This is a special heart felt appeal. I made a number of picaxe projects with my son during his middle school years around 10 years ago. I now need to make just one more. My son passed away during the middle of last year aged just 22 and I've taken it on myself to make one last project with him. The project being his headstone. I'm teaching myself how to carve granite but I'd like to put some technology in to the project too as I believe my son would appreciate that. Part of the headstone will have a couple of 'rivers' (river table idea) made from tinted blue resin to represent the tears of different family members. I'd like to make the rivers light up when it's dark. I've no problem with the programming side of this and the use of an LDR etc. What I'm not sure about is how to charge some kind of cell and how to incorporate the circuit to do this. I know I could perhaps use some kind of garden light but I'm worried about the quality aspect there and how long it might last for. I need something that can physically last as long as it can. I'll need to power between 4 - 8 10mm LED's which I suppose would be best in parallel. I'll slowly ramp these up over maybe a minute rather than coming on full like a switch. But what size cell/battery will I need, assuming these will remain on up to 12 hours in winter. What size solar cell will I need to charge up the battery/cell. I'll also need a schematic of the electrical circuit to do the job as I'm not experienced at that kind of thing. The other consideration, as I think I implied above, is that this needs to have a life as long as it can. I'm in my early 60's so I'll be able to maintain it for maybe another 20 years but there may not be anyone around to do that after I'm gone. Could something be designed to be expected to last say 100 years, or more even if over designed. In summary, I'm okay with the programming but I need help with the circuit and specification of components needed please. (The attached file is the concept sketch which shows the 'rivers' at the sides in what I'm calling 'wings' - these have been included due to the beautiful but exposed nature of the grave location, approx S39° 30.300' E176° 51.400'.
Electronics Help Needed Please (for a very special project)
- Thread starter kiwiowl
- Start date
erco
Senior Member
So very sorry for your loss, how moving that you want to make and maintain such a thoughtful memorial to your son.
I am no expert on solar power, but I share your concerns on long term parts longevity out in the elements. Your comparison to garden lights is reasonable imo. I can see the LEDs lasting for decades, but the solar cell and rechargeable battery will inevitably degrade over time. I would use premium components and allow for periodic replacement after 5-10 years, obviously using a weatherproof, tamper-proof access panel on top with the solar cell.
Circuit-wise, I will defer to other experts here. But I would offer tbat a reasonable place to start might be to dissect some good quality garden lights. The solar cell charges the battery and doubles as the light sensor to switch the LEDs on, no LDR required. Since this sounds like a.long term development project, you have time to test various solar cell, battery and LED configurations.
God bless you and your family at this most difficult time.
I am no expert on solar power, but I share your concerns on long term parts longevity out in the elements. Your comparison to garden lights is reasonable imo. I can see the LEDs lasting for decades, but the solar cell and rechargeable battery will inevitably degrade over time. I would use premium components and allow for periodic replacement after 5-10 years, obviously using a weatherproof, tamper-proof access panel on top with the solar cell.
Circuit-wise, I will defer to other experts here. But I would offer tbat a reasonable place to start might be to dissect some good quality garden lights. The solar cell charges the battery and doubles as the light sensor to switch the LEDs on, no LDR required. Since this sounds like a.long term development project, you have time to test various solar cell, battery and LED configurations.
God bless you and your family at this most difficult time.
Last edited:
Sorry for your loss - and you idea for remembrance seems poignant.
From my limited experience with solar power 1) solar-charged lights in the equipment shed out back, now in their 6th year of use and 2) 6-year-old solar-charged backup power for 8 to 14 hours for our frequent but usually short (under 8 hours) power outages. The shed uses a 12 volt LiFePO4 battery; the solar backup system uses a bank of AGM batteries.
First, you need to know what you area's highest and lowest temperatures will be - this will determine which type of battery is needed. Most likely that will be LiFePO4 (lithium iron phosphate) or LTO (lithium titanate oxide).
Second, you need to know the power required by the PICAXE when idle - it DOES use some power when just waiting. The solar charge controller will also use some power during the dark hours and that must be included.
Third, you need to measure how much power the LEDs will use per hour and design for worst case of having long winter nights and consecutive cloudy / rainy / foggy / whatever weather that might block the sunlight from your solar panel(s) for perhaps a week - yes, I'm a pessimist about weather and solar power but designing for the worst case means things are more likely to continue working. Figure on 16 hours without ANY sun on most winter days and that your most likely solar power hours will be the two hours on either side of solar noon (NOT noon clock time). That can be determined here for the US and their maps may cover more than that: https://www.esrl.noaa.gov/gmd/grad/solcalc/ There should be similar sites for the rest of the world.
