Digital Safe Project

[JasonMI]

Hi. I'm new here so please forgive my unfamiliarity with the nuances of this site. (Any guidance is appreciated!) I'm looking to design an electronic keypad wall-safe with some additional "non-standard", but relatively simple, features not typically available in current products. I am in management and marketing and although I aced calculus 20 years ago, have no engineering experience/education, so this is way beyond my ability. Depending on the time/cost I would possibly like to hire someone to design the circuit board for me.

To give you a good idea of what the project entails here a few (but not all) of the features I am looking incorporate...

One of the features I would like to modify from standard is to make it hard wired to AC, with a battery back-up that uses standard AA(?) NiMH rechargeable batteries. The AC needs to power the safe and (smart) charge the batteries.

An LED light on the front of the safe needs to alert the user to depreciated battery capacity (similar to PC UPS battery back-ups), whether it is currently being powered by AC or the batteries. If this is too difficult or costly then a "battery-low" light would suffice.

It needs LEDs to light the safe interior while the door is open, with a small switch inside the door to disable the lights if the owner so desires.

Because the owner of the safe may not be able (or want) to hard wire, the batteries should be able to power the safe for 12 months while being accessed 2x every day (with interior lights disabled). This is similar to the battery life on existing safes.

I would also like to incorporate door contact sensor so that the owner can connect the safe to a home security system. I know this alone would have nothing to do with the control board, but could it be dual-purpose, and control the interior lights (using the same power supply as the board) as well?

Lighted keypad. (illuminates on first press)

Confirmation LED on front of door near keypad that comes on while a key is pressed, and lights solid for approx 5 seconds upon user entering a valid access code and the safe unlock handle being released by the electro-magnetic bolt.

LED that blinks, notifying owner of invalid access code attempt. Notification clears upon next valid access.

Then, of course, the standard stuff to be expected, such as the user being able to change the access code or reset to default (by holding a reset button located inside the safe for a predetermined length of time), timed lock-out function for invalid access attempts, etc. (specific access code programming procedures, lockout duration, etc. can be discussed when the time comes to actually begin the design process)

Additional note: the safe will have a physical key override (in case of both power sources failing or there is a need to reset a forgotten access code) and it's almost unnecessary to state, but it has to be super reliable.

I don't know if there are existing boards and parts that could be used/modified/pieced together, or if it would be better to design from the ground up. Due to the nature of the project there will be physical space limitations, so the smaller the better. I've also wondered about the hard wiring aspect as far as whether or not an AC adapter of some type would have to be used or if there is an "on-board" solution, as well as if there are any safety issues with putting an AC adapter inside a wall or enclosed safe.

I have heard that having a PCB of low to low-medium complexity can range from $500 to $1500. Is this fairly accurate? How do I find someone reliable?
Any advice on where to begin would be great!

 

[Mr RB]

Hi JasonMI, welcome to the forum and thank you for providing a comprehensive design spec for your project.

One bit of information you left off, is this a single item build for your own use or is it to be eventually manufactured? That changes some of the design decisions especially regarding which parts to use. Also, does it need to be presentable in a box, or will you be ok with a PCB and wires hanging off it, for you to install?

It looks like the same basic functionality as most safes, which can be bought very cheaply (small ones) with working keypad locks. You could even take the lock out of a $100 office safe. Or, if you were thinking of manufacturing then you should aim to add features that no cheap electronic safe has.

 

[JasonMI]

Thank you for the welcome.

It is something that I would like to manufacture as a product available for sale. It does not need to be presentable in a box, a board with wires hanging out of it would be expected. The plan to is to make a prototype for testing functionality and then several display models for presentation to retail stores merchandise selection people, as well as sales floor display models.

Existing parts manufactured for other safes will not do. The basic functionality would be very similar to existing safes, except that I want to change the programming in regard to setting access codes and when the lockout function occurs. One of the major features that differentiates it from existing products is that it will be hard wired to AC with the battery back up, as well as a couple of other minor features (minor from a technical standpoint, but huge in regard to functionality and marketing) that I will discuss with a designer if/when I find one.

There are also physical features of the safe that are major improvements over existing products that don't apply to the electronics.

Thanks again!

 

[Mr RB]

...
It is something that I would like to manufacture as a product available for sale. It does not need to be presentable in a box, a board with wires hanging out of it would be expected. The plan to is to make a prototype for testing functionality and then several display models for presentation to retail stores merchandise selection people, as well as sales floor display models.
...

