Jump to content

Arduino Puzzle Cache


31BMSG

Recommended Posts

10 minutes ago, 31BMSG said:

Has anyone used an Arduino board to create a puzzle cache?

 

There are lots out there. One cacher not far from where I live has hidden a bunch of them, using sensors to detect temperature (both hot and cold), knocking (using a piezo sensor), and more. There's also one that I haven't found yet that needs to be physically brought to the right location before you get the coordinates to the final.

 

I think it's reasonable to assume that there are puzzle caches out there using every possible Arduino sensor in some fashion, plus a lot more variations. An Arduino is a relatively inexpensive way to create a good gadget cache.

Link to comment

I've made two. IMO the main issues to sort are how you'll power it and how you'll keep it dry, or cope with water ingress.

One of ours is built into an ammo tin (flooded when a newbie closed the lid incorrectly, survived as the electronics were high mounted), the other is inside a plastic lock'n'lock style in a hollow log, no probs so far...

  • Upvote 1
Link to comment
6 hours ago, lee737 said:

I've made two. IMO the main issues to sort are how you'll power it and how you'll keep it dry, or cope with water ingress.

One of ours is built into an ammo tin (flooded when a newbie closed the lid incorrectly, survived as the electronics were high mounted), the other is inside a plastic lock'n'lock style in a hollow log, no probs so far...

 

I have made some caches containing active electronics. Water, of course, is the greatest problem. Devices inside permanently closed PET-preforms (https://coord.info/GC3A4YW) have succeeded to work  8 years already. They are powered up with a medium size lithium battery. The setup is estimated to work at least 10 years even though it consumes small amount of power all the time. Modern microprocessors have many ways to reduce power to extreme minimum but it takes some work to make them work properly.

 

My last published cache (https://coord.info/GC804JG) contains a motion sensor to activate electronics. Electronic parts are inside a standard junction box and it is powered with AAA-batteries. Quiescient current is less than 1uA. Batteries will die faster than lithium ones and the case is not pressure tight as PET-preforms. Let's see how long it will survive. It propably get muggled before batteries die :)

Edited by arisoft
  • Upvote 2
  • Helpful 1
Link to comment

Thanks for the replies, comments about reliability and maintenance issue were what I was looking for.

 

Water is scarce in this area but my initial plan is to house the components in a waterproof container inside an ammo can, I already have permission to place a cache in a spot that's semi-protected from the elements. Since all of my ammo can caches have always been closed correctly, I didn't consider someone leaving a can open, thanks.

 

My initial plan is to have the electronics housed under a locked partition inside the ammo can but with a servo needing to be powered I'm thinking the cacher should bring their own power supply. The cacher would have to capture a randomly generated 5-6 digit number flashed for a split second and then enter that number on a keypad to unlock the hinged partition. First thought was to use a photocell to determine the ammo can being closed before re-locking the partition, but with the possibility of the can not closing correctly maybe a proximity sensor under the latch would be better. Hmm, I could also use the proximity sensor to upload the latch status to the cloud. I see endless possibilities to jump down the rabbit hole with this project.

 

 

Link to comment

Newest arduino caches being hidden in my vicinity have "USB" in their titles. You'll have to come with your power bank. They seem easier to maintain than battery powered caches.

 

One of them, located in a wonderful garden in Germany, leads you to grab the arduino-thing at posted coordinates, power it up, find a way to climb several meters higher (there is a publicly accessible observation tower nearby) and read the final coordinates that are then displayed.

Link to comment
20 hours ago, 31BMSG said:

Has anyone used an Arduino board to create a puzzle cache? I've been studying the web site but would appreciate hearing the pros and cons of such a cache from someone that's actually built one.

 

I've hidden several caches using Arduinos. But mine are primarily multicaches and not mystery caches.

Link to comment
16 minutes ago, bflentje said:

 

I've hidden several caches using Arduinos. But mine are primarily multicaches and not mystery caches.

I saw some of your caches while reading through the threads cerebus1 provided earlier. Do any of your multicaches require the movement of something from one of the stages, such as an RFID chip, to the final to open the container? This idea crossed my mind but I'm wondering what the probability would be of the chip being returned to its proper location.