Fourth, how much sun will you get each day? This site works for the US/Canada (if it's up): solarelectricityhandbook.com/solar-irradiance.aspx
There is also a site that gives solar irradiance for Europe/Asia but I don't have that link at hand.
Fifth, you'll need to how many (how few?) useful sun hours your location will have during the worst months of winter so you can determine how many hours the LEDs will be lit and thus how much battery capacity you will need - for the LEDs when lit and for the PICAXE 24/7/365. Regardless of battery type, plan your design around not using more than 50% of the battery's capacity. Why? Batteries of ALL types lose some capacity as they age and you want this to continue working for perhaps 50 years?
When you know how much power (volts * amps * hours for watt hours), you can determine how big a battery you will need. I can't help much with the numbers because I have no idea how much current the LEDs need or how many hours/day they will be lit on the worst case days. If you're in northern Scotland, you'll have much less sun than on a beach in Spain.
Current production batteries would be LiFePO4 or LTO, with the LTO having a slight advantage in low temperature operation (remember all the electric cars that couldn't be charged in the snow in Chicago in January of 2024?). The LTO is more expensive for the same capacity (amp hours: AH) but they could have a life of 20,000 charge/discharge cycles - that's a little over 54 years at one charge/discharge cycle per day. The highest
number of cycles I've seen on an LiFePO4 battery is 12,000 cycles - that's a little over 32 years. When sodium ion (Na) batteries become commonly available (something other than $15 samples from the manufacturer with a $150 shipping charge), they have even better cold weather performance.
"Good" solar panels - those from a known maker - typically have a warranty of something like 80% of original output after 20 years. Plan for the solar panel(s) to age and include that "fudge factor" in your "How big?" calculations - you want this to keep working for a LONG time.
My mental design for the solar power would have a battery sized to provide two weeks of power with NO sun. Because you need a solar charge controller of some type to manage the charging of the battery and a multi-cell battery LiFePO4 or LTO battery would need a BMS, you need to put some serious thought into what voltage the battery should be. A single LiFePO4 cell is 3.65 volts when fully charged and 2.5 to 3.0 volts is considered discharged. Obviously, you will need at least 3 volts for the PICAXE and whatever voltage is needed by the LEDs.
You also need to think of how you will mount the solar panel(s) - maybe one or two 20 watt panels and remember I'm designing for worst case winter - and how you will weatherproof the electronics, the wiring and the connectors. Also think of ways to theft-proof the box containing the electronics, the battery and the solar panel(s). If the headstone faces North, the solar panel(s) could be epoxied to the back and facing South for optimum solar exposure. However, that would have the panel(s) vertical and possibly not at the optimum angle for your location on the Earth - but they would be almost theft-proof. If this is a family cemetery, you could probably put up a pole for mounting the solar panel(s) in the clear, facing South and at the optimum angle for your location.
This is going somewhat off topic as it's more about solar power than the PICAXE so PM me if you have questions. I have some spreadsheets for a small solar system and one of those might be adaptable to your needs. Include your general location, how many LEDs, what their typical current is and what voltage you plan on using for the PICAXE. Remember that the most common commercial battery packs are 12 volts and UP and that's what the BMS units are designed for so your battery might be 12 volts and everything would be powered by a voltage regulator or, even better, an efficient buck regulator - and you have to add those losses to the amount of battery capacity and size of the solar panel(s).
I'm not looking for an inexpensive way to do this - I'm looking for a way that lasts almost forever.
From my limited experience with solar power 1) solar-charged lights in the equipment shed out back, now in their 6th year of use and 2) 6-year-old solar-charged backup power for 8 to 14 hours for our frequent but usually short (under 8 hours) power outages. The shed uses a 12 volt LiFePO4 battery; the solar backup system uses a bank of AGM batteries.
First, you need to know what you area's highest and lowest temperatures will be - this will determine which type of battery is needed. Most likely that will be LiFePO4 (lithium iron phosphate) or LTO (lithium titanate oxide).
Second, you need to know the power required by the PICAXE when idle - it DOES use some power when just waiting. The solar charge controller will also use some power during the dark hours and that must be included.
