Last Friday, I posted a thread on CandlePowerForums. This long-awaited thread was the product introduction of the Aviatrix, the first aftermarket LED-Ring drop-in mods for the Surefire A2 Aviator... posted on the heels of several weeks of forum chat in the general flashlight section about the potential for such a product.
The product introduction thread I posted last Friday made several statements regarding the new product, including product description, price, initial quantity offered, and proposed delivery schedule. After posting the thread, I wasn't quite sure what to expect... but nothing prepared me for the response I received! I thought that maybe I would sell the first production run (25 units) in a week or two, and that the product would gradually pick up momentum as postive feedback was posted by CandlePowerForum members. Call it the miscalculation of the century - I was stunned. The first production run sold out in less than 2 days!!! I was pleased, shocked, grateful, and astonished... all at the same time!
OK... here's the gist of it all. After receiving each Paypal payment, I replied to the buyer with an email. In each email, I thanked the buyer for their purchase, and indicated my intention to keep them informed regarding the progress of the mod, and to let them know when their (much appreciated) order shipped. By Sunday afternoon, I was kind of in a state of shock - I very honestly didn't expect the product to take off with the momentum that it did. In any case, my confidence in my ability to fulfill the orders and deliver product was never in question - I was just glad that I had stated 7-10 days for the expected order fulfillment time! . Here's the point I'm trying to make: I am writing this post to keep all Aviatrix buyers up-to-date as promised, and to announce that I am unable to meet the 7-10 day delivery schedule. I would be unable to keep it even if I don't count the weekend days in the 7-10 day shipping timeframe (which I wasn't, BTW... I can't work ALL the time! ). This announcement is NOT due to manufacturing, technical, or component availability issues... there is no outside force that I can lay the blame - nor would I do so even if there was one. The delay is completely self-imposed, as I will now explain.
After the orders began pouring in, I got thinking. While thinking is often not the best use of one's time, I ignored my better judgement and proceeded to some pretty intensive thinking nonetheless. This is the curse of an over-active brain: once it gets an idea, it doesn't stop until the idea is either proven unrealistic or is realized! And the ideas that I started pondering are ideas that will benefit you all... if you can see it in your conscience to repress your eagerness long enough to hear this out.
During the first batch of orders, I received several requests for LiIon rechargeable battery support. I was a bit dismayed by this, because I knew that the circuit I had designed would absolutely MAX OUT at 8.5V - even if the rechargeables were 0.1V over their typical "full" mark, it would destroy the circuit. I was concerned enough about this that I made a special post (#38) in the Aviatrix product introduction thread, and warned potential customers that this was NOT a good option to request, unless they were already using LiIon rechargable batteries most of the time already, and not just asking for it as a "someday" extended feature that might be "nice to have just in case...". I began questioning this stance somewhere around the 4th order for rechargeable support, wondering if I had approached the design from the best possible engineering viewpoint.
This was not all that was on my mind. When the design was initially realized, I knew that I had chosen the ONLY microcontroller on the planet that was capable of performing all of the required functions in the VERY, VERY limited amount of physical space I was limited to working within. Yes... I can say that with authority, having an intimate day-to-day knowledge of electronics - being an electro/mechanical engineering consultant as my chosen profession. As a consequence of using this microcontroller, I was limited to 100mA of output current for LED drive - which had to be split between the two LEDs to be switched On/Off. This gave me 50mA per LED, which is enough to get a very good brightness out of them... but NOT enough to reach the full potential of the THC3 LEDs I had chosen for the main modded emitters. THC3 LEDs are meant to be run typically at 80mA, and run at a maximum of 100mA. This kept bothering me. The first prototype had performed MARVELOUSLY well even though I had limited the current to 50mA per LED, but I kept wondering what it would be like if I could switch in the FULL power allowed - and get nearly twice the brightness! This kept nagging and nagging at me until I figured out how to solve it.
So... flash back to last Sunday. I've sold out the initial production run of 25 units, and already have the blank printed circuit boards and parts to build them. All that's needed is time... a week worth of soldering an Aviatrix here and an Aviatrix there, and by the end of the week (Yesterday) I should have them all done and ready to ship. Right? ........Wrong!.