Link to comment
4 minutes ago, 31BMSG said:

I saw some of your caches while reading through the threads cerebus1 provided earlier. Do any of your multicaches require the movement of something from one of the stages, such as an RFID chip, to the final to open the container? This idea crossed my mind but I'm wondering what the probability would be of the chip being returned to its proper location.

I've seen a few locally where an RFID card was needed, it was housed in on container very close to the main box.  One used a simple puzzle to get the lock combo for access to the card, which was then used a couple of feet away to open the main cache box.  That made it easy for the finder to put it back correctly.

Link to comment
6 hours ago, ottieolsen said:

We've made several Ardiuno/Smart caches in Western Kansas. unfortunately, the biggest issue seems to be cachers not reading the cache page to simply understand it....

There's an old saying (from Ikea perhaps:D) "When all else fails, read the instructions". Obviously, some cachers have never bought Ikea.

Corollary - If that fails, use a bigger hammer.

Additional corollary - If that doesn't get it working, you have an electrical problem.

  • Funny 1
Link to comment

I have a couple of arduino caches, but not actually in the field yet due to issues trying to get a locking mechanism (or more accurately, unlocking) to work... because I don't want to just output a numeric code and use a combination lock (in fact one of the caches just wouldn't work with anything but an electronic lock triggered by the arduino).  Presently I have a solenoid latch, but the problem with it is that it's not strong enough to pull back if there is any force/friction on it... so still looking for a better solution.  The other problem is that the locks I have seen all require 12V and use substantial current, meaning I need to install into the cache a decent battery, such as a UPS battery, which I will need to recharge from time to time...

Link to comment

I ordered an Arduino starter kit yesterday so I'm a few weeks away from experimenting. I took an old house lockset apart and I think I'll try controlling that with a 5v stepper motor as a first step. Thanks for bringing up the force/friction situation, I'll make sure to check that aspect.

Link to comment
3 hours ago, funkymunkyzone said:

I have a couple of arduino caches, but not actually in the field yet due to issues trying to get a locking mechanism (or more accurately, unlocking) to work... because I don't want to just output a numeric code and use a combination lock (in fact one of the caches just wouldn't work with anything but an electronic lock triggered by the arduino).  Presently I have a solenoid latch, but the problem with it is that it's not strong enough to pull back if there is any force/friction on it... so still looking for a better solution.  The other problem is that the locks I have seen all require 12V and use substantial current, meaning I need to install into the cache a decent battery, such as a UPS battery, which I will need to recharge from time to time...

I picked up a little door latch solenoid from eBay - rated at 12V, but reliably activates at above 8V, draws 350mA or so - I've been testing it with hundreds of cycles over the past 2 months, the 6xAA alkalines have barely been drained at all....

Link to comment
5 minutes ago, lee737 said:

I picked up a little door latch solenoid from eBay - rated at 12V, but reliably activates at above 8V, draws 350mA or so - I've been testing it with hundreds of cycles over the past 2 months, the 6xAA alkalines have barely been drained at all....

 

Does your testing involve the latch being mounted on a door that has some force on it creating friction when the latch is trying to pull open?  Eg spring loaded door, or someone pulling on the door...

 

Link to comment
1 minute ago, funkymunkyzone said:

 

Does your testing involve the latch being mounted on a door that has some force on it creating friction when the latch is trying to pull open?  Eg spring loaded door, or someone pulling on the door...

 

To a degree. With a door striker, if you pull on the door hard enough, enough friction will be produced between the striker and the latch to prevent it activating. My door isn't spring loaded, so can't comment on that bit.

Link to comment
7 minutes ago, lee737 said:

To a degree. With a door striker, if you pull on the door hard enough, enough friction will be produced between the striker and the latch to prevent it activating. My door isn't spring loaded, so can't comment on that bit.

 

I'm keen to hear your thoughts on how to pull back the latch and then have the door open, or indicate to the finder that they can pull the door now.  I mean there are options such as use a microswitch to detect when the door is actually pulled open, and then the voltage can be dropped again on the latch, but all the while the latch is pulled, it's draining the battery (ok if it's the finder's battery, but not so much if it's a built in battery that requires charging/replacement periodically).  Another option is another solenoid that pushes the door a moment after the latch is pulled.  I've tried one that pulls the door to allow the latch to pull out, and then release and spring the door open, but still have some teething issues to get it to reliably work (my door is hinged at the bottom so the door should fall open once unlatched and pushed)

 

Link to comment

The initial plan is that the door is horizontal resting on it's own weight against a stop, I had not considered someone pulling up on it before/as it unlocks. I'm going into this expecting several failures in design, I learn better that way. Until I actually get some parts I'm only brainstorming but I appreciate the input, these give me things to look for in testing.