Ahah, prototypes for manufacturing make a big difference!

Your plan looks professional, and will save you money as the initial function test prototype does not need to be cosmetic which should be cheaper and easier to modify (ie does not need specialty PCB made).

However after that you are really going to need an engineer with professional experience bringing a consumer product to market to be sold to the general public. There will be special requirements for safety compliance and also he would need experience working with your CADCAM guys so things are shaped and positioned etc to work with the injection molded enclosure with the keypad and display etc.

Maybe someone will speak up if they want to give it a go, and have the skills and experience. It might take a while before everyone sees your thread.

...
Existing parts manufactured for other safes will not do. The basic functionality would be very similar to existing safes, except that I want to change the programming in regard to setting access codes and when the lockout function occurs. One of the major features that differentiates it from existing products is that it will be hard wired to AC with the battery back up, as well as a couple of other minor features (minor from a technical standpoint, but huge in regard to functionality and marketing) that I will discuss with a designer if/when I find one.
...

I agree it's best not to discuss all your special value features until your designer guy is locked into an NDA.

Are you set on using a keypad? I think they are very un-secure, in this day and age anyone can buy tiny spy cameras and motion triggered video spy cams etc, so all they need to do is get a camera somewhere near your safe and record the button sequence as someone keys it in. If you have special features of how to deal with this then it's best not to discuss it but I thought you should be aware of that issue. It's an issue that will probably cause the phaseout of keypad type locks anywhere where security is important.

 

[JasonMI]

That is an interesting concern. I'm always open to suggestions. Finger print scanner is a possibility, however I feel consumer grade is not reliable enough.

 

[Mr RB]

I like the idea of having a level of separation between what the user touches and what the code actually is. So the user looks into a little window (where only he can see the display) and each number is randomly assigned to a symbol (the keypad has only symbols) and after each digit it recrambles the number->symbol screen. So an observer could only see them enter the symbols, but that would be different each time.

Or, you could have a little window again and the only entry control is a wheel. The user turns the wheel up or down to select their digit (or text character) like on a dial safe, but again the "digits" on the wheel are shown in a scrambled order on the display and it rescrambles after each digit so nobody could hack the code by observing the wheel being turned.

 

[blueroomelectronics]

Also check your patents. Scramble keypads are Patented.
**broken link removed**

 

[JasonMI]

Thanks Bill. Scramble keypads are really cool.

Does anyone have a rough idea of what would be a reasonable amount to pay for such a project?

 

[KeepItSimpleStupid]

You can always ask http://www.inventgineering.com/ Do a search for contract; electrical engineering.

There was a place somewhere that basically did the design, and the everyting they did became yours, so you were free to manufacturer anywhere.

Power throgh a "firewall" might be a problem in itself. One it could reduce the overall rating or might have to be special from the get go. Class II wiring, basically under 24 VAC 100 W and power limited <100 W doesn't generally need protection (e.g. conduit). What comes to mind is ceramic spacers and a threaded Stainless or titanium alloy rod for power and something welded/silversoldered on the outside for a cover, to minimize penetrations. Titanium has about the same thermal conductivity as stainless, alloys about 1/2. A ceramic step colar would be used for the penetration. Ceramic because of low heat conduction and stainless because of low heat conduction. Class II is like doorbell and home termostat wiring.

In a design I did for woek, I was able to get qty 1 (not this one) from a major supplier. **broken link removed**

For another system I designed, which used UL listed fire stuff, had about 8 channels of 24 VDC or 12 VDC backed up class II power using PTC's. Fuses were optional. One issue I had, was that "some strobes" from the same manufacturer would not start. We had some before and after a re-design where the peak load was lower in the new design.

There is a spacial transformer that's called "energy limiting" and can be found for 24 VAC at around 40 VA. This type of transformer cas witstand a direct short on the secondary with no ill effects, except it may get a little warm. So, fused, PTC thermister, or energy limiting is OK. 24 and 16 VACtransformers can be found to mount to round outet boxes.

Once that penetration is made, the fire rating of the safe would have to be tested agian.

This seeming simple modification to the safe, but's you on the liability side and te safe would have to be tested again for the fire rating.

You might also, be able to use only two penetrations and get the "tamper" switch out through some sort of modulation or lack thereof. e.g. lack of a 1000 Hz modulation on the incoming power is picked up as tamper. Lack of 1500 Hz is picked up as "low battery". You cut wire, tamper and low battery can go off.

