It’s been a while since the last post. Unfortunately a period of illness has prevented me from writing more. All is well again and so I thought that now was the time to put pen to blog so to speak.
As I write this I am sat on a Continental flight from Manchester to Newark en-route to Fort Lauderdale where with luck I’ll get some diving in – whilst putting more dive hours on the light. I may end up repeating some elements of the prior post which I’m trying to fully recollect as I write this So please bear with me!
The dive light as must be clear by now is fully finished and working like a charm! So what makes it tick, how was it made and just how versatile and bright is it?
Taking the last part of that question first. I have it set up with a 50 watt globe although the electronics will cope with a 100 watt globe. How long it will burn is more a factor of the battery technology and capacity you choose and the level you run the globe at. The electronics provides for 5 power levels: off/low/medium/high and full.
These broadly equate to O% ,25%, 50%, 75% and lOO% .
So far the only change I am likely to make is to the 25% level. This is really intended to provide sufficient light to be able to get from the shore to the waters edge. I think 25% is a little too bright and so I am going to adjust this down so as to further conserve battery life for when its really needed.
Whilst mentioning batteries, I have so far built two complete systems, both 50 watt. One utilises NimH batteries (16.8 volt nominal at 4500mAH) and the other uses a pair of old (ex notebook computer) but perfectly good Lithium Ion packs with an original capacity of 4500mAH.
What you choose to use depends on how long you want it to run for, any size limitations and of course most important! How deep your pockets are!
Lead Acid are the cheapest but also the bulkiest and heaviest. NiMH are a good compromise in that they are not too expensive, and deliver a better power to weight ratio when compared to Lead Acid. Finally Lion are the most expensive but you get the most power to weight ratio out of all three.
So what makes it tick?
I spent a long time looking at a broad range of microcontrollers and eventually (and not surprisingly) settled on a PIC micro from Microchip. The range of available devices is significant. There are a range of software development tools readily available and in keeping with the ethos of this project, the PICS are cheap!
Being someone who likes a challenge I decided to develop the entire thing in pure assembler. Including the generation and contral of the output PWM (pulse width modulation). Now that last bit was my downfall… wasted a lot of time banging my head! Eventually, reason set in and I settled upon the PIC 12F683 which has a hardware pwm generator on-board.
I am making the electronics available as a kit just as soon as I can sort out a suitable manufacturer for the boards but for the present at least I’m not prepared to place the code in the public domain. Many hours (long hours) have gone into the development and testing of the final software, so when I make the kits available, the code will be protected.
The particular PIC used has a rather nice hardware pwm module on board although programming it was an interesting experience, the first time around at least.
The module allows the provision of on on/off duty cycle over the complete range of
O – 1OO% which is just what the doctor ordered.
I elected to overdrive the MR16 globe beyond it’s designed for 12 volts. Yes this will (and does) reduce life from typically 2000 hours down to around 200 or so. More than enough for our purposes, and of course they are very cheap to replace. This is compensated for by the significantly higher and whiter levels of light that are generated. My lion version drives the globe at 17 volts (1OO% target value).
When the battery voltage exceeds the target I adjust the duty cycle downward to equate it to the 17 volt energy level. This ‘adjustment’ process could be real-time but I saw little real advantage in doing so. Instead I recalibrate the levels for each half voltage change in battery level from values above the target voltage. This adjustment derives from the onboard analogue to digital convertor which is constantly assessing battery condition. This ‘recognition’ of battery condition results in a recalibration of duty cycle for each of the selectable ‘levels’which translates to brightness that remains constant and true to the level you have currently selected.
Indication of level is signalled to you by way of a piezo transducer and Morse code. On each change you hear a confirmation [off, low, medium, high and full].
The levels transition one to the other very smoothly by drifting the pwm duty cycle toward its ‘new’ target. The levels cycle up and down [ off /low/medium/high/full/ high/medium/ low and off ] which is my preference following trials with other combinations.
Finally I constantly assess the battery condition and protect it by shutting down the light following a series of audible warnings that this is about to happen, thus giving the diver ample time to switch over to his/her backup. Although the battery housing sports a very bright LED that remains on from the moment the battery is connected until it is once again disconected this, whilst useful (i.e. you can read instruments by it) can no way be considered a reserve.
In my next post I will endeavour to include some photos of the finished article.