EasilyLed,
Perhaps I can provide one perspective on your question. As you may know, we recently announced our second titanium flashlight run. The first one was quite the learning experience even though we only made a handful. It took a long time before we were willing to look at titanium again.
There is little doubt that working with titanium is much more expensive than working with aluminum. This is due to higher material cost, higher machining costs and additional insulation requirements because there is no insulating anodize coating the surface. One benefit of titanium in this regard is that there are no anodizing costs and no post anodize operations to be performed. However, the net effect is that a titanium flashlight will be significantly more expensive to manufacture.
When you significantly raise the price of a flashlight by going with titanium, you dramatically reduce the possible market size. Markets are price sensitive, no matter what you sell. This forces you into a much smaller production run, which in turn raises the price due to dropping economies of scale. Producing one of something can easily be 100x to 200x more expensive than the per piece cost when producing a thousand of the same thing. The cost difference is even higher if you add in the cost of R&D (research and development). This is in addition to the costs previously mentioned.
So, let's go back to your original premise - the modular light. There are several items that might be considered candidates for "modules". The LED. The reflector. The power supply. The battery compartment. The button and perhaps the switch. Then comes the question whether or not modules should be swappable without tools.
In the case of the LED, swapping without tools becomes difficult if you want to ensure a good electrical connection and a good thermal path. The same probably holds for the power supply. Most of the other "modules" can probably be designed to thread together and thus swapping modules without tools is practical for those other modules.
The main reason to swap the LED is to avoid obsolescence. LED technology is progressing at roughly 20 to 25% per year. Although one year's increase is not enough for the average person to see, two or three years can produce a substantial increase. Due to the thermal, electrical and optical issues involved, I would argue that the LED "module" should not be considered a user-serviceable part. This is why we are planning to offer an LED upgrade when sufficiently good LEDs become available.
The main reason to swap the electronics is to add features to the flashlight or to make the electronics more efficient. Generally, the original design will have set the maximum power due to size and thermal considerations - something that swapping modules cannot fix. We don't consider the power supply a user-serviceable part. We are probably unique in the flashlight industry because we have a fully software controlled power supply and user interface. We already have the highest power efficiencies in the industry so there is probably no need to worry about increasing the power supply efficiency. And since we can upgrade the software, we can always make improvements to our existing power supply and user interface design without changing the hardware. I.e., we can add all of the latest features to the older power supplies. Whenever you get a new (higher output) LED, you automatically get one or more additional brightness levels, the latest features and a fully recalibrated flashlight.
Our design is capable of swapping reflectors, swapping battery compartments and swapping buttons (by swapping the switch cap). The singular issue is this: are there enough customers willing to pay the price for these modules to make it cost effective to manufacture them?
I submit that our design is fully capable of doing exactly what you are asking for. I also submit that the market may not be large enough to justify manufacturing the other modules.
A few years ago a CPF member was going to manufacture an extra battery tube for one of our products. We were willing to help him with technical information so that everyone would be sure that his custom part would work. He started an interest thread and got around 30 hits, as I remember. He then went out for quotes to manufacture the part. He was floored by how much it was going to cost. It came out to over $50 each for 50 units, as I remember. And that was in aluminum. Needless to say, the project got scrapped.
If I was assured that I could sell a reasonable number of accessory parts for a profit, I would not be reluctant to make them. The reluctance is in taking the risk of manufacturing parts that no one will buy.
On a side note, as others have mentioned, the thermal characteristics of titanium will not be a significant limiting factor in a good design. Even if a pure titanium design cannot be accommodated for thermal reasons, a hybrid design using a copper or aluminum spreader next to the LED can fix a marginal design and still allow for a full titanium exterior. Our titanium heat sink has no problems dissipating the heat.
Henry.
