Pictures...

Some pictures of the bar with a few of the LEDs turned downward to eliminate the bleeds and a few left upward to show the difference...

Diagnostic Red.
Diagnostic Green.
A quarter of the bar (1/2 of 1 Board Module wired)...
This is actually diamond but we just wired the LEDs incorrectly...
Some more...
...and more...
The LED lit liquor shelf... just a random picture.

Boards Finished

We finally finished populating and programming the boards. I had some stumbling points, but as always, the DDF team came up with the solutions...


Hey Guys--

I just finished populating my second board and am having a problem. I power it, OK. Program it, works. Verify it, OK. Connect it to USB, I get a COM port. Load up Clint's software and the config file. That works. I can even set diagnostic modes. Whenever I try to run fire or any scheme, it runs for what looks to be 3-10 cycles then stops. The RX light goes solid, then flashes for a short duty cycle about every half second or so with the TX light completely off. Any ideas?

-Jim Edwards

Hey Jim,
...Did you follow the oscillator calibration procedure (outlined in the DDF documentaton)? You might try going through that again. Also, which version of the DDF board do you have, and if you have version 2.05, how are the power jumpers set?[I have SJ1 jumpered up top and SJ2 jumpered to the left and two rev 2.05 boards.]

Uninstalling/reinstalling the FTDI driver shouldn't change anything, neither should switching the USB ports.

If redoing the oscillator calibration procedure doesn't solve the problem you could probably take it across the river sometime this week for one of us to look at.

Cheers,
Scott

First off, you could have some problems with that power supply, since the board can pull as much as 4A (with all channels on). That said, you really should set the jumpers for USB power mode which powers just the logic with the USB bus and yields better reliability especially with an insufficient power supply. I recommend downloading the newest firmware release from the project website (web.mit.edu/storborg/ddf). The latest firmware should handle missing or non-responsive drivers just fine (and even tag them as bad). Also, there's a PDF guide to flashing the board, which includes a description of the oscillator calibration procedure. This procedure may help, though the problem you are describing is not the typical presentation of a poorly calibrated oscillator.

Let me know if any of that helps.

Cheers,
Grant


I should have the LEDs wired by the end of the week, I figure connecting them and ordering them correctly should be done by the end of the weekend...

Stay Tuned!

UPDATES!!!

Back in business! I found the short between the 3.3V and 5V lines that was overheating the FTDI chip. (It was between the I²C address lines on one of the LED Driver ICs). I can only test a few of the LEDs since I haven't finished putting the connectors on the wires yet. The table is finished though... pics to come soon. Also, board two is going much smoother with the proper equipment. Don't try this with a RadioShack soldering iron... even a cheap-o other brand one... It makes too much difference with a good magnifier (or microscope), a good soldering iron, and a good copper braid to ignore. Anyway board 2 should be up and running by the end of the week, as should the rest of the wires.

If all goes well, this will be finished before September 1st, at least up until minor troubleshooting, so it can go in our new apartment.

Stay tuned!

Diagnostic Blue (Turning into Red)
Diagnostic Green
Diagnostic Red
Diamond Pattern

Updates

Sorry I haven't updated in a while... I got frustrated with the circuit board overheating so I began fabricating the coffee table and mount the resistor packs on the second circuit board... Stay tuned for pictures and updates coming within the next day or two...
Clint updated his software to v5, haven't been able to check it out yet, I can't wait...

What do you mean interactive? What the heck are reed switches?

By interactive I mean placing a glass down, or a coaster will...

or...

Decision

After a lengthy discussion with some friends, I decided to go with my idea and...

MAKE THE SURFACE INTERACTIVE

I'm going to go ahead and announce it, although I clearly hinted at it earlier as some have mentioned in emails when I posted my Digikey order with the reed sensors and magnets. So I plan on incorporating reed sensors into the indivuidal cells and making it interactive. While the DDF board allows for a sensor input as does the IC driver, I think reed switches would be a very clever addition.

