Saturday, October 31, 2015

2015-10-30 (F) Weekly Summary

I was all set to take pictures of this project and be finished. The more I looked at it the less I saw that was actually complete. Despite wanting to be done with this project it wasn’t finished. Sure, most everything worked but it didn’t feel complete. Instead of taking a few pictures and adding labels I wrote up a list of all the things which needed to be done before the project could be called complete. This project has gone on a long time and I want to be done but here the difference between having a half-baked product and something that I’m proud of. At the same time this is a gamble, I’m 95% done with this but there’s still time to give up.

First shot. Not going to be used for the final instructions

The first problem to tackle was to make another proximity sensor. The old one had short wires which made it impossible to aim correctly. Liquid tape had been used to protect the back of the original proximity detector but the parts were so inexpensive that it made more sense to keep it for future use and build a new one with a long cord. Ethernet wire salvaged from a cable which was easier to bend. Wires were soldered to the back of the protoboard like before but the wires going to the screw terminals were arranged more neatly than before to save room in the enclosure.


New emitter-receiver tube

Neat wire layout

Before liquid tape was applied the program was tested. Connections all seemed good but it took two days of debugging to get the sensor to work. The proximity sensor couldn’t tell the difference between an obstruction and empty air. Feedback told me that there was always an obstruction. Resistance was added and a variable resistor was even tested but nothing gave reliable feedback. The problem turned out to be light leakage at the BACK of the tubes. Light from the emitter was sneaking out behind it and affecting the receiver a couple millimeters away. It only took a scrap of paper to block the offending light beams and confirm the light leakage problem.

Proving the light leakage problem with a scrap of paper

Cleaning up the hardware on the wrist mounted rig was more than just a little polish. It involved taking out all the nuts and bolts then swapping spacers, replacing ordinary nuts with nylon insert lock nuts, trimming long bolts and finally reassembling everything in the correct order. Fortunately the reassembly went quickly since the parts all fit neatly and were finally sized appropriately. One problem with reassembly was putting the controller back into the enclosure. The power would fail when it was pushed into the bottom. Naturally this seemed like a shorted connection but it could not be replicated outside the enclosure. The problem was the exposed reset button the Arduino was being pressed against the bottom of the enclosure. This switch was removed.

Cleaned up wrist mounted rig

Shortened bolts and properly mounted servo horn

Reset switch removed

Necessary changes to the program were written down but haven’t been done. For example, the weight of the keyboard causes the servos to move out of position while resting. This can be remedied by continuously supplying the servos with the correct position.

Servos slipping while resting with keyboard in position

A large white foam board was set up to take pictures. White was used because it will be easiest to remove. My cell phone camera was tricked into overexposing the photos so the background was a brilliant white while the subject was in focus and a reasonable tone. A worksite light and a fluorescent light were used to blast the white foam board and subject with light.

Photo area

Isolated picture of keyboard

Labels were added to the isolated photos. Those labels were taken from a comprehensive list of the parts used in this project. The part list included names for each part printed or purchased.

Labeled keyboard picture

Labeled wrist mounted rig

The rest of the weekly summaries have been arranged by date.

First time here?

Completed projects from year 1.

Completed projects from year 2.


Disclaimer for http://24hourengineer.blogspot.com/

This disclaimer must be intact and whole. This disclaimer must be included if a project is distributed.

All information in this blog, or linked by this blog, are not to be taken as advice or solicitation. Anyone attempting to replicate, in whole or in part, is responsible for the outcome and procedure. Any loss of functionality, money, property or similar, is the responsibility of those involved in the replication.

All digital communication regarding the email address 24hourengineer@gmail.com becomes the intellectual property of Brian McEvoy. Any information contained within these messages may be distributed or retained at the discretion of Brian McEvoy. Any email sent to this address, or any email account owned by Brian McEvoy, cannot be used to claim property or assets.

Comments to the blog may be utilized or erased at the discretion of the owner. No one posting may claim claim property or assets based on their post.

This blog, including pictures and text, is copyright to Brian McEvoy.

