So, I'm working on a simple circuit that seems to be working so far, but is different enough from my past experience that I want to make sure I'm not doing anything wrong.

I've got an a-stable 555 timer, adjustable around 6-40 Hz, whose signal I want to connect to three separate outputs, and I want those three to be individually activated with 5v signals.

I've hooked the switchable activation signals (5VDC) to the bases of three nte2361 npn transistors. I've hooked the shared 555 output to the emitters of all 3 transistors, and I've hooked the three output connections to the collectors. The outputs connect to a separate circuit which supplies 3.3V on those lines and reads signal when they are grounded out. So, any time a 5V activation line is on at the base and the 555 signal is low (ground) the transistor is active and connects the output to ground through the 555.

In theory this all seems right to me, but it seems like I always see transistors controlled by their base, not their emitter. In this case, I'm kind of using both, but the emitter is the one changing constantly. Is there a reason not to do this, or is this fairly common and it's just my inexperience that makes it look strange?

Also, if I've got 3.3V at the collector and the emitter cycles between ground and 5V, will the reverse voltage condition created when the emitter is at 5V be a problem for the transistor? I was considering putting a diode between the 555 and the emitters to prevent current from going the wrong direction, but I'm not sure it's necessary, or even that it would do what I'm thinking in this scenario.

Thanks,
Eric

 

I just saw one like this today or yesterday. Yes, you can switch a transistor from the emitter. Think about a cascode amplifier.

The typical reverse voltage of an emitter to base is about 5 volts, so it won't run backwards on 1.7 volts. You're OK. Check your transistor datasheet to find the exact reverse breakdown voltage from emitter to base.

Of course, if you provided a drawing, I would be more sure.

 

What do you expect the transistor to do? Why is it there?
You are switching the emitter between 0 and 5 volts but holding the collector at 3.3 volts?
What is the transistor base connected to?
Where are the resistors?
So far, I don't understand why a diode couldn't do what you want in place of the transistor if that is what you are doing.

 

The transistor is intended to provide a ground path for the 3.3V signal only when:
1) the external 5V activation signal is present AND
2) the 555 output is low

This circuit is designed to simulate the open collector output of 3 flowmeters, but only to simulate flow on any given flowmeter when the corresponding water valve output is active. It's a way to test some water control electronics without needing a flowmeter and plumbing to all be connected. With this circuit I should be able to adjust the 555 for a reasonable range of simulated flow rates, and then have it simulate the electrical signal that the external board would see with water flow, only when the external board has called for a valve to open and expects to see water flow.

The 3.3V signal from the other board reads 3.3V when open. If you ground out the signal, the external board recognizes that as a flowmeter pulse, and an almost immeasurably small amount of current flows to ground.

Here's the schematic I should've have posted up front (except I hadn't drawn it yet!) Sorry for the delay. Also, I forgot to update part numbers in this schematic. The transistors are the ones described above, not the 2N2222 listed below:

 

Each of your three transistors is essentially the top of a two-transistor AND gate where your 555 acts as the bottom transistor (for all three AND gates at the same time).

In your case, it would be known as a two transistor AND gate with an OPEN collector. In your case, the OUTPUT connects/disconnect ground to a load that is connected to Vcc.

However, I would still like to see what you have between your transistor and your 3.3v source. You should check that your little 555 timer can sink the load coming from the three 3.3v collector connections.

 

Load shouldn't be a problem. Hooked amp meter in series from 3.3v source to ground and the current is too low for me to accurately measure, less than 1 uA. Don't know if it's a regulated current source or maybe a digital pin with a very high value pull up resistor, but a dead short on those 3.3v lines yields almost nothing.

Having said that, better safe than sorry and I may want to use this circuit for other things sometimes, so I'm thinking maybe I'll add 150 ohm, 1/4w resistor on each [EDIT: ouput, where the 3.3v supplies are.] Should be just right if I connect an led with 5v on it, and would also limit an unloaded 5v to 33mA per transistor, 100mA max on the 555. I'm using the regular 555, not CMOS, and it's rated for around 200mA, so I think that would cover my bases. Pretty much always dealing with 5v, or occasionally 3.3. Sound right, or am I missing something?

Thanks for the clarification on AND circuits. Feel much better knowing this isn't something way off base I'm attempting. Still super new with solid state components and trying to learn as fast as I can!

 

Interesting how I see it like a cascode amplifier and Gopher sees it like an And gate. Both are correct. It's just a matter of which one makes more sense to the Thread Starter.

 

I haven't had a chance yet to look up cascade amplifiers, but I definitely will. In the meantime, I've added the output resistors and a single LED indicator, along with putting in proper part #s for the transistors. Let me know if you see anything else that could be improved. Thanks for all the help!

 

That's, "cascode".

 

Doh! Thanks, that might've been a confusing Google search.

 

The good news is that the breadboard version of the circuit as drawn in post 8 works perfectly when connected as described to the water control electronics it was intended to test.

The less-good news is that I'd also like to play with this timer circuit in other applications, and I'm getting mixed results. Basically:

1) If I hook an LED up to +5VDC on one side and the output (top side of R6) on the other, and then alternately provide 5VDC or an open-circuit to the valve-control input (left side of R1,) I get the expected behavior - with 5V trigger, the LED pulses in sync with 555 timer output, and without 5V trigger, no LED.

2) If I hook up an Arduino pin (set to input with 10k, maybe 20k?, internal pullup active) to the output (top side of R6) and then provide the same 5V or open to the valve-control input (left side of R1,) then I get highly erratic results on what the Arduino reads. If the 5V signal is present, the Arduino reads pulses in sync with the 555 exactly as intended, but if the valve-control input is left floating, then sometimes the output is silent as intended, sometimes it pulses in sync with the 555, and sometimes it pulses on and off faster than the Arduino can count!!!

Do I need a pull-down resistor on the base of the transistors to ensure that they stay low in the absence of a 5V signal? Is the transistor somehow staying active on its own, or is something else happening here?

Is this leakage current? Is there a certain amount that can always leak through, and that amount happens to be enough to trigger the Arduino input, but not enough to light an LED?

I'm really hoping this just needs pull-down resistors. If so, let me double check the proper application: I assume it would go from ground to the junction in between the base and the base resistor, not before the base resistor? I've seen references elsewhere to a 10:1 rule of thumb: the pull-down resistor should be ~10X the value of the base resistor. So in this case I'd want around 22k?

Thanks for all your help so far!

 

Do I need a pull-down resistor on the base of the transistors to ensure that they stay low in the absence of a 5V signal? Is the transistor somehow staying active on its own, or is something else happening here?
Is this leakage current? Is there a certain amount that can always leak through, and that amount happens to be enough to trigger the Arduino input, but not enough to light an LED?

Er, yes. Always a good idea to have a pull-down resistor. 22k should be fine.

 

Sweet, thanks! What can I say, I still have a LOT to learn. As always, thanks for the help. Love these forums!

 

Thanks again for all the help! I've updated my schematic and put together the final version of this circuit. The only difference between the schematic and the real one is that I used two separate 10k pots instead of a single 20k pot. This way allows me to change frequency either quickly or with great precision. Everything works perfectly and I'm very happy!