It's more the ratio of the diode voltage drop : VCC that causes the shift; the diode forward voltage only changes a small amount relative to the change in VCC.

 

<zombie mode> Neeed whole circuit! NEED BRAINS!..

__________________

 

The OP's cct? I bet it's just a resistor from Vcc to pin 7, resistor from pin 7 to 2 & 6, capacitor from 2 & 6 to gnd. The basic astable in every 555 datasheet.

 

I've rebuilt the frequency generator circuit and it's working pretty much to spec, although I am using different value for R1, R2 and C this time. I did try to reproduce the behaviour I was seeing earlier, but couldn't. As others have commented, the frequency is pretty stable across the range of voltages, so I am going to put my earlier experience down to poor circuit build - I am using a breadboard so maybe one of the connections wasn't quite tight enough, or it could have been a dud cap, or.....

I'll install Pspice so I can draw up the circuit next time, but for anyone who is curious, it is the standard 555 astable circuit I am using.

The reason for building a freq gen is that I want to drive an IR diode in such a way that it's signal can be picked up by a 'tuned' IR receiver. The component I am using for the receiver is the RPM7100 series from Jaycar
http://jaycar.com.au/products_uploaded/ZD-1952.pdf
and what I was hoping to be able to do is to detect the presence/absence of the IR beam using this receiver. When an IR beam with matching frequency is detected, the output of this receiver should go low, otherwise it should be high (around 5 volts).

However, I have hit a couple of challenges...

1) reading the datasheet for the receiver, its not clear to me what frequency it will detect at - the datasheet describes 4 different frequencies. The only identifying marks are on the back of the receiver - 1B34 04A.

Question: Can someone tell me categorically what frequency this IR receiver is tuned for?

2) I am using a BC549 NPN transistor with the o/p from the 555 wired to base, and an IR LED with (current limiting) series resistor connected between +5v and collector. According to my theoretical knowledge, this should cause the IR diode to switch on/off at the same frequency as the o/p from the 555.
However, I am seeing a drop in the frequency when measured at the LED, for example, if I measure the o/p of the 555 I get 36KHz, but when I connect the o/p to the transistor base, and then measure at the collector, I get readings around 28 - 30 KHz, and it is quite unstable.

Question: I'm sure there's an easy explanation for why the frequency will vary in this way. Do I need to compensate by increasing the 555 frequency, so that it will be say, 36KHz, when measured at the LED? What about the way the frequency shifts - is there something else I need in my circuit to enable a constant frequency output from the LED?

 

Do you have a series resistor from 555 pin 3 to the base of the BC549.??

__________________
Eric " Good enough is Perfect "
I will NOT answer PM's requesting technical help, please use the Forum
PIC tutorials: Nigel's www.winpicprog.co.uk/ Bill's: www.blueroomelectronics.com/

 

The LED frequency will be the same as the 555 output frequency. Use a CRO to view the waveform if you don't believe it. Or replace the LED with a resistor and measure the frequency at the collector (pointless, I know, but you can do it as an exercise).

It also looks like the carrier has to be switched on and off roughly 600us/600us (see page 2 of datasheet). This is sometimes done to weed out noise IR sources. You'll want another 555 to create this pulsing effect.

As to the centre frequency of the device: no idea. Sorry. You could find it yourself by decreasing the transmitter power to quite low (so you don't saturate the device - should give better readings) and sweeping the frequency - record the freq that the signal is detected, and the freq that it stops being detected at; the centre frequency should be in the centre of those 2 values.

 

From ericgibbs
No, I have just wired the o/p direct to the base. Should I need a resistor there?

From dougy83
Ermm, that's how I know the frequency is different - I am measuring it with a CRO
Just to clarify, I am seeing a difference in the frequency between the following 2 readings.
a) the unconnected output from the 555 (i.e. a wire taken directly from pin 3 and NOT connected to anything else
b) the collector of the transistor (which is also the negative lead of the LED) when the 555 o/p is connected to the transistor base.
It looks as though the resistor, LED, transistor 'group' somehow affect the input signal frequency, but how/why I have no idea.

The point you make about switching the carrier is something I had missed So, it looks like I will need to do 2 things to generate a suitable signal
1) generate pulses at the required frequency, say 37KHz
2) use another circuit to break switch this on for 600us then off for 600us
Hmmm, this simple(?) project is getting more and more interesting by the minute...

