Jack Crossfire's blog View Details
Posted by Jack Crossfire | Jan 09, 2016 @ 03:37 PM | 2,401 Views
Canon EF 50mm mechanism (1 min 34 sec)

So after powering up the lens with a minimum of components, the motor turned briefly then stalled again. After manually turning it, it turned more consistently. It never stalled again.

The internet can't agree on what type of motor it is & might have confused it long ago with the stepper motor in the F1.8 or what they term as a "micro motor". The motor impedances are over 300k ohms. Steppers have much lower impedances. The motor looks like the other ultrasonic motor documented on the internet. Canon would get sued for printing USM on it if it was a stepper. So rest assured, it's a USM.

The only other teardown showing any detail of a USM motor was


Its wire bundle went to a soldered point which failed. The 50mm's wire bundle went straight in the motor with no soldering.

Based on the scarce documentation, USM motors work more like human muscles than what's marketed as muscle wire. They contain thousands of tiny ratchets which move the payload. In the lens, the tiny ratchets are on the end of a large piston shaped thing. Most of the volume of the motor is a pancake of ceramic & metal alloys dedicated to just generating the vibrations. Only the very end of the pancake converts the vibrations to motion.

Because the tiny ratchets are always engaged with the payload, the motor has very high torque. Turning the stiff lens...Continue Reading
Posted by Jack Crossfire | Jan 08, 2016 @ 10:33 PM | 2,345 Views
So someone at the Goog some time ago introduced an algorithm which finds related channels to your gootube channel, probably based on browsing history of your viewers. To this day, computers can't recognize similarities in video content, only the valuation of ages old click counting has been increased to prevent a recession.

It popped up Casey Neistad some time ago, but never paid attention, figuring he was another weirdo who subscribes to the heroineworshipper channel. In fact, the concept of subscribing to a channel is still a bit offputting. Why couldn't they call it following or favoriting? How many parking spaces on 237 were filled to come to this decision?

It turned out he was one of the 1st reviewers of the hoverboard, previously was the Twit photo guest who was sponsored by gopro, a very well known videographer who never finished highschool, & lives in a huge 9 figure apartment in Manhattan.

Neistad videos are boring as batshit, but the technical quality is always spot on. Focus is always sharp. Editing is perfectly timed. Lighting is always bright. It's an enormous investment to make daily videos that well produced.

Sometimes in the drivel, there is something you've never seen before. Sometimes he has a novel solution to a common problem, like getting a hoverboard up a curb. The most novel visual at this time is seeing him talking to a camera while riding an electric skateboard. There used to be very pooly lit drive time rants from Dave. Then...Continue Reading
Posted by Jack Crossfire | Jan 08, 2016 @ 10:13 PM | 2,300 Views
After 8 years & hardly any use, the mighty 50mm died. Its autofocus died, while everything else still worked. The focus ring still moved, but the motor was dead. The internet abounds with failed autofocus tales, manely emphasizing broken mechanics. The gears break. The cylinder gets bent. Regarding the motor, the internet is silent. The motors are accepted to be without failures in the 30 years they've been around.

It might be good enough without autofocus to live with. In the days when foot transportation abounds, the normal lens camera of choice has been the phone. The DSLR is rarely used for anything. There's hardly any need for a 50mm lens except for the irrationality of a camera which can't take normal photos.

The only smoking gun is the motor is very hard to turn. It could never turn on its own unless it had a very tiny planetary gear in the motor casing. The 50mm f1.4 only arrived because it was the old man's lens. A better investment would be something lighter like the f1.8 or cheap wide zoom because 50mm is never used....Continue Reading
Posted by Jack Crossfire | Jan 03, 2016 @ 06:49 PM | 2,898 Views
Another holy grail in robotics is the ability to transport a DSLR. Unfortunately, the 1st experiment in this took it on trails, which failed miserably. On pavement, the vehicle was barely manageable. It had a constant steering oscillation, either due to the top heavy payload, the padding, or the Futaba servo setting up a sympathetic oscillation with the gyro heading. No amount of D or P gain made a difference.

Suspect harder suspension would make a big difference for these payloads. The wheelie bar was useless at preventing backflips.

Once on the trail, backflipping during the ascent abounded. Wheel hubs slipped inside the tires. Despite another dry December, the trail found a way to be packed with mud. As altitude increased, the mud became impassible. Walked with the lunchbox over most of it. Finally, the battery died after only 3.6 miles.

Driving a DSLR up this trail will never happen, because of the limited battery range. It might happen with the gopro, when dry. Driving a DSLR on the usual paved trail might happen with stiffer suspension.
Posted by Jack Crossfire | Jan 01, 2016 @ 05:12 PM | 3,706 Views
These Tower Hobby servos date back to the DVD robot of long ago. They were $6 in those days, but should have held up. It went hundreds of miles, then snapped during a mild crash after 1 mile. There are no metal gears for standard servos. Now going through all the Futaba servos in the apartment until they need new gears. The Hitec servos won't fit in the lunchbox's servo saver. Only Futaba & generic Towers will.