Problem 1: SOLVED

After emailing Atmel Support, I got two responses, one from an email I thought went unnoticed and one from a Tech Support form on their website:

RESPONSE 1:
Dear Customer,

The procedure for manual firmware upgrade described in AVR Studio help should get the new firmware into the device. It is important that you follow the procedure exactly. Please do not use a USB hub without external power supply, as these some times does not fill the current de3livery requirements. The USB port should be able to deliver 200mA.

Problems

But First... Some Pictures...

 My AVR programmer...
...the contents.
Working!
Reading the chip!
Writing OK!
Verifying FAILED!
Leaving Programming Mode FAILED!
This advanced money saving system has helped me save over 300,000 dollars. ;-)



So by accident I clicked "update" in the AVR studio... this screwed up the whole deal. The programmer got stuck in an update loop and the manual update instructions are no help,
only because they just don't work. I emailed tech support @ amtel -- see if I get any help...

So just messin around I plug in the USB cable into the board-- it doesn't work-- because I didn't install the software, presumably-- so I unplug it and as I do so I put a fingertip
on the FTDI chip... OUCH! Its buring hot. I take out the USB cable quickly and then the barrel connector and it quickly cools. I wondered what made it overheat so I plug in the 5V
barrel connector (that now fits, with the new jack!) and its temperature shot up there... unplugged it and let it cool then tried the USB... that didn't make it shoot up. Weird. I
email Scott @ MIT to see if he has any ideas...



Since I can't program the ATMEGA and test the first board I guess I'll get starting on populating the second... although that provides no luminary gratification... oh well.

Clint's Floor! (and my new car...)

Clint is almost done his floor. I'm really impressed-- one of the projects that made me decide to make mine-- check it out!

Change of Plans...

No longer are we doing a bar... It will now be a living room table / coffee table... basically this is for two reasons.

  1. Although changing a dancefloor into a bar is no awesome feat of genius, it was an original idea [[here]]... and doesn't serve our purpose to replicate existing designs anymore...
  2. No one wants to stand up to play with an interactive bar... but play with an interactive coffee table while sitting down? That might just work...

Help From The Masters...

Scott Torborg from MIT/DDF Hall of Fame writes...

Hey Jim-

I noticed your blog posting about "Mysterious Components" on the DDF board. Here's an explanation:


X1 is an external ceramic resonator for the AVR microcontroller. There's no reason to use it unless you want to run the AVR faster than 8MHz.



R47 is a 1M stabilization resistor for X1. Basically, it provides a large resistance that serves to help the oscillator stabilize faster on startup. It is rarely necessary but may be with some ceramic resonators.

Updates

Whole Board. Notice how I had to "depolarize" the polarized headers on half one of the quadrants because I flipped the connectors around by accident before i soldered them. I could have left them polarized the wrong way, but chose to keep it consistent.
The last two halves of quadrant connectors and left unfinished so I can finish the SMD passives I can't figure out.
USB and Power Connector mounted. Still figuring how I should bridge the solder jumpers...

How Small is Small?

Corner View (12/16 Resister Packs per quadrant)
A little blurry but look at those resistor packs!
Whole Board View

Surface Mount Soldering

Board with 16 LED Drivers and 8 Resistor Arrays (Red)
...again.
LED Drivers Mounted. Notice the terrible soldering job on the resistor arrays (they were done first). However, they were all tested to be good joints with no bridges.
The parts for the hot air rework pencil I attempted to make off of tips from engadget. It was really just useless, it didn't ever heat the air enough to melt the solder.
After I decided on the Flood and Suck Method [infidigm.net] I bought the desoldering pump from RS. Didn't work too well, the high pitch made it hard to fit a soldering iron and the pump in the same space.
This worked much better. Had it laying around after the hot air pencil idea didn't work.

Matrix Fabrication and Parts Arrival

Some pictures for your viewing pleasure...

Bar Project Background

We are making a bar. I will not take credit for the LED surface... it is not my idea. It was MIT's idea (see link below). Further, modifying the dancefloor into a bar was not my idea either. That was the idea of the creator of the Disco Bar (also, see link below). I'd like to credit all the people who devised and developed the idea so if I somehow forgot you please let me know.

Syndicate content

User login