Saturday, January 1, 2011

Quick Xbox controler hack

Recently I found an old Xbox controller in my projects bin. Not having an Xbox, I decided to do something that was last popular five years ago, and make it work with USB.

Xbox meets USB
Tinned and ready to go. Xbox controler cable on the left, USB + heat shrink tubing on the right.

For those of you playing at home, the Xbox controller is already a USB device. All you have to do is lop the existing end of the cable off, put the end of a standard USB cord onto it, and your done! The wires inside of my cord and controller were even color coded correctly, so all I had to do was match colors. There will be one extra wire coming from the controller, ignore it.

Soldered and shrink-tubed, if you were particular about things you could replace the shielding, but I didn't feel the need to.

I installed the XBCD drivers, plugged the controller in and all the functions were recognized. Since then I've been using it to control Orbiter, for which the controller works fine, but my reentry skills seem to leave a lot to be desired. So far all my attempts have been more meteor than space shuttle.

Saturday, December 25, 2010

Fixing a Sticky SRAM X7 Rear Derailer

I have a SRAM X7 rear derailer on my Big Dummy, and over the past few months it has stopped dropping into the highest gear. I dumped it in the parts washer a few times and that would clear up the problem for a short while, but sooner or later it would start hesitating again. John had a similar problem, and SRAM replaced his derailer, but I wanted to fix mine. Once you start drilling parts off your derailer, it’s not likely that SRAM will take it back, so don’t give them a hard time if you mess it up.

I replaced the steel pins in the derailer with brass tubes. This reduced friction on the parallelogram enough that even off the bike the derailer’s motion was noticeably smoother. On the bike, the derailer feels lighter and drops faster than before, and is perhaps only a little less crisp. The tubing does introduce a bit more play into the derailer, but this is only noticeable on the bench, and certainly isn’t large enough to impact shifting accuracy.

Step One: Break It More.

The derailer is held together by four semi-tubular rivets, which also function as the pivots for the parallelogram. The shop heads are easy to drill out with a 3/16” bit, and the ridges on the shaft behind the factory head will prevent the rivet from spinning if you use a light touch. Remove the bottom pin in the lower knuckle by drilling out the shop end and tapping it out with an awl and hammer. This pin also holds the spring and once removed unloads the derailer, making the rest of the operation much easier.

With the first pin removed, test the rotation of the other three pivots independently. When I tested mine the top plate was binding at both pins, but the bottom plate rotated freely. I decided to leave the bottom plate rear pin in place, and replace the other three pins. Drill out the pins that you need to replace, and set them aside. At this point the spring should be free, and the top plate of the parallelogram can be removed. This is as far as the derailer has to be disassembled.

Step Two: Make New Pins.

The pins are 3.969mm (5/32”) in diameter, and their length varies by their location in the derailleur. Rather than measuring the length of the removed pins, which may have been too short in the first place, and now had one end drilled down, measure the outside dimension of the plates at the pivot. The top plate is not square to the axis of rotation, so measure at several points for each pivot and use the highest value. Using the high value ensures that the new pin will clear the ends of the plate.

My local hardware store, and most hobby shops, carries round stock and tubing in small sizes. To replace the pins I chose 5/32” brass tube, and to secure it M3 30mm bolts with a pair of washers and a hex nut. If you chose to replace the bottom plate rear pin, note that it is considerably longer than the rest of the pins, and a bolt longer than 30mm would be required. Brass has the advantages of being cheap, available, and very easy to cut and work. The downside of the tubing is that it adds the risk of the tube crimping under load, but with the close tolerances on the plates and the knuckles should support the tube and prevent deformation. An option if you are able to make accurate cuts in round stock would be to use a circlip on either end, which would be simpler and allow the use of steel pins.

Cut the tubing to the width of the pivot plus a millimeter or so. So long as the length of the tube is greater than the width of the pivot the length does not have to be particularly precise. Since the pins length is specific to the pivot, mark the pins after cutting them. For 5/32” brass tube fine tooth hobby saws cut it well. Remove burrs and square the end of the tube with a bit of 200 grit sand paper, and remove any interior burrs with a #2 X-Acto knife.

Cut and deburr a short (15mm to 20mm) piece of brass tube to hold the spring in the lower knuckle during reassembly.

Step 3: Reassembly.

Reassembly has to occur under tension for the last two pins, as it is very difficult to install the spring after the pins. Attach both plates and the spring to the upper knuckle, and spread the plates apart. The exposed hook on the spring will loop around the lower knuckle bottom pin, slide it into the notch in the lower knuckle and slide the short piece of tube you cut earlier into the lower pivot hole until it is securing the spring and centered in the knuckle.

Pull the knuckles apart until you can align the top plate pivot holes and the lower knuckle. Slide the new pin for the pivot in. This will constrain the movement of the lower knuckle and make it easier to slide the lower pin in. Slide the lower pin in, which will push the brass tube holding the spring in place out, and transfer the spring to the new pin.

Once all the pins are in place, center them up on the knuckles and then install the washers and bolts. Apply some blue Loctite to the bolts, then thread on the nuts and tighten them down.

Test the movement of the derailer, the spring should drive it to its closed position from any point without hesitation or binding. There should be only a small amount of sideways play in the parallelogram. If everything checks out, install the derailer on the bike and adjust the shifting.


With the weather I haven’t gotten to out on any long rides, but so far the derailer is performing much better than before. We will see how well it holds up, but for now it was a fun project and improved the shifting for a very low cost. There are a few more photos of various parts on my Flickr.

