0
\$\begingroup\$

I have some questions about a circuit I'm using for an upcoming art piece.

I've been using the following circuit schematic that discharges current from a wound coil every time a magnet pendulum swings above the coil to keep it in 'perpetual' oscillation. It works just fine as it is and it pulls about 100 mA for the short duration of the discharge pulse.

I wanted to make around 50 of these circuits on the same power supply that are each 'enabled' using simple voltage controlled switches via an Arduino (5 V logic signals.) I've designed a simple low-side (edit: high-side) switch using an NPN and PNP transistor in order to be able to control the 9V (VCC) source from the 5V logic signal.

While testing this with two circuits (see diagram) and two Arduino logic signals (one for EN1 and one for EN2,) it works as intended as long as one circuit is disabled and another one is enabled.

When both are enabled, magnet swings over either coil will trigger coil discharges (LEDs light up) across both circuits. I think intuitively it makes sense because when both switches are open, current is pulled from the 9V source to both of the circuits since they're in parallel and it doesn't know how to discriminate .

enter image description here

I'm trying to figure out a way to isolate the circuits from one another in order to use the same source. My plan B is to use separate voltage sources (batteries) for each circuit but I'd much rather be able to use just a few large power supplies that can be shared throughout the 50 circuits.

Does anybody see any glaring errors or misunderstandings? Does anybody have any advice on what I might be able to do to?

\$\endgroup\$
8
  • 1
    \$\begingroup\$ 9V batteries aren't particularly stable. What happens if you add a large (4700µF or more) capacitor to the 9V supply? You can use multiple smaller ones too if you don't have such a big capacitor. \$\endgroup\$ Dec 8, 2021 at 12:20
  • \$\begingroup\$ Hey Thanks for your comment! Yeah, I've noticed that when hookup up the batteries, over time they start to behave pretty inconsistently and I have to reset the potentiometers to tune it in. also they have very bad mAh specs and won't last that long. Will the large cap in parallel with the 9V supply isolate the circuits so that they don't trigger each other? \$\endgroup\$
    – melonhead
    Dec 8, 2021 at 12:29
  • \$\begingroup\$ The large capacitor(s) will be able to buffer the current surges from the individual circuits, preventing the battery voltage from dropping momentarily. It may or may not be enough to prevent false triggering of your circuits; you'll have to try it out. Make sure that each circuit is attached to the capacitor via its own wires, your circuits should not share the same supply wires as that will also create voltage drop. Thicker wires may help too. \$\endgroup\$ Dec 8, 2021 at 12:34
  • \$\begingroup\$ I can't see any low side switching - it all seems to be high side. As well, the leds might like the reverse voltage pulse when the current is interrupted. If there is more than -5V, the led will die over time. \$\endgroup\$
    – Kartman
    Dec 8, 2021 at 12:53
  • \$\begingroup\$ @JonathanS. ok just tried that and it actually seems to be working! I'm trying to understand the physics of what's going on here. So when a load demands a large current in a short amount of time, power supplies have a hard time supplying it and keeping the voltage steady so sometimes the voltage will drop momentarily. This voltage drop at the source will cause a current to be output into the loads (in this case any circuit that is switched on)? \$\endgroup\$
    – melonhead
    Dec 8, 2021 at 13:02

2 Answers 2

1
\$\begingroup\$

You need to limit the base current of the PNPs. Add a R in series with the base - perhaps 470 Ω.

\$\endgroup\$
1
  • \$\begingroup\$ Hey thanks for the suggestions - ive added a current limiting resistor :) \$\endgroup\$
    – melonhead
    Dec 8, 2021 at 23:02
0
\$\begingroup\$

Your circuits draw a lot of current when they operate, in fact the current consumption is not really limited by anything (apart from the transistors' beta). 9V batteries are quite bad at delivering large currents; their voltage drops dramatically when they have to. This voltage drop influences all the other circuits connected to the same battery.

Capacitors, on the other hand, are very good at delivering large current pulses for a very short amount of time (but they don't store much energy). However, because your circuit only operates for a very short amount of time, the capacitor is able to supply the necessary energy for that short pulse of power consumption. This means that the battery doesn't have to supply these pulses anymore so that it's voltage doesn't drop as dramatically. That, in turn, eliminates the possibility of the circuits influencing each other.

Keep in mind though that the circuit, as you've drawn it, has absolutely no protection against overcurrent and probably operates its transistors with more current than they're specified to withstand. In other words, it might randomly fry itself, in particular transistor Q1. You could try adding a 22 Ohm resistor in series with Q1's emitter. That should also lower the power that's being wasted in the circuit, extending battery life. (If 22 Ohm doesn't work, try 10 Ohm.)

Additionally, LED1 might get damaged by excessive reverse voltage as Kartman noted. To prevent that, you can add a regular silicon diode (i.e. 1N4148) in series with the LED. (LEDs only light in one direction; in the other direction they block current from flowing. However, they're only able to block a certain amount of reverse voltage and 9V might be too much. Regular silicon diodes are much better at blocking reverse voltage.)

\$\endgroup\$
1
  • \$\begingroup\$ Hey many thanks for your comments. I did indeed make the additions and tested them to make sure they work for 3 of the circuits. Also I'm planning on using a few large 9V sources so that i don't have to have 50 9V batteries \$\endgroup\$
    – melonhead
    Dec 8, 2021 at 23:01

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.