I couldn't let it go. I just couldn't build and ship a product that was not the BEST that I could offer, even if it cost me money (and believe me, this is starting to cost me a LOT of money... if not in components ordered, then in time away from my other contracted projects!).
I began the new design Sunday evening, not sure if it was even possible. I did research to find a voltage regulator that would solve the problems with rechargeable battery voltage limits, and I figured out a way to supply FULL POWER to the LEDs without stressing or damaging the microcontroller chip. Now, I just needed to see if I could take the new design (with all the added parts) and route a new PCB using my PCB design CAD system. Three days... three agonizing days of constant sitting EVERY HOUR for 14-16 hour days at the computer screen, laying traces, ripping them up, re-routing, moving components, starting over, you get the idea. Sheer computer assisted TORTURE trying to make it all fit into that impossibly small space I had to work with inside of an A2 Aviator bezel.
Finally, and with much pleasure, I'm here to report success! I finished the new PCB design on Wed. evening, sent it to the circuit board production plant, and waited 2 days (ordering the new parts the board required in the meantime). Today (Friday) morning, I picked up the new Aviator circuit boards and took delivery of the overnight package containing the electronic components I'd need to build them.
Just a few hours ago, I finished soldering the first prototype of the "New Improved" Aviatrix together. I chose a 1 Green/2 White LED configuration for this build... as I wanted to see what the beam pattern of that particular configuration would look like (it's quite nice!). I photographed it from every angle with a friend's borrowed digital camera, with the help of my beautful lab assistant "Henny". I have tested the new electronics (starting on a solderless breadboard shortly after completing the new PCB layout on Wed.!) to make sure there were no problems. Here are the photos... all are thumbnails, so click the thumbnail for the full-sized (1024x768) images:
Here is what YOU get for being so patient during the extra 5 days of delay waiting for your Aviatrix:
* Ability to use rechargeable batteries, any voltage/charge level you want - without fear of harming the circuitry. No more "special" mod needed to allow rechargeable support.
* Ability to drive the LED emitters to their FULL CAPABILITY - no compromise in brightness.
* Improved mechanical layout, no possibility of shorting to the bezel metal if the solder-mask coating should ever wear through from repeated installation/removal.
* Better location of the microcontroller (now fits UNDER the board on the same side as the LEDs, instead of on the top where it was occasionally interfering with the incandescent bulb ring, causing it to press further into the spring-loaded A2 electronics section behind the bezel.
* Better upgrade ability - this is the most important feature addition that I'm now going to explain in detail:
Because every LED on the new Aviatrix design is now an independently controllable output (from a software standpoint...), I can take this SAME BOARD and so some REALLY neat future versions of the Aviatrix with it. I hinted at this yesterday when I revealed my ideas for the Aviatrix Bloodhound model (red/blue with flashing alternation for hunters) in the main thread... but I have a LOT more ideas where that came from, all of which I'll reveal in due time. The point is this: If you buy an Aviatrix today, it will NOT become an obsolete product as I introduce new models and versions - for a small fee I will be able to upgrade your Aviatrix circuit board if you ever decide that you want one of the new Aviatrix models that I introduce, instead of having to buy a whole new Aviatrix from scratch. I will be able to do it ALL with this board... with a simple microcontroller swap-out.
If you've managed to read this far without falling asleep, I just want to say one more thing. You, my customer, are THE most important thing to me. I know many of you will read this with a cynical, cliche'd "yeah... right... typical marketing BS" mental response - but I want you to know that this is ABSOLUTELY the TRUTH. Without the support of my customers, I wouldn't be able to do ANY of this, and there would be no Aviatrix mods for me to sell or for you to enjoy! I really DO want YOU to have the very best thing that I can possibly produce for you, because only if YOU are happy will I get good word-of-mouth and possibly future sales. These are the reasons why I delayed everyone's Aviatrix production and delivery for a week - I'm hoping that after reading all of this you will forgive me and understand the spirit in which it was done.