Third, you need to measure how much power the LEDs will use per hour and design for worst case of having long winter nights and consecutive cloudy / rainy / foggy / whatever weather that might block the sunlight from your solar panel(s) for perhaps a week - yes, I'm a pessimist about weather and solar power but designing for the worst case means things are more likely to continue working. Figure on 16 hours without ANY sun on most winter days and that your most likely solar power hours will be the two hours on either side of solar noon (NOT noon clock time). That can be determined here for the US and their maps may cover more than that: https://www.esrl.noaa.gov/gmd/grad/solcalc/ There should be similar sites for the rest of the world.
Fourth, how much sun will you get each day? This site works for the US/Canada (if it's up): solarelectricityhandbook.com/solar-irradiance.aspx
There is also a site that gives solar irradiance for Europe/Asia but I don't have that link at hand.
Fifth, you'll need to how many (how few?) useful sun hours your location will have during the worst months of winter so you can determine how many hours the LEDs will be lit and thus how much battery capacity you will need - for the LEDs when lit and for the PICAXE 24/7/365. Regardless of battery type, plan your design around not using more than 50% of the battery's capacity. Why? Batteries of ALL types lose some capacity as they age and you want this to continue working for perhaps 50 years?
When you know how much power (volts * amps * hours for watt hours), you can determine how big a battery you will need. I can't help much with the numbers because I have no idea how much current the LEDs need or how many hours/day they will be lit on the worst case days. If you're in northern Scotland, you'll have much less sun than on a beach in Spain.
Current production batteries would be LiFePO4 or LTO, with the LTO having a slight advantage in low temperature operation (remember all the electric cars that couldn't be charged in the snow in Chicago in January of 2024?). The LTO is more expensive for the same capacity (amp hours: AH) but they could have a life of 20,000 charge/discharge cycles - that's a little over 54 years at one charge/discharge cycle per day. The highest
number of cycles I've seen on an LiFePO4 battery is 12,000 cycles - that's a little over 32 years. When sodium ion (Na) batteries become commonly available (something other than $15 samples from the manufacturer with a $150 shipping charge), they have even better cold weather performance.
"Good" solar panels - those from a known maker - typically have a warranty of something like 80% of original output after 20 years. Plan for the solar panel(s) to age and include that "fudge factor" in your "How big?" calculations - you want this to keep working for a LONG time.
My mental design for the solar power would have a battery sized to provide two weeks of power with NO sun. Because you need a solar charge controller of some type to manage the charging of the battery and a multi-cell battery LiFePO4 or LTO battery would need a BMS, you need to put some serious thought into what voltage the battery should be. A single LiFePO4 cell is 3.65 volts when fully charged and 2.5 to 3.0 volts is considered discharged. Obviously, you will need at least 3 volts for the PICAXE and whatever voltage is needed by the LEDs.
You also need to think of how you will mount the solar panel(s) - maybe one or two 20 watt panels and remember I'm designing for worst case winter - and how you will weatherproof the electronics, the wiring and the connectors. Also think of ways to theft-proof the box containing the electronics, the battery and the solar panel(s). If the headstone faces North, the solar panel(s) could be epoxied to the back and facing South for optimum solar exposure. However, that would have the panel(s) vertical and possibly not at the optimum angle for your location on the Earth - but they would be almost theft-proof. If this is a family cemetery, you could probably put up a pole for mounting the solar panel(s) in the clear, facing South and at the optimum angle for your location.
This is going somewhat off topic as it's more about solar power than the PICAXE so PM me if you have questions. I have some spreadsheets for a small solar system and one of those might be adaptable to your needs. Include your general location, how many LEDs, what their typical current is and what voltage you plan on using for the PICAXE. Remember that the most common commercial battery packs are 12 volts and UP and that's what the BMS units are designed for so your battery might be 12 volts and everything would be powered by a voltage regulator or, even better, an efficient buck regulator - and you have to add those losses to the amount of battery capacity and size of the solar panel(s).
I'm not looking for an inexpensive way to do this - I'm looking for a way that lasts almost forever.
Greetings -that lat./long. is ~Taupo,half way up NZ's North Island.
