A long time ago, I saw a circuit using a silicon controlled rectifier
as a latching relay to control a bilge pump. It used two float switches
(one mounted higher than the other) and both had to be up to start the
bilge pump. The pump continued to run until the lower float switch
fell.

The advantages were that the back flow from the hoses did not cause the
pump to start again, and the low current demand allowed the switches to
last longer.

I cannot relocate the circuit,

Does this ring any bells? Anyone know where to find the diagram?

The only circuits I can come up with have the full pump current going
through the lower switch, unless I go to a mechanical relay.

"Offbreed" . wrote in message . A long time ago, I saw a circuit using a
silicon controlled rectifier
as a latching relay to control a bilge pump. It used two float switches
(one mounted higher than the other) and both had to be up to start the
bilge pump. The pump continued to run until the lower float switch
fell.

The advantages were that the back flow from the hoses did not cause the
pump to start again, and the low current demand allowed the switches to
last longer.

I cannot relocate the circuit,

Does this ring any bells? Anyone know where to find the diagram?

The only circuits I can come up with have the full pump current going
through the lower switch, unless I go to a mechanical relay.

Quickly, this isn't going to be any help, but once an SCR fires, (in a
DC circuit), it isn't easy to halt it's conduction without interrupting the
current flow by introducing another switching device.... are there trick
circuits that fool the SCR into turning off?
The other (non-helpful, I know) question is that if there is backflow
from the discharge hoses, shouldn't there be checkvalves to prevent seawater
from entering (especially during a moderate heeling turn?)
I'm wondering about the same pump control techniques, but this is at my
house when we have more than a little rain. Like this past week!
Old Chief Lynn

Lynn Coffelt wrote:

Quickly, this isn't going to be any help, but once an SCR fires, (in a
DC circuit), it isn't easy to halt it's conduction without interrupting the
current flow by introducing another switching device....

Well, yes, that can be a problem, but I think SCRs can be shut off by
changing the bias of the gate circuit. They have to have a minimum
current flow to work. Trouble is, all my semiconductor experience was
with transistors, diodes, and ICs, and I cannot find much about their DC
application on the net. (And whatever twit named a series of boats "SCR"
is begging for a kick in the shin.)

The other (non-helpful, I know) question is that if there is backflow
from the discharge hoses, shouldn't there be checkvalves to prevent seawater
from entering (especially during a moderate heeling turn?)

The check valve would have to be right at the pump, or close to it, to
keep the water flowing back out of the hose from floating the switch.

I'm wondering about the same pump control techniques, but this is at my
house when we have more than a little rain. Like this past week!

Should work fine in a sump, where the hose emptying back into the sump
would be even more a problem.

You are talking about a house, so you have a bit more room for spare
parts. Here's what I figure for a latching relay circuit (if you don't
need it, sure as anything someone else is going to ask):

Need:
Pump, relay with at least two normally open switches, two float
switches, bunch of wire.

Radio Shack has at least one 12VDC relay, and at least one 120V relay
that will work. Watch out, they look exactly the same and you have to
check the label on the package. (They are DPDT.)

Place the upper float switch in parallel with one of the normally open
switches. One side of them goes to one side of the power, the other goes
to one side of the relay coil. The other side of the relay coil goes to
the lower switch, and the other side of the lower switch goes to the
other side of the power supply from the first set of switches.

The remaining normally open switch controls the pump.

I think the Radio Shack relay is a generic automotive module, and I know
I've seen bases for it around somewhere, so you could solder the float
switches, power supply, etc to the base and just plug in a replacement
relay when the old one goes bad.

Of course, it would be handy to add a third float switch even higher
than either of the first two, and wire it to an "OH,****" alarm G.

Check out this link for an explanation of how an SCR works.


http://www.allaboutcircuits.com/vol_3/chpt_7/5.html
The Silicon-Controlled Rectifier (SCR) - Chapter 7: THYRISTORS - Volume
III - Semiconductors




Chuck



Offbreed wrote:

Well, yes, that can be a problem, but I think SCRs can be shut off by
changing the bias of the gate circuit. ...

chuck wrote:
Check out this link for an explanation of how an SCR works.


http://www.allaboutcircuits.com/vol_3/chpt_7/5.html
The Silicon-Controlled Rectifier (SCR) - Chapter 7: THYRISTORS - Volume
III - Semiconductors

Good page. Thanks. I had class time on SCRs, but never messed with them
in the shop and that page brings it back very nicely.

Valid summary?: The only ways to shut down the SCR involve reducing the
current flow through the SCR, and the simplest is to have the shut off
switch shut off the full current and accept the reduced life on that
switch.

I suppose momentary short around the SCR, or a pulse, would also work,
but I don't see any simple and reliable way to do that. None that don't
make me cringe. Another reason to get the extra switch and make a high
water alarm.

In article ,
Offbreed wrote:

Well, yes, that can be a problem, but I think SCRs can be shut off by
changing the bias of the gate circuit.

I don't think the above is the case for SCR's. Once triggered into
conduction, they stay in conduction untill the current flow thru the
device is interupted external to the device. All the SCR's that I
have ever used seem to work this way. You mileage may vary.....

Me

Me wrote:
In article ,
Offbreed wrote:

Well, yes, that can be a problem, but I think SCRs can be shut off by
changing the bias of the gate circuit.

I don't think the above is the case for SCR's. Once triggered into
conduction, they stay in conduction untill the current flow thru the
device is interupted external to the device. All the SCR's that I
have ever used seem to work this way. You mileage may vary.....

