Group Actions
Engagement
  • Author Author: hifromkatie
  • Date Created: Date Created
  • Views 862 views
  • Likes 1 like
  • Comments 4 comments
Related
Recommended

Jacket Progress

hifromkatie

Everything is now sewn on to the jacket:

 

 

 

I've carried out an LED test to check that I'm ok powering from  3.3V and it seems that everything is ok, so that's saved routing another power line down the length of my arm.

 

 

A quick overview of what's on the jacket and what it's for:

 

The FLORA:                          Main processing unit

The FLORA GPS module:      Current location

Compass:                              Current facing

colour sensor:                         Program destination

8 LED pixels:                         8 compass points for direction in 'find me home' mode, 8 speed indicators, 1 mph steps, rolls around to display the next 8mph once the first 8 has been exceeded.

 

+ late addition....

light sensor:                         I was planning to route a standard switch into the cuff to switch between direction and speed modes, but then whilst I was working out the route for it I realised I could use the light sensor as a switch. This can then be run on the SCL and SDA line which is already down the sleeve for the compass and colour sensor.

 

 

 

Issues:

 

The first Issue i've found now it is connected is that the first neopixel sporadically flashes, from a quick probe I think there may be an intermittent connection between the output data line and the power rail, so I will check this out next.

 

The second issue is that the GPS wasn't getting a fix at first, I wired it up on the jacket with crocodile clips and found that the GND connection isn't all that reliable, so I'm going to reinforce this will a few more strands to beef the connection up, I may do this to all the power rails as I'm running quite a lot quite a long way! My jacket sleeve is much longer than any track that I've routed before, and the thread is much thinner than I would use as standard track on a PCB.

 

My third issue is that comms with the other sensors is a little iffy, sometimes I can get data returned, but the majority of the time I can't. also the colour sensor LED is a bit flickery, this all indicates that there is a wiring issue, similar to above. I'm not really sure what to do about this, I knew at the start that I was pushing my luck with conductive thread going down the length of the sleeve, I will try to add some more threads to the connections to reduce the resistance of the connections to try to improve the connections.

 

I've got other issues and tweaks, but I'll progress to those once the basics are sorted.

  • in reply to zengirl2

    The light sensor illuminates the first LED when running, the issue with the loose conenction is the last LED.

     

    I was really hoping to be further along, but unfortuantly I've had a terrible week at work this week, so I'm much further behind unfortunatly image Still going to be working on it though, as I want it finished as soon as possible.

  • in reply to shabaz

    Good idea regarding changing down the speed, I'll see if it's possible once i've got a moment, I hadn't thought of it.

    The comms is I2C, Ideally I'd like the connections much shorter, but unfortunately they would make it a little impractical for the original purpose, so I would like to do as much as I can to try to make it work this way.

     

    The LED issue is  a loose connection on the final LED, as pressing on the connection helps it, it should just be a simple tightening of the connection.

  • Katie,

     

    I can't speak to the light sensor library program, but I know the color sensor program blinks the first neopixel as a warning to let you know it is about to take a reading.  Could this be an explanation?  As far as your idea of adding more thread for the long track on the sleeve -- I definitely would do that.  I once had a switch fail on a costume because the distance was too great -- probably about 5 yds.  I didn't have time to do extra thread in that case, but I'm sure that would have probably solved the issue. 

     

    Keep at it -- it sounds cool and I think your idea of using the light sensor as a switch is brilliant. 

  • Hi Katie,

     

    Interesting project. From your earlier photo the wiring to the 8 LEDs looked nice (not that I know anything about sewing :-)

    I suppose, if the remainder LEDs work and just the first one flickers, it proves that the wiring through the arm is good, and maybe the power rail to the first LED is not making good contact to it? It's just a suggestion, it sounds like you know what your're doing anyway.

    As for any I2C type comms (I'm guessing this is what the sensors use?), that is hard over a distance, I've sometimes seen issues over 20-30cm, if it is at high speed. If you can slow down the I2C comms bit periods then that will help. I couldn't find the schematic for the Flora board, but sometimes it can help reducing the pullup resistor values (e.g. down to about 1.5k resistance). But, you probably want to try a slower bus speed first (or maybe Adafruit may already have information on how long the I2C can run for using the Flora board and typical wiring to the supplied sensors in the kit).