Friday, October 30, 2015

2015-10-29 (Th) Wrist Mounted Chording Keyboard

Isolated drawings had labels added. Label names were taken off the parts list. No attempt was made for uniform sized text but the largest text that could fit was used. Images were edited in GIMP and saved as JPGs.

Some of the images had to be displayed in a large format in order for the text to be legible.









Parts list with links

 To do:
  • Program changes
    • Servo shouldn't rest
    • Make sure travel limits are within the bounds of the physical movement
    • Lights transition as servo moves
  • Write instructions
    • Compress and link to all files
      • OpenSCAD files
      • STL models
      • Arduino code 
    • Name everything and use name uniformly throughout instructions 
    • Make diagrams with every part labeled 
    • Schematic for servo controller and proximity switch
The rest of the posts for this project have been arranged by date.

First time here?
Completed projects from year 1.
Completed projects from year 2.


Disclaimer for http://24hourengineer.blogspot.com/
This disclaimer must be intact and whole. This disclaimer must be included if a project is distributed.
All information in this blog, or linked by this blog, are not to be taken as advice or solicitation. Anyone attempting to replicate, in whole or in part, is responsible for the outcome and procedure. Any loss of f

functionality, money, property or similar, is the responsibility of those involved in the replication.
All digital communication regarding the email address 24hourengineer@gmail.com becomes the intellectual property of Brian McEvoy. Any information contained within these messages may be distributed or retained at the discretion of Brian McEvoy. Any email sent to this address, or any email account owned by Brian McEvoy, cannot be used to claim property or assets.
Comments to the blog may be utilized or erased at the discretion of the owner. No one posting may claim claim property or assets based on their post.
This blog, including pictures and text, is copyright to Brian McEvoy.
2015-10-27 (Tu)

Thursday, October 29, 2015

2015-10-28 (W) Wrist Mounted Chording Keyboard

A photo area was set up with a large piece of white foam board and two strong lights. The primary light was a 500 watt halogen work light with warm light. The secondary light, for fill lighting, was a full spectrum light. Identical work lights have been used in the past. Isolated pictures were desired for the photos so labels could be attached.

Photo area

Some of the pictures had the background removed. Taking the pictures with high light intensity made removing the background easier. The photos were over-exposed to wash out the white background so it could be easily cropped away. Over-exposing was done on a camera phone by holding a dark piece of material over a portion of the photo area so the camera would adjust the settings for the dark area then pulling the material away before taking the picture. Manual settings on a digital camera would have made the procedure easier by adjusting the f-stop and shutter speed.







Lists of parts were made for each portion of the project, Keyboard, wrist mounted rig, wrist controller and proximity switch. Each list was a compilation of the parts which would be printed and the parts which should be purchased. Some items, like switches, can be substituted to the maker’s liking. If someone wanted to use the code for the keyboard with a desk mounted keyboard it would work fine and could still utilize the Bluetooth if desired.

Parts list with links

 To do:
  • Program changes
    • Servo shouldn't rest
    • Make sure travel limits are within the bounds of the physical movement
    • Lights transition as servo moves
  • Write instructions 
    • Take photos of parts to be isolated
    • Isolate photos
    • Compress and link to all files
      • OpenSCAD files
      • STL models
      • Arduino code 
    • Name everything and use name uniformly throughout instructions 
    • Make diagram with every part labeled 
    • Schematic for servo controller and proximity switch
The rest of the posts for this project have been arranged by date.

First time here?
Completed projects from year 1.
Completed projects from year 2.


Disclaimer for http://24hourengineer.blogspot.com/
This disclaimer must be intact and whole. This disclaimer must be included if a project is distributed.
All information in this blog, or linked by this blog, are not to be taken as advice or solicitation. Anyone attempting to replicate, in whole or in part, is responsible for the outcome and procedure. Any loss of f

functionality, money, property or similar, is the responsibility of those involved in the replication.
All digital communication regarding the email address 24hourengineer@gmail.com becomes the intellectual property of Brian McEvoy. Any information contained within these messages may be distributed or retained at the discretion of Brian McEvoy. Any email sent to this address, or any email account owned by Brian McEvoy, cannot be used to claim property or assets.
Comments to the blog may be utilized or erased at the discretion of the owner. No one posting may claim claim property or assets based on their post.
This blog, including pictures and text, is copyright to Brian McEvoy.
2015-10-27 (Tu)

Wednesday, October 28, 2015

2015-10-27 (Tu) Wrist Mounted Chording Keyboard

Any bolt which had to be a hinge point couldn't be tightened so the bolt had to be a locknut. A pair of jam nuts would also work. Up to this point ordinary nuts had been used because they could be moved easily while a locknut requires tools. Since all the hinges were confirmed the nuts could finally be replaced with lock nuts.

