Sunday, October 21, 2012

Power Wheels Robo Magellan platform?

My 5 year old friend Adam has promised me his Power Wheels jeep to convert in to a robotics platform - no bribe and not sure how he even got the idea.

I did promise to drive him around in it when I convert it to R/C steering control first.  It has a decent amount of space for equipment without modifying the child compartment.

Lots of possibilities like after converting it to remote control, setting it up for Robo Magellan (although to big for competition) would be fun challenge.  Could then plop in my future kids and drive them around the park based on the GPS;  Technology enabled yes, however will that be wife approved?

7-bit vs 8-bit addresses in I2C

Discovered that my I2C device address was incorrect (at least my understanding).  After getting the compass working, I modified the Quickstarter code a but and hooked up the RGB LED Backpack.  No success, but then reviewed the Arduino  Logic Analyzer capture again.

Found that my 'address' appears to be 0xFE - not 0xFF as expected.  Also noticed that when it works, the LA tool I2C decoder is in 'READ' mode.  That little bit didn't make sense.  I changed the address in my modified compass code and the RGB LED Backpack came to life.

After a very confused Google search session, I contact Pierce at Logos Electromechanical asking puzzled questions.  He clued me in on some subtleties regarding addresses.  The key was this very clean high-level I2C tutorial that he shared.  THANKS!

Turns out the problem was that when I used the bit-bang method, it wasn't accounting for the 'WRITE' bit.  So when the L-A tool picked up the address it showed the address's 2-8 bits and the 'write' bit.  This accounts for the 0xFF = 0xFE on LA.  I need to do some more binary math to convenience myself... 

I2C with Compass module via Parallax Kickstarter

As mentioned on a prior post, lack of process understanding toward a working Propeller I2C interface and the RGB backpack is highlighting how little I know of I2C in general.  My career focuses on clinical application interfaces - so this is all in the same space right?  Nope!  My application interfaces are way up in the stack, no where near bit-banging stuff.  In the quest for understanding -- many mental cob-webs have been disturbed regarding byte and bit shift manipulation with a smattering of HEX based math.

This Parallax Kickstarter has been a good way for me to get back to the basics and provide that I can build an I2C interface and capture the output on the logic analyzer. Also this gave me a proven path to build up some logic glue around it with a NTE490 based level shifter based on the Philips I2C application note.

The other advantage of the kickstarter is that it right to the point in terms of basic functionality. Several I2C objects are out there, they look good, however setting them properly has been a challenge given that I don't know the basics.  Nice job Parallax!  There is a side effect as well given the examples range across 3 platforms.  I am brushing up on my C for Arduino and future work with the Prop GCC compiler.

Saturday, October 20, 2012

RGB Backpack - I2C on Propeller

Working with the Parallax Propeller Professional Development board and the Logos Electromechanical RGB Backpack.  Attempting to recreate the I2C connectivity I accomplished with the Arduino.  So far - NO LUCK.

At a high level I have attempted to modify a couple different I2C objects from the exchange, and use demos to communicate with the backpack - however the backpack pulls the SDA low and holds it there.

Without the backpack attached, my logic analyzer did report valid I2C data coming out of the Prop and decode it as I expected based on what I sent; this was uni-directional however - no ACK.

BTW - the Propeller also introduced the need for a level shifter, Frys had the MOS-FETs required so that looks good.

Next Steps - Focus on I2C understanding and perhaps other hardware to interface.  Checking out the 'kickstarts' examples next...

Tuesday, October 16, 2012

RGB LED Backpack - Logos Electromechanical

Playing with my old neglected Arduino Diecimila board. Had to refigure out how it worked (still managed to get it going without reading) after spending LOTs of time working on Parallax Propeller boards...

Built the Logos Electromechanical RGB Backpack tonight.  Required basic soldering and building a wire harness (had parts on hand - nice!)

Logos EM provided a demo Arduino file to test out the backpack.  After relearning how to compile and upload (stumbled on picking the right board config for the Ardino - forgot about that requirement), otherwise it worked first time.  PIN-Outs matches on my cable, soldering was good.

Noticed that Logos EM is using the WSWire lib, ended up flipping it back to Wire lib to avoid modifying additional hardware or code - looks like pull-ups R aren't enabled otherwise.

Its been while since building hardware on the StempleLABs bench, so this was a nice evening project.

Monday, September 17, 2012

Early 2009 iMAC clean install

Good thing this wasn't a production environment - but then it would not be as much fun!!

I was stuck in Lion given that my primary productivity application 'Paperless by Mariner Software' was not and still is not ready to support Mt. Lion.  Considering another solution for PIM/Scan file management.  Bummer.

I decided to move that into a VM for Lion and do a clean install of Mt Lion on the iMac.  Took a LONG time cause I didn't go from point A to point B given my curiosity.  In this case curiosity did kill the cat, the big one - Mt Lion!  Recovered by going all the way back...  OUCH.

BTW another change for change sake was switching from VMWare Fushion to Parallels.  After seeing all of the reviews of Parallels 8 and the speed, I found that Parallels had a deal to switch from VMWare for the same cost as the VMWare 5 upgrade.  So far so good.

