Above, the Luminiser lantern being powered by a candle (left), and that candle alone, right.
In my just-prior post, I worked out the basic efficiency numbers for the Luminiser candle-powered electric lantern. It’s vastly more efficient than, say, a mantle-based oil lamp, such as an Aladdin (r) lamp.
I was so struck by how well the thing worked …
Scratch that. I was so struck that the thing worked, at all, that I neglected to show any numbers on operating costs. Let me fix that now.
If you are “on the grid”, nothing is as cheap as plugging an LED lamp into the wall. No surprise there. Not by a longshot.
But suppose that, as a moral issue, you would not allow the general use of electricity in your home. You live in a home that is not merely “off the grid”, but one that is purposefully and thoughtfully un-electrified. For the sake of argument, let’s say you would selectively allow battery-powered devices, when useful and necessary. A flashlight, for example, might be OK, but a battery-powered television would not. But you had to use disposable (alkaline) batteries, for such devices, because there’s no place to charge your rechargeable batteries.
That’s all by way of setting up the comparison. How would the operating cost of this candle-powered lantern stack up against that of a standard battery-operated lantern or flashlight using cheap, disposable AA alkaline batteries?
Turns out, depending on what you burn in your Luminiser, it’s either vastly cheaper or merely a lot cheaper, than producing the same amount of light with disposable AAs.
I didn’t expect that, and I find it kind of interesting. Despite the seemingly Rube Goldberg nature of this device — you use the heat of a little oil lamp (“oil candle”) to run a thermo-electric generator, to power some LEDs — the running cost of this is vastly lower than using disposable AAs in a flashlight.
Here are the results of my cost calculation. Assuming I haven’t slipped a decimal point somewhere, the Luminiser powered with ordinary gas-pump K1 kerosene costs about 3% as much to run as a battery-operated lantern powered with disposable batteries.
Description of the calculation follows.
A few key details
This calculation assumes the following prices, current as of January 2024:
- 33 cents per AA battery, based on a box of 60 currently at Home Depot.
- $5/gallon for K1 kerosene (roughly the U.S. national average right now).
- $15/gallon for Kleen Heet deodorized kerosene (Home Depot price).
- $30/gallon for paraffin oil (the finest fuel for flat-wick oil lamps), based on the current ACE Hardware price.
For the output of the Luminiser, I’m just accepting the manufacturer’s specs of 200 lumens, for 8 hours, using one 44 milliliter oil candle.
The only hard-to-pin-down unknown is how many lumen-hours you can squeeze out of the typical disposable alkaline AA battery. This is hard to pin down from manufacturers’ published data for many reasons, not the least of which is that they’ll lie. But in addition, modern flashlights contain circuits that will turn down the brightness if they are left on. And, they’ll get dimmer as they run, in any case. Manufacturers tend to publish data on maximum brightness, and then on run time, where (unstated) the run time is mostly at some much lower brightness. This means you can’t just multiply published lumen numbers by published run time numbers. That will typically vastly overstate actual light output.
In my post on candles and lanterns I used an example of a real-life device that produced about 300 lumen-hours per AA battery. That was a marine distress signal, and likely had been optimized for long battery life. Similarly, this Nitecore flashlight works out to about 250 lumen-hours per AA battery, on low.
The figure of 300 lumen-hours for a typical AA alkaline battery is consistent with a typical AA alkaline battery capacity of 3 watt-hours of energy, and an overall energy efficiency of LED/driver circuit of 100 lumens per watt. The AA alkaline capacity figure is pretty much a known, and the lumens-per-watt figure is at the high end of the current crop of off-the-shelf hardware-store lights. (E.g., 90 lumens per watt for these LED bulbs (Home Depot reference).
Close enough for this kind of calculation.
Addendum: Re-using/replacing the oil candle.
Edit 1/22/2024: One day later, and this is obsolete. See next post for making the permanent refillable replacement for these.
Above: Original oil candle, 3/16 twist drill, glue syringe, and tiny drill (for air hole).
Below: Luminiser burning with factory-original candle, and with refilled candle.
The key to operating this cheaply is to use some sort of re-fillable oil lamp to power it. As shipped, the device comes with a small disposable oil lamp (“oil candle”). That’s engineered to work correctly with this device, but is an expensive way to produce light. To run it cheaply, you need to a way to use off-the-shelf kerosene or lamp oil to power this.
I did the obvious thing and demonstrated that I can, in fact, refill the little disposable oil lamp that comes with the light. At least once. Drill a hole just big enough for a glue syringe, drill a second smaller hole for to release air, fill the syringe with lamp oil, and inject in into the oil candle.
That works fine. Light might be a touch dimmer, consistent with using lamp oil (paraffin oil) for the refill, which by reputation will not burn as hot as kerosene. But if it is dimmer, it’s not dimmer enough to matter.
Lamp oil and kerosene have high flash points, so I’m not terribly worried about the little open holes in the shoulder of the oil candle. Other than as a spill risk. Pretty sure the plastic enclosure (and the plastic oil candle itself) would melt before it got hot enough to flash over the raw lamp oil.
But the wick on these disposable “oil candles” does not appear to be adjustable. Or, at least, not without a lot of effort. So this looks like it may work once or twice, but not indefinitely.
In the long run, I’m probably going to adapt one of my small (night-light-sized) oil lamps for this purpose. These lamps are just a few inches tall, and take a round cotton-cord wick instead of a traditional flat oil-lamp wick. They can produce a flame that’s about the size of the flame produced by this oil candle. So, by inference, they should be just about exactly hot enough to run this device as the oil candle does.
The Luminiser seems like a robust device, in terms of fuel source. People have run it successfully using a variety of setups for candles, for example. Separately, I got it to run by simply sitting it on top of the chimney of one of those miniature oil lamps. It’s no surprise that refilling the disposable oil candle with lamp oil works well.
At this point, I’m sure I can find a setup that is both convenient and works well. But I need to work up something a little more permanent, and a little less hazardous, than any of these makeshift solutions. I should probably also muck about with the electrical side a bit. For example, see if I could I gin up a USB charger circuit, and splice it into this. But that’s for another day.
I’m not usually one to fawn over technology. But I am reminded of Arthur C Clarke’s dictum: Any sufficiently advanced technology is indistinguishable from magic. I mean, I know how it works — as discussed in the last post, it’s a TEG. But at a gut level, you feed this gizmo a little tiny candle flame, and it spits out enough light to read by. Not magic, but it sure looks like it.
