Electronic Design - Project Management

OH.... I forgot to add extensively test (thrash) production prototype samples... to prove a robust design. Just after completed production prototype design.

(perhpas even an extensive field trial).

JG

I think you want sales and marketing up front BEFORE writing any specs or doing any work - the purpose of designing a product is to make money, to make money it has to sell, therefore sales & marketing (and customers) has to like the product! (Boss'es *like* sales: They are people who *make money* while all those engineers are just pissing it away on stupid stuff, so you want sales people on your team to placate the boss).

During development you will need to work in parallel with sales & marketing - they f.ex. can test features on "tame" customers before they are designed in and prioritise what the customers actually want.

You will probably want several prototypes, not just one.

You will want to reduce cycle-time as much as possible. Time is more precious than money!

Cool. You presuade a boss who knows little about the technology to sign off on every step, so you can blame him when bad things happen.

John

Evidently there is a big difference between requirements (what you want it to do) and specs (what it does). While cost is an obvious factor, reliability, maintainability, power consumption, size are all characteristics that might be the most important consideration in different situations. I guess if its cheap enough, it doesnt need to be reliable or maintainable, just buy some spares. (This doesnt work with airplane engines). Want to tell us what you're building so we can give you our ideas of how much we'd pay for it?

[snip]

Some might disagree, but I have found that, for new designs where there is a lot of creativity involved, it is better to keep the sales/marketing people away from the process until you have a working prototype or a pretty good idea of how one would be built. There should be many iterations of the prototype by a single or small group of individuals so the cost of development at that stage is limited to the resources required in that small group.

Much of the expense of product development comes from endless meetings, trips, customer conferences - essentially too many cooks in the kitchen who have no real bearing (at that stage) on the the success of the product, but simply waste company resources.

Once you have something that you are certain is inline with what Sales, Marketing, customers, etc. will want, and you have the major kinks out,

*then* it is ok to start bringing people on board. At that point, your the role should director to servant, taking into account the remaining demands, tweaks, features,...striving to accommodate each.

Too often, the R in R&D is hindered by unnecessary bureaucracy.

-Le Chaud Lapin-

That's for sure. I had so many issues with Motorola sales people trying to sell stuff I was still working on that I had the secretary bar the sales people from access to my labs ;-)

...Jim Thompson

--
|  James E.Thompson, P.E.                           |    mens     |
|  Analog Innovations, Inc.                         |     et      |
|  Analog/Mixed-Signal ASIC\'s and Discrete Systems  |    manus    |
|  Phoenix, Arizona            Voice:(480)460-2350  |             |
|  E-mail Address at Website     Fax:(480)460-2142  |  Brass Rat  |
|       http://www.analog-innovations.com           |    1962     |
             
I love to cook with wine.      Sometimes I even put it in the food.

(Snip process)

You may also have to placate many agencies who require a cut of the action up front:

FCC, UL, VDE, CSA, TUV and sometimes ATF.

FCC in the US: Federal Communications Commission: Makes sure your product does not interfere with radio communication more than necessary.

UL in the US: Makes sure you are not overfunded.

VDE provides UL-like services in Europe.

CSA provides UL-like services in Canada.

TUV provides UL-like and FCC-like services in Germany.

ATF is used in your auto transmission.

--Winston

On Sun, 03 Dec 2006 13:08:18 -0800, Winston wrote: [snip]

And regulates your drinking and smoking while testing your firearm ;-)

...Jim Thompson

-- | James E.Thompson, P.E. | mens | | Analog Innovations, Inc. | et | | Analog/Mixed-Signal ASIC's and Discrete Systems | manus | | Phoenix, Arizona Voice:(480)460-2350 | | | E-mail Address at Website Fax:(480)460-2142 | Brass Rat | |

| 1962 | I love to cook with wine. Sometimes I even put it in the food.

Yes, I forgot to mention that bit.

