SparkFun Forums

[bigglez]

Greetings Andy,

(1) The LED ballast resistors (100 ohms) will only yield
27mA peak for the LEDs. A value of 15 ohms will put
you at 106mA peak (desired).

Thanks Peter! quick question though: how did you calculate these numbers? I would have thought at 100 ohms I'd have I = 5v / 100ohms = 50mA (still not enough however).

Remember the LEDs are 8:1 MUX'd, so the peak current
is eight times the average. For 100mA peak the total
resistance is (in words) "Supply minus LED drop, divided
by desired peak current".

(5 - 2.2)/0.1 = 28 ohms.

But, there is equivalent resistance in the FETs so the
next step is to find "28 ohms - Fet resistance".

The TPIC is 7 ohms, the other FET is 200mR and can be
ignored. So 21 ohms (use 22 ohms standard value).

I Spiced the circuit and looked at a 5% loss in supply
voltage and worst case tolerance on the other components,
that brought me to 15 Ohms. You can pick the mean
average and use 18 Ohms. 15 to 22 Ohms is not going to
make much difference, and is within Abs Max for all
components.




Comments Welcome!

[arader]

got it, thanks!

one last quick question: Since I'll be planning on daisy chaining the boards, I'm going to have 4 connectors on each pcb (top, left, bottom, right). These are J1 - J4 in the schematic. The left most backpack will be connected to the powersupply, the rest will use one of the edge connectors for SPI and for power. Since there are potentially 4 different places power can reach the board, should I use 4 47uF caps? or can I just do with 1 central one and tie all of the power connectors to that?

[bigglez]

Greetings Andy,

Since there are potentially 4 different places power can reach the board, should I use 4 47uF caps? or can I just do with 1 central one and tie all of the power connectors to that?

Not critical. You can place the 47uF anywhere,
as long as it is across the power and ground.
(obviously, you'll have the other caps near
the supply pins on each driver IC and the AVR).

Skipping ahead a bit, you'll need to widen the
power and ground traces on the PCB, to reduce
the voltage drops and inductance. Better yet
make one side of the PCB into two planes (Vdd
and Ground) and hog out signal traces as required.

On a side note, hav you defined the data protocol?
I'm wondering what traffic is on the bus between
the various modules?

Comments Welcome!

[arader]

Skipping ahead a bit, you'll need to widen the
power and ground traces on the PCB, to reduce
the voltage drops and inductance. Better yet
make one side of the PCB into two planes (Vdd
and Ground) and hog out signal traces as required.

the design of the PCB is definitely going to be tricky, especially since I need the pcb to be smaller that 31.7x31.7mm. I already did a quick layout and it's fairly obvious I need to place components on both sides of the board. Do you think something this small would benefit from a 4 layer board?

On a side note, hav you defined the data protocol?
I'm wondering what traffic is on the bus between
the various modules?

I haven't designed the details yet, but I have a few ideas basically my goal is so that each node knows its row/col values, and can either request or be given that data

edit: I'm going to be ordering parts today, is there anything at all missing from the schematic? any last minute changes?

edit 2: Here's the parts I'll be ordering from digikey:

296-4620-1-ND - IC 8BIT SHFT REG TRI-ST 16-TSSOP
296-21143-5-ND - IC 8BIT SHIFT REGISTER 16-TSSOP
296-17619-1-ND - IC ANALOG MUX/DEMUX HS 16-TSSOP
FDC6306PCT-ND - MOSFET P-CHAN DUAL 20V SSOT6
RR08P10.0KDCT-ND - RES 10.0K OHM 1/16W .5% 0603 SMD
RR08Q15DCT-ND - RES 15 OHM 1/16W .5% 0603 SMD
PCC2367CT-ND - CAP CERAMIC 10UF 4V X5R 0603
490-3907-1-ND - CAP CER 47UF 6.3V X5R 1206

I'm mainly concerned about the caps, will they work fine?

[bigglez]

RES 15 OHM 1/16W .5% 0603 SMD
CAP CERAMIC 10UF 4V X5R 0603
CAP CER 47UF 6.3V X5R 1206
I'm mainly concerned about the caps, will they work fine?

Once again, you have to pick components with higher
ratings that your intended circuit.

The resistor will dissipate I^2 * R = 0.1 * 0.1 * 15 = 150 mWatts
peak, or 150/8 average power. You should use 250mW
minimum for these.

Edit: 150/8 = 18.75mW, so 1/16 W (62.5mW) resistors
are not too small. 1/10 W (100mW) or 1/8 W (125mW) would
be a better (practical) choice.

The 10uF ceramic cap has to have at least a five
volt rating, and I'd sleep better at night with 2X
the working voltage. Try an electrolytic 10uF 16V
SMT part such as Digikey PCE3061CT-ND

Same for the 47uF Cap, try Digikey PCE3064CT-ND

The ceramic caps are okay, but cost a lot more.
The small electrolytics have good performance.
Alternatively, a Tantalum cap such as Digikey
478-1739-1-ND

Comments Welcome!

[arader]

I'm moving a discussion out of the PCB forum back to here, the original discussion can be found here

Why would the data flow in both directions? Doesn't
new data arrive and flow from module to module?

In most cases it won't, but I really want my design to be software limited rather than hardware limited. For example, I could implement a very interesting Game of Life if I allow two way comms.

Instead if each backpack only talks to the 4 directly connected to it, I can get something that more closely relates to the internet.

The internet?

It's just an example what I mean is that there's no single point of failure, and that all nodes are completely identical.

Part of the problem is that you are attempting a
complex design for your first project. Also, I'm
unfamiliar with the tools you are using, but it
looks as if there is no ERC and DRC functions,
that would catch silly (and obscurely silly) errors.

That's pretty much the norm for me, start out with what I can't do and find out what I can. I've already played around with much of this stuff elsewhere, this just happens to be my first time wanting to use a custom PCB for it. I've downloaded the freeware version of Eagle, hopefully it will help some.

If a traditional prototype (with hand wiring) is not
your cup of tea, consider making a two sides over-size
prototype to prove the concept. Sounds as if you'd need
four (or more) modules to work out the comms?

Once you have these working together spin the PCB
layout (adding more layers) to reach the final size.

This is definitely a good idea. I'm going to play around in Eagle for a while and see what I can come up with. If it looks like making the small pcb will be complicated and error prone, I'll scale up in order to at least have a working prototype.

Thanks!

[tdg8934]

I just bought 2 of these boards (that the price keeps going up):

http://www.sureelectronics.net/goods.php?id=147

I have been able to do some basic LED scrolling but really need some help.

The link has another link to a manual - if anyone wants to help.

I am using a Parallax SX-28 micro-controller with SX/B (Basic like language).

http://forums.parallax.com/forums/defau ... 7&m=273480

Thanks!