Link to comment
17 minutes ago, funkymunkyzone said:

I'm keen to hear your thoughts on how to pull back the latch and then have the door open, or indicate to the finder that they can pull the door now.

 

Flash some LEDs, play a sound? Mine has the latch activate for a second or so, for battery preservation.

Link to comment
6 minutes ago, funkymunkyzone said:

 

I'm keen to hear your thoughts on how to pull back the latch and then have the door open, or indicate to the finder that they can pull the door now.  I mean there are options such as use a microswitch to detect when the door is actually pulled open, and then the voltage can be dropped again on the latch, but all the while the latch is pulled, it's draining the battery (ok if it's the finder's battery, but not so much if it's a built in battery that requires charging/replacement periodically).  Another option is another solenoid that pushes the door a moment after the latch is pulled.  I've tried one that pulls the door to allow the latch to pull out, and then release and spring the door open, but still have some teething issues to get it to reliably work (my door is hinged at the bottom so the door should fall open once unlatched and pushed)

 

In theory a stepper motor with an arm would move the latch into the unlocked position, once the door is off it's stop a microswitch would indicate the door is open and the stepper would return the latch to it's locked position. Just as you can pull the door on your house closed with the keyset locked the finder would push the door against the stop until it latched. In the event the cacher failed to push the door into the locked position something (spring/foam) mounted on the inside of the ammo can lid would ensure the board is locked against it's stop. The stop can be adjusted so there is no friction on the latch but that doesn't account for the cacher pulling on the board while it's unlocking, it's also going to use power until it's returned to the lock position.

 

I also considered something linear that would pull the latch into a detent unlocked position, much like automotive door locks, and then power down until it's commanded to lock, I found a 7v linear actuator but at $70 USD that's a bit out of budget. I just need to get some parts and start experimenting.

Link to comment
On 12/20/2019 at 12:21 PM, 31BMSG said:

I saw some of your caches while reading through the threads cerebus1 provided earlier. Do any of your multicaches require the movement of something from one of the stages, such as an RFID chip, to the final to open the container? This idea crossed my mind but I'm wondering what the probability would be of the chip being returned to its proper location.

 

No, never done this. And I am pretty sure I never would either because the chip card would never consistently make it back to its home in my experience. MaNY cachers can't even rehide a cache back in the proper location. To trust them to move the key element of a puzzle back would be asking the impossible.

  • Upvote 1
  • Helpful 1
Link to comment
19 hours ago, lee737 said:

We have two which, upon successful completion of the puzzle/task, tell you the location of the cache - far easier than actuating doors! After the trials of the 'door one' I'm working on, I think I'll head back to screens and coordinate instructions...

 

I hear you.

 

The problem I have is that one particular design I have is a pair of caches that will be located some distance from one another (perhaps different ends of the country, or even different countries) that have to be opened together at the same time - I have a cunning algorithm that can make this happen, without relying on keeping a clock running or having the boxes communicate with each other in any way nor any other external system.  Anyway, I need the doors to the cache to electronically open otherwise all that will happen is previous finders will give the code/location to get to the final and the "field puzzle" element will simply be skipped.

Link to comment
On 12/23/2019 at 6:46 PM, funkymunkyzone said:

 

I hear you.

 

The problem I have is that one particular design I have is a pair of caches that will be located some distance from one another (perhaps different ends of the country, or even different countries) that have to be opened together at the same time - I have a cunning algorithm that can make this happen, without relying on keeping a clock running or having the boxes communicate with each other in any way nor any other external system.  Anyway, I need the doors to the cache to electronically open otherwise all that will happen is previous finders will give the code/location to get to the final and the "field puzzle" element will simply be skipped.

Sounds like quantum entanglement <g>.

If you are going linear, had you considered a design that uses a short throw at the 'locking business end' and a levered mechanical advantage for the solenoid which would travel further?  Take a 1" solenoid throw down to 1/8" unlatch motion - with that kind of  leverage you've got an 8X mechanical advantage to work with that would go a long way towards overcoming your frictional issues.