--Winston

snip

You seem to have pretty much covered all the bases but you have forgotten one extremely important factor. Team work. You are in a business and your job is to help it be as successful as possible. The only way to do this is to work in a team with others outside R&D. There are people who will tell you to keep sales and marketing away until you have a prototype. They are wrong. It is only your innovations that will enable your companies product to differentiate itself from its competitors and to pick the best innovations you need a long term dialogue with S&M. The same applies to manufacturing. Your prototype is useless if it uses parts or processes production cannot handle. You need to know what they can do and you need to be able to discuss new parts and processes with them at the earliest possible time. R&D cannot and must not work in isolation.

Ian

Here's some of my stuff:

What do you think something like these are worth?

John

Do they come with Monster Cables ?:-)

...Jim Thompson

--
|  James E.Thompson, P.E.                           |    mens     |
|  Analog Innovations, Inc.                         |     et      |
|  Analog/Mixed-Signal ASIC\'s and Discrete Systems  |    manus    |
|  Phoenix, Arizona            Voice:(480)460-2350  |             |
|  E-mail Address at Website     Fax:(480)460-2142  |  Brass Rat  |
|       http://www.analog-innovations.com           |    1962     |
             
I love to cook with wine.      Sometimes I even put it in the food.

LOL

Rene

Well, we do put SMBs on the front panels, and make a bit selling BNC-SMB cables on the side!

John

Jim Thompson wrote: I had so many issues with Motorola sales people

Andy writes:

When I worked at Bendix Avionics the sales people won a big contract for developing a new ARINC navigation system for commercial aviation. No engineering input was required prior to getting the contract. ( Go figure )

The QUALITY guys were in my office in about a week, wanting parts lists and schematics so they could price the item out and get the MTBF numbers...

Hell, I hadn't even figured out IF the product could be designed yet, much less worked out how big it was, or how the design would go. .... OR where the hell we were going to hire all the engineers we would need to do various pieces I couldn't do myself ...... :>))))

The solution was to give them some crap that would occupy their time until I could figure out my next move.... Good luck. You get pretty good at gernerating crap after a while......... Especially since they send the "new" guys in quality to get this stuff, and they would accept ANYTHING just to get the hell out of my office...

Somehow, it all comes together in the end. We ain't stupid, but we ain't God, either...

Project design generally goes a lot faster if the designers don't have to give viewfoil presentations to management every morning detailing their progress.....

Andy in Eureka, Texas (retired EE)

[Snip]

Exactly.

You are making the mistake of trying to impose something on other people in an effort to make your own life structured and meaningful.

They've been f****ng it up for years.

Be a good chap and keep it to yourself.

Everyone else is downloading Ring Tones.

DNA

Thanks guys.

The product is an active RFID tag... Yes! there are a few about... but we think we have a twist that is designed specially for an end application.

RFID active tags are little monsters that sit dormant then scream out their tag number when woken up / activated.

The trick is to wake them up without any energy being used by the tag (because they are battery powered). And the wake up detection mechanism must also be very low current (to achieve long battery life 2~5 years)

I am thinking of using the Integration Associates IA4420 and IA4220 and IA4320 transceiver, receiver , transmitter chips from Integration Associates.

JG

If you put'em on eBay you'll be able to rapidly find out. :-)

Most of the stuff I make would probably have I lear 1 and possible 2 zeroes dropped from the price tags if they were put on eBay...

What "wake up" distance do you need? Presumably the advantage of active tags is the much greater read/write distance, but even passive tags can often manage to get a handful of microwatts out of the transmitter at, e.g., 10', which is plenty to switch a FET and turn on your microcontroller and then verify whether or not anyone's "really" talking to you.

Assuming you're not super space constrained (i.e., the tag is closer to those for box cars than the ones for candy bars), with a reasonably high-Q front-end filter I'd willing to bet you could get enough power out of a "wake up" pulse on a specific frequency as your link budget normally allows communication over anyway to completely eliminate any "sleep" current.

(On the other hand, often you can run the battery power management numbers and find that an average current draw of, e.g., 10uA is well under the battery self-discharge rate anyway, so you can just wake up the CPU every now and again and check for "pings." This is probably cheaper...)