Versatile developement board system design

[MrDEB]

Been working on several different configurations for a versatile development board that will accept most any micro but using as little pcboard real estate as feasible. This would do Adruino and Pic using same options such as LCD, LED matrix, keyboard, and most any other suggestions.

 

[misterT]

I use the largest chip in my favorite family when developing. When the product is "done" then you can make intelligent choices about what uC to use in the long run.
If you program in asm, then you may not have the luxury to choose completely freely..

 

[JonSea]

I hope the board design has more thought put into it than the survey options...since you haven't actually provided any choices.

 

[MrDEB]

Am trolling for suggestions of what people use. For instance I have several LCD's wired up with ribbon cable and 10 pin icd connectors. Add to this several boards with push-buttons etc. Even have several daughter boards for the TAP - 28. The board I am looking to maybe put together would be compatible with the TAP-28 which is what I presently use mostly but maybe a board with more bells n whistles?
Unlike some of the boards on the market that have a SMT chip already soldered to board.
As for choices, I need suggestions.

 

[kubeek]

I am not quite sure how you want to plug all those different micros into that development board. Will it have a header for arduino compatible boards? Will it have a dip socket? How about those micros that come in tqfp packages?

I think you should first throughly think about what you expect from your board, and go from there.

 

[JonSea]

I don't understand why you'd use a 10 pin cable to connect a character LCD. Most of them have either a 16 pin SIP connector or a 2x8 connector.

I don't think you need to support the TAP-28 in any way. There aren't many boards to plug into it and I doubt you'd have much crossover anyway. It's a totally different philosophy than what you have in mind.

 

[MrDEB]

Reason for 10 pin is that's all that is needed. I even wondered if an edge card connector would work?
As far as the TAP 28, I feel its a good step in the right direction but I want to have something to support the 18, 28 and 40 pin chips plus be able to use all the options that are available such as the LCD, led bi colored matrix, PWM, a keyboard and be able to use either the Adrunio or pic micros all in one inexpensive package. As far as supporting the SMD chips, that could be intergrated into this project as well. How? havn't really thought about supporting them but?

 

[kubeek]

This is how the arduino connectors looks like https://www.gammon.com.au/images/ArduinoUno_R3_Pinouts.png on the ardino uno.
Are you sure you mean arduino connector, and not the bare ATmega chips in dip package? And if it is so, are you sure that ATmega has the same power and ground pins as PICs do?

 

[JonSea]

Most development boards supporting HD44780 character LCD displays have either a 16x1 or 8x2 connector so the display can stack right on the board. It's true that most of these displays will operate in the 4-bit mode, but this is not universally the case. And most displays also need two pins for backlighting – 16 pins total.

I'm sure in your wisdom and with your Yankee sense of thrift, wasting 6 header pins is unwise, but the added cost of a special cable and adapter board that won't support a backlit display is false economy.

I'm sure your boards will truly be something else!

 

[MrDEB]

The LCD in my preliminary design is a break away pcboard that the hobbyist solders their LCD module to plus one resistor for the back light. Then configure a ribbon cable. Only 4 pins are not used.
The adrunio will in preliminary designs use the USB converter break out board that Sparkfun sells. The voltage issue will be addressed as well.
As I mentioned, I have only just started putting ideas on paper etc. Maybe bootloader would be better?? but looking at ease of use and cost.

 

[JonSea]

Ahhhhh.... I see. Advice and suggestions are only as good as what you want to hear.

Developing a dev board is certainly an ambitious undertaking for you.

 

[MrDEB]

After I come up with a method to allow more room on the main board I am rethinking the LCD set up. Ribbon cable or mount directly on to main board. I am just mulling over ideas. two main reasons for considering ribbon cable was freeing up board space and versatility in LCD placement perhaps even offer both options. This way the user could use the breakaway board for ease of using an LCD elsewhere. I for one have several wired LCDs and contemplated a pcBoard instead of point to point wiring.
Then the option of a serial LCD which was a passing thought.

 

[JonSea]

The advice you'll been given and ignored before is more useful now than ever.

You need to stop, quit drawing schematics, sampling hardware and drawing circuit boards and define what you're trying to do, including:

1. What is the target application? A general purpose development board to try out things? A board to be used for an application?

2. What makes this board different or better than the alternatives? Yes, you can pack it with features, but why should a person want your board instead of existing boards, perhaps designed by someone more knowledgeable in the field?

3. Who is the target audience of the board, and how will you reach this audience? If your audience is readers of ElectroTech, you need to think about how your talents are perceived here.

3a. What features are important to the target audience? What do they absolutely need? What features are desirable? What don't they care about? Tied to this to some extent is what programming languages does your target audience use? Are there any hardware features important to that language?

4. How will you supply this board? Assembled? A kit with parts? A bare board? If you're planning on a bare board, you need to design around parts that are easily available from many sources, including sources not in the US.

5. How will you fabricate the board? How much are you willing to invest in the hopes that someone buys these?

6. How will you license this/these boards? Creative Commons? Open Source? Will you be willing to let other people fabricate their own boards?