The trick with servos is the screw needs to go down past the case. That should transfer the load from the screw to the case, rather than from the screw to the gear to the case. Unfortunately, servo horn screws are yet another extremely rare hobby screw size with no more local shop. It would be more effective to eliminate the "servo saver" so the lone existing screw from 1988 can reach far enough.

After some grinding & 2 stripped screws, managed to get a custom horn loosely fitting where the servo saver went. The screws would never be a perfect fit. These servos arrived from China with only the Futaba S3003 sticker but nothing else matching the photos, no screws or attachments. Only stray Sanwa mounting screws & Sanwa horns from 1988 fit.

The screws are 1.6mm inner diameter, 2.2mm outer diameter, at least 9.5mm long, 0.825mm thread spacing.

After test driving in manual mode, it was time to drill out the Chinese gear & use a standard screw size to get the steering horn flush. It drifted quite a bit.
Posted by Jack Crossfire | Dec 31, 2015 @ 02:49 AM | 4,734 Views
A week after XMas, the mighty Team Associated 1/18 scale 14T pinion & 10T pinion arrived. Installed the 14T & experienced no change in efficiency compared to the plastic 16T. The plastic with heatshrink & glue showed no signs of coming off, so it might have been good enough. The 14T was cheap enough to not be worth the trouble of using the 16T. The change in efficiency was insignificant enough to not bother with the 10T.

What did show signs of improvement was tuning the steering to use more damping. The steering on the Ruckus oscillated quite a bit without a damping constant. Once the damping was jacked up, it drove straighter & power usage dropped from 230mAh/mile to 200mAh/mile on 1 test drive. The best explanation was the brushed servo took a lot of power to oscillate. It was a case of making a big difference in power usage by merely changing a software parameter.
Posted by Jack Crossfire | Dec 27, 2015 @ 09:23 PM | 2,456 Views
Hauled 2 batteries to do 10 miles + 3 miles with cargo. Decided to press to 13 miles in the end without stopping to pick up cargo. Massive discounts at Target meant not running home, but driving some cargo at 13 min/mile & carrying the rest for another 1.4 miles. Didn't have any flips. Hauling 2 batteries, the 1st battery died after 11.5 miles & took 4430mAh, so a slight reduction in range from 12 miles with no cargo.
The 2nd battery took 2300mAh to go 3 miles on account of cargo.

The next challenge would be hauling a DSLR 20 miles, but the mileage will probably be devoted to 12 mile grocery hauls for some time. Cold weather is unsticking hot glue everywhere.
Posted by Jack Crossfire | Dec 27, 2015 @ 03:05 AM | 3,002 Views
Expanding the cargo capacity began with a container that was too small to reach all the mounting holes. Even if it was expanded as originally planned, it would interfere with the radio, so the easiest solution was aluminum extender rods. They probably weighed as much as any attempt to expand the container.

The plastic container was trimmed & lights installed. A common string proved effective at securing a test payload, though hard to get off.

The 1st drive was 3 miles to walmart & 3 miles back with the maximum payload at 6 - 6.66mph. It backflipped when starting, so the string had to be undone & the payload had to be shifted over the front wheels. Don't think the wheelie bar would help because the objective is to keep the payload from falling out. Didn't try any sharp turns at speed. It was pretty unstable. It took 3541mAh to go 6.4 miles, so quite an uptick with the payload.
Posted by Jack Crossfire | Dec 25, 2015 @ 06:25 PM | 2,766 Views
The answer is no. Self balancing hoverboards don't really self balance or hover & they're extremely rare except by 1 guy on 2nd street. The range goes up, every time they're advertised, 1st 5 miles, then 10 miles, then 20 miles. Like most breakthroughs in the economic boom, they're not real. There's no information on how they work & no reviews by anyone who isn't a reseller, but the advertisements reveal some clues. Sellers are careful to place them upright on the ground to appear self balancing, but sometimes tip them over to show the battery indicator. There are no closeups of the electronics & no-one has any idea how they work.

The user must actively shift on top of the thing to stay balanced. The user presses 2 buttons on the foot rest to activate the motor. Pressing 1 button causes it to go forward until the acceleration is straight down. Pressing the other button causes it to go backward. Judging from what look like I2C lines on a blurry photo, there's an MPU6000 on each footrest, so they do sense true tilt angle.

They're 1 of the growing number of things for which no-one has ever made a firmware hack because no-one knows how. The 1st step in using 1 for hauling cargo is making it balance itself. This seems possible with the existing hardware. They only omitted it because it probably was easier to control if the user had to press a button to activate the motors & specify the direction.