 

Link to comment
59 minutes ago, ecanderson said:

Sounds like quantum entanglement <g>.

If you are going linear, had you considered a design that uses a short throw at the 'locking business end' and a levered mechanical advantage for the solenoid which would travel further?  Take a 1" solenoid throw down to 1/8" unlatch motion - with that kind of  leverage you've got an 8X mechanical advantage to work with that would go a long way towards overcoming your frictional issues.

 

The solenoid locks are all one unit and electrically and mechanically they are extremely simple - chuck 9V to a relay that lets through 12V high current from the battery, lock pulls open - nothing complex.  Unfortunately with simplicity is, well, failure to work. lol

 

Link to comment

Ah - must have missed that you were working with an existing locking assembly.  Thought yours was home brew solenoid design.

 

If I were doing this from scratch, I'd consider having the previous finder / closer supply most of the energy.  Think of a mouse trap where energy is stored in closing the door against a spring, and is then relatively easily released with minimal solenoid force to a latch release to pop the door open again. A low friction trigger would require a single quick pulse to do the job.

  • Upvote 1
Link to comment
4 hours ago, ecanderson said:

Ah - must have missed that you were working with an existing locking assembly.  Thought yours was home brew solenoid design.

 

If I were doing this from scratch, I'd consider having the previous finder / closer supply most of the energy.  Think of a mouse trap where energy is stored in closing the door against a spring, and is then relatively easily released with minimal solenoid force to a latch release to pop the door open again. A low friction trigger would require a single quick pulse to do the job.

I appreciate the nudge in this direction! Having the current latch in hand was the only reason I was focusing on it, I can easily change directions and I think I just did. After your mousetrap example I did some internet research and I think a rotary latch will work perfectly for what I had in mind, one pulse and it remains unlocked. I do plan for the cacher to supply power for both the logic and the servo/solenoid/stepper as well as something mounted on the bottom of the ammo can lid to ensure the door is pushed into the locked position when the lid is closed. That is IF the cacher decides to actually to close and lock the lid. The way it's been going in this area I wouldn't be surprised if someone left the ammo can open to avoid locking the inner door.

 

 

Link to comment

I have a few of these. All are multis, since clever electronic lock mechanisms are hard to build and easy to break.

 

I use sensors, LEDs, 7-segment displays, LED displays... but no motors or solenoids. Sorry, one with a motor, but not for opening a hatch but for lowering a bison tube down a pipe. Also, I always use small, cheap Arduinos or ATinys, not full size Arduinos.

  • Helpful 1
Link to comment

@31BMSG

Regarding power use.  If you supply your own power, there's no need to keep the logic alive 24/7, even if a lower power design is employed.  A single 'start' button push (tied to the power source) can be used to either enable the logic by supplying power to the board (if you can get it booted fast enough) or instead, a small latching circuit that will keep the power to the board enabled through a 5V SSR until the board logic or a timer circuit  resets the latch to drop out the SSR and open the entire circuit from power (except for the 'start' button).  That may already be old hat for you, but if not, it's one way to increase your battery life dramatically.  All the battery has to do is keep the logic live long enough to deal with the puzzle, plus one high current but very short shot to the solenoid to pop open the door. 

 

Being in Texas, you have both an advantage and a hit as regards environment.  You don't have the major cold that saps the ability to produce the juice, but you do have the hot that substantially increases the self-discharge rate, particularly for some battery chemistries.  You're going to need to pick your battery wisely.

 

Rechargeable lithium ion and standard lead-acid cells really do NOT like to be discharged deeply,  NiMH (and NiCd, if you can still find them) has one of the fastest self-discharge rates.  Lead types and lithium ion hold their charge longer. Lithium ion loses about 5% in the first 24 hours, but drops to 1~2% per month after that, making it a good candidate for longer term use between charges.  A factory fresh lead acid loses about 5% per month.  Both of the nickel chemistries (with some notable exceptions designed to minimize self-discharge as a trade off for total capacity [e.g., Eneloops])  lose 10~15% in the first 24 hours, then another 10~15% per month. 

