Advice on moving from a hobby project to a commercial product

[j_omega]

I have completed a open-source design (minus gerber files) for an aircraft engine monitor. It's a very low-cost solution relative to comparable products on the market. I'm not looking to sell large numbers of these, but I would like to make the board available for sale to other electronics enthusiasts who would like to build their own. However, I have discovered that there is a very large gap between a completed project and actually selling units. Any advice on how to bridge this gap?

[ataradov]

Selling in low quantities is easy - Tindie.com is the answer here.

Certifications and all that stuff is hard, and only people in that field will know what is required. This does not looks like flight critical piece of equipment though.

[j_omega]

I forgot to mention that this is only for experimental aircraft. No flight cert needed.

[Mr. Scram]

I forgot to mention that this is only for experimental aircraft. No flight cert needed.

That's important. Aircraft technology is one of the most regulated categories out there and this will certainly hamper any new product brought to market.

Of course, only selling items for experimental aircraft will limit your market significantly. You should also still take care to limit your legal liabilities.

[iwtommo]

Having not done this before (but having a few projects i foresee being marketable) i think the first decision you need to make is the size of your first production run.

Too large production run - expensive, stock sat around for ages, money spent is also an opportunity cost
Too small - Run out of stock, insufficiently optimised production process (with the frantic rush to build more units)

Im sure at some point you will have to deal with all the legal particulars - registering as a company, hiring an accountant, ect ect

Again, i am still learning all this myself, so will keep an eye on this thread 

[sokoloff]

Couple of things from a nerd a pilot/owner of a certified aircraft:

1. Look in the relatively new FAA NORSEE policy, which might allow you sell some of these units into the certified market as well, as "Non Required Safety Enhancing Equipment".

Some feature considerations to make it appeal to the widest audience:

2. To facilitate installation in 6-cylinder turbo aircraft, add another K-type channel for turbine inlet temperature and code to treat that as "first-class citizen". (There are some twin-turbo aircraft with two TIT channels needed as well if you have enough ADCs or add an analog switch or additional ADCs.)

3. To facilitate installation in inter-cooled aircraft, add another (likely K-type) air temp sensor, such that a user could use the carb temp channel and this new temp sensor to have all of OAT, CDT, IAT and calculate intercooler efficiency numbers to track intercooler cleaning requirements.

4. Consider a caution and warning light output. Some users may want a CHT caution at say 390°F but a CHT warning at 420°F.

5. The "onboard memory" is ambiguous on the github page. I want at least 2 second (ideally 1 second) logging of datastreams for at least 25 flight hours (more if it's hard to download the data).

6. Make it clear that downloading datalogs is possible.

7. It's unclear whether you have fuel totalizer functionality (ded reckoning of fuel onboard), but that's extremely handy (and, on balance, safety enhancing by a significant margin).

8. Consider adding RS-232 or ARINC429 input channel to take inputs from a serial connected GPS unit (dist and time to next waypoint, dist and time to end of flight plan, current lat, long, alt). The first two are for fuel totalizer usage and the last three for data logging.

9. You might be able to get by with one fewer button using a short press to advance and a long press for previous. This is fairly standard and might be easier to use in turbulence than buttons that are too close together.

[jbb]

Im sure at some point you will have to deal with all the legal particulars - registering as a company, hiring an accountant, ect ect

Having a limited liability company could be a very good idea.  Make sure you operate it properly to avoid a) trouble with the tax department and b) somehow breaking the limited liability bit.

I suggest you should look at the following physical aspects:

• ElectroMagnetic Compatibility (EMC).  This would normally consist of emissions testing (make sure device is not a radio jammer) and immunity testing (device doesn't crap out when exposed to modest quantity of applied radio noise or Electrostatic Discharge (ESD))
• Test the device out to the rated operating temperatures to make sure it operates. Startup at very low temperature can be an issue with crystal oscillators.
• The power input to the device should be able to handle some spikes and surges from the aircraft power rail (no idea what standards apply)
• The input channels to the device should be able to handle ESD and - where reasonably possible - mis-connection.
• Try to avoid parallel connections of your device with any truly critical sensor.  Even if your device fails horribly the plane should keep working :-)
• Make sure your micro's got a good bit of flash and RAM left for extra features or fixes

Then software aspects:

• Make sure the device detects and shows faults e.g. sensor wires falling off, result out of range etc.
• Be careful with buffer length checking in serial receive functions and data output functions.  You don't want to overreach your RAM and corrupt other variables.
• Use a watchdog timer (or similar) to reset the device if something goes wrong with the firmware.  I would much rather have an instrument that reboots again and again than quietly sits there with a 'stuck' display.
• The device should display the serial number and firmware version number on boot up (important if you ever want to certify a design)
• You might want to make sure the device can be updated without trashing the settings.  (Alternatively, make it possible to export the settings, update the firmware and then import the settings)

Then management aspects.  These are helpful for managing a product / project, and essential if you would ever like to certify the design:

• Maintain specifications
• All units should ideally have a serial number and hardware version number listed on the panel (not sure if front or back) and programmed into the device.  This is a bit tricky with kit sets...
• Run a software issue tracking system for bugs and features
• Document your testing, and note down exactly how you tested things. For extra credit, do test plans.

[max_torque]

Sorry, but no way i the world would i sell any complete product into any aero market that wasn't fully certified.

Unless you are going to make millions of $$ the risk to reward is skewed completely the wrong way!

(making a "kit" of parts available for self assy is an option, but even then i'd be wary of what i might be letting myself into)

[David Chamberlain]

And no sooner...

https://www.tindie.com/products/j_omega/aircraft-engine-monitor-kit-6-cylinder/

[james_s]

It should be possible to register as an LLC and also have business insurance. I'm not an expert on such matters but I believe that limits your own personal liability.