regulatory approvals and the beauty of DIY / open source

General Discussion - Everything else for V1 goes here

Moderators: seaton, strogg

regulatory approvals and the beauty of DIY / open source

Postby philtulju » Tue Sep 25, 2007 6:52 am

Hello!

I have been following the discussion on flickr and this group now, discovered the thread a couple of days ago. A huge amount of interest! Very exciting.

I am a professional engineer who has been working in the US and now China for about 6 years. I have broad electronics and mechanical design experience: computers, embedded systems, optical communications, and wireless stuff.

I am interested in helping out and contributing to the discussions and the design. I am not co-located probably with anyone, but I should be back in the US next year sometime.

Anyhow, I wanted to share some thoughts and some of my experience.

Something that has been mentioned many times has been about regulatory approvals. This can mean FCC/CE certfication, UL listing, etc. etc. There are several general sets of rules and also many one-off regulatory bodies, depending on where you are in the world.

Reasonably, some people have suggested that pre-approved modules with modular certification are one design approach to avoid trouble. However, these modules are usually expensive and are certainly not open designs, and then your project is at the whim of some suppier that may not be around for the long term.

My philosophy on open source is that it is the best way to make a project immortal: How many more-or-less similar radio receivers, MP3 players, web browsers, or whatever have been designed in this world? Rather than continually reinventing the wheel, open source allows lots of creative thought to be put on improving a particular project, rather than have a dozen secret half-solutions all rot when their passionate inventor moves on to something new. While almost nobody compiles it from source code, witness the success of Firefox. A hardware project is the same way. Most people will buy kits or even completed modules, rather than build it themselves, but still everyone benefits from the open, well-documented design.

The open design is also a way to have a key competitive advantage over the commercial solutions that are available, e.g. Pocketwizard, Skyport, radiopopper?, et al. These companies have good products that are well designed and have all the necessary legal approvals to be sold as products in certain countries. However, getting all of these approvals takes a lot of time and a lot of money. While they may be laughing all the way to the bank with their newfound Strobist windfall, make no mistake - a lot of their multi-hundred dollar price tags go to testing labs and regulatory bodies.

But witness the Gadget Infinity / Cactus etc. products - they are examples of how simple and cheap a solution can be from a parts perspective. And they avoid all of the pesky expensive certifications by not having any. Ebay and direct importation from web sites have allowed a lot of products commercial success, even though they would not be legal to sell openly in domestic retail channels.

I don't want to get too political, but I think this is fine, for a couple of reasons:

1) While I feel bad for companies who are very conciencious, the market reality is that almost all of the "FCC/CE approved" products on the market have fudged, cheated, or downright lied to get their papers. If you were to spend the time and money to test almost any off-the-shelf electronic or radio product, you would find that with scrutiny, the device did not meet all of the regulatory requirements. It is too difficult to test everything, too easy to cheat or trick a lab, or even pay them off for approval, esp. in the case of international testing labs. Even if you make a really clean-as-a-whistle design, you have to compete in cost and features with others who do not.

2) Because the developed world has largely taken a "fingers in the ears" attitude towards the practices above, we already have an RF environment with lots of spurious emissions. But cheap, smart silicon and lots of great math and communication theory have made wireless communications super reliable nonetheless. New modulation schemes and coding techniques are robust against such interference. Thus, a rigid frequency plan is not needed any more.

These regulatory rules are largely obsolete. They were developed decades ago in an era of very dumb transmitters and receivers, and largely in a pretty non-democratic way. Again, maybe too political, but they were architected with one-to-many broadcasting in mind, like commercial TV and radio stations, and not with the more peer-to-peer architecture of wifi and the like. Commercial, military, etc. interests win out, surprise surprise. The real innovation in wireless communications is all happening in the tiny wireless ghettoes of the "free" ISM bands - this is why so many great encoding schemes have had to be invented. And they work: it has been a while since the vacuum made the TV go out, or your phone kept picking up KLOV or your neighbors steamy phone calls.

Thus, I think of certs much in the way I think of patents - a business or political tool for established companies. Not a good choice for the individual inventor, hobbiest, or even small entepreneur.

So let's not let these rules slow down or compromise the design. Let's design a good radio, from the bottom up, and make sure it works really well. Let's architect it so that it could be certified in as many countries as possible. Then, when the project is a wild success and 10000 piece orders are coming in, some entepreneur will spend the capital to get it certified, and can sell it properly. In the mean time, kits or individuals building devices will meet the demand, and stay under the radar.

Sorry for the long post. I hope it is worth the time to read it.

philtulju
philtulju
 
Posts: 6
Joined: Mon Sep 24, 2007 4:38 pm
Location: Shanghai, China

Postby philtulju » Tue Sep 25, 2007 7:11 am

God, I feel like a windbag, but I forgot to mention - this "don't ask, don't tell" model is already used successfully by lots of products besides Gadget Infinity / Cactus trasmitters: FM radio transmitter kits, high power walkie talkies and "long distance" cordless phones, video senders, etc.
Maybe the HAMs want to strangle me, but I think it is as close to a victimless crime as it comes - and these devices are far, far more interfering than a trigger could be, or would want to be.
philtulju
 
Posts: 6
Joined: Mon Sep 24, 2007 4:38 pm
Location: Shanghai, China

Postby kwanyin » Fri Dec 07, 2007 5:57 am

I've been thinking about this FCC stuff. I find myself agreeing with the above points. It makes practical sense.


1, Since we are 'hobbist' working in the RF ghettos, we'll go under the radar of the FCC and avoid costly UL types of test.

2, We will be working in established RF range where devices like this are often being used. I don't believe we'll take down a plane's guidance nor will we induce an heart attack in a pacemaker with such a quick and 'weak' single.

3, Durration time. Our trigger burst will be far less than 1/60 sec. If there is interference in voice communications, it's not like we're going to jam medical, fire, police communication with such a short trigger duration. Maybe we can open up a few garage doors?

4, Off the shelf components should have FCC approval.

So, what's the harm in just building our devices for 'personal use'?


Devil's advocate:
There is one story that comes to mind. In the early trials of the Apache helicopters, they were being taken down by garage door openners.

The smart people who designed it, used an ingenus idea to work their controlling surfaces. The used radio singles and relays. This avoid projectiles cutting wires and hyrdrolic lines. This took off lots of hyrdrolic weight...fluid, valves, pumps, etc. It was a great idea!

The one fatal flaw was a garage door openner could interfer with the control surface.
kwanyin
 
Posts: 3
Joined: Sat Sep 29, 2007 6:15 pm


Return to General

Who is online

Users browsing this forum: No registered users and 1 guest

cron