I misread what someone wrote about a home brew phone circuit, and got
the gate current mixed with the holding current. The holding current in
his circuit was not enough to keep the SCR conducting so he had to keep
a bias on the gate, with the first SCR he tried, one I have on hand, so
I thought I might be able to use it on the pump.

My error.

"Offbreed" wrote in
oups.com:

I cannot relocate the circuit,


It's quite easy, actually.....

SCR cathode to 12V negative bus

SCR gate to 10K ohm 1/4 watt resistor to limit its gate current to a
safe level. The other side of the resistor hooks to the higher float
switch, whos other lead is hooked to +12V. The resistor limits gate
current to 12/10000 = 1.2ma. The anode of the SCR is hooked to one lead
of the lower float switch. The other lead of the lower float switch is
hooked to the negative lead of the pump motor. The positive lead of the
pump motor is hooked through the appropriate inline fuse directly to the
house battery + post to prevent some idiot from turning off the bilge
pump! NO BILGE PUMP SWITCHES PLEASE!

As the water rises, the lower switch closes, powering up the SCR anode,
but without a gate + the SCR doesn't fire. As the water flips the upper
switch, the +12V applied to the resistor brings the gate current up
above the microamp-level firing current and the SCR gates on. Once
fired in a DC circuit, the SCR will not open just because the upper
switch opens and it loses its gate current. Current runs the pump until
the water level in the bilge drops to open the lower switch which shuts
down the pump, and ungates the SCR because it has no Anode current, now.

The circuit won't cycle, again, until the lower, then the upper switches
both close to gate the SCR, again, as the cycle repeats at a much slower
rate.

Use a powertab 8A SCR and just screw it into the wood somewhere way
above the waterline, even out of the bilge if you can to prevent
corrosion. Run a gate lead out of the bilge to the 10K resistor
soldered directly to the SCR gate lead to support it. As this gate lead
will be hooked directly to the 12V plus, make sure the power to the
upper switch hooks to the motor side of the bilge pump fuse for safety.

Now you have a true hysteresis bilge pump. AT 4A, the SCR barely gets
warm

In an emergency, just jumper out the SCR and you have a simple bilge
pump with the lower switch.

Larry wrote:
"Offbreed" wrote in
oups.com:

I cannot relocate the circuit,


It's quite easy, actually.....

SCR cathode to 12V negative bus

SCR gate to 10K ohm 1/4 watt resistor to limit its gate current to a
safe level. The other side of the resistor hooks to the higher float
switch, whos other lead is hooked to +12V. The resistor limits gate
current to 12/10000 = 1.2ma. The anode of the SCR is hooked to one lead
of the lower float switch. The other lead of the lower float switch is
hooked to the negative lead of the pump motor.

Here you have the entire load going through the lower float switch, and
I was hoping to avoid that. One of the advantages of the original
circuit was that the lower switch was not exposed to the sparking that
would cause.

The positive lead of the
pump motor is hooked through the appropriate inline fuse directly to the
house battery + post to prevent some idiot from turning off the bilge
pump! NO BILGE PUMP SWITCHES PLEASE!

As the water rises, the lower switch closes, powering up the SCR anode,
but without a gate + the SCR doesn't fire. As the water flips the upper
switch, the +12V applied to the resistor brings the gate current up
above the microamp-level firing current and the SCR gates on. Once
fired in a DC circuit, the SCR will not open just because the upper
switch opens and it loses its gate current. Current runs the pump until
the water level in the bilge drops to open the lower switch which shuts
down the pump, and ungates the SCR because it has no Anode current, now.

The circuit won't cycle, again, until the lower, then the upper switches
both close to gate the SCR, again, as the cycle repeats at a much slower
rate.

Use a powertab 8A SCR and just screw it into the wood somewhere way
above the waterline, even out of the bilge if you can to prevent
corrosion. Run a gate lead out of the bilge to the 10K resistor
soldered directly to the SCR gate lead to support it. As this gate lead
will be hooked directly to the 12V plus, make sure the power to the
upper switch hooks to the motor side of the bilge pump fuse for safety.

Now you have a true hysteresis bilge pump. AT 4A, the SCR barely gets
warm

In an emergency, just jumper out the SCR and you have a simple bilge
pump with the lower switch.

Okay, many thanks. I added the technical part of what you added in the
correction to make it a single post (that part about the DC motor
shutting it's self off would have driven me nuts). :

Oops....I forgot to mention a problem with DC motor driving with this
latch. DC commutator motors don't draw current through the whole
rotation, just when the commutators make contact. So, you'll be feeding
pulsating DC to the SCR, which WILL ASSUREDLY make it unlatch! The cure
to the problem is to put a 100 ohm, 5 watt resistor right across the
motor, so that when the motor commutator opens, the SCR has a holding
current through the 100 ohm resistor to keep it latched, insuring a
steady flow of holding current until the lower switch opens.



Eh, well, sexy voice, sexy body, and sense of humor are a combination to
be treasured. Your friend was very lucky.

Offbreed wrote in news:3MqdnaLc4IVazCLeRVn-
:

Here you have the entire load going through the lower float switch, and
I was hoping to avoid that. One of the advantages of the original
circuit was that the lower switch was not exposed to the sparking that
would cause.



Won't matter. You'll use a Rule float switch that has a mercury switch in
it so it won't ever pit up or corrode. Mine never has....

There's not that much current, only 4-5A, anyways.