Self balancing boards don't stay still if the...Continue Reading
Posted by Jack Crossfire | Dec 24, 2015 @ 11:47 PM | 2,342 Views
The lunchbox had some mounting holes in a more convenient place that would allow a larger payload. Rigged a quick cargo platform with headlights out of cardboard.

Hauled 1 qt of egg nog back, proving the lunchbox's ability to move significant cargo. Battery died .75 miles from home, after going 10.25 miles at 6mph. It only made 4.75 miles with the cargo. It still worked in 45F, but the voltage dropped all the way down to 5V because unlike the analog gyro of old, the MPU6000 needs only 3.3V. Didn't have any flipping issues. Helas, duct tape didn't stick to the payload, allowing it to shift forward onto the front wheels. Payload on the front wheels is very inefficient.

A tupperware with partitions would work better & allow more cargo than a flat piece. It needs only mounting points for the bolts & 2 holes for the shock absorbers. Tupperware with holes for string would keep the heaviest pieces over the back wheels.

The largest payloads of interest:
1 qt milk carton: 9"x3"x3"
soap refill: 9.5"x5"x4.25"
mane cleaner: 9"x3.75"x2.125"

The lunchbox dimensions dictate a 9.5"x8.5" area. A standard tupperware could be cut up & resized. A metal cage would work better because there wouldn't be any drag, but would be heavier.


cutting a 4.25" segment out of both & joining to make 4"x10"x8.5"
Laundry detergent & mouthwash would have to be in the smallest size. It would finally achieve its original goal of moving DSLR gear, water, extra batteries for very long runs.
Posted by Jack Crossfire | Dec 23, 2015 @ 06:48 PM | 2,690 Views
The vehicle undergoes extensive testing before its 1 way trip to the city. The heatshrunk pinion gear continued to hold up for 12 miles with the full gopro, 4Ah 7.4V battery, & headlights on. Used the 16 tooth pinion, going 9-10min/mile on flat ground. It won't go out until installing a properly fitting 14 tooth pinion.

Current use with the least efficient configuration has been 220-240mAh/mile. A 4.3 mile drive used 1Ah. A 7.75 mile drive used 1747mAh. The 7.75 mile drive had 1.4 miles at 17min/mile.

Suspension still seems to work with this payload. When rolling, the wheels can move in & out, finding an equilibrium. On the bench, various wheels move all the way out & stay there, making it tilt.
Posted by Jack Crossfire | Dec 22, 2015 @ 03:48 PM | 3,150 Views

It was a good day, to be sure. Quite a thrill to see a rocket return from space to its launch pad, as it happened for the 1st time. Restarting a 1st stage engine 4 times & throttling it down enough to land is quite remarkable. The extra weight of the longer stage no doubt helped.

Humans dreamed of flyback boosters for decades, but how big of a success was it? Old timers remember the days when another vehicle was mostly recovered on every flight.

The shuttle had to throw away its mane fuel tank but recovered all 5 engines. Falcon 9 has to throw away its 2nd stage & 1 of 10 engines. We can estimate each vehicle throws away an equivalent fraction of its cost. The shuttle boosters required a large boating operation & several days to recover. Falcon 9 returns directly to the launch pad. Returning to the launch pad was once considered the hardest goal, so who would have thought it would be the 1st success.

Falcon 9 can only be reused when enough upmass is thrown away to allow it. So far, SpaceX has always tried to maximize the amount of upmass by packing as many payloads as possible, only using government flights to try landings. No matter how much the shuttle lifted, it was always recoverable. The question remanes whether SpaceX will budget more landing reserve now that they proved it was possible.

The shuttle boosters had to be completely rebuilt. The reusability was merely of the raw metal. The orbiter required months of...Continue Reading
Posted by Jack Crossfire | Dec 20, 2015 @ 10:22 PM | 2,641 Views
Going 3 miles with no camera or headlights took 500mAh at 12V. This also proved the boards all need to be conformal coated. Drilling out the pinion gears with a 3/32" & putting in heatshrink with a drop of glue on the end seems to be the temporary solution. Also pulled the gear mesh apart & converted to 2WD, which cooled down the motor. Didn't notice any handling issues.

It definitely needs a bigger battery. It's either a 2nd board for electronics or reinforcing the existing board, but the existing battery compartment is too small....Continue Reading
Posted by Jack Crossfire | Dec 20, 2015 @ 01:15 AM | 2,493 Views
After much software calibration, the Ruckus drove briefly, then died. The motor heated up, softening the heatshrink, which allowed the pinion to pull off axially. So using heatshrink to expand the shaft didn't work. The tacon 1/18 motors have a 2mm shaft. The pinion is a hair over 2mm, too little to keep the heatshrink intact. All that held the pinion in place were remaneing bits near the ends.

