# switching circuit with multiple power sources

Discussion in 'Electronic Basics' started by Jim Alexander, Aug 1, 2005.

1. ### Jim AlexanderGuest

I have three digital cameras on which I need to remotely trigger the
shutters simultaneously. Triggering a single shutter is normally done
with a remote that simply shorts two pins in the remote cameras connector
together (the shutter pin gets shorted to ground; there's actually also
an autofocus trigger pin, but a circuit that works for the shutter should
also work for the autofocus trigger).

I don't want to risk just connecting all of these cameras in parallel
directly to a single switch since I don't know anything about the
internal electronics except what I can read off a multimeter, and so
I'm not at all sure that shorting their internal power sources together
is safe (and the cameras a very expensive, so I can't just try it and
hope for the best).

So my thought was to connect each pair of terminals to its own NPN transistor,
and connect all three base terminals together, and connect the bases to yet
another power source (probably consisting of a battery and a resistor) through
a switch. Pressing the switch would allow current to flow into the base
drive all three transistors to saturation, current would flow across the remote
terminals, and a picture would be taken. Schematically, it looks something
like this:

/-- shutter 1 (V1)
--------- T1
| \-- ground 1
|
/ | /-- shutter 2 (V2)
----Rb---------/ ----+-------- T2
| | | \-- ground 2
---bat---- |
|A /-- shutter 3 (V3)
--------- T3
\-- ground 3

Now my basic semiconductor electronics knowledge is pretty rusty. I know
how to pick the right base resistor when I am trying to drive a single
transistor to saturation, but it seems to me I will need more current
to get all three transistors saturated. The transistors are not really
connected in series or parallel since their E and C are connected to
3 independent power supplies, and their B terminal gets fed by a fourth
power supply. Is it even kosher to not have the base and emitter connected
to seperate supplies? Every switching circuit I've ever built used a
single supply, but I don't know whether this is a necessity.

Can anyone tell me if there are any serious problems in how I am trying to
accomplish my goal? Or can anyone suggest a better way? Note that one
reason I want to use transistors is I want the cameras triggered as close
to possible to simultaneously, so switching needs to be fast. Also,
the resistance across the switch needs to be negligible, so I don't think
a 4066 IC would work for me. Any ideas would be appreciated.

2. ### Dan HollandsGuest

It seems to me a much simpler solution is to just use a 3 pole switch or
relay

--

Dan Hollands
1120 S Creek Dr
Webster NY 14580
585-872-2606

www.QuickScoreRace.com

3. ### petrus bitbyterGuest

Jim,

You don't want to connect the switches in parallel but this way you do not
fuly separate them either. More important, the resistance of a fully
saturated transistor cannot be neglected. Usually it's higher then the
onresistance of a 4066. Nevertheless I think a 4066 will do fine. I'd give
it a try.

As for the current required to drive a transistor into saturation: Three
transistors require three times the current that one requires. This current
depends highly on the specifications of the transistor involved. You will
need the datasheet to find out. You'll also have to find out the plus and
minus of each camera. Tie the minusses to each other (including the minus of
your battery) and tie the plusses to the collectors of the transistors.

cam1 cam2 cam3
| | |
| | |
| | |
| | |
___ |/ ___ |/ ___ |/
+-|___|--| +-|___|--| +-|___|--|
| |> | |> | |>
| | | | | |
minus----------------+-----|----------+-----|----------+
_/ | | |
+bat-o/ o-+----------------+----------------+

created by Andy´s ASCII-Circuit v1.24.140803 Beta www.tech-chat.de

If you really need hard contacts, you can try three (reed)relay, one relay
with three NO contacts or a simple one throw, three pole switch.

petrus bitbyter

4. ### petrus bitbyterGuest

Jim,

You don't want to connect the switches in parallel but this way you do not
fuly separate them either. More important, the resistance of a fully
saturated transistor cannot be neglected. Usually it's higher then the
onresistance of a 4066. Nevertheless I think a 4066 will do fine. I'd give
it a try.

As for the current required to drive a transistor into saturation: Three
transistors require three times the current that one requires. This current
depends highly on the specifications of the transistor involved. You will
need the datasheet to find out. You'll also have to find out the plus and
minus of each camera. Tie the minusses to each other (including the minus of
your battery) and tie the plusses to the collectors of the transistors.

cam1 cam2 cam3
| | |
| | |
| | |
| | |
___ |/ ___ |/ ___ |/
+-|___|--| +-|___|--| +-|___|--|
| |> | |> | |>
| | | | | |
minus----------------+-----|----------+-----|----------+
_/ | | |
+bat-o/ o-+----------------+----------------+

created by Andy´s ASCII-Circuit v1.24.140803 Beta www.tech-chat.de

If you really need hard contacts, you can try three (reed)relay, one relay
with three NO contacts or a simple one throw, three pole switch.

petrus bitbyter

5. ### Jim AlexanderGuest

[ my reply was accidentally mailed rather than posted - sorry! ]

]It seems to me a much simpler solution is to just use a 3 pole switch or
]relay

I should have said this needs to be a momentary push-button switch. I
don't think such switches are made to guarantee simultaneous (say within
1 ms) contact across all of the poles. If you press the switch off
center, for instance, the contacts on the side you are pressing on will get
connected first. How much a delay this causes depends on the construction
of the switch and how hard you are pressing.

I am also considering driving the switches with the output of a timer chip
so that I can control precisely how long the terminals are shorted. The
cameras can be set to take multiple exposures if the shutter switch is
held down. That's a lot more difficult to accomplish using a mechanical
switch.

As for relays, electromechanical relays aren't fast enough, and reed or
solid-state relays with multiple poles don't seem to be very easy to
come by (and are certainly going to be a lot more expensive than low-power
transistors).

6. ### Bob MonsenGuest

Sorry, your picture isn't going to work. For the transistors to be
turned on, a voltage must be applied between the base and emitter. Thus,
the grounds all need to be connected together. In your picture, the
transistors probably won't turn on, because the camera ground will move
wrt the base of the transistors.

Thus, you need to connect up ground 1, 2, and 3 to make it work
properly. You also need to connect the - terminal of the battery to that
same ground. Then, when you apply the voltage to the base, the
transistors will turn on for you.

If you are afraid of ESD, for example, and don't want to connect the
grounds and possibly mess up the electronics in the cameras, one way
would be to use optoisolators. For example, a 4Nxx series optoisolator
could be used, one per camera, with a reasonably large current (maybe
20mA) pushed through all the LEDs at once, in series, with the button.
That would give you a (nearly) simultaneous activation of all of the
output transistors, but no danger from ESD or ground loops.

http://www.fairchildsemi.com/pf/4N/4N25.html

They are pretty cheap.

--
Regards,
Bob Monsen

If a little knowledge is dangerous, where is the man who has
so much as to be out of danger?
Thomas Henry Huxley, 1877

7. ### Rich GriseGuest

....

determined that a transistor switch will do the job, then use three
optoisolators. You drive the 3 leds in series, and each one sits
at its own camera and the transistor turns on and shorts (closes)
the switch.

Good Luck!
Rich

8. ### John FieldsGuest

---
I just posted a circuit for you on abse under the same subject, but
it has an error which I'll fix this afternoon. Basically, what
happens now is that in CONTINUOUS mode the outputs to the cameras
arent de-skewed. The fix will have them de-skewed in both MOMENTARY
and CONTINUOUS modes.

9. ### quietguyGuest

I would have thought using 3x optocouplers driven by a pulse generator would be
the way to go

David