BTW, how do you guys insert bits of other peoples appends so that they are referenced correctly? I am cutting and pasting bits of your appends, then highlighting and selecting the 'Wrap quote tags around' icon, but there must be a better way that also gives the quoter details, and the link?

 

hi,
The BC549 HFE is about 200 at 100mA, with a Vbe on of approx 0.7V

So the series base resistor should be:

Rb = [Vout-0.7v]/0.005 where the 5mA is required base current.

So if Vout=10, then Rb=1.8K [ this will give a base current of *10 the minimum required and will, drive the BC549 fast into saturation.

Note: have you measured the Vmin out from the 555..??

Can you confirm the supply voltage and the Resistor in series with IR diode and type.

__________________
Eric " Good enough is Perfect "
I will NOT answer PM's requesting technical help, please use the Forum
PIC tutorials: Nigel's www.winpicprog.co.uk/ Bill's: www.blueroomelectronics.com/

 

Without a series base resistor for the transistor then the output of the 555 and the base-emitter of the transistor are probably blown up.

__________________
Uncle $crooge

 

I've downloaded and installed LTSpice as per a previous members suggestion. I've uploaded a visual of the circuit, which now has 2 555 chips - the first just gives a frequency of around 36KHz, while the second one turns this on and off at a frequency of around 833Hz. If this signal is correctly generated, then I expect to see the ZD-1952 output go low....that's the theory anyhow.

The circuit diagram (with some comments) is Circuit.jpg. I also ran a simulation and by measuring at point A, got the output shown in SimulationTrace.jpg. Pretty cool, but only theoretical of course!!

So then I went and built my circuit using the same spec components I'd used in the model. The results are not too discouraging, but certainly not as accurate as those in the model.

I measured the frequency output at 4 places as shown in the circuit diagram. The results of those measurements can be found in their respective jpeg files, e.g. CROTraceA, CROTraceB etc.

I guess the most obvious differences are

1) both frequencies are lower than expected. Instead of 833Hz, I am seeing around 755Hz, and the expected 36KHz is actually 34KHz. I suppose the latter is close enough though.

2) The duty cycle of the 833Hz frequency is nowhere near symetrical.

There is also quite a bit of frequency still coming through during the 'off' period - this can easily be seen in the screenshot signalTransition.jpg.

Anyhow, although I am still frustrated in my attempts to get the ZD-1952 recognise my signal, this has been an interesting learning experience.

I suppose the bit I find most challenging/frustrating, is that I don't know how to easily vary the frequency...at least, not without going through another calculation for R1, R2 and C1...which takes me way too long. I have experimented with a variable cap, but this resulted in the frequency changing in one direction, only to come down and go in the other, and only slightly. Apart from getting hold of a sophisticated signal generator, I'd appreciate any advice on how to 'tune' my circuit so that it emits the signal carrier expected by the ZD-1952.

For anyone interested in a bit more detail, here are the measurements I am getting on the output of the ZD-1952.

1 - IR diode not connected, i.e. no pulsed IR - 5.05V
2 - IR diode connected - 3.1V

So, it looks like the receiver is picking up the pulse, but I still get the same reading (3.1V) even if I cover the LED (using a heatshrink tube).

I then experimented with a couple of remote controls from DVD, Topfield, with the IR diode still connected.
2.8
4.5
3 - Sony DVD remote button pressed - 2.5V
4 - Topfield remote button pressed - 3.8V

I then repeated tests 3 and 4, but this time with the IR diode disconnected.
3 - Sony DVD remote button pressed - 2.8V
4 - Topfield remote button pressed - 4.5V

All up, a bit of a confusing picture. Seeing as how the voltage dropped considerably when I connected the diode (test 2) I am wondering if in fact the receiver is working, and even when I cover the diode with heatshrink, the IR may still be 'escaping' and being detected.

Attached Thumbnails

         

   

 

I agree, some cheap ceramic capacitors can have a huge voltage coefficient.

Try replacing the capacitor with a higher quality film capacitor and it should be more stable.

__________________

I do not answer private messages asking for help because no one else can: benefit from advice I may give or correct me if I'm wrong.

Please ask on the open forum if you have a question and I'll be happy to help, if I know the answer.