Continue to Site

Welcome to EDAboard.com

Welcome to our site! EDAboard.com is an international Electronics Discussion Forum focused on EDA software, circuits, schematics, books, theory, papers, asic, pld, 8051, DSP, Network, RF, Analog Design, PCB, Service Manuals... and a whole lot more! To participate you need to register. Registration is free. Click here to register now.

Dual color dot matrix display...should I make a kit?

Status
Not open for further replies.

Los Frijoles

Full Member level 1
Joined
Oct 5, 2006
Messages
98
Helped
9
Reputation
18
Reaction score
5
Trophy points
1,288
Location
Provo, Utah, United States
Activity points
2,252
dot matrix display how

I just finished my hardware designs for a dual color dot matrix display which is controlled by a 8-bit bus. As I was looking at the price to fabricate it (I am using olimex.com), it occurred to me that it would be nice if I distributed this as a kit so that people could have the same thing I came up with without the development cost (I've already spent probably $50-$60 working on this).

Basically the cost breakdown is as follows:
- ~$60-$70 for the PCB and breakout board
- ~$50 for parts
Which gives a total cost of: ~$120

Now the main question is: Assuming this gets off the ground, do you guys think that the average hobbyist would be willing to buy something like this? Being a hobbyist myself I know that $120 is quite a chunk of change for the average shoestring budget. It would come pre-programmed with the microcontroller already on it. Being a kit it would have to be assembled by the buyer rather than myself, so I might was well say right off the bat that the board is almost entirely surface mount with 0.4mm being the finest pitch parts used. I would assemble it myself, but then I would have to factor in labor costs and that would make the cost of the board go up.

The full physical and technical specs:
- 6.3" x 3.05" with two mounting holes on either side of the board .4" from the bottom and .1" from the side.
- Power and control supplied by a 41-pin Hirose DF9 connector. A breakout board would be supplied with the kit since I fit that into the PCB panel.
- Since LED rows are the only things on at the same time it should draw 800mA-1.2A at the most. Worst case scenario (and defying the hardware) it would draw 4A with all the LEDs on.
- Controller is a PIC18F4550 using the PLL to get 12MIPS
- Frame speed is going to be either 30fps or 24fps
- A bootloader of some sort (I still have to write the software) for programming over the 8-bit bus. It would probably hook up to the parallel port on a computer, but not many of those are around anymore.
- Each LED (red & green) will have 4 levels of brighness, so that gives a total of 16 "colors" per pixel.
- Control is done over an 8-bit bus
- All chip components (resistors, capacitors) are size 0805
- Diodes are SOD-123
- IC sizes used are TQFP-44 and TSSOP

1/10th scale demo for seeing how bright it is/isn't: https://www.youtube.com/watch?v=CC-wYQkWSWk

I will have virtual pics of the board as soon as I make the 3d model for the dot matrix modules.
 

hobby project colour dot matrix

Los Frijoles said:
Now the main question is: Assuming this gets off the ground, do you guys think that the average hobbyist would be willing to buy something like this? Being a hobbyist myself I know that $120 is quite a chunk of change for the average shoestring budget.
This is always a tough question to answer - is there a need in the market for this product, and if so how large of a market. I don't really understand what the intended use of this device is, so I can't comment on the market potential. Before sinking a lot of time into it, I'd survey the market to see what similar devices are available and get an idea of the market size. To be successful you'll have to be price-competitive. You also have to consider how you intend on marketing it.

Los Frijoles said:
...so I might was well say right off the bat that the board is almost entirely surface mount with 0.4mm being the finest pitch parts used. I would assemble it myself, but then I would have to factor in labor costs and that would make the cost of the board go up.
I'll tell you from personal experience that surface mount kits are hard to sell. Most people buy kits over assembled units to save money. Surface mount devices are hard to solder and the majority of people woun't but them - again, I know this from personel experience. You should see if you can make this with thru-hole components.

BTW - You may want to look into using "Logic Level MOSFET's".
 

how to make matrix display

Well, I know there is a market out there for dot matrix display modules for experimentation and such because there are several vendors on ebay who seem to sell that sort of thing regularly. However, I don't know much about what exactly people want in a display. I am using this for a dot matrix clock and I just thought it would be nice if someone was thinking of designing something like this and doesn't want to re-invent the wheel.

As for through hole components, its not happening. I tried and failed with my smaller design (half the column number), so there is no way this larger design is going to fit on one board with all those components. Not to mention that I have resorted to mounting some of the chips beneath the dot matrices (as in between the bottom of the matrix and the top of the board) rather than on the bottom of the board. Although that will make debugging hard I made sure that the chips I did that with were not as likely to need debugging as other chips on the board such as the microcontroller.

Here is the eagle layout:
**broken link removed**
As you can see, it is extremely dense (0.01" routing grid) and even then I had a hard time getting everything to fit. Even at 175dpi not all the connections are shown. The lines of holes beneath LED1-10 are spaced a 0.1" to give a bit of a scale.
 

eagle dot-matrix display

Hi,

It sure is a great looking project - well done.

However as you have already worked out, the cost of the parts is well over $120 but that doesn't include anything for phone calls, delivery charges etc, never mind your time for preparing the kits or building them - and then of course Pofit - no point in doing something if you don't make money.

If it was me I would carry on building a prototypes and put your basic plans, less the program code, on the web, and just sell the programmed chip.
If you sell loads then you know there is a market, so you can then look at selling a better updated version as a full kit / readymade.

The other thing, do finish the project as many employers would be well impressed if you have done all of that yourself.

good luck
 

Status
Not open for further replies.

Part and Inventory Search

Welcome to EDABoard.com

Sponsor

Back
Top