Those are but a few of the questions you should have answers to before you draw out a schematic or lay out a board.


Consider the TAP-28 board for a moment.

Why? Because at that time, printed circuit boards were expensive. The cheapest I could make a small quantity of boards was $75 for three. In order to get boards at a reasonable cost, I had to make 24+ at a time, which cost over $100. This is not the case any more. Ten boards for less than $20 delivered.

What's different? The TAP-28 was designed to support applications with a cheap board – TAP = ThrowAwayPIC – a board cheap to assemble with features to make it usable in the real world. It doesn't have bells and whistles, but it provides a basic building block for creating some amazing stuff.

Components required? There's nothing on the TAP-28 board that a reasonably stocked hobbiest won't have. And most of the parts can even be purchased at RadioShack.

So how much did I invest in the TAP-28? The first batch of 20-some boards was over $100. The second batch was about $250 for a hundred boards. The design and documentation took weeks of my time. The results? I had cheap boards for my projects. Hundreds of boards have been shipped around the world. I made back my monetary investment after only a couple years and enough "profit" to have a cheeseburger on the way to the post office when I had boards to ship.

Read that carefully. To build boards for $2.50 a piece, I had to buy 100 at a time. A twenty dollar bill will buy 10 today and get you change.

Oh yeah, what documentation are you planning on providing? If someone is going to use your board, you'll have to provide legible schematics. Seems like you've had some problems with this in the past as I recall?

You may create something amazing, but only if you stop and think before you start to do.

 

[AtomSoft]

i saved the above comment and love the info. As i also plan to create a dev system.

 

[MrDEB]

At the moment I am looking over my ideas and looking at parts availability.
I posted this thread to get some input on what people are using, doing and desires for a board.
I have a list of wants, part numbers and design ideas that take up 3 pages in my notebook already. Been looking at price as well that's why I am trying to cut down on PCboard real estate. I look forward to your suggestions and the cost of what it took to get the TAP28 to market. Interesting reading for sure.
Thanks for your insight.

 

[AtomSoft]

I cant wait to see what you come up with. While this doesnt really apply ... i like small boards with basic needs on them like mines:
https://atomsoft.wordpress.com/2014/04/06/pic18f4xj5x-explorer-v2/

But i do plan on making another. Larger. More pins etc. I would love something like MikroE development boards

 

[JonSea]

As I already stated, my reasons for developing the TAP-28 were largely selfish – I desired reasonably priced boards. To archive that goal, I provided a service to share the wealth with the community. I got cheap boards, and the community got reasonably priced, well designed and documented boards. I didn't market them as a money-making product. All told, I've made a few hundred bucks on an investment of a few hundred bucks over a couple years. If you factor in the time I've spent designing the boards, creating the documentation and so on, it's measured in cents per hour.

But keep in mind too, I had a specific design goal in mind – an application board that wasn't available in the market place. What advantages do you expect your board to have over everything else in the marketplace? There are an increasing number of PIC boards that support Arduino shields, designed by people with considerable knowledge. What will make yours different or better than the rest?

 

[3v0]

When I was on a project waiting for the hardware types to get the boards built and running I would get a simple development board as close to the final hardware as possible. No point in getting a complex board and then have to configure it.

Intel used to build big complex development boards and a series of processor specific daughter boards to plug into it. Did not care for them.

These days I just slap a uC onto a solerless BB and go from there if there is hardware I need to understand prior to doing a layout. Much of the time I just start on a PCB layout. Use OshPark.

IMHO development boards make a lot more sense back in the days of the Z80. Now all the IO is on the microcontroller chip. Why bother.

 

[AtomSoft]

A dev board is still useful now. Not so say your wrong 3v0 but i still use a couple microchip dev boards.

Technically the Arduino is a dev board and its very popular.

JonSea... "What advantages do you expect your board to have over everything else in the marketplace? "

This isnt the most important thing to be honest. Unless he plans to sell millions... which i doubt it the case. I dont think he is trying to TOP or be #1 in the market. He to me, is trying to make an affordable development system for users. You can make arduino clones and still sell out.

What can make a CLONE better is being made in USA over Italy... people love USA stuff... and it could be the exact same thing. So it doesnt really have to be the best thing. It just has to be the next thing or cheaper or simpler.

If you have code... people prefer something with lots of code and documentation. I can not stress that enough.

I just got that STM32-Nucleo and noticed its mbed... mbed is an online compiler... which is ok but not my style. Now i have to figure out what compiler i can use. I work with crossworks for arm... i have to search for info on that... why... because its not clear in the documentation.. but it only cost $12... so its a win!

 

[NorthGuy]

Technically the Arduino is a dev board and its very popular.

It is, but people don't use it as a development board. They assemble it with shields and directly use in their devices. When they need a next device, they buy a new board. That's why it is so popular,

I think the best development board is something that you can stick in a breadboard and then build circuits around it. This way you can test how parts work. Then you pint your own PCB. It is very hard to build anything close to the final product around development board.

To me, the best board would be very minimalist, but with a very good set of connectors, so that you can easily connect practically anything without wiring it on the breadboard.