Here, I just picked one: http://www.altronix.com/products/product.php?name=fireswitch108
Battery backup. I actually used one from ALL_SAF. For one design it was a 3A unit, the other designe used like a 10A unit. Many times, these are designed for lock systems, so the fire alarm can open the locks when the fire alarm goes off. Powering strobes is another application and so is constant power, all class II circiuts.

So, it's like back-up power, UL listed, class II, mounted elsewhere. Someone might think it's a tamper wire. Tamper is best wired with a constant current (2 wires), so wire breaks can be detected, but a modulated power supply would work. Safe internal batteries are possibe as well.

Right now, don't mind me. I'm brainstorming from an entirely differnet perspective.
Penetration, Batteries/charger in a UL listed fire alarm rated locked cabinet that already incorporates varius tampers. Some sort of class II wiring linking the two systems which could report any number of things: Most important: power wires cut.

It was my experience that the SLA 12V batteries were basically replaced every 3 years.

 

[blueroomelectronics]

Why not use induction coupling like they use on some cell phone chargers. It could be in the back or bottom of the safe.

 

[KeepItSimpleStupid]

Induction coupling through magnetic steel?

 

[JasonMI]

Bill, could you send me a link to an example of a phone charger induction coupling?

 

[alec_t]

If you are making a commercial product then you will have to do a thorough search of patents to ensure you are not infringing anyone's valid patent for any of the features you intend to include. If any feature is patent-protected you will probably have to negotiate for a licence to use it and pay royalties.

 

[KeepItSimpleStupid]

Induction coupling through metal wont work.

There happens to be an article on how to build a system in te August Issue of Nuts & Volts Magazine. p36, "Build your Own Induction Charger", Mthew Bates. Aug 2013; www.nutvolts.com

 

[JasonMI]

OK, here is where I am...
I found a friend of a friend that is an electronics hobbyist. We are going to use an Arduino (tentatively) board to get the prototype functionality working. From there we will work on power options (smart charging batteries, hardwired to transformer, probably similar to that used for doorbells). Once we have taken it as far as we can go along this route, the next step would be to start cold-calling PCB design/manufacturing companies in SE Michigan, and inquiring about designing a dedicated board to do what we have done using the "turnkey" board.

Does this sound like a good idea?
***Still wondering about anyone's opinion in regard to expected cost?***

[note: I have found illuminated vandalism-proof keypads via alibaba out of China. I will have to get more info to know if they will interface with the turn-key board we choose. Through my business, I have contacts in Hong Kong that can arrange for me to have samples sent, that way I won't have to buy the normal required minimum quantities for prototyping. Also, I am not worried about spy cams seeing codes, since this will be a product targeting home users. I will put a warning/disclaimer in the manual about such possibilities.]
https://www.alibaba.com/product-gs/916691993/anti_destruction_high_quality_light_keypad.html
https://www.alibaba.com/product-gs/895185409/vandalism_metal_blue_light_keypad_with.html

 

[Mr RB]

That sounds like a good cheap way to test the basic functionality and see if you want to go further with the project.

As for "PCB design/manufacture", PCB manufacture is a simple task that needs big expensive machines and lots of speed and volume. PCB/product design (done right, for commercial apps) is a slow, iterative task, that requires lots of time and effort from someone who has experience in bringing a complete product to market. It's unlikely you can get both in the same company, and if you did it would be a worse result than getting a specialist in each.

If you want a rough idea of the work, getting the arduino working is maybe 5% of the job, designing a dedicated PCB and refined software maybe +10%. Getting a good PCB, that works with your mass manufacturing PCB parts population etc (you need to talk to them) and works with your connectors, molded enclosure, etc adds maybe another +30%. Then the last 65% will be parts ordering, liasing with all your manufacturers, CAD/CAM, injection molding, branding, assembly etc.

People think once they can make a working prototype then they have a product. The reality is, that getting a working prototype on an arduino board represents (at most!) maybe 5% of the design work of bringing a product to market, and maybe 1% of the total money and total process.

 

[KeepItSimpleStupid]

RB:

I like your thinking.

@OP
In production, your aim is low cost in volume and somebody how to put both together. packaging and assembly. While development could very well be an Auduino, the final project could be a PIC or maybe even an FPGA. If the features are small, maybe even a PICAXE.