 

  • Helpful 2
Link to comment
On 12/26/2019 at 5:40 AM, ecanderson said:

@31BMSG

Regarding power use.  If you supply your own power, there's no need to keep the logic alive 24/7, even if a lower power design is employed.  A single 'start' button push (tied to the power source) can be used to either enable the logic by supplying power to the board (if you can get it booted fast enough) or instead, a small latching circuit that will keep the power to the board enabled through a 5V SSR until the board logic or a timer circuit  resets the latch to drop out the SSR and open the entire circuit from power (except for the 'start' button).  That may already be old hat for you, but if not, it's one way to increase your battery life dramatically.  All the battery has to do is keep the logic live long enough to deal with the puzzle, plus one high current but very short shot to the solenoid to pop open the door. 

 

Absolutely.. A clearly marked "on/off" switch takes care of this.

 

On 12/26/2019 at 5:40 AM, ecanderson said:

Rechargeable lithium ion and standard lead-acid cells really do NOT like to be discharged deeply,  NiMH (and NiCd, if you can still find them) has one of the fastest self-discharge rates.  Lead types and lithium ion hold their charge longer. Lithium ion loses about 5% in the first 24 hours, but drops to 1~2% per month after that, making it a good candidate for longer term use between charges.  A factory fresh lead acid loses about 5% per month.  Both of the nickel chemistries (with some notable exceptions designed to minimize self-discharge as a trade off for total capacity [e.g., Eneloops])  lose 10~15% in the first 24 hours, then another 10~15% per month.

 

I mostly use Li-Ion, but one cache is powered by a 9V alkaline. All of them hold their charge for months, and I mean many months.

I was surprised that this was so easy. I thought the outdoors moisture would kill the battery in a matter of days, but it just runs and runs. Of course, there are no connectors exposed to the outside, everything is inside the box, but the box is not super tight.

 

I also have one powered by a power bank, as Tungstène mentioned. This was not without problems though; with simple power banks it works just fine, but smarter power banks detect that too little power is drained from my low power circuit so it turns off! I had to add an extra resistor just to waste some energy!

Link to comment
On 12/26/2019 at 3:40 PM, ecanderson said:

Regarding power use.  If you supply your own power, there's no need to keep the logic alive 24/7, even if a lower power design is employed.  A single 'start' button push (tied to the power source) can be used to either enable the logic by supplying power to the board (if you can get it booted fast enough) or instead, a small latching circuit that will keep the power to the board enabled through a 5V SSR until the board logic or a timer circuit  resets the latch to drop out the SSR and open the entire circuit from power (except for the 'start' button).  That may already be old hat for you, but if not, it's one way to increase your battery life dramatically.  All the battery has to do is keep the logic live long enough to deal with the puzzle, plus one high current but very short shot to the solenoid to pop open the door. 

 

 

Further to this - I've been using the power circuit described in this link - https://makerself.wordpress.com/2014/12/23/power-circuit-redesigned/

I haven't deployed this live in a cache yet, but have one working well on the bench, and another home-project using the same - this page goes over it well. With lithium batteries you should get battery life near to battery shelf life (especially with the low find counts around here!)..... 

Link to comment
On 12/27/2019 at 12:56 PM, Ragnemalm said:

 

Absolutely.. A clearly marked "on/off" switch takes care of this.

 

 

I mostly use Li-Ion, but one cache is powered by a 9V alkaline. All of them hold their charge for months, and I mean many months.

I was surprised that this was so easy. I thought the outdoors moisture would kill the battery in a matter of days, but it just runs and runs. Of course, there are no connectors exposed to the outside, everything is inside the box, but the box is not super tight.

 

I also have one powered by a power bank, as Tungstène mentioned. This was not without problems though; with simple power banks it works just fine, but smarter power banks detect that too little power is drained from my low power circuit so it turns off! I had to add an extra resistor just to waste some energy!

I don't even trust the use of an On/Off switch.  Rather, use a momentary push button type switch that triggers the latching of power  (I use an SSR rather than a 5V relay due to lower power required to enable it, but either could be made to work) to the circuit.  The circuit itself can then decide when to drop power out from the SSR based upon time or activity of the finder.   To initiate things again, the button would need to be pushed.  In the interim, zero power drain.

