Winter is here, and studs are slow....

Took the Big Dummy out for the first spin with its new sneakers. 19.1 miles, and the incredibly fast 11.7 mph average speed. It turns out the studded tires really suck the energy, particularly on pavement. For some reason they don't seem to be any worse than normal tires on hard pack snow, though. They did work well- I didn't have any issues on any of the icy sections, even on climbs. You could feel some small amount of slipping, but nothing like what you'd get with normal tires.

I met John on the paved trail on the west side of 494, then we headed over to Minnetonka to have some coffee. All in all, a fun ride.

Funny how the 29'er makes even the Big Dummy look like a kid's bike!

Surly Big Dummy all dressed up for winter

I just installed some Nokian Mount and Ground 1.9" 160 studded tires on the Big Dummy. I'm hoping to keep riding in to work, no matter the conditions. I've never tried studded tires before. These things look like they'll make a huge difference. Not only will the studs help, but the rubber is also super tacky and there are wide gaps between the lugs for shedding snow chunks.

I took the liberty of putting the bike on my bathroom scale- standing it up on the rear wheel. With the studded tires and light setup it clocked in at a round 50 pounds. Ouch. No wonder why it was a little tough to get out of the work stand! At least it'll be great training for spring.

Riding to work in the winter beats driving even more than in the summer, at least in my opinion. Every ride is an adventure. Plus, there is nothing like riding home with the trail lit up by your lights and being surrounded by twinkling reflections from each falling snowflake.

DIY Wooden Beer Case

I'm on the Instructables weekly mailing list, and a couple months ago I saw a neat project someone had built- home made wooden beer bottle cases. I thought this would be a great Christmas gift for my brother, who is an avid home beer brewing enthusiast. Shh, don't tell....

The instructable project was nice project, but I thought for a gift I needed to kick it up a notch or two. For one, the referenced project used simple screw and glue construction to join the four sides of the case together. That works great, but isn't the prettiest. I thought I'd try dovetailed joint construction instead. I've never tried it, but I thought it would be a great skill to learn. It also a much nicer looking and stronger joint.

I picked up a General Tools dovetail jig from Home Depot. It's pretty basic, but much cheaper than a nicer setup from Rockler. Since the sides of the crate are 12" tall, you have to reposition the jig three times per side. A bit time-consuming, but not a big deal. Once the board is clamped in the jig, you use the included router bit to form the "tails" and "pins" of the joint. I chose to use a half-blind dovetail, with the exposed tails on the handle side of the beer crate.

I sized the crate to fit 12x standard beer bottles. I discovered shorty Sierra Nevadas don't fit, but anything "normal" will fit fine. The sides are constructed from 1"x12" pine, and the bottom, sliding cover, and internal dividers are made from 1/4" birch plywood.

The internal dividers are cut from 1/4" birch plywood ripped to 6" tall strips, cut to length, then notched so they lock together. I didn't glue them together or to the box so they can be easily removed if necessary.

Crate 1 of 2 ready for finishing!

To finish it off, I'm planning on using T-shirt iron-on transfer paper to print out some custom brewery decals and iron them on to the wood. That should really add some nice personalization to the project. There is a nice article also on Instructables about this technique. 

Getting more life out of a 1997 Niterider Digital Head Trip

Now that I've gotten into electric RC airplanes, I have plenty of decent LIPO batteries and their assorted chargers around. This greatly simplifies making a LED bike light- now I just need the light itself, and the associated electronics to drive the light. I'd like to break this project into two components:

1) Build a device to allow me to drive my existing old-school 6V Niterider digital head trip lights off of 7.4V and 11.1V LIPO batteries. This will include a voltage regulator and a low voltage cutoff to prevent the LIPOs from being over-discharged and damaging the batteries.
2) Source appropriate LED lighting system to drive with the LIPOs.

So, my old Niterider is a 6V model, with maximum 15 watt power usage. (from my memory of the advertisements, not actual measurements)

Using P=V^2/R, and V=IR:

R = .42 ohms
I = 2.4A

A possible method to both regulate the voltage and protect the battery packs would be to use a UBEC. Hobbyking has one potential model: http://www.hobbyking.com/hobbycity/store/uh_viewItem.asp?idproduct=2160. This product can output either 5.0v or 6.2v depending upon jumper setting, plus it is capable of 4A current draw and has a low voltage warning buzzer. And it only costs $11.25.

This seems like a good solution to me. Cheap, easy and simple! 

As part of an order from Hobbyking for some RC airplane parts, I added a UBEC to try with my existing bike lights. I ended up ordering a different unit than I specified in the previous post- I went with this one instead: TR-UBEC15. It was a couple more dollars, but it has a much higher current capacity as well as being in stock.

Converting the UBEC to power the Niterider Digital head trip was easy- but ultimately unsuccessful.

