those that is advertised to never need fluids, won't leak,
etc....are these appropriate for this application?

Thanks in advance....de Denton WB7TDG

In article , "denton"
writes:

Hi all...
I am currently using a homebrew 35 amp power supply that has a 50 amp
blocking diode in series with the B+ line going to a lead acid deep cycle
battery to feed two hf rigs and one vhf/uhf rig.

The power supply has no real current limiting and I want to insert a bit of
resistance in the B+ line to keep the current from exceeding 8 amps or so on
battery charge up. Can anyone give me an idea of what value would be
appropriate.

If you want to limit the charging current to a maximum, then you
need a current-limiting regulator circuit with some form of current
sensing in series with the battery line.

A series "resistor" can be made from standard magnet wire with the
value of resistance taken proportional to the resistance per whatever
length is stated. Copper draw dies and general manufacturing of
magnet wire, plus slight changes in dimension due to winding it all
up on some kind of form, allow about a 10% tolerance on such a
"resistor." A very small value resistance is hard to measure in the
home workshop environment so a length of magnet wire is a
reasonable substitute.

Any common op-amp can be used to sense the voltage drop across
a milliOhm resistance (the magnet wire "resistor"). The voltage drop
will be in the milliVolt range but the op-amp can amplify that to drive
a meter or whatever. If put in the return lead rather than the hot lead,
you avoid most of the "swing" in voltage possible in the hot lead. With
only milliVolt drops in the sensing resistor, there isn't a great deal of
upset with to the charging voltage. That sort of current measurement
can be used both for visual indication of charge current and for coupling
to the current regulating circuit.

["B+" is the very old term for plate voltage supply for tubes, ain't really
appropriate for these new-fangled contraptions known as
semiconductors...:-) ]

If the charging circuit is basically a constant-current supply (which can
be set), then the output voltage will be set by the battery being charged.
The charging can be halted by limiting the maximum voltage supplied
to the battery(s). That's the opposite of what is normally done; i.e., a
constant-voltage supply with a high current limit. Just as feasible in
practice to have either constant-voltage or constant-current. Some of
the older power supply manufacturer's data will explain various
regulators' configurations. I used an old Kepco softcover book back
in the late 1960s as a model for a "universal" bench supply.

It isn't clear how you have the arrangement of charger, rigs, and
blocking diode(s). There's all sorts of variations possible on those.
Any blocking diode will always have some voltage drop across it
in forward conduction. For a silicon diode that is in the neighborhood
of 0.6 VDC; at 8 A that is 4.8 W of loss across the diode. That will
also affect the charge voltage. If you "float" the charger across the
battery, then you have to account for that and keep the radio primary
input current from affecting the charger.

You aren't accomplishing much but just sticking a resistor in series
with a charger supply. If the series resistance is high you create a
quasi-constant-current supply with a wide variation of output voltage
(called "compliance" in constant-current circuits). If the series
resistance is too low, there's little compliance and voltage is basically
that of the basic charger rectifier and capacitor.

I am also thinking of replacing the deep cycle battery with a sealed
battery...one of those that is advertised to never need fluids, won't leak,
etc....are these appropriate for this application?

Those should work fine - provided - you stay within battery
manufacturer's ratings.

Uninterruptible power supplies exist in the tens of thousands out
there in cyberland with little problem. Their internal batteries are
sealed and the chargers float across the battery regardless of the
load being on or off.

Len Anderson
retired (from regular hours) electronic engineer person

In article , "denton"
writes:

Hi all...
I am currently using a homebrew 35 amp power supply that has a 50 amp
blocking diode in series with the B+ line going to a lead acid deep cycle
battery to feed two hf rigs and one vhf/uhf rig.

The power supply has no real current limiting and I want to insert a bit of
resistance in the B+ line to keep the current from exceeding 8 amps or so on
battery charge up. Can anyone give me an idea of what value would be
appropriate.