Decided to widen the pinion hole with a 3/32" bit to 2.38mm. That allowed slightly more heatshrink in. Then added a drop of CA glue to the end of the shaft. The next step would be a 7/64" bit widening it to 2.75mm & even more glue. The next step will be an assortment of Team Associated 10T -14T 1/18 pinions which have a set screw which The Goog confirmed works with the 2mm shaft. http://www.amazon.com/Team-Associate.../dp/B000BP2CX2

The short test drive used 250mAh at 12V. With a finite number of turnings of the screws allowed, decided it was time to convert to 2WD to try to increase the range. This required removing the front differential to remove the driveshaft. Then, disconnected the front wheels from the differential while letting bits hang. This required bending plastic fork shafts with a finite number of bendings allowed. The front differential has to remane in place to connect everything else together.
Posted by Jack Crossfire | Dec 19, 2015 @ 05:37 PM | 2,556 Views
Transporting groceries by RC has been an ongoing idea, but never yielding a practical solution. Strapping to the roof is very top heavy. It couldn't turn. Stopping, turning the wheels, & pulsing the throttle made it flip over. It would need a 2 handed control to navigate every sharp turn. A trailer is most likely, but cumbersome. It would be more compact than a bike.
Posted by Jack Crossfire | Dec 19, 2015 @ 03:33 AM | 2,205 Views
It was decided the mounting was too flimsy to put a battery on top. The old battery compartment was the only place it could go, but only after it was already ground away. Enough of it was left to reasonably secure the battery, but a zip tie will be required. Another attempt to recycle old battery cells failed. The old 3S 1Ah would stand in until a properly sized battery could come in from China.

Massive amounts of software have to be calibrated. Bluetooth has to be reconfigured. Then, once the range is known, the front wheels will be freewheeled to try to increase the range using 2WD mode.
Posted by Jack Crossfire | Dec 06, 2015 @ 11:01 PM | 2,435 Views
It took a full day to troubleshoot the MPU6000. This oldie but goodie is the chip in ardupilot/PX4/Pixhawk/whatever it's branded as today. The current installation wouldn't take any write requests when it powered up. It required a few read requests before I2C started working. Then it would take write requests.

It took another day to swap electronics installations, calibrate & debug some more of the IMU once in the vehicle. Building the new IMU software stack for the truck is quite an undertaking. It's once again not looking like magnetic heading is going to be accurate enough. It's within 10 deg at most. Then it was back to commuting.
Posted by Jack Crossfire | Nov 29, 2015 @ 11:47 PM | 2,661 Views
The smaller remote control was finished, after a last minute decision to upgrade the radio. It was easy to make the antenna always perpendicular to the vehicle. This should greatly increase range. The battery was shrunk with an LP2989 voltage regulator. Unfortunately, the output voltage fluctuated from 3.3 to 2.8 with changes in input voltage. Putting in the bypass capacitor lowered the voltage to 2.2, so that was a cactus voltage regulator. Current is always 60-100mA.

The new driving computer was finished, with a last minute decision to omit GPS. It would be a lot of work to program a map. It would start with drawing a route in Goog earth, still the only program which allows drawing a route. The turns would still require a lot of manual override, with the GPS errors. Quite a few steps remane, to make the computer use the MPU6000 with optional AK8975, install it in the truck, make it use the serial port instead of SPI, send realtime video through the serial port instead of wifi.

Everything needs to be conformal coated.
Posted by Jack Crossfire | Nov 27, 2015 @ 04:26 AM | 2,568 Views
The Bezos mobile was a long time coming. Behind the hype over amazon.com's neverending losses was an obscurity called Amazon Web Services. Unlike the headline store, AWS was churning out billions in passive income from nearly the beginning. Almost the entire internet has been hosted on AWS servers. Today, AWS contains most of the computing power in the world, an obscene amount of clockcycles which can't possibly all be dedicated to web pages.

While never announced, AWS probably makes a lot of money from government code breaking entities. It allowed Bezos to fund his own space program in complete secrecy, with no official government aid.

Going vertically to 62 miles & recovering all the components was quite a stunt, but without much point. It's been done horizontally, forever. Spaceship 1 went 7 miles higher. A fully recoverable system using today's technology can't be scaled to orbital flights, whether vertical or horizontal. Some expensive parts must be discarded.

No details of the Bezos mobile have ever been released, but it's probably similar in efficiency to the Delta IV. All Kiwipedia could gleam is the Bezos-3 engine uses the mane combustion chamber to feed the turbopumps instead of using a separate gas generator. It's the 1st time a cryogenic engine has emerged from purely private funds. With the debut of government money, he plans on developing another cryogenic engine using liquid methane.

Reading up on the Bezos mobile revealed a nugget about...Continue Reading