EEVblog Electronics Community Forum

Electronics => Beginners => Topic started by: Ale2.0 on September 29, 2020, 08:41:00 pm

Title: Pnp Transistor does not cunduct completely
Post by: Ale2.0 on September 29, 2020, 08:41:00 pm
Hello there

I have a pnp transistor turning the reset pin of an ne 555 to ground. It actually switches but theres a residual voltage of 600mv. If I use the npn beside it doesn't behave like that and actually has 0v.

I replaced everything, Transistor, ne555. I lowered the resistances (values written with pencil are the current ones)

But nothing helped. I also added a resistor to ground with 20k (from pin 4 (reset)) but this did not help either.

Would appreciate any help as I am running out of ideas.

Ps. The circuit dims a waveshare lcd backlight with pwm based on a LDR. It also has the ability to turn off this backlight. Npn for the raspi and pnp for the original signal of the Display. Although it does also not work if I directly connect the pin of j2 to ground.

Thank you beforehand and greetings from Switzerland
Ale 

[attach=1]
Title: Re: Pnp Transistor does not cunduct completely
Post by: xavier60 on September 29, 2020, 09:12:10 pm
A PNP transistor would be operating in Common Collector mode, also called Emitter Follower.
BJTs need about 600mV Base to Emitter to turn on, passing Collector to Emitter current.
With the Base pulled down to 0V, the Emitter of the PNP has to be about 600mV above the Base voltage.
Title: Re: Pnp Transistor does not cunduct completely
Post by: Ale2.0 on September 29, 2020, 09:20:28 pm
Thank you for your answer.

Now I see where I went wrong... Is there any simple way to alter my circuit so it would work? The thing is already soldered so adding as few components as possible is preferred :))
Title: Re: Pnp Transistor does not cunduct completely
Post by: xavier60 on September 29, 2020, 09:31:10 pm
The residual 600mV shouldn't prevent the 555 from being reset. What amplitude is the PWM signal?

Edit: Also are you inserting the PNP correctly? Collector to ground.
Title: Re: Pnp Transistor does not cunduct completely
Post by: Ale2.0 on September 29, 2020, 09:54:33 pm
Hello

The pwm has an amplitude of around 4.8V. When the 600mv are applied to the reset pin the pwm stops but the output still is at around 3V (stable). Which is interpretet as high by the display and so switches it to full brightness.

And yes I think I inserted it correctly. Collector connects to gnd and the emitter to the 47k to +5v.
Title: Re: Pnp Transistor does not cunduct completely
Post by: xavier60 on September 29, 2020, 09:55:55 pm
I was surprised when I looked at a 555 data sheet. Looks like 600mV isn't low enough to guarantee a reset.
Title: Re: Pnp Transistor does not cunduct completely
Post by: Peabody on September 29, 2020, 09:58:03 pm
The NPN transistor is the right part to use for this purpose.  To get the same result from a PNP, you would have to bring its base 0.6V below ground, which you can't do.  The NPN only needs its base to be at least 0.6V above ground, which you can easily do.  You would use a PNP if you were switching high-side power to a device.  Use NPN for grounding things.
Title: Re: Pnp Transistor does not cunduct completely
Post by: xavier60 on September 29, 2020, 09:58:32 pm
If you short the reset pin to ground, does the output then go fully low?
Title: Re: Pnp Transistor does not cunduct completely
Post by: Ale2.0 on September 29, 2020, 09:59:02 pm
Yes it does.
Title: Re: Pnp Transistor does not cunduct completely
Post by: xavier60 on September 29, 2020, 10:02:52 pm
I have never liked using a 555's reset pin, now I know why, https://www.electroschematics.com/quirky-555-timer-reset-function/ (https://www.electroschematics.com/quirky-555-timer-reset-function/)
All I can suggest for now is to stick to NPN BJTs and add an extra inverter stage.
Title: Re: Pnp Transistor does not cunduct completely
Post by: Ale2.0 on September 29, 2020, 10:04:17 pm
Hmm that's exactly what I wanted to avoid.   :P

Thank you anyway and have a nice night.
Title: Re: Pnp Transistor does not cunduct completely
Post by: mikerj on September 30, 2020, 10:02:22 am
Rather than using a transistor to buffer the signal, you may be able to to drive the reset pin directly from the source via a schottky diode assuming whatever supplies this signal isn't very high impedance.  If this is a through hole construction you can easily replace the transistor with a diode.  A bit of a kludge but may give a quick fix.
Title: Re: Pnp Transistor does not cunduct completely
Post by: Ale2.0 on September 30, 2020, 11:27:35 am
Hm

I'll look at it but as this circuit will run 24/7 I don't really want to use a solution like this.

It seems like I'll have to build an inverter but I'll post the final solution here.