If you want to limit the charging current to a maximum, then you
need a current-limiting regulator circuit with some form of current
sensing in series with the battery line.

A series "resistor" can be made from standard magnet wire with the
value of resistance taken proportional to the resistance per whatever
length is stated. Copper draw dies and general manufacturing of
magnet wire, plus slight changes in dimension due to winding it all
up on some kind of form, allow about a 10% tolerance on such a
"resistor." A very small value resistance is hard to measure in the
home workshop environment so a length of magnet wire is a
reasonable substitute.

Any common op-amp can be used to sense the voltage drop across
a milliOhm resistance (the magnet wire "resistor"). The voltage drop
will be in the milliVolt range but the op-amp can amplify that to drive
a meter or whatever. If put in the return lead rather than the hot lead,
you avoid most of the "swing" in voltage possible in the hot lead. With
only milliVolt drops in the sensing resistor, there isn't a great deal of
upset with to the charging voltage. That sort of current measurement
can be used both for visual indication of charge current and for coupling
to the current regulating circuit.

["B+" is the very old term for plate voltage supply for tubes, ain't really
appropriate for these new-fangled contraptions known as
semiconductors...:-) ]

If the charging circuit is basically a constant-current supply (which can
be set), then the output voltage will be set by the battery being charged.
The charging can be halted by limiting the maximum voltage supplied
to the battery(s). That's the opposite of what is normally done; i.e., a
constant-voltage supply with a high current limit. Just as feasible in
practice to have either constant-voltage or constant-current. Some of
the older power supply manufacturer's data will explain various
regulators' configurations. I used an old Kepco softcover book back
in the late 1960s as a model for a "universal" bench supply.

It isn't clear how you have the arrangement of charger, rigs, and
blocking diode(s). There's all sorts of variations possible on those.
Any blocking diode will always have some voltage drop across it
in forward conduction. For a silicon diode that is in the neighborhood
of 0.6 VDC; at 8 A that is 4.8 W of loss across the diode. That will
also affect the charge voltage. If you "float" the charger across the
battery, then you have to account for that and keep the radio primary
input current from affecting the charger.

You aren't accomplishing much but just sticking a resistor in series
with a charger supply. If the series resistance is high you create a
quasi-constant-current supply with a wide variation of output voltage
(called "compliance" in constant-current circuits). If the series
resistance is too low, there's little compliance and voltage is basically
that of the basic charger rectifier and capacitor.

I am also thinking of replacing the deep cycle battery with a sealed
battery...one of those that is advertised to never need fluids, won't leak,
etc....are these appropriate for this application?

Those should work fine - provided - you stay within battery
manufacturer's ratings.

Uninterruptible power supplies exist in the tens of thousands out
there in cyberland with little problem. Their internal batteries are
sealed and the chargers float across the battery regardless of the
load being on or off.

Len Anderson
retired (from regular hours) electronic engineer person

Use ohm's law. The voltage drop across the resistor will be the supply
voltage (13.8) minus the discharged battery voltage (10 volts) or 3.8
volts. R = E / I so 3.8 / 8 = 0.475 ohms. So a 0.5 ohm resistor. The
power disipated is I*I*R or 8*8*0.5 = 32W.

The battery voltage will quickly climb and the charging current will
drop accordingly so the battery takes longer to charge this way but it
should be fine if you normally "float" it across the supply to keep it
charged *and* it will protect the battery and the supply from over
current when it is discharged.

Use ohm's law. The voltage drop across the resistor will be the supply
voltage (13.8) minus the discharged battery voltage (10 volts) or 3.8
volts. R = E / I so 3.8 / 8 = 0.475 ohms. So a 0.5 ohm resistor. The
power disipated is I*I*R or 8*8*0.5 = 32W.

The battery voltage will quickly climb and the charging current will
drop accordingly so the battery takes longer to charge this way but it
should be fine if you normally "float" it across the supply to keep it
charged *and* it will protect the battery and the supply from over
current when it is discharged.