Still coping files back from the Time Capsule and setup iLife and other things however the now clean install appears to have been worth the extra hassle. The iMac had successive installs of Leopard, Snow Leopard, Lion and now a clean Mt. Lion.

Saturday, April 28, 2012

Parallax projects ready for show and tell

To get ready for the Parallax Expo 2012, I pulled all my boxes of projects and parts. I had no idea I had this much stuff. Parallax is a great source for parts that are plug and play plus lots of documentation! Also they are located in my back yard so its a good combo! Thanks Parallax!!

Fluke 233 Power Saver Hack

My first Fluke, got the wrong one last year. Turns out this flexibility of this meter--which appealed to me--is the down fall for light use. I discovered that between uses, the batteries would die. This seemed odd since the $3 meter I bought from Harbor Fright is still using the batteries it came with several years later.

While researching this, found out that the meter runs RF and or IR data links even when off. This translated into power drain.  The meter uses 5 AA batteries.  At times I wouldn't use it but once on one set of batteries. This was a bummer.  Considered selling, but there are lots of these meters for sale. Also noticed that the place I bought this meter no longer carries it.  I didn't do enough research before plunking down he cash on this tool.

Per my research, Fluke recommends putting paper between the contacts of the battery to prevent the power drain.  This is a no go given that it takes screw driver to open the port. I did the next best thing - hack on a couple switches. It takes two switches given that the screen has a batteries and the base unit too.  The batteries are in series with + to - connector.  I cut one set of the connectors and soldered in switches. Shrink wrapped the switches and hot glued them on--total hack but works great so far.

Tuesday, March 20, 2012

StingRay IR Line Follower Kit demo

Upgraded the StingRay and tested the IR line follower with Parallax demo code. Adjusted only a couple things to get it to proof of concept working. Looks like this will be a slick hardware add-on that will allow for lots of software tweaks! NICE.

Madera, CA FIRST Robotics FRC Event

Looking forward to another volunteer opportunity with the FRIST FRC program. April 4-7, 2012 Madera, CA will be hosting the FRC event. I am signed up and confirmed for field assemble, robot inspecting, field repair/reset and field disassemble. Going to be a long couple of days! I had a blast working the Sacramento Regional.

This will be the first year for the event at Madera, hope to get a chance to influence some of the processes they will use for years to come. I compiled a bunch of notes from the Sac event and started a draft for a robot inspectors' handbook. I suspect one may already exist, but if not I intend to make it available for utilization to the program. My talents tend to be in abstracting technical to non-technical with strong communication skills. I found that the whole FIRST program is very appealing because it allows me to put my MBA in practice in a very technical way...

Sunday, March 11, 2012

Parallax Expo 2012 - Robot Line-Following Contest

Getting ready for my 3rd Unofficial Parallax Expo April 13 & 14. Looks like the name changed and its out of beta testing to become the Official Expo this year as two day event with several contests. This will be the first robot related event I will be competing in - 200 ft line following contest.

After considering a Lego bot, I decided to add on to the StringRay.  Picked up this IR tracker unit after scoping out a couple options from both Parallax and Pololu. The unit pictured appears to be the most robust in terms of features and development assistance capabilities. Also the most expensive, however it will cut the hardware development time to near ZERO and allow me to focus on writing the Prop Spin code; still very new to me so the less time tweaking hardware the better in order to be ready for the event.

I am waiting to hear back in my StringRay's RC radio for manual control will be considered 'illegal' for the event. It is how I manage the kill switch and camera controls.

When I purchased this line-follower unit, it was the last one in stock. I suspect I am not the only one going after this. Parallax does a good job of promoting its products and sponsors a lot of Free events like this one. I am sure it pays back in product sales in the long run. I have a nice little collection of Parallax stuff in my stock pile...

Tuesday, March 6, 2012

Perception is a funny thing... Windows 8

I tend to watch and manage perceptions very closely with regard to the services I manage in my profession. Its not often when I step back and think about my own perceptions.

Having read several early notes about the MS Windows 8 preview, I downloaded it, created a new VM and fired it up. No issues initially but found the UI on a non-tablet device to be disorienting.  I am still getting use to the mouse scroll button going the opposite way on OS-X Lion (even though I am use to it on the iPhone).

During the install I was finding myself concern with the personal info required to install Windows.  It wasn't until the next day that I figured out what it was trying to do; getting all APPLE'ISH.  Perception is funny because when I bought into the MAC stack several years back I wasn't that concerned about getting an iTunes account.  But when Microsoft does it, better watch out!?!  Why is that?

Apple has done a very good job appealing to the consumer, guess that is why at work I have a Windows box.  After getting past Microsoft's attempt to FRIEND me, I immediately started to wounder how the Metro thing would really work in the enterprise environment. For heath care I suspect it would be great given the tendency to run a single application as a primary tool. Some touch screens would make it easier on the desktop.

Since I didn't have a touch screen on the desktop, I had to fumble around a new user interface trying to figure out how to close an app (drag down from the top).  Disappointed to report I didn't figure it out without Google... It always humbling when learning a new platform--its a good reminder to have patience when support others who don't sit in front of a computer for a living.