Disassembled rig with lock nuts being attached

A bolt cutter was used to trim the bolts. Some of the bolts were long enough to cause binding while other were simply longer than necessary which didn't look good. While the bolts were off the longer spacers were placed next to the keyboard holder.

Rig with trimmed bolts and longer spacers

Servo horns were held in place by friction alone since they would occasionally skip or need to be repositioned. Once they were run a few times to confirm they would work where they were attached small screws were inserted through the horns and into the gear of the servo.

Servo with screw holding the horn in position

When the controller was put into the enclosure it failed to function. Troubleshooting showed that it was not a matter of wires shorting. The problem was an issue of the reset button being pressed to the bottom of the enclosure. A soldering iron was used to remove the switch entirely.

 Arduino with removed reset button

Hook and loop fasteners were attached to the bottom of a portable USB charger and a USB cable. The hook side of the fasteners were used so they could easily attach to the wrist band being used.

 Hook and loop fasteners on the charger and USB cable

A video was made which showed the unit performing. Unfortunately the weight of the keyboard pulled the levers out of place while they were disabled. This can be fixed in the program by continuously sending servo pulses rather than sending periodically. The red and green lights change instantly but it would be preferable if they faded into one another as the servo moved. No use for the blue light yet but it could pulse slowly when the machine is resting.

Video demonstrating the weight of the keyboard


 To do:
  • Trim bolts on wrist mount (3)
  • Use locknuts on hinge points (2)
  • Insert screws into servo horns into servo (2)
  • Install longer spacers for hinge-side keyboard end 26.5mm long (2)
  • Velcro or magnets for attaching keyboard to holder
  • Add velcro to a battery pack
  • Program changes
    • Servo doesn't rest
    • Make sure travel limits are within the bounds of the physical movement
    • Lights transition as servo moves
  • Write instructions 
    • Compress and link to all files
      • OpenSCAD files
      • STL models
      • Arduino code 
    • Name everything and use name uniformly throughout instructions 
    • Make diagram with every part labeled 
    • Schematic for servo controller and proximity switch
The rest of the posts for this project have been arranged by date.

First time here?
Completed projects from year 1.
Completed projects from year 2.


Disclaimer for http://24hourengineer.blogspot.com/
This disclaimer must be intact and whole. This disclaimer must be included if a project is distributed.
All information in this blog, or linked by this blog, are not to be taken as advice or solicitation. Anyone attempting to replicate, in whole or in part, is responsible for the outcome and procedure. Any loss of f

functionality, money, property or similar, is the responsibility of those involved in the replication.
All digital communication regarding the email address 24hourengineer@gmail.com becomes the intellectual property of Brian McEvoy. Any information contained within these messages may be distributed or retained at the discretion of Brian McEvoy. Any email sent to this address, or any email account owned by Brian McEvoy, cannot be used to claim property or assets.
Comments to the blog may be utilized or erased at the discretion of the owner. No one posting may claim claim property or assets based on their post.
This blog, including pictures and text, is copyright to Brian McEvoy.
2015-10-27 (Tu)

Tuesday, October 27, 2015

2015-10-26 (M) Wrist Mounted Chording Keyboard

Proximity detector hardware was still malfunctioning yesterday. Reliable reads couldn’t be taken. The problem was assumed to be a matter of the infrared emitter broadcasting excessively strong. An adjustable resistor was added to the circuit to change the resistance for testing. There was already a 910Ω resistor in series with the emitter when a 10KΩ resistor was added. Some adjustments were made and it seemed to improve but there was no consistency.