 

Yeah, having to add a parallel load resistor to get the power bank to operate isn't ideal.  Reminds me of the fun some folks have when trying to swap out for LED lights on their vehicles.  Not enough load, and the BCM thinks the 'bulb' has burned out.

Edited by ecanderson
  • Helpful 1
Link to comment

I have published one Wherigo that used a synchronization of between the WIG cart and an Arduino.. Upon completing the Wherigo portion, you were given a code based on the current time/date. This code was good for a 15 minute window. The final was in a large Tough Box that was magnetically locked. To release the two mag locks, the code was entered on an exterior key pad. There were additional steps after opening the door, going back and forth between the WIG cart and the Arduino data collection systems.  

Monster Hunter 01 Small.jpg

Monster Hunter 12 Small.jpg

Monster Hunter 13 Small.jpg

  • Love 1
Link to comment
On 12/31/2019 at 10:57 AM, ecanderson said:

I don't even trust the use of an On/Off switch.  Rather, use a momentary push button type switch that triggers the latching of power  (I use an SSR rather than a 5V relay due to lower power required to enable it, but either could be made to work) to the circuit.  The circuit itself can then decide when to drop power out from the SSR based upon time or activity of the finder.   To initiate things again, the button would need to be pushed.  In the interim, zero power drain.

 

Yeah this is something I thought of too as a good way to manage battery life, if I ever have to go down that path of internal power (replaced/recharged by me as CO).

Link to comment
On 12/22/2019 at 8:23 PM, funkymunkyzone said:

 

I'm keen to hear your thoughts on how to pull back the latch and then have the door open, or indicate to the finder that they can pull the door now.  I mean there are options such as use a microswitch to detect when the door is actually pulled open, and then the voltage can be dropped again on the latch, but all the while the latch is pulled, it's draining the battery (ok if it's the finder's battery, but not so much if it's a built in battery that requires charging/replacement periodically).  Another option is another solenoid that pushes the door a moment after the latch is pulled.  I've tried one that pulls the door to allow the latch to pull out, and then release and spring the door open, but still have some teething issues to get it to reliably work (my door is hinged at the bottom so the door should fall open once unlatched and pushed)

 

Since Chinese customs keeps refusing to let the shipper send my electronics to the US I've been experimenting with other things, the latch is one of them. I found a latch that I think will work for this application, it works on the same principle as a car door latch but is only 2.5 inches long. When the latch is released it pushes the door open nearly 20mm and until  you use a lot of force, pulling the door doesn't seem to effect the tension required to release it. Right now I'm waiting on a 5v solenoid as well as a 5v servo to test this with, until then I still don't know if they have the power to pull the release.

 

If interested this is an R4 micro concealed rotary latch manufactured by Southco, there are numerous configurations but the part number I received is R4-10-21-201-10. This is a US company but have distributors worldwide, the Australian service center's phone number is (61) 3 9330 5000. Shop around if you can, I found the same part number priced in the US from $9 to $27.

  • Helpful 1
Link to comment
1 hour ago, 31BMSG said:

Since Chinese customs keeps refusing to let the shipper send my electronics to the US I've been experimenting with other things, the latch is one of them. I found a latch that I think will work for this application, it works on the same principle as a car door latch but is only 2.5 inches long. When the latch is released it pushes the door open nearly 20mm and until  you use a lot of force, pulling the door doesn't seem to effect the tension required to release it. Right now I'm waiting on a 5v solenoid as well as a 5v servo to test this with, until then I still don't know if they have the power to pull the release.

 

If interested this is an R4 micro concealed rotary latch manufactured by Southco, there are numerous configurations but the part number I received is R4-10-21-201-10. This is a US company but have distributors worldwide, the Australian service center's phone number is (61) 3 9330 5000. Shop around if you can, I found the same part number priced in the US from $9 to $27.

Nice work. I'll have a look for that one.

Link to comment

Join the conversation

You can post now and register later. If you have an account, sign in now to post with your account.
Note: Your post will require moderator approval before it will be visible.

Guest
Reply to this topic...

×   Pasted as rich text.   Paste as plain text instead

  Only 75 emoji are allowed.

×   Your link has been automatically embedded.   Display as a link instead

×   Your previous content has been restored.   Clear editor

×   You cannot paste images directly. Upload or insert images from URL.

Loading...
×
×
  • Create New...