Monday, June 7, 2010

Silca Pista

Somebody at my shop threw an old Silca Pista into the metal recycling bin.

I took it back out.

The hose was cracking at the ends, the chuck needed a new washer and isn't schrader compatible, and the gauge was smashed. Oh, and the pump couldn't go over about 30 psi without leaking around the plunger seal. With most pumps this would mean that it was indeed trashed, but unless you've hit it with a car, Silca pumps just don't die.

I cut down the hose (reusable spring hose clamps made this a snap), replaced the chuck, and installed a new spacer and leather washer kit on the plunger. I don't care about the broken gauge, but I could have replaced that as well.

Total retail cost for all the parts I used: $14.

The pump now works wonderfully, is compatible with presta and schrader, and can hit over 200 psi. And if any of the parts fail, they are generally available at either a local shop or on order from QBP.

As a side note, the barrel of the pump is made from nicer tubing than most bikes. Including all of my bikes. Awesomeness. I'm very tempted to mount it under the boom tube of the Sea Goat as an over-sized frame pump.

Update 1/23/2011 : Replaced the smashed gauge with one from a different pump that failed. Yay for common parts standards!

Monday, March 22, 2010

The Electric Electra

This repair was interesting. I thrive on oddball requests, but sometimes bikes come in where one look sends me running the other direction. Sometimes, I don't run quite fast enough.

I got called down to do a brake adjustment on the bike, and immediately noticed that it had a hub motor, and that the axle was significantly larger than the 9x1 QR that the fork was designed to take. To make it fit, the dropouts had been filed out, and the torque on the anti-rotation washers had distorted them further (750W motor). The brake adjustment was needed because the wheel had slipped down in the dropouts, which I pointed out. I also recommended that we not do any further work on the bike unless the safety issues were cleared up.

The owner agreed, and we worked up an estimate for fixing most everything. The existing fork would have to stay, as there isn't a 26" fork with an extended threaded steerer that fits a 14mm axle currently available. The wheel could be spaced out to 100mm to fit the fork instead of the fork being forced to fit the 80mm hub. The rear rack had failed, and would have to be replaced. The rear wheel would be rebuilt with a new rim and spokes to take the weight of the rider and the battery pack.

I figured that this would take about a day of work and billed it accordingly, plus an allowance for parts. At the end of the first day I realized I'd made a bit of an error in the time estimate.

I had made 2 trips to the local Ace, one for hardware to adapt the mount for the battery to the new rack, and the second to find 14mm, 1.5 pitch nuts to use as spacers on the axle. I'd designed and produced the mount, declared it fit for service and mounted the battery and the rack. This took a few pretty enjoyable hours of measuring, scribing and drilling 16 holes accurately with a hand drill. The results were worth it however, as it came out looking very nice, and is extremely sturdy. The front wheel was an entirely different can of worms.

There was a nipple that had come off rolling around in the wheel, it was horribly out of true, and the electrical connection terminated at the motor controller. To work on the wheel a good half of the electrical system had to be removed and the wires punched out of the molex-style connector (to pass through the 14mm nut). The front wheel took more than a few hours of work, and although it tickled my problem solving side, the design is incredibly heinous.

The second day was spent building the rear wheel (fun), and putting everything I'd taken apart back together (frustrating). All told, I spent 15-20 hours on this one repair. At the end of it I was confident that it would hold up to any reasonable road ride.

When he picked up the bike, I found out that he'd been riding an incredibly heavy and overpowered bike over curbs. Which perhaps explains why everything was destroyed.

Tuesday, March 16, 2010

The Great White (Bike)

One of the best parts about working in a bike shop is that I get to meet people that are awesome, and who I might otherwise never cross paths with. Recently, a Davisite welded up his own longtail cargo bike, and I've gotten to put on a few of the finishing touches, such as reaming out the seat tube and facing and tapping the parts that need it. Which in and of itself is always fun, but that I'm doing it on a bike that someone built and cares about, makes it much more rewarding.

It's due back in so that we can see it all built up, hopefully I'll be there to take a picture or two and post them up here.

Tuesday, March 2, 2010

Tire Tire Liner Fail.

A very nice couple from West Sac stopped by the shop and asked us to replace their tandem's rear tube, which had sprung a leak on the ride over. I was summoned forth to complete said task, and while they were getting entered into the system I chatted with the husband. He mentioned that it was surprising that the rear had gone flat, as he'd "put two tire liners in there."

Sadly, doubling up on tire liners is not a particularly good idea, and so I figured that a pinch flat was likely the cause of his leak. I was correct, but I'd underestimated the extent of the problem.The entire tread of another tire had been stuffed in there as well. Since the actual tire is an old Schwinn 26" the tire they stuffed in there was slightly too small, and because it was in one piece (unlike commercial tire liners) it couldn't expand to seat properly. So it absolutely chewed the tube to pieces as it moved around on each rotation of the wheel. To make matters worse, the second liner had deteriorated and cracked lengthwise in several places, which was also wearing through the tube.

I took out all the extraneous liners and put in a new tube, which worked fine. I explained what had caused the flat, and he said, "So my fix was the problem...Oh man, my wife's gonna kill me!"

Thursday, January 14, 2010

What to do with 75 pounds of oranges?

In a word: Marmalade. Evil Mad Scientist Labs cooks up a batch, and gives tips on scaling the recipe. The real gem is their parting line, "You may recognize our technique as one common in mathematics. We have reduced a difficult problem (what to do with 75 pounds of citrus) into a problem whose solution is well known: what to do with many jars of marmalade. QED."

Reminds me of Cooking for Engineers, and Cooking with Engineers.