Liquid tape had not been applied to the back of the sensor so to test the thought that it could have been a matter of light leaking out the back a small bit of cardboard was inserted between the emitter and receiver. Results were instantly visible. The testing program was intended to turn the RGB LED red when an obstacle was detected and blue the rest of the time. As soon as the cardboard was put between the emitter and detector the unreliable reads disappeared.


More testing was done with the adjustable resistor in place. When the emitter was turned down very low only a reflective surface could trigger the detector but the original resistor by itself provided very reliable results returning 94% of attempted reads. The threshold in the program was set to 90% to account for noise and mechanical differences of other people’s hardware.

Testing the light leakage with a piece of cardboard

Liquid tape applied to back of emitter-receiver tubes

Importance of blocking the light leaking was not apparent until these experiments. If a divider had been modeled into the design earlier some of this testing may have been unnecessary but the liquid tape does a suitable job. Testing was done after applying liquid tape and it was successful. Most recent code for the servo controller is available below.


To do:
  • Trim bolts on wrist mount (3)
  • Use locknuts on hinge points (2)
  • Insert screws into servo horns into servo (2)
  • Install longer spacers for hinge-side keyboard end 26.5mm long (2)
  • Velcro or magnets for attaching keyboard to holder
  • Add velcro to a battery pack
  • Write instructions 
    • Compress and link to all files
      • OpenSCAD files
      • STL models
      • Arduino code 
    • Name everything and use name uniformly throughout instructions 
    • Make diagram with every part labeled 
    • Schematic for servo controller
The rest of the posts for this project have been arranged by date.

First time here?
Completed projects from year 1.
Completed projects from year 2.


Disclaimer for http://24hourengineer.blogspot.com/
This disclaimer must be intact and whole. This disclaimer must be included if a project is distributed.
All information in this blog, or linked by this blog, are not to be taken as advice or solicitation. Anyone attempting to replicate, in whole or in part, is responsible for the outcome and procedure. Any loss of f

functionality, money, property or similar, is the responsibility of those involved in the replication.
All digital communication regarding the email address 24hourengineer@gmail.com becomes the intellectual property of Brian McEvoy. Any information contained within these messages may be distributed or retained at the discretion of Brian McEvoy. Any email sent to this address, or any email account owned by Brian McEvoy, cannot be used to claim property or assets.
Comments to the blog may be utilized or erased at the discretion of the owner. No one posting may claim claim property or assets based on their post.
This blog, including pictures and text, is copyright to Brian McEvoy.
2015-10-26 (M)

Monday, October 26, 2015

2015-10-25 (Su) Wrist Mounted Chording Keyboard

A second emitter-receiver tube was printed. This tube was identical to the first. Instead of desoldering the first tube and attaching a different cord a second was made entirely new so the first could be reused in another project in the future. Protoboard was attached to the wires and soldered in place then cut to size.

New Ethernet cord was salvaged which felt more flexible. The sheathing over the wires was softer and the insulation over the copper was more pliable. Strangely, the wire inside turned out to be solid as opposed to stranded. Stranded wire tends to conduct better and be more flexible.

Salvaged Ethernet wire and soldered protoboard

When the wires were soldered in place the sheathing was pulled over the wires to hide as much as possible and pull the wire out of the other end so it could be attached to the screw terminals in the control enclosure. Wires were attached neatly and orderly this time. Last time they were attached with numerous bends and unnecessary lengths. Care was taken to arrange the Ethernet cord so it could exit the enclosure in line with the row of terminals.

Wires soldered to protoboard

Sheath pulled over exposed wires

Wires arranged neatly in terminal strip

To do:
  • Reprint emitter-receiver tube
  • Solder connections for new emitter-receiver tube with flexible wire
  • Trim bolts on wrist mount (3)
  • Use locknuts on hinge points (2)
  • Insert screws into servo horns into servo (2)
  • Print longer spacers for hinge-side keyboard end 26.5mm long (2)
  • Velcro or magnets for attaching keyboard to holder
  • Add velcro to a battery pack
  • Write instructions 
    • Compress and link to all files
      • OpenSCAD files
      • STL models
      • Arduino code 
    • Name everything and use name uniformly throughout instructions 
    • Make diagram with every part labeled 
    • Schematic for servo controller
The rest of the posts for this project have been arranged by date.