Very sorry to hear of your loss but (in light of my experiences) if the memorial is in a public cemetery best first consult with the authorities regarding anything of this nature! Some have strict rules about headstones... Stan. - Wellington,NZ
Very sorry to hear of your loss but (in light of my experiences) if the memorial is in a public cemetery best first consult with the authorities regarding anything of this nature! Some have strict rules about headstones... Stan. - Wellington,NZ
Last edited:
PhilHornby
Senior Member
If I drop those coords. (aka -39.505000, 176.856667) into Google Maps, I get Western Hills Lawn Cemetery in the Napier area...
I bow to your cartography. Such a formal resting place's rules certainly should be investigated. See => https://www.napier.govt.nz/services/napier-cemeteries/western-hills-cemetery/ (Some decades back we lost a daughter & found our desired headstone did not fit the cemetery's criteria ...).
At a technical level I too endorse LiFePO4 batteries but caution on the longitivity of PVs in such a high sunshine (& UV) region like Napier. Stan.
At a technical level I too endorse LiFePO4 batteries but caution on the longitivity of PVs in such a high sunshine (& UV) region like Napier. Stan.
Last edited:
Thanks for your concern. I've already got clearance from the Sextant for the design. The thing is, the cemetery is 'locked' when it's dark, not sure many people will see the blue glow, but that's not the point.
Thanks for your extensive reply and the time you have spent to write it. It will take me a while to read and digest this, but thanks again.
Hi,
A very thought-provoking project on which a (large) book could be written, to cover all the technical aspects alone. So I'll start with just a few "bullet points" (or Chapter headings) and you can ask for further details on any particular features of interest.
No, an operational lifetime of 100 years cannot be "expected", but there is plenty that can (and needs) to be done to give a reasonably extended lifetime. Many electronics components have a typical functional lifetime of perhaps 20 years, with batteries in particular, generally less than that. Even the "Data Retention Time" of the Program Memory in a PIC(axe) chip is specified as "typically 40 years" (TRETD in the Data Sheet) with no Max or Min limits quoted.
Perhaps you should be looking at (some of) the methods used in the design of space satellites, for example the duplication of ALL* hardware modules, each cross-linked with the (automatic) ability to detect and isolate faulty units. Strictly, sensors (excluding cameras) should be triplicated so that a "majority decision" can be made if two sensors disagree. But, as said above, a LDR is not needed because the output from the Solar Panel(s) can indicate the ambient light level directly**. Even satellite lifetimes are normally quoted in years rather than tens of years, noting that the "deep space probes" that have run for around 50 years don't use solar panels nor storage batteries, but nuclear reactors! However, a look through the "Picaxe in Space" ($50 Sat) thread on this forum (which ended here) might be rewarding. See also This Thread which ran a PICaxe and flashed a LED on 3 AA Alkaline cells for 11 years!
** PV panels basically have a "Constant Voltage" output (except in darkness or when overloaded, of course), around 500 mV for each "pane" (silicon diode) connected in series, where the maximum available output current increases proportionally with the light level. Typical Efficiency is up to 20% (nearer 10% for Amorphous Silicon types) from a maximum summer solar illumination of 1000 Watts/square meter at right angles to the panel surface, or around 400 W/m2 for direct sunlight in winter (due to a longer path though the atmosphere). A good quality "roof" panel could last for about 25 years (to around a 25% reduction of output) but it might be difficult to find small, high quality panels. The low-cost "resin" encapsulated panels have been reported to go "cloudy" (or milky) within just a few years, so this material might be best avoided.
It may depend if you have any control over the orientation of the headstone, but I suggest vertically mounted PV panel(s) facing North. The reason is that maximising the Winter power output is far more important than achieving a high average annual value. In Winter, the sun's maximum elevation (at Latitude 40) is about 27 degreees above the horizon (i.e. 90 - 40 - 23 degrees Earth tilt), and lower when away from noon or from north. Thus the optimum slope is about 20 degrees from vertical, but the Cosine of 20 degrees is 0.94 and the Sine 0.34, so a Vertical panel will "lose" only about 6% of the available light energy compared with a Horizontal panel's 66% (or worse, taking into account reflection off the acute-angled glass surface(s) ). And perhaps as important is that a Vertical panel will restrict the high Summer energy where the problem could be over-heating, or over-charging the battery. Note that disconnecting (or short-circuiting) a PV panel can't change the amount of energy entering the panel, so if the avaiable electricity is not "used" it just adds to heat build-up in the panel/electronics.
