Firstly it’s a big thanks to the Moderators for tolerating a Thread that has really nothing to do with
Vintage Radio Repair and Restoration even though it’s in the
Homebrew Equipment Section – hopefully others may find this useful?
Observant Forum Members will have noted that this Thread has been running for six months now - it has been stop/start with lots of periods of inactivity due to other more pressing matters. Before Posting anything on the Forum I did have a go at trying to reverse-engineer and recreate a commercial strobe that I had disassembled but very quickly encountered stability issues with my early attempts to "
tame" the HEF4060BP – see early Posts above.
After various false starts and trips up several culs-de-sac the Project really took off and I embarked on a long and often arduous journey to reach a successful outcome. It's been really educational (for me) but probably a bit tedious for the Forum as I've not only discovered the joys of crystal oscillators, binary counter/dividers and square waves but I've also had cause to use my Topward function generator and ageing Farnell analogue oscilloscope. Add to that I've learned lots about voltage regulators and transistor driver stages. Most of this has only been possible with advice, guidance and encouragement from several Forum members which has proved invaluable - can't thank them enough.
In particular, a huge shout-out to
David G4EBT,
Radio Wrangler and
Tony Duell.
A quick summary which of course doesn’t fully convey the much wailing and gnashing of teeth at various points: -
1. Started off looking to build a 300Hz strobe to check the rotation speed of my hi-fi turntable. Tried building a circuit based on the Product Data Sheet using a 4.9152 crystal and an HEF4060BP binary counter/divider but this proved too unstable and was not fit for purpose.
2. Enter
Radio Wrangler with his suggestion to try the Qantek QX14T50B4.915200B50T crystal oscillator which is accurate to within 50 parts per million (0.005%). I also replaced the HEF4060BP with a 74HC4060 and these introduced tremendous stability to my circuit. I had a rock steady 300Hz on Pin3 (Q13) of the 74HC4060. I could have stopped there as I had a working strobe at 300Hz and which performed very well with my strobe disk - but I did appreciate that this was a symmetric square wave with a 50% duty cycle. From here it becomes mainly an educational project although I will return to this later. This is the waveform on my analogue 'scope coming from Pin 3 of the 74HC4060:-
3. Now I needed to convert this to an asymmetric square wave with a duty cycle of 1/16th on and 15/16ths off - I did of course appreciate Tony Duell’s advice that there " ... is nothing magical about that figure" in Post #61 of my Thread but it seemed like a good figure to aim for?
Radio Wrangler also provided guidance on the pin connections for various ICs to perform this conversion but I found that I could only get a stable waveform when using
Tony Duell’s instructions for the 74HC08. This is the waveform on my analogue 'scope coming from Pin 6 of the 74HC08:-
So far so good and looked promising to me. Once again I could have stopped there as I had a working strobe at 300Hz and the asymmetric waveform with the shorter on time did seem to give a much sharper focus to the lines on my strobe disk - but maybe that's just expectation bias or “
audiofoolery” on my part! The LED was, however, a lot dimmer, so I needed to add a transistor stage – the ICs would need a 5V supply so some form of voltage regulation for that part of the circuit was required as power was coming from a PP3 9V battery.
4. David G4EBT is very much my Forum Mentor, Tutor, Guide, Educator and Instructor with seemingly endless patience and he was once again of enormous help at this stage and provided me with advice on building a voltage regulator module using an LM7805.
As a brief aside I have to mention that I did order some 7805 voltage regulators from a supplier who shall remain nameless and this was duly fitted into the circuit only to find excessive heat and some smoke from this component. After checking the text on the component (jeweller’s loupe is handy) I discovered that I had in fact been sent L7905CV voltage regulators which are different, especially in their pin configuration!!! The supplier was quick to replace these at no cost but the lesson to always read the small print cannot be overstated.
Later on
David checked the total consumption of the strobe and a 78L05, rated at 100mA, was deemed more than adequate so was used as a replacement.
5. David G4EBT also sent me details of a standard transistor module using a 2N3904 and I added these two modules to my circuit. Wired up the transistor stage and checked the voltages to all the ICs and the feed to the LED etc. Double checked that I was still getting 300Hz from Pin 6 of the 74HC08 and that I was also getting 300Hz at the Base and Collector of the 2N3904. Amazingly it all seemed to work and this is the waveform on my analogue 'scope at the Base of the 2N3904 - as you can see it is set at 500mV/DIV so I reckoned we were getting just less than 1 Volt: -
6. On the Collector of the 2N3904 I seemed to be getting the Gain I was expecting but the asymmetric square wave had been “inverted” somehow – this confused me for some time! The ‘scope was set at 2V/DIV so we had about 6.0V: -
I had no idea what was going on due to my limited knowledge of such things but
Tony Duell took the time to provide a detailed and eloquent explanation which I would summarise as follows: -
“That’s what I'd expect. It's a common emitter amplifier, so when the output of the 74HC08 goes high, it forward biases the base-emitter junction of the 2N3904. That junction, like just about any forward-biased silicon diode junction, will drop 0.7V or so.
Look at the transistor. When the output of the 74HC08 goes high, it sends base current into the 2N3904, forward biasing the base-emitter junction. This allows the collector-emitter current to pass, so the collector goes low in voltage. Think of the collector-emitter path in the transistor as a switch which turns on when the base current is applied. The voltage across a closed switch is small. So the collector is at a similar voltage to the emitter when the base current is flowing. So it goes low ANY common emitter (or common cathode) amplifier inverts the signal.
Now consider the LED and its series resistor. For the LED to emit light there must be a voltage across it. One side of the LED is at +9V or so. The other side goes to the collector of the transistor. So the LED is on when the collector is at a low voltage. The LED is on therefore when the output of the 74HC08 is high.”
As I said earlier this has been a very useful educational Project for myself and I had to read
Tony Duell’s explanation several times to fully understand what was going on!
7. The circuit schematic went through many iterations, mainly suggestions and improvements by
David G4EBT, including pencil and paper versions, Microsoft Paint versions and finally a version I struggled to complete using Eagle software - it almost got the better of me but my "stubborn" gene kicked in and I persevered. Of course I have to mention a basic connection that I had omitted and this error was picked up by
David!!! But we finally ended up with a circuit that was to all intents and purposes complete - so at this point I had a Working Prototype, albeit on a couple of linked Breadboards. Hardly elegant but it worked!
At this point it’s over to
David G4EBT again who converted the circuit schematic into a PCB layout complete with tracks, holes and pins – quite amazing as I’m still battling with Eagle software to get to the stage of having a viable board. A couple of final tweaks have removed the reverse polarity diode as the board is powered by a PP3 9V battery and this helped shave some length off the PCB.
David used the UV dry film negative photoresist method of making PCBs to produce a number of boards, a couple of which he sent to myself. Not to mention a package full of useful components required to complete my very own PCB – very generous.
(Continued in next Post)