First time here?
Completed projects from year 1.
Completed projects from year 2.


Disclaimer for http://24hourengineer.blogspot.com/
This disclaimer must be intact and whole. This disclaimer must be included if a project is distributed.
All information in this blog, or linked by this blog, are not to be taken as advice or solicitation. Anyone attempting to replicate, in whole or in part, is responsible for the outcome and procedure. Any loss of f

functionality, money, property or similar, is the responsibility of those involved in the replication.
All digital communication regarding the email address 24hourengineer@gmail.com becomes the intellectual property of Brian McEvoy. Any information contained within these messages may be distributed or retained at the discretion of Brian McEvoy. Any email sent to this address, or any email account owned by Brian McEvoy, cannot be used to claim property or assets.
Comments to the blog may be utilized or erased at the discretion of the owner. No one posting may claim claim property or assets based on their post.
This blog, including pictures and text, is copyright to Brian McEvoy.
2015-10-25 (Su)

Sunday, October 25, 2015

2015-10-24 (Sa) Wrist Mounted Chording Keyboard

Pictures were going to be taken in order to label all the parts but too many problems needed to be fixed before the project was acceptable. Everything worked but too many details were not acceptable. The day was spent scrutinizing details of the project looking for things which were necessary to make it function well. Small details, like long screws, were ignored before because they were not integral to the basic functionality but now shortening those bolts is important.

Since there are more parts to this project than most previous projects new photos will be taken to help someone recreate this project instead of only reusing files from the blog. A single .zip file should be able to contain all the relevant files such as the chord diagram, Arduino code, and printer files.


Keyboard

 Distance sensor, servo controller, wrist mount actuator

To do:
  • Reprint emitter-receiver tube
  • Solder connections for new emitter-receiver tube with flexible wire
  • Trim bolts on wrist mount (3)
  • Use locknuts on hinge points (2)
  • Insert screws into servo horns into servo (2)
  • Print longer spacers for hinge-side keyboard end 26.5mm long (2)
  • Write instructions
    • Compress and link to all files
      • OpenSCAD files
      • STL models
      • Arduino code 
    • Name everything and use name uniformly throughout instructions
    • Make diagram with every part labeled 
    • Schematic for servo controller
The rest of the posts for this project have been arranged by date.

First time here?
Completed projects from year 1.
Completed projects from year 2.


Disclaimer for http://24hourengineer.blogspot.com/
This disclaimer must be intact and whole. This disclaimer must be included if a project is distributed.
All information in this blog, or linked by this blog, are not to be taken as advice or solicitation. Anyone attempting to replicate, in whole or in part, is responsible for the outcome and procedure. Any loss of f

functionality, money, property or similar, is the responsibility of those involved in the replication.
All digital communication regarding the email address 24hourengineer@gmail.com becomes the intellectual property of Brian McEvoy. Any information contained within these messages may be distributed or retained at the discretion of Brian McEvoy. Any email sent to this address, or any email account owned by Brian McEvoy, cannot be used to claim property or assets.
Comments to the blog may be utilized or erased at the discretion of the owner. No one posting may claim claim property or assets based on their post.
This blog, including pictures and text, is copyright to Brian McEvoy.
2015-10-22 (Th)

Saturday, October 24, 2015

2015-10-23 (F) Weekly Summary

Last week started with a dose of variety. I attended the second annual IOTHackDay. The team I joined was building a robotic xylophone. We started building weeks before so that coding could take place at the IOTHackDay but the day of the competition we still hadn’t gotten motion from the twenty mallets which each struck a single key. In the end we played music robotically but didn’t win the competition. Fortunately really cool projects won and we had so much fun.

Our team and the xylophone

Top down view of robotic xylophone

Panorama of the convention

Testing on the servo controller side of the wrist mounted chording keyboard continued. A program was written to confirm that the new RGB LED was functioning and the distance sensor worked. No problems with the RGB LED but the distance sensor didn’t work right away. A couple changes to the program reassigned the correct pins and ultimately it was a missing resistor which caused the malfunction. Suitable resistors were stocked and installed.

