Glowing Fitzroy Storm Glass





Introduction: Glowing Fitzroy Storm Glass

About: I like to tinker and experiment with electronics, robotics, programming, and photography. Along with my latest interest in Steampunk.

The Fitzroy Storm Glass is a weather instrument that became popular in the 1860's for predicting the weather.  It is a sealed glass vial that contains chemicals that crystallize in various formations depending on changes in the weather. Admiral Fitzroy, Captain of the HMS Beagle, conducted a study and perfected the formula during the years of 1831 to 1836. The exact origin of the instrument is unknown and believed to have been invented around the 1700's. Because the instrument is somewhat fragile, I wanted to make something that would allow hanging on a wall and safely illuminate for observation. The hanger is made to prevent the storm glass from tipping over. White leds with a variable dimmer using a LM317 are used for illumination. I got the storm glass from Klockit

Step 1: General Layout

The parts for the wooden hanger is made from 1/2 inch hardwood plywood. Three pieces are used. A back plate, base shelf and a piece midway to stabilize the tube on the base.

Step 2: Base and Shelf

Two pieces are used for the base and shelf. All 1/2 inch hardwood plywood. The base has a portion bored out to hold the base of the storm glass. I used a 2 inch hole saw to bore half-way down then a 1-1/2 inch flat spade bit to take out more area.  A screw driver or chisel can be used to clean out the rest. The top shelf, uses a 1 inch bore saw to make the hole.

Step 3: Holes for Leds

Drill 3 holes in the base for the leds.  Wire the leds in parallel and attach a paired lead to run to the led dimmer. Use hot melt glue to hold leds and connections underneath the base.

Step 4: Back Panel Holes

Four holes need to be drilled on the back plate to mount the base and shelf. Use 1-1/4 flat head screws and pre-drill holes slightly smaller than the screw's diameter to prevent the wood from splitting. Countersink the holes to keep the screws flush on the back. Attach a screw, then carefully square-up the assembly before attaching the second screw. This will allow accurate and square placement of each piece.

Step 5: LED Controller

I used a LM317 regulator to control the brightness of the leds. The controller is helpful when photographing crystal growth. The wide range of contrast, makes photographing the crystals a challenge. The controller was built with a plastic box with a two-conductor cable connecting to the storm glass holder. A 9 to 12 volt wall-wort style power supply was used for power.  Current can be adjusted from 1 to 20 milliamps for the leds. Series operation of the leds is possible, but will require at least 15 volts for the power source. The balance between the leds seems to be fine when running in parallel. Follow the schematic and build the circuit on perf-board and place inside a suitable box.

Step 6: Schematic

The parts in the schematic are not super critical.  The values for R1 and R2 set the brightness. The capacitor C1 can be any value from 10uf to 1000uf as long as the rated voltage is higher than the power supply. Using connectors for J1 and J2 makes it easier to build. Use what you have.

Step 7: Crystal Growth

I made a movie of the crystal growth before I started making the holder.  Having a controlled lighting source such as this, would have been helpful.

Step 8: Conclusions

Different woods could be used for the holder. Lowes had some red oak 4" x 2' x 1/2 inch thick hobby boards for less that $4. I used the plywood because I had some left over from an earlier project. You can apply a stain and finish to complete the project.

Admiral Fitzroy made these observations as regards to predicting the weather:
The formula is also mentioned at the above site.

    If the liquid in the glass is clear, the weather will be bright and clear.
    If the liquid is cloudy, the weather will be cloudy as well, perhaps with precipitation.
    If there are small dots in the liquid, humid or foggy weather can be expected.
    A cloudy glass with small stars indicates thunderstorms.
    If the liquid contains small stars on sunny winter days, then snow is coming.
    If there are large flakes throughout the liquid, it will be overcast in temperate seasons or snowy in the winter.
    If there are crystals at the bottom, this indicates frost.
    If there are threads near the top, it will be windy.

Over the weeks I have been observing the changes of the crystals along with the weather. The wider the temperature changes are, the more changes I see in the crystal growth. If the temperature drops rapidly, crystals will start to snow inside the glass. Placing the storm glass outside in a protected porch would show the most effects from the ambient weather.  Can you predict the weather with this device? Perhaps, but I would still check with weather reports on TV or the internet.



