I hate electronics!
Jeff Dege
Posts: 85
I hate electronics!
I'm working on a really simple little project. A BS II and a 418MHz RF receiver chip in a Radio Shack project box driving a couple of relays.
I lay it all out on a breadboard, and it works perfectly.
I start building it on a perfboard - one piece at a time.
I'm using one of those perfboards with the individual copper solder pads around each hole. Most folks solder leads together or use bits of wire. Me, I use wirewrap. It's not anything like permanent, given that I'm not using special wirewrap sockets. But it lets me test the connections and the circuits, and back out easily if it's not working. I solder the connections only when I've tested it.
First piece - the header that will connect power, on-LED, and on-off switch (which will be mounted on the project box, not the board, and which will be connected to the breadboard for now), and the voltage regulator. Plug everything in, wirewrap, test that we have continuity where we want it and not where we don't, turm on the power and test that we have the input voltage where we should have the input voltage and the regulated output voltage where we should have the output voltage, then solder everything and test again.
Next step, the socket for the BS2 and the header for the programming cable. Test that the BS2 runs. Solder the joints.
Third, the 418MHz receiver. Test that the BS2 can see the signals. Solder the joints.
Finally, wire up the relays, with their driving transistors, clamping diodes, etc. Test - and nothing works.
Unwrap all the wire, lay out the relay circuits on the breadboard and use clip jumpers to connect them to the circuit that is already on the perfboard. Test that the complete circuit is working.
Move one component at a time from the breadboard to the perfboard, using jumper clips to make the necessary connections. Test to ensure that everything is still working after every step.
Eventually, the perfboard is complete. Only the off-board components - that will be mounted directly on the project box - are still on the breadboard, and they're connected through the on-board connectors that will be used in the box - all the clip jumpers have been removed.
And everything works fine.
So go back and solder up all the joints.
Power up and test - and not even the power LED lights up.
ARGGHHHH!!!!
I'm working on a really simple little project. A BS II and a 418MHz RF receiver chip in a Radio Shack project box driving a couple of relays.
I lay it all out on a breadboard, and it works perfectly.
I start building it on a perfboard - one piece at a time.
I'm using one of those perfboards with the individual copper solder pads around each hole. Most folks solder leads together or use bits of wire. Me, I use wirewrap. It's not anything like permanent, given that I'm not using special wirewrap sockets. But it lets me test the connections and the circuits, and back out easily if it's not working. I solder the connections only when I've tested it.
First piece - the header that will connect power, on-LED, and on-off switch (which will be mounted on the project box, not the board, and which will be connected to the breadboard for now), and the voltage regulator. Plug everything in, wirewrap, test that we have continuity where we want it and not where we don't, turm on the power and test that we have the input voltage where we should have the input voltage and the regulated output voltage where we should have the output voltage, then solder everything and test again.
Next step, the socket for the BS2 and the header for the programming cable. Test that the BS2 runs. Solder the joints.
Third, the 418MHz receiver. Test that the BS2 can see the signals. Solder the joints.
Finally, wire up the relays, with their driving transistors, clamping diodes, etc. Test - and nothing works.
Unwrap all the wire, lay out the relay circuits on the breadboard and use clip jumpers to connect them to the circuit that is already on the perfboard. Test that the complete circuit is working.
Move one component at a time from the breadboard to the perfboard, using jumper clips to make the necessary connections. Test to ensure that everything is still working after every step.
Eventually, the perfboard is complete. Only the off-board components - that will be mounted directly on the project box - are still on the breadboard, and they're connected through the on-board connectors that will be used in the box - all the clip jumpers have been removed.
And everything works fine.
So go back and solder up all the joints.
Power up and test - and not even the power LED lights up.
ARGGHHHH!!!!
Comments
Don't sit on your parts too long with the soldering iron- and after you are finished, beep out the board, checking for cold joints.
Ryan
▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
Ryan Clarke
Parallax Tech Support
RClarke@Parallax.com
Take a day off and then just trace all your power.
▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
"When all think alike, no one is thinking very much.' - Walter Lippmann (1889-1974)
······································································ Warm regards,····· G. Herzog [noparse][[/noparse]·黃鶴 ]·in Taiwan
I've found that if I try to work when I'm frustrated I tend to do really stupid things.
A couple of hours later, I brought out the multi-meter and tested everything around the power supply for continuity. Everything looked good. So I turned on the power and checked the voltage levels. The unregulated external voltage does show up on the voltage regulator's input pin, but the output pin shows 0V.
I'd figured I'd applied to much heat and fried my 7805.
Since I'm not using wire-wrap sockets, I tend to make most of my connections on the pins that are most suited for wire-wrap. Through-hole resisters, transistors, diodes, etc., have long leads, and it's easy to wrap multiple connections. IC sockets, relay pins, etc., have very short leads, and are sometimes touchy to wrap even a single connection.
A 7805 has long leads - and their square, so wirewrap sticks tight and doesn't unwind. So I'd stacked a lot of connections on its ground and output pins. And then soldered the whole thing, besides.
Coming at it unstressed, I realized I didn't need to undo all those connections. I added a second 7805 on an open area of the board, made a couple of wirewrap connections to unregulated power and to ground, powered it up, and saw 4.98V on its output pin. So I connected its output pin to a convenient pin (not to the first 7805, but to the pin that was wired to the first 7805 that was easiest to add a new connection to), and powered things up, and everything worked great.
So I soldered these last joints, flipped over to the component side, clipped the leads to the first 7805, and removed it.
I only had to clip two leads. The third lead - the output voltage lead - has snapped. Don't know how, or why, or when. Never had a lead break like this before. But I figure that I created - and then fixed in place - a gap when I was soldering.
Altogether it took maybe ten minutes to identify and to fix the problem. Next step is drilling holes in the project box for the switches, LEDs, etc.
Lesson learned?
When you're agravated and frustrated - walk away. Odds are that whatever problem you're having, you'll miss seeing the easy fix, if you try working in that mood.