PIC driven transistor - smoothed signal edge - why?

[flotulopex]

Hello,

I'm driving a transitor with a PIC; it is meant to control an IR led for a remote with a 38kHz carrier freq.

Depending on what PNP or NPN transistor I use, the final signal edge is smoothed (green arrow).

Clearly, the signal is much better with a PNP transistor (BC307b graph).

Since both transistor suppose to be "complementary" and have identical specs, I don't understand why the edge (BC237 graph) is smoothed that way while using a BC237 NPN transistor.

Any clue? Is this due to the IR-Led's capacity?
<img src="http://www.picbasic.co.uk/forum/attachment.php?attachmentid=2672&stc=1&d=121353203 4">

[Acetronics2]

Hi, Flotul

Never heard of transistor parasitic capacities ??? ...

Search Keyword is " Totem pole " ...

Also, try to use a // capacitor ( 1nF) on your 1K2 res ... or a BAT85 ( Schottky ) diode ... in the right way to "pull off" electrons from the trans. base !!!

You might see a litte enhancement.

so, might we say it is a " Flatul Hence " ??? ... LoL !!!

Alain

[Darrel Taylor]

Hi Roger,

I don't think the Voltage waveform at the collector matters, since the LED operates from Current.

According to the datasheet, the LED has 25pf capacitance. So when you turn off the transistor, that voltage can only dissipate thru the relatively small leakage currents.

To allow the LED to discharge faster ... Placing a resistor across the LED would probably improve your waveforms, but it won't change the function and will only consume more current.

A better indication of the Drive Current would be to measure the voltage across the 33 ohm resistor. Which should be very "square".

DT

[mister_e]

What kind of o'scope is this?

Sound card based?

The second example show that OR your base resistor is a bit too low OR the source can't give enough current to your transistor base OR you o'scope can't draw the wave properly.

[ardhuru]

Perhaps a resistor between the base and the emitter (around 2.2K) might help the transistor snap faster.

[dhouston]

What kind of o'scope is this?

It looks like a Protek 220 ISA bus scope card which has very high sampling rates. If I'm right, the scope is not a limiting factor.

And 21 cycles at 38kHz is ~550uS which is a typical burst duration for several IR protocols.

[Acetronics2]

Hi, Dave

This scope ... has already shown its limits with Flotul :

We were offered a wonderful sample of demodulated "HF" ( LoL ... better to say Mid Freq ) signal, some posts ago ...


Now, this waveform is perfectly correct and true ... and shows bipolar switching limits with unipolar supplies ...

I also have samples of that ( but taken on an ... analog scope !

Alain

[flotulopex]

My scope is a "cheap" HANTEK DSO-220 USB modell

I couldn't try Alain's suggestions up to now; don't have any Schottky diodes neither 1nF capacitors... will be bought and tested by tomorrow evening.

My initial question was more about to understand the difference between both transistors since nothing else changes in the circuit.

so, might we say it is a " Flatul Hence " ???

I wish this was true: I would bother you All again with a methane powered PIC project

[dhouston]

My scope is a "cheap" HANTEK DSO-220 USB modell

Interesting. It looks like it was copied from the Protek 220 which was an ISA card DSO capable of 20 million samples per second. Of course, the sampling rate will be much slower using USB. The GUI is identical to Protek's with the exception of the colored knobs.

[Acetronics2]

My scope is a "cheap" HANTEK DSO-220 USB modell

Let's be honest ... and use the tools for what they are intended to do ...

I couldn't try Alain's suggestions up to now; don't have any Schottky diodes neither 1nF capacitors... will be bought and tested by tomorrow evening.

Amateur ...

My initial question was more about to understand the difference between both transistors since nothing else changes in the circuit.

This is exactly the same feature : junction capacities to discharge ... a strong cut off is obtained by polarizing the base to - 2 to 3 volts ...
just look at old "logic" schemes !!!

I wish this was true: I would bother you All again with a methane powered PIC project

Ah, la vache !!! ...

have a Good evening ...

Regards

Alain

[flotulopex]

Amateur, amateur...

In the early 1900's, the very best professionals & engineers built the Titanic