Engineering Design Projects

[leenremm]

Hello all,

As a soon-to-graduate electrical engineer, I am very familiar with a basic electronics, micro-controllers, software, sensors and actuators. Furthermore, I have a range of hobbies such as flying delta-wing gliders and small e-sky helis, paragliding, rock-climbing and mountain-biking to mention a few.

Along with a few engineering friends, I would like to get involved in designing simple electronic circuits, whether they are for flying, swimming, walking, running, or even sleeping.

Could someone give me some guidance as where to begin? Where can I go to find people (and pilots) with their small electronic requests?

Does anyone have small electronic-design projects that we could help you with?

Is there a website or company which facilitates these kind of projects for young engineers?

Thank you for your time!

[ghoti]

You have made a good start. Follow your passion and get business cards to seed the world and start your own business, on paper at least to start.
Bill, Registered Profeessional Engineer (retired)

[leenremm]

You have made a good start. Follow your passion and get business cards to seed the world and start your own business, on paper at least to start.
Bill, Registered Profeessional Engineer (retired)

Thank you Bill. It is my passion, and i do intend to start a business out of it. A few friends and myself have already designed a few small products (conceptually), and we are all searching for more projects, funded by individuals or even small organizations...

Let me know how if you hear about any projects coming up!

[JohnMuchow]

You have drive and initiative...that puts you ahead a of a lot of ther people! Having started my own business in 1992, I had a few thoughts (all my own very personal opinions)....

- Take a look at other devices that are commonly bought/sold or brought up on various forums here (and on any number of other forums). Read the questions brought up in Nuts & Volts and Servo magazine (or any equivalent SA magazines). What are people looking for? What would they like done differently in existing products? Where is there a niche or place for you to get started in? Spend a couple of hours a day looking around.

- Pick a few small devices/projects and actually build and test them. And I mean test them HARD!!! At least a couple of each should be tested to destruction to see if their limits are within your expectations. Why? Well, why should anyone hire you if you have no real world experience designing products.
The differences between what works on paper and what works in the "real world" are often huge. I have made a number of mistakes over the years desiging the products I've sold. Each one of them was a complete surprise as the design on paper "worked" perfectly. It's only when the device got hot, or was used beyond its recommended limits, or was connected incorrectly, etc.
A very large portion of my design time is figuring out what the user could do differently than what the directions for the product say to do. Another chunk of time is spent figuring which of these scenarios is worth incorporating protection against. For example...the firmware to control a new device we're working on is approx. 200 lines long. Another 800 lines of code (and some hardware) is required just to prevent major damage to the device if the user does any number of other things that are "wrong".

- Get yourself a web site and show off your design builds. Tell users what you do and how you do it. It doesn't have to be fancy (actually, IMHO, it shouldn't be) but everything better be spelled correctly and the photos better be good. A great site goes a long ways towards getting you business.

- What are people complaining about and how can you help? Modify every device you can find to make it better. If a device overheats, mount a better fan or heat sink. If an LCD screen is too hard to read, add LEDs or a backlight. And then take lots of pictures and put it on the web site. This will let users know what you can do for them and give you more experience. And it's less expensive than designing a new device.

- Be prepared to spend hundreds of hours looking for products, customers, reading datasheets and blowing up components. For this one new product, we spent close to 2000 hours over 2 years just to get it to the preproduction stage. But, we had very, very big demands for it and were prepared to work for the benefits we know we'll get when the unit is released for sale.

- Keep thinking of ways to help let users know that you have the experience they're looking for, that you'll do what you say you will and than you'll do it on time and on budget. You'll be amazed at how many people don't do what they say they will and are late and overbudget. Reliability and honesty are HUGE!!

- Find the flying clubs in your area, put together a good tool kit, soldering equip. wire, and some battery packs and chargers/inverters and head out to their fields. Help them repair anything that's broken, charge their batteries for them, do whatever you can to help. Then give your new smiling friend your business card. :) This not only gets the word out about you, but it lets you hear about what they need. You get to see what others are flying and once you talk to them, they'll usually be happy to vent about one or more things that angers them about a product. This is a perfect way in for you to bring up your servivces. Then, whatever you do, do it damn well and do it fast. Your customers will remember you and the word will slowly spread.

Wow. long post. :)
Just some random thoughts that might give you a few ideas.

John

[Acetronics]

For example...the firmware to control a new device we're working on is approx. 200 lines long. Another 800 lines of code (and some hardware) is required just to prevent major damage to the device if the user does any number of other things that are "wrong".


Hi,

The one who wrote those lines really knows about his subject !!! Idiotproofing is an art ... ;)

I also think last paragraph is as important as all previous ones together ... : show you have the knowledge !!!

good luck

Alain

[Comatose]

You're looking for small-quantity custom-designed stuff? Like, say, a guy who says "i have this battery and this system and need a protection system for it. I'll need five total"?

