Monday, September 24, 2007

Trust



Engineer Alan - he who did all the hard work for our boat conversion - was so impressed I'd built my own VCS electronics for the boat (electronics that makes the generator speed up or slow down to keep it happy and the drive batteries) he handed me this thing hoping I could find out why it wasn't working.

Isn't it pretty? And oh what trust!! I know what it's supposed to do and he gave me a bit of paper also said with just enough technical data to mean I knew what batteries to connect to see what happened....

The thing connects to the radio control receiver for a model boat. The power for the boat's motor is connected on the right - between 2 and 12V (volts) so the black and red wires above on the right are from my 9V battery. The yellow wires on the right are what would go to the motor - in my case, a test meter.

On the left, the germanic coloured wire came with the unit. Red is the plus from the receiver, black the minus, yellow the control signal. I think the wording on the bit of paper says that supply voltage should be between 3.6 and 6V. But the wording might mean the control signal should be in that range. I've no definite idea.

Oh, I know the big white rectangle on the right is a relay that switches the output to reverse the motor. I'm pretty sure the black thing to its left is a power transistor to provide the variable power/voltage for the speed of the motor. Black blob down centre is a simple transistor drives the relay. White thing top left is a quartz crystal so this tiny circuit generates a very high frequency signal - I would have to assume, a reference signal. I can't work out from the lettering the frequency - and haven't yet tried attaching my oscilloscope to see - it's probably around 2 MHz (two million cycles per second.) I have traced the coding on the black "chip" on the left - it's a microprocessor which thereby might do almost anything. It's in my Maplin's catalogue as ideal for radio controlled models' motor speed. Um, yes, so that fits. Haven't been able to find any detail although they did have a diagram with pin connections implied it has a frequency "in" (that fits with the quartz crystal being there).... I suspect very strongly it compares it's local frequency with the input frequency and so generates a voltage output proportional to the difference (which drives the motor)... and also a signal to show if the input is of lower or higher frequency, causing the relay to switch over and reverse the direction of the motor being driven.

Oh, from the actual wiring tracks, I'm pretty sure black blob down centre, which is definitely a simple transistor, (I've got loads of identical!) merely amplifies the output from the microprocessor to drive the relay, just as I'm pretty sure the black thing to the left of the relay merely amplifies the output from the micropocessor to control the speed of the motor. ...

But how to find what's wrong (if anything?) Well, I suppose I could check there is an oscillating signal with my oscilloscope but actually I think it would be more sensible to demand to borrow the input equipment and see what happens and get my oscilloscope out then. Because I suspect there is nothing wrong with this circuit. Why?

Well, for a start, all the solder joins are good - usual reason an electronic circuit fails is a faulty join. Secondly, none of the components look anything but pristine - components that have got too hot show it and even smell of it! Thirdly, for connecting my wires I did a very naughty test... simply disconnected and reconnected the input voltage to find the output sometimes suddenly switched to "full ahead", sometimes switched to "full astern", sometimes switched to "off". The circuit was definitely active and the relay definitely clicking. Alan had thought it might be the problem... but with the best will in the world I don't think Alan has the foggiest idea about frequency-comparison circuits... which I think this thing is supposed to do. No, I don't know for sure but I can't think why else it has an on-board frequency regulator - the quartz crystal. The only other reason it might have one is if the input signal is digitally encoded- perfectly possible.... I've no idea, but I'd think unlikely. Digital encoding takes a microprocessor with more legs on it. I know. As I said, what trust from Alan I might have any idea! Well, I suppose he's right I do have idea... but, goodness me, no instant solutions.

Funnily enough, most of my tests and thoughts simply pure logic and not especially electronic. Quartz oscillator, ergo, high frequency being generated. Why? Must be to compare or time signals coming in. Equally, logic, device converts input into speed and direction, therefore there will be an output for speed and another one to reverse direction... trying to work out how much more like solving a crossword puzzle rather than technical electronics - I have this information, what can I work out from it?!

I must get the other bits off Alan... I suspect that if I have a look at the signal the receiver produces with my silly-scope all will be revealed! Oscilloscopes reveal almost anything if suitably connected - not so good about digitally encoded signals because setting them to reveal the pulse sequence is tricky.. get it right, oh, yes, there it is. Huge fun to see it working.

Ah well. I guess I just ask Alan to borrow the other bits without explaining my deductions. I think, actually, the diddy circuit pictured is probably OK and the input was wrong or the connections failed or a battery was flat. You'd be amazed how often little devices fail because nobody realised a battery was flat or a plug not fully in place. "Solid state" electronics, of itself, rarely fails unless it does so immediatly. The bits where humans interface and mechanical things happen - that's usually the problem. I've blown up many devices on that basis.

As a codicil, I'm curious, for this entry, "Return" seems to be producing double-spacing. So I've left out my P codes and wait to see if this spaces OK for being "published."

1 comment:

  1. Weird. This time the paragraphs spaced without adding coding!

    ReplyDelete