Arduquee, a Arduino powered Marquee

[galannthegreat]

Oh, I know why it blocked that link, I put in my postal code (I don't know why they want that, marketing maybe) and it saved to my IP.

[madsci1016]

Thanks for the emails guys. I plan on having the evaluation kits together in a few weeks, but don't hold me to it. I'm still trying to find a cheap and reliable source for the LED matrices.

Assembled a V2 Slave board. Have a look.



More pics on my website.

Not sure I like the Cap on the back, but i don't know what else to do with it.

[Admin]

Not sure I like the Cap on the back, but i don't know what else to do with it.

You have plenty more room - get several SMD caps of smaller value and put them in parallel.

[madsci1016]

You have plenty more room - get several SMD caps of smaller value and put them in parallel.

I thought of that but i was trying to keep SMD to a minimum to make it easier for people to solder it together. (and maybe price? would several SMD caps cost more then a single 10uF cap?)

Opinions needed. Leave the big cap the stick out a little, but is easier to solder, or change to many SMD caps?

[galannthegreat]

They would probably be cheaper if you were to order per reel.

[Admin]

I think mouser.com can tell you whats cheaper

[madsci1016]

Opinions needed. Leave the big cap the stick out a little, but is easier to solder, or change to many SMD caps?

Still need thoughts on this though. Also, i'd have to go through another design and verification cycle if i changed to SMD caps.

[madsci1016]

I think mouser.com can tell you whats cheaper

You know how much I love Mouser.

Also, did you see that digi-key added sort by price now?

But anyway, back on topic. It's more a question if you guys think that cap will really be an issue for people. It is stopping the whole board from laying flat, but does that matter?

[galannthegreat]

Could you move the smaller coupling capacitor over a bit and fold the electrolytic over?

[madsci1016]

Could you move the smaller coupling capacitor over a bit and fold the electrolytic over?

I'm already using a very short 10uF cap, so that wouldn't buy much more.

[madsci1016]

Hmm, how about Multi-layer Ceramic caps?

http://search.digikey.com/scripts/DkSearch/dksus.dll?Detail&name=445-1371-1-ND

Can anyone with experience tell me if they equal electrolytic in surge suppression?

[chelmi]

Thanks for the emails guys. I plan on having the evaluation kits together in a few weeks, but don't hold me to it. I'm still trying to find a cheap and reliable source for the LED matrices.

I bought one from these guys http://www.sureelectronics.net/goods.php?id=230 a year ago (they used to sell they individually) and they were good. Shipping takes some times but it not as bad as Futurlec. I also bought a pack of 100 LEDs and they all work perfectly.

[madsci1016]

Dang, same ones I bought off ebay, but half the price! Thanks chelmi.

That makes the most expensive part a shift register, at least on the slave boards anyway.

My next problem will be getting more SMD Atmega328s. Glad I bought a handful from Sparkfun when I did.

[Admin]

Hmm, how about Multi-layer Ceramic caps?

http://search.digikey.com/scripts/DkSearch/dksus.dll?Detail&name=445-1371-1-ND

Can anyone with experience tell me if they equal electrolytic in surge suppression?

Ceramics are much better than electrolytics for surge suppression, but the disadvantage is ceramics need to be much bigger to hold the same charge. Tantelum capacitors have the best of both worlds, but they cost much more. I use tantelums on my Axon II.

[madsci1016]

, but the disadvantage is ceramics need to be much bigger to hold the same charge.

Mmm, well I found 10uF SMD ones for 6V, so that should cover it. I just ordered a handful from Digi-key to try out.

[madsci1016]

Grrr.

Apparently I missed the fact the ebay matrices are cathode columns, my design (and the $5 sparkfun matrices) are cathode rows. 

Anyone know of a cheaper source of LED matrices that are cathode rows? Also, anyone want to buy 20 LED matrices?

I emailed the supplier directly to see if they are willing to sell to me directly, but i doubt it (yep, no dice). If i have to change my design, it will be a real challenge. I will have to start looking at different shift registers with higher current ratings. I was able to make cheap shortcuts with the SFE configuration.

[Admin]

If it makes you feel better, it took me 7 prototypes and quite a lot of mistake money just to get the Axon designed. Then another 3 prototypes for the Axon II.

The Sabertooth motordriver took 27 prototypes

[madsci1016]

If it makes you feel better,

It does.

My first brush with redesign during prototyping. This is life I guess. It sucks that this will mean I'll have to increase final prices a bit to get back dev costs.

27? really? Dang, well at least the end result was a rock solid motor controller.

[GearMotion]

Also, anyone want to buy 20 LED matrices?

Only if they are ridiculously cheap. :-D

[madsci1016]

I'm going to  have to redesign. No way will I produce a product that ties the customer to an over priced single source for LED matrices. (I really don't like Sparkfun)

Stay tuned guys, it might be another month before I have kits ready.

[Admin]

Don't you just hate free-market competition?

http://www.sparkfun.com/commerce/news.php?id=452

Controlling RGB LED Matrices

[madsci1016]

Don't you just hate free-market competition?

Not when the competition is $60 per 8x8 matrix when mine will be 1/3 that price. 

I've been aware of the SFE designed boards, but they are over-engineered with a Atmega chip for each matrix, and overpriced.

My design is a single Atmega for the whole 'chain', only costing $6-8 in parts per board, not counting the LED matrix itself, but those can be had on eBay cheaply.

Thanks for the link though, I'm going to add a comment to the post.

[madsci1016]

Update guys:

Got the PCB in for the design that goes against the spec sheet current limits (and better judgment) and wouldn't you guess it? It doesn't work! Well, it lights, but it's obvious the IC is limiting current because of dim rows that vary brightness based off how many are lit in a row.

O well, I had to confirm the cheapest option doesn't work.

But never fear! I found a high current sourcing IC, and the design is being fabbed right now! Stay tuned.

[madsci1016]

Still waiting on PCBs for new design.

Update though, tried both my ideas for PWM color control, and both failed. The AVR isn't fast enough for more then one matrix. O well.

This means only three colors available on the bi-color screen, and 7 on the future RGB matrix.

Hope this isn't a turn off for too many potential customers.

[madsci1016]

Early Arduquee on my costume:

http://www.billporter.info/led-lab-coat-3-0/