Certified Robot Inspector for Sac FRC Regional 2012

Looking forward to volunteering at this year's FRC (F.I.R.S.T. Robot Competition). I have read about this event in SERVO the past couple years, however seeing one of these incredible robots first hand at Robogames got me really interested.

Given this is my first FRC event, I have a lot to catch up on.  Spent most of Saturday reading the various guides and rule books. After being selected for something technical I got to run through a certification exam (complete with a certificate). I was a little concerned that I might have been selected to volunteer as a food vendor or parking lot attendant... I would have done it either way.

After finding out about the certification requirement, I thought perhaps I was in over my head. Of course after doing homework it all was a lot more clear. The robots all share common sub systems that are configured in standard ways.  The fun begins when the participants design the platforms for the sub systems and the various end effectors attached. Here is the description of the job duty.

I would love to get my hands on a KOP (kit of parts). Lots of hardware goodies and software; heavily subsidized by various sponsors. FIRST was around when I was in school, but not at my school.

I am looking forward to this because it will continue to expose me to more of the concepts I picked up from the Sierra College CIE program. Now that the MBA is complete I need to finish the CIE AA.  I suspect given the Junior College crunch right now getting enrolled will be harder!

Monday, March 5, 2012

StempleLABS Stingray robot - part 1 - Introduction

This Stingray robot chassis and drive system is a Parallax kit that I purchased at the Propeller Expo 2011. It came with the now discontinue Propeller Robot Control board.  This included regulated power, motor speed controller and other peripherals.  The kit was built right away and I tested with various demo programs, but the it sat on the shelf with my other bots (all in various stages of form and function).

A couple weeks later I added the PING))) kit for Stringray (3 front facing pings) and again run demo programs and minor coding.  It was hungry for AA batteries so I didn't get very far...

Blade 450 flys again, needs more tweaking

Of course after my first crash, in order to maintain enthusiasm for this hobby, I considered the challenge of taking it apart and back together again. Turns out that was the easy part. With new parts things were slightly out of adjustment. I started to adjust only to find other things were not out.

Back to reading about theory again. Figured out specifically how the swash mix works, and how to tweak the gyros. I turned off the gyro and attempted to drive around on the ground. Quickly I gained a sense of how critical that technology really is and I developed a new sense of how complex the PID loop is that is in use. Looking forward to understanding more about flybar-less setups in the future.

Current challenge is to tweak out the vibrations I get on initial spool up. With the slightest tap on the rudder it comes out of it for a smooth hoover.  However how left stick on the rudder causes the bird to drop like a rock. I suspect my HEAD SPEED isn't were it needs to be. My pitch is likely off also. I need more tools! A contact less RPM gauge. Its on the wish list. I used a pitch gauge and swash plate leveler. Now that I understand how these tools work, I have seen online others that would appear to work a lot better then my set!

For now, more simulator time. Good news is on the sim I am running circuits and hovering upside down...

Final parts replacement tally:
 - Flybar
 - Rudder blades
 - Main rotor blades
 - Rotor shaft
 - Blade grips
 - Spindle
 - Head dampeners
 - Drive gear

Sunday, February 26, 2012

Large motor test rig

Built a motor test rig to experiment with a powerful Sabertooth 2x25 motor controller by Dimension Engineering.  I suspect that I have selected a motor controller that is larger then required for these motors.  So far this controller is very solid. 

The motors come from old ride on scooters, no specs.  Haven't done any real investigation on capabilities yet.  On the rig now it is controlled by a 5k pot in analog mode.  The goal is to test with each mode.

The power supply is rated at 5amp max and already tops out if significant changes in current occur.  The rig is left over scrap plastic.  Lots of rough cuts to in the motors mounted.  I don't plan on running these with much load so there aren't any vents.  This may change in the future.

Still grounded... Blade 450 heli

After this crash I replaced about $30 of parts. I don't have the pitch angle tool yet (china eBay order = slow). I adjusted the blade tracking but it wasn't enough. I need to retrim out the whole head.

Couldn't get off the ground and still have lots of vibration. Guess I will be logging more sim time and read J. Salt's ebook on setup tips. Swash Plat Setup Guide

Flight 36 resulted in this crash.  Lost orientation.
Damaged parts removed

@StempleLABS on twitter

Setup a twitter account for StempleLABS.  Looking for follow electronics and robot related twitter feeds. is back

In 2010 I was late to renew the domain registration, a domain parking company stole it.  They wasted their money because I waited them out over 2011.

I was running my own server with the domain registered through a dynamic DNS service.  Since then with so many cloud based servers and cheaper domain management, I got out that.  Now this domain is registered for $10 and hosted through Google's blogger service.  Very cheap for me and significantly less support and maintenance on my part is required.  I am going to miss my owner server, however my interests have changed--been there done that--so looking for new challenges.  It saves some $$ on the utility bills.

So far Blogger has been very slick to work with.  Its quick and appears to allow as much customization as one wishes to invest the time is configuring...