As said above, the Lithium Iron Phosphate (LiFePO4) cell technology is currently, probably the most appropriate for long life. It's nominal voltage of 3.2 volts is more than enough for a PICaxe, which can also easily measure its own internal supply voltage (using the CALIBADC10 command) and can control the (overall average) power drain to keep the battery voltage within (my recommendation of) a range of 3.0 to 3.5 volts, without any other hardware. Again, the battery capacity should be at least a week, to accommodate winter light levels and the number of cycles over a period of tens of years. It may even be appropriate to include a yearly cycle component, perhaps with a secondary battery optimised for a smaller number of deep cycles.
Personally, I would consider a few modifications to the concept. Firstly could the "tears" actually "fall" downwards ? Cycling the LEDs (i.e. switching on one or two at a time) will reduce the power drain considerably and "flashing" a LED is much more visible than a steady dim light. Also, and in view of your comment in #8, might there be a "daytime" mode? Of course the power consumption will be much higher, but no battery is required to store the energy and there is a good correlation between the amount of power supplied by the PV panels and the electricity needed to make the LEDs visible at the corresponding ambient light level. One of the tricks to obtain good visibility in daylight is to ensure that the background (i.e. LED Off state) is as "black" as possible by reducing (or at least diffusing) all reflected light. This can include adding a dark cover filter, because reflected light undergoes two attenuating passes through the filter, whilst the LED output only one.
* Implementing the duplication of PV panels is quite easy because each can be connected via a forward diode (which may be already present within the panel) to prevent a fault loading the other PV(s). Duplicated LED(s) (with their own drivers) don't even need to be electrically connected; they can be simply located close to each other. Ideally, duplicated PICaxes should run separately-developed software, since "bugs" can be the greatest cause of a system failure. But the greatest problem is the batteries which probably have the lowest reliability and often fail to a short-circuit (or high leakage). Thus they must be capable of being (automatically) switched away from both the charging source and the load, to allow these to continue with any duplicated unit(s).
Particularly if you don't want the complication of a fully duplicated topology, you might construct two totally independent "day" and "night" systems, sharing only the PV panel(s) and removing the potential unreliability of a battery from the daylight system. Of course the software versions would be totally different, with the daylight system re-starting every morning, avoiding any risk from a memory leak or the need for additional "watchdog" monitoring. A truly "belt and braces" approach !
Cheers, Alan.
A very thought-provoking project on which a (large) book could be written, to cover all the technical aspects alone. So I'll start with just a few "bullet points" (or Chapter headings) and you can ask for further details on any particular features of interest.
No, an operational lifetime of 100 years cannot be "expected", but there is plenty that can (and needs) to be done to give a reasonably extended lifetime. Many electronics components have a typical functional lifetime of perhaps 20 years, with batteries in particular, generally less than that. Even the "Data Retention Time" of the Program Memory in a PIC(axe) chip is specified as "typically 40 years" (TRETD in the Data Sheet) with no Max or Min limits quoted.
Perhaps you should be looking at (some of) the methods used in the design of space satellites, for example the duplication of ALL* hardware modules, each cross-linked with the (automatic) ability to detect and isolate faulty units. Strictly, sensors (excluding cameras) should be triplicated so that a "majority decision" can be made if two sensors disagree. But, as said above, a LDR is not needed because the output from the Solar Panel(s) can indicate the ambient light level directly**. Even satellite lifetimes are normally quoted in years rather than tens of years, noting that the "deep space probes" that have run for around 50 years don't use solar panels nor storage batteries, but nuclear reactors! However, a look through the "Picaxe in Space" ($50 Sat) thread on this forum (which ended here) might be rewarding. See also This Thread which ran a PICaxe and flashed a LED on 3 AA Alkaline cells for 11 years!
** PV panels basically have a "Constant Voltage" output (except in darkness or when overloaded, of course), around 500 mV for each "pane" (silicon diode) connected in series, where the maximum available output current increases proportionally with the light level. Typical Efficiency is up to 20% (nearer 10% for Amorphous Silicon types) from a maximum summer solar illumination of 1000 Watts/square meter at right angles to the panel surface, or around 400 W/m2 for direct sunlight in winter (due to a longer path though the atmosphere). A good quality "roof" panel could last for about 25 years (to around a 25% reduction of output) but it might be difficult to find small, high quality panels. The low-cost "resin" encapsulated panels have been reported to go "cloudy" (or milky) within just a few years, so this material might be best avoided.