Step 1: Cut off stock connectors off of UBEC and solder on EC3 (to connect with my LIPO batteries)

Step 2: Cut open old, worn out Niterider NIMH battery pack to recover the cord:

Step 3: Solder the NiteRider power cable to the UBEC output. 

The NiteRider plug is a "N" style plug, available at your neighborhood RadioShack. It's also center positive. In my cable, the white wire was the center positive, and the black was the outside ground. The red wire was unused. The colors may vary, so be sure to use a voltmeter to double check which lead is correct.

When everything was soldered and insulated with heat shrink- I did a quick short check then fired it up. It worked just fine- unfortunately the UBEC didn't deliver 6V- it was closer to 5.8V. The NiteRider Digital head unit didn't like that, and registered it as a flat battery and went into the low power "limp home" mode.

Bummer. Next time I place a Hobbyking order I'll try another UBEC, and hopefully I can find one with a specified higher voltage output. This might be a good option- 7.2V output- will overdrive the bulb slightly but should avoid triggering the head unit low voltage error. 

FYI, I finally ordered and soldered together the new UBEC (which works). See the post HERE.

In the mean time, I made a simple charge adapter to allow my RC battery charger to charge my old NiHM NiteRider batteries. It just a Type N power socket (from RadioShack) connected to an EC3 connector.

It works great. The big advantage is that the charger is programmable- so I can have it run multiple discharge-charge cycles automatically to cycle the battery pack. This is particularly important since the packs have been sitting around half-discharged for the last year or so.

Lake Level Sensor- Success!

I was able to get the lake sensor system to successfully push data to google docs: LINK

The “lakelevel” is the distance in inches from the sensor to the lake surface. I need to change this to lake height above sea level in feet and inches. I also need to add some sort of smoothing function to remove ripples and waves from the output. The water temp is in degrees F. (not working correctly at the moment) The photocell is relative sky brightness- 0 is no light, 1000+ is full sunlight. Air temp is in degrees F, and air RH is relative humidity.

I’m having some sort of hardware issue with the lake temperature probe- you’ll notice the anomalous reading in that column. Right now it’s sending data to google docs every 10 minutes. Once I figure out the water temperature probe, I’m going to add some data visualization -graphs over time + a graphical current status icon - then attempt to send those to the Mooney lake association website.

Also, the current code is a resource hog on the host PC. I need to dramatically reduce CPU use and then export the application as a stand-alone executable. Right now it's running from the Processing development environment. 

It's been a very interesting project- a great excuse to dive a bit deeper into hobby electronics and programming. 

Lake Sensor Project: Software

I'll admit that my software experience is pretty limited- mostly some classes way back in college. Not that this sensor package is anything terribly complicated by any standard, but it's been a nice learning experience. It's also been challenging, especially with a new crying baby and lack of sleep making concentrating on a new subject difficult!

So, at the moment the sensor package is transmitting data just fine. Originally I wanted the sensor package to transmit a packet of data only when requested. This worked fine when I had the PC and the Arduino directly connected via a USB cable, but this didn't seem to work with an XBee wireless module sending the data through the ether. I'm sure this is an easy problem to solve, but my limited web surfing didn't track down an obvious solution. To work around this I decided to have the sensor package continuously transmit data, at least for now.

The picture above is the raw data being fed into my laptop, after been wirelessly sent from the lake sensor to the XBee module. As you can see, it's just a list of numbers currently, delimited by commas. Here is what they are:

ultrasonic sensor, ping time to lake surface and back in milliseconds,  water temperature in degrees C (obviously not right, something needs to be fixed there), photocell resistance (lower numbers = darker), air temp in degrees C, relative humidity.

The next steps are:

1. write some code to import this data into a usable form (probably using the free language Processing). Make Magazine seems to have a great tutorial on this.
2. convert data into proper form- millisecond ping data into lake level in feet above sea level, temps converted into degrees F, etc.
3. send the data into a google docs spreadsheet
4. send to Mooney Lake association website

Lake Sensor Project: Hardware Construction

Bare PVC Electrical Box

Drilling holes for ultrasonic distance sensor

Ultrasonic sensor, DHT22, and plug for thermistor installed:

All sensors are fixed in place and holes sealed with hot glue. The box should be essentially water tight.

Sensors wired to external daughterboard (for mounting pull up, pull down resistors for various sensors)

Daughterboard connected to XBEE shield via ribbon cables

Boards installed in box w/ self adhesive velcro. 

The hardware is essentially done. Eventually I'd like to set up a battery and solar powered system. Having an extension cord run down the hill to the dock is a little kludgy. Ideally I'd use some old RC airplane lithium polymer batteries that can't supply enough current to fly aircraft well any more, but would still work fine for this application. However, LIPO batteries need specialized charging circuitry that would require more research than I have time for right now.