And none of this explains why inverting a transistor makes the circuit work for my daughter?

I would expect the wrong package orientation for the collector and emitter.

 

I've checked and double-checked that.

But also, why would it work by inverter EITHER of the transistors?

It's a right old conundrum.

 

The LED isn't in parallel to the buzzer.

Are you sure. I see the LED and buzzer, both connected to ground (blue) at bottom of picture and I see both LED and buzzer connected to collector of PNP transistor.

Also is Item 11 a crystal or a capacitor as the symbol is a crystal?

 

Item 11 is a piezo buzzer, probably acting as a capacitor since the speaker is responsible for the sound - a bit strange I agree.

Take a look at the most recent picture, I think the first one was confusing making it look like the horizontal and vertical blue wires were connected when they're not.

View attachment IMG_2987.JPG

 

What is the big red thing in parallel with LED.

 

That's a loud speaker, 32 ohms impedance.

 

Remove LED for now. Second, Item 11 should be a capacitor as others have shown.

 

32 ohm speaker just affects volume
LED clamps voltage across speaker to 2V or so and has no current limit resistor. Improper use of LED.

 

That's what's in the booklet for the electronics kit, it simply lights the LED when you push the button.

With or without the LED makes no difference, that is until I invert the transistor.

 

You do not match the picture shown in post 1. The emitter of the PNP connects to V+ and V+ connects to R34 then to R32 then to base of NPN transistor. Cap also connects to NPN base. 34 should be varaibale resistor as shown in original photo.

 

Yes the only thing I did was remove the variable resistor as I'm not interested in affecting the tone.

As a reminder, this is what is in the booklet.

Other tone generator experiments also did not work until inverting a transistor.

 

Sorry, that 1st picture is wrong 53 and 54 connected in series from V+ to base of NPN transistor. Given how the book used the LED, I wouldn't trust. Go with the picture someone posted that I have been re-posting.

 

You are on Experiment 23 and I really question the validity of that circuit. Now look at Experiment 24 which closely matches what myself and others have been saying. Try implementing Experiment 24 first and let's see if we can get that one working. Although the use of a buzzer (BZ) in the feedback path is strange as opposed to capacitor. But after some Googling, two-terminal piezo acts like a capacitor. Do not assume the book is a correct design, this is a kid's toy and I doubt the designs have been properly vetted..

 

circuit is a known oscillator and it works, it used to be very popular. but it is not perfect, it is very sensitive to values, supply voltage etc. even change in load can change frequency a lot. i am not sure why that is offered as a training example. specially with the way LED was used.

 

We tried the next one, and same result - LED lights but no sound - however, inverting the green PNP makes it work with the variable tone. And inverting the red NPN also makes it work.

Without inverting anything the LED brightness changes ever so slightly.

 

Also swapping out the speaker (32 ohm) for an 8 ohm works.

 

(1) You should also be able to swap the LED (26) for a basic wire (3) and it should still work.

I assume by work, moving the variable resistor (53) changes the tone and the flash rate of the LED. The LED is actually flashing a a high rate.

(2) Also try swapping out the buzzer (11) for a 10uF capacitor, this should also work.

I hate to go there, but if you invert/flip the PNP transistor, then maybe its been assembled wrong. Those transistors and their equivalents can have two different pinouts, maybe "factory" changed parts.


The BC557 Comparison: If you ever swap a 2N3906 for a BC557 (a very common European equivalent PNP), the pinout is exactly reversed. The BC557 is C-B-E. If you plug a BC557 into a circuit designed for a 2N3906 without flipping it 180°, the circuit won't work.

 

Would the LED still work? Because it does light up fine, and does change brightness with the variable resistor.

 

The LED is not needed. Could be replaced with wire. Are you able to look "inside" your PNP module and see the transistor and if there are any markings on the case.

 

yes, green PNP is S9012, red NPN is S8050, from all data sheets I can find, pin 1 is E, and they're oriented correctly to match the plastic triangle their mounted on, showing the arrow on the emitter.