Stangest thing

C

ccdubs

Member
Hi all,
I have the wierdest thing happening with a vary simple circuit that uses a 08M.

The circuit is mains powered so I have a 1.5VA 230-24V transformer, full bridge rectifier and 1000uF smoothing cap feeding a LM7805 voltage regulator which powers the 08M.

The 08M is programmed to pulse an output pin which drives a buzzer at 24Vdc through a BC337 transistor.

When I turn the power on, the circuit takes a while for the cap to charge and then springs into life and runs well.

Here's where it gets strange. I wanted to measure the voltage across the piezo buzzer but as soon as the probe touched the solder terminal of the buzzer (24V rail) the buzzer stopped. I took the probe off and it started again (yes the meter was set to volts and not amps).

I then tried to measure other voltages (eg chip supply)...exactly the same thing. As soon as ONE probe touched a solder joint the buzzer would stop. I disconnected the probe from the meter and again same thing. This seems to happen when I touch almost any of the solder joints.

I tried a different probe...same thing. I even tried with a long piece of wire.....same thing.

What is going on here? Can it be RF?

I should also mention that the circuit only works reliably (ie, buzzer buzzes) if I remove the current limiting resistor to the base of the BC337 and replace it with a link.
 
premelec

premelec

Senior Member
If you haven't already done it put bypass capacitors on _both_ sides of the regulator as well as at the PICAXE socket - something may be oscillating - or not :-( It also sounds like it could be a floating input pin picking up something... they all need to be tied down...
 
C

ccdubs

Member
I will try as you suggest. Why do the input pins need to be tied down if they are unused? What effect could they have?
 
B

BeanieBots

Moderator
Get your base current limiting resistor put back ASAP. You will be over stressing the PICAXE output which will destroy it.

Floating inputs draw excessive current.
I'm guessing that you have some form of activation input. If you have a pushbutton or similar, it will need a pullup (or down, depending on your design) resistor. Otherwise that input will also be floating when the button is not pushed.

Also make sure that you have a pulldown resistor on the download input. This is the one most people forget. It is ESSENTIAL for reliable opperation, not simply a nice to do or good practice.

 
C

ccdubs

Member
I don't have an activation input. The circuit pulses the buzzer as soon as it is live.

I will try a pull down resistor on the download pin and let you know what happens.

While the answers here all seem to have merit, I still don't understand how touching a disconnected probe to a solder joint can effect the cct.
 
B

BeanieBots

Moderator
If you do not have a pull down on serin, then THAT IS your problem, gauranteed.
A floating input is exactly that, it flaps about in the breeze. Touching just about anything onto anything else will effect the 'breeze'.
With serin, each time it 'flaps' up, the PICAXE thinks it is getting a new program and stops what it is doing to get ready for the data.
 
C

ccdubs

Member
Do I need a resistor? Or can I just tie it down to the 0V rail. A resistor causes a bit of a space issue.

I don't need to program the chip in situ as it is mounted onto a 8 pin base.
 
M

moxhamj

New Member
Officialy you use a resistor to pull serin down, but I've built many circuits with it just tied to 0V with no problems.
 
C

ccdubs

Member
Hi all,

I have fixed the cct by tying the serial in to 0V. I have also replace the current limiting resistor to the BJT and it works well.

Thank you to all those who contributed!
 
You must log in or register to reply here.
Top