It may depend if you have any control over the orientation of the headstone, but I suggest vertically mounted PV panel(s) facing North. The reason is that maximising the Winter power output is far more important than achieving a high average annual value. In Winter, the sun's maximum elevation (at Latitude 40) is about 27 degreees above the horizon (i.e. 90 - 40 - 23 degrees Earth tilt), and lower when away from noon or from north. Thus the optimum slope is about 20 degrees from vertical, but the Cosine of 20 degrees is 0.94 and the Sine 0.34, so a Vertical panel will "lose" only about 6% of the available light energy compared with a Horizontal panel's 66% (or worse, taking into account reflection off the acute-angled glass surface(s) ). And perhaps as important is that a Vertical panel will restrict the high Summer energy where the problem could be over-heating, or over-charging the battery. Note that disconnecting (or short-circuiting) a PV panel can't change the amount of energy entering the panel, so if the avaiable electricity is not "used" it just adds to heat build-up in the panel/electronics.
As said above, the Lithium Iron Phosphate (LiFePO4) cell technology is currently, probably the most appropriate for long life. It's nominal voltage of 3.2 volts is more than enough for a PICaxe, which can also easily measure its own internal supply voltage (using the CALIBADC10 command) and can control the (overall average) power drain to keep the battery voltage within (my recommendation of) a range of 3.0 to 3.5 volts, without any other hardware. Again, the battery capacity should be at least a week, to accommodate winter light levels and the number of cycles over a period of tens of years. It may even be appropriate to include a yearly cycle component, perhaps with a secondary battery optimised for a smaller number of deep cycles.
Personally, I would consider a few modifications to the concept. Firstly could the "tears" actually "fall" downwards ? Cycling the LEDs (i.e. switching on one or two at a time) will reduce the power drain considerably and "flashing" a LED is much more visible than a steady dim light. Also, and in view of your comment in #8, might there be a "daytime" mode? Of course the power consumption will be much higher, but no battery is required to store the energy and there is a good correlation between the amount of power supplied by the PV panels and the electricity needed to make the LEDs visible at the corresponding ambient light level. One of the tricks to obtain good visibility in daylight is to ensure that the background (i.e. LED Off state) is as "black" as possible by reducing (or at least diffusing) all reflected light. This can include adding a dark cover filter, because reflected light undergoes two attenuating passes through the filter, whilst the LED output only one.
* Implementing the duplication of PV panels is quite easy because each can be connected via a forward diode (which may be already present within the panel) to prevent a fault loading the other PV(s). Duplicated LED(s) (with their own drivers) don't even need to be electrically connected; they can be simply located close to each other. Ideally, duplicated PICaxes should run separately-developed software, since "bugs" can be the greatest cause of a system failure. But the greatest problem is the batteries which probably have the lowest reliability and often fail to a short-circuit (or high leakage). Thus they must be capable of being (automatically) switched away from both the charging source and the load, to allow these to continue with any duplicated unit(s).
Particularly if you don't want the complication of a fully duplicated topology, you might construct two totally independent "day" and "night" systems, sharing only the PV panel(s) and removing the potential unreliability of a battery from the daylight system. Of course the software versions would be totally different, with the daylight system re-starting every morning, avoiding any risk from a memory leak or the need for additional "watchdog" monitoring. A truly "belt and braces" approach !
Cheers, Alan.
Thanks Alan for your extensive answer. A lot here for me to digest and think about. Time is on my side so that's a good thing.
Just in case you hadn't thought of luminous paint.
The other solutions proposed look a bit daunting.
The other solutions proposed look a bit daunting.
gertvdwalt
New Member
I might be late to the conversation but can I suggest using super-capacitors as the energy storage? When used at lower rated voltages they last a long time. They are also quite easy to charge.
Tritium illumination will gradually dim by 1/2 every 12.33 years but they are available in blue and have the significant advantage that there are no batteries or electronic components that could fail prematurely.
Last edited:
Tritium is very commonly available for use in things that you want to "glow in the dark".
Do a google search for "tritium markers" to see what I mean.
I own one of the small ones attached to my keyring that must be about 10 years old and it still stands out clearly in a pitch black bedroom.
However, it does occur to me that these will not be anywhere near as visible on moonlit nights so if kiwiowl is interested in this he would need to buy one of the brightest and test it to see if it was bright enough for him.