    • Casting Contest

      Casting Contest
    • Oil Contest

      Oil Contest
    • Make it Move Contest

      Make it Move Contest

    We have a be nice policy.
    Please be positive and constructive.




    Any chance of this freezing solid if left outside.

    Was looking for some pictures of each “prediction” levels?


    Nice instructable. Have you had the Fitzroy long enough to form an opinion of it? I am interested in getting one if it works reasonably well.

    5 replies

    If it gets warm enough during the day, most of the crystals will be dissolved and the solution will be supersaturated. If the temperate drops fast enough, such as if a storm was coming in, crystals will fall out a "snow". This is shown in my video. So perhaps it may work to predict a approaching storm. During the summer, when its hot, the crystals should stay well dissolved. I eventually plan to do a long term time lapse video of the glass along with the sky. To predict the weather, I'll trust weather reports. The Fitzroy weather glass, along with a bit of history, still presents an interesting thing to watch.


    I was reading that Fitzroy storm glass was also highly affected by humidity. Apparently it was a mandatory piece of equipment in lighthouses in the British Empire for a long period of time.

    I can't see how humidity can effect the instrument since its in a sealed glass tube. More likely the effects of weather that comes along with different levels of humidity is affecting the glass. It is interesting to watch the changes by way of time lapse photography. Right now I have a time lapse session running. The weather has been warm, so not a lot of crystal has occurred.

    Thanks, I didn't really expect it to predict the weather, but I think it would be interesting to watch the crystals form and dissolve.

    I received a Storm Glass as a gift last Christmas - exactly like the one pictured above. Its kind of a gimmick-y thing. It really only responds to temperature, as you might imagine something inside a sealed glass cylinder might. It does re-crystallize. It was initially clear, now is visibly yellowing. I guess that means its predicting .... I don't know what. My Oregon Scientific internet-based weather display is my go-to device for weather predictions.

    I originally read how to make one in a 1912 copy of The Book of Knowledge, the children's encyclopedia. It was about 45 years ago when I was about 15. Here is a photo of the page with the instructions.


    You may have better luck with the LEDs in series. You will need to guarentee a higher minimum input voltage but you will need 1/3 the current and the LEDs will be guarenteed to share (some have variations in their forward voltage and hog most of the power when in parallel). The voltage may end up being fairly high if white or blue LEDs are used.

    If you are having issues with your LEDs not being equal brightness, but can't supply the 16 or 17V required by a series configuration, put a 22 ohm resistor in series with each one to help balance them out.

    1 reply

    Don't need any luck. I was surprised I haven't seen a current hogging problems at all with this circuit. Its true for the best performance, the leds should be in series. I want to use white leds for illumination, so I have to live with the high voltage drop. I didn't want to use anything over 16 volts for the power due to availability. The total current is only 20 milliamps and I don't need to make it less. Most of your multi-led flashlights on the market (and there are millions) just use parallel leds and a series limiting resistor. Using 22 ohm resistors for balance is an excellent idea.

    I was reading reviews of this same Fitzroy (same company as the one you ordered) and the chief complaint with them seems to be that once they crystallize initially, they don't continue to change. So, I was wondering, have you experienced such problems with yours?

    1 reply

    Most people will store their Fitzroy storm glass in the house. Two hundred years ago, we didn't have air conditioning and temperature controlled environments. So in the home, the storm glass doesn't see changes in temperature. I have mine hanging on the wall near a window. During the day, most the crystals will dissolve back. Then overnight and in the morning, the glass is full of growth. The more extreme the temperature change, the more interesting the glass becomes.

    I have to make one of these...Very good!!...

    How is it that I have never seen one of these in my entire life? Thanks for the introduction, that thing is memorizing!

    1 reply

    I remember reading about it 45 years ago when I was a kid. I think it was Scientific American in the Amateur Scientist section. I wanted to make one, but couldn't get all the chemicals. Now the storm glass is pretty common on the internet. The crystals are camphor and the other chemicals effect the solubility, depending on temperature or possibly other factors.

    Cool 'ible and MANY thanks for this piece of history and tech behind the glass!
    I read the wiki-article and i find those glasses fascinating :)

    Probably a silly question but I'm new to reading schematics. On R2 the leg on the right goes back and points to the resistor. How does that connect? Is R2 a potentiometer? It looks like it is in the pic of the control box. Thanks.