I pass on that sort of thing all the time, as its not what we're set up to do efficiently.

One very important thing to learn early on is the difference between something that works and something that works well. For example, the very first ParkBEC prototype worked. It was another 1000 hours of design time before we considered it commercially releasable. On the "quantity five" type projects, its hard to get someone to foot the bill for those extra 1000 hours to make it work well.

[JohnMuchow]

I totally agree with you Comatose.
We recently finished a short run of a tiny voltage-cutoff device to protect 12V SLA batteries and wouldn't have been able to come close to the client's budget unless the qty. was greater than 25pcs. and the design was just a modification of part of one of our standard products.

Though I think that leenremm might find that there's a market for "one-offs" (quantity = 1) by working with individual users who have specific needs for an electronic device. But, that word, "efficient", you brought up is critical IMHO. I don't think leenremm can make a profit on creating devices like that. We'll, not if he includes labor as part of his costs. I certainly do. :D

I think that leenremm and his friends can learn a lot doing small jobs and later move up to larger and larger quantities, or perhaps standard product offerings. But that will probably be many hundreds or thousands of hours (of unbillable time) later. They'll have to decide whether it's worth it or not.

Hmm...or perhaps just do it the way I got started? Create devices for themselves, then for their friends who like what they see, then a short run for the increased demand for a few devices (or just the one most popular), then start their own company and rake in the millions of $$.

OK...that last part may or may not happen. :D :D

John

[BushmanLA]

I so badly want to work making small fun RC and robotics projects when I finish my EE degree. The bad thing is I live in Louisiana and the only jobs are oil field related and the economy sucks in general (but the crawfish and Cajun music are great!).

I dream of making useful bits of kit like Sparkfun and Dimension do. If I have to keep working on directional drilling tools for a living I'll puke.

[ghoti]

I can't resist adding this rule for a successful business:

Always give your customer a little bit more than he expects.

[Dan Baldwin]

I can't resist adding this rule for a successful business:

Always give your customer a little bit more than he expects.

I second that. You have to anticipate what the customer actually needs, NOT what he is asking for. Several times, the customer told me exactly what they wanted, I programmed a device to do exactly what they said, and I got "Well why doesn't it do (something else)?".

Dan

[JohnMuchow]

LOL Dan! You're giving me all kinds of flashbacks....

Customer: What do you mean it doesn't do XXX?!?!?
Me: You never requested that feature.
Customer: But it's really, really important!!
Me: You never mentioned it before, ever.
Customer: But it's really, really important!! Can't you just quickly programn that feature in?
Me: It needs additional hardware to support it and you've already fielded all the units.
Customer: But it's really, really important!! I think it should be really easy to do.

I don't remember much after that other than grabbing a bottle of tequila, the customer's address, and a shotgun. :cool:

[Dan Baldwin]

The customer ALWAYS thinks that it should be easy to do.

Dan

[dirtybird]

If you look back at your post where you designed a cut off unit for charging A123 batteries. You put off a customer asking for a 10 cell cut off. A123 batteries require high voltages to work. There is a problem finding components for them because lipolys are expensive at high voltage and few people use them. There is an open field for ESC's and chargers that work up to 65 volts. Get busy and design something for us.

[Dan Baldwin]

There is an open field for ESC's and chargers that work up to 65 volts. Get busy and design something for us.

I respond better to requests than to demands.

Dan

[dirtybird]

OK. Please get busy and design something for us.
BTW you make more money when there is a demand for your product.

[Comatose]

I second that. You have to anticipate what the customer actually needs, NOT what he is asking for. Several times, the customer told me exactly what they wanted, I programmed a device to do exactly what they said, and I got "Well why doesn't it do (something else)?".

Dan

Oh, God! I have a $11,000 invoice that's going to go unpaid til the end of time because the thing did exactly what they asked for. This is completely true in custom work - you need to really grill your customer. Think about it, if they knew how to do it, why would they be having you do it? Given that, why expect they'll know what they really need?

[Dan Baldwin]

OK. Please get busy and design something for us.
BTW you make more money when there is a demand for your product.

I don't make any money on the DIY projects that I've posted, although they do take a great deal of my time. Even when I sell PC boards and components for these projects, I only ask enough to cover my costs. I do it to help fellow flyers out. My day job is building industrial equipment and controls.

There are several people flying with 10S and larger A123 packs, and charging using DB terminators. They just break the packs into 5 or 6 cell packs for charging.

I have already committed to a couple of projects, so I won't have any time in the near future to take on a new DIY terminator project, although I will look into making the non-DIY terminator handle higher cell counts if I get a chance. Perhaps the OP of this thread would like to take on this project.

We should probably take any further comments/criticisms of my terminator designs to my DIY terminator thread instead of crashing someone elses thread. http://www.rcgroups.com/forums/showthread.php?t=795122


Dan