RGB LED testing

Terminals with added resistor

A suitable enclosure was printed then a lid was reprinted because it didn’t fit the first time. The enclosure was really snug but everything fit without damaging the wires while installing. Tension was high enough that there was no need to secure the electronics in the box.

Electronics inside the enclosure

Testing on a fully functional device started. Using a proximity detector instead of a physical switch seemed to be worth the effort. Freedom from a plastic rod rubbing on the back of my hand validated the work spent making a contactless switch. Perhaps a future project would be a more discreet version which could sense multiple hand positions.

Video of contactless switch testing

Finishing touches were put on the emitter receiver tubes including some protoboard and a layer of liquid paint. The liquid paint was primarily for cosmetic reasons but does provide physical protection. A micro USB port was glued to the lid of the controller enclosure for power only. Since the devices can all work well with 5V USB power makes sense.

Emitter receiver finished up

USB power port being glued in place

Two other instances of variety came in the form of a repair job to the 3D printer where a faulty connector was replaced and a new car purchase. It was a busy week. It’s always a busy week. So, I guess it was a regular week. That hardly seems to need pointing out.

Printer repair

New car

The rest of the weekly summaries have been arranged by date.

First time here?

Completed projects from year 1.

Completed projects from year 2.


Disclaimer for http://24hourengineer.blogspot.com/

This disclaimer must be intact and whole. This disclaimer must be included if a project is distributed.

All information in this blog, or linked by this blog, are not to be taken as advice or solicitation. Anyone attempting to replicate, in whole or in part, is responsible for the outcome and procedure. Any loss of functionality, money, property or similar, is the responsibility of those involved in the replication.

All digital communication regarding the email address 24hourengineer@gmail.com becomes the intellectual property of Brian McEvoy. Any information contained within these messages may be distributed or retained at the discretion of Brian McEvoy. Any email sent to this address, or any email account owned by Brian McEvoy, cannot be used to claim property or assets.

Comments to the blog may be utilized or erased at the discretion of the owner. No one posting may claim claim property or assets based on their post.

This blog, including pictures and text, is copyright to Brian McEvoy.

Friday, October 23, 2015

2015-10-22 (Th) Wrist Mounted Chording Keyboard

A micro USB port was attached purely as a power supply port. Soldered wires on the breakout board broke while testing so a firm mounting location seemed important. Screws were used to mount the USB port for the keyboard but required a standoff and placing the device closer to an edge would have eliminated the need for standoffs. In this case the port was glued to the lid of the enclosure. If any problems arose from this configuration the port and lid could be replaced as a unit and the wires running from the USB port could be taken out of the screw terminals. Since no more testing would be done the curing time of the glue was acceptable while screws or bolts would have been necessary to continue testing.
Soldered wires on breakout board

 Using glue, E6000, on the lid and USB port

Port held to lid for glue to cure

Construction was nearly completed. An assessment needs to be taken to recognize what problems still need to be addressed. For example: the cord between the emitter-receiver tube is too short.


Downloadable Files:
To do:
  • Write instructions
    • Compress and link to all files
      • OpenSCAD files
      • STL models
      • Arduino code 
    • Make diagram with every part labeled 
    • Schematic for servo controller
The rest of the posts for this project have been arranged by date.

First time here?
Completed projects from year 1.
Completed projects from year 2.


Disclaimer for http://24hourengineer.blogspot.com/
This disclaimer must be intact and whole. This disclaimer must be included if a project is distributed.
All information in this blog, or linked by this blog, are not to be taken as advice or solicitation. Anyone attempting to replicate, in whole or in part, is responsible for the outcome and procedure. Any loss of f

functionality, money, property or similar, is the responsibility of those involved in the replication.
All digital communication regarding the email address 24hourengineer@gmail.com becomes the intellectual property of Brian McEvoy. Any information contained within these messages may be distributed or retained at the discretion of Brian McEvoy. Any email sent to this address, or any email account owned by Brian McEvoy, cannot be used to claim property or assets.
Comments to the blog may be utilized or erased at the discretion of the owner. No one posting may claim claim property or assets based on their post.
This blog, including pictures and text, is copyright to Brian McEvoy.
2015-10-21 (W)