BritishIdeas Interesting Tech Projects

I know this is a tech blog, but I can’t resist posting this to stop some unsuspecting person from being ripped off…

We have a 2004 Honda Accord. The dealership says we need a new cabin air filter and the cost is $89.99 + tax. I decided to look into doing it myself.

A quick search on Google revealed this video on You Tube showing a 13-year old boy doing it in about two minutes with zero tools. Well, he used his fingers, does that count?

Looks pretty easy. Next a search to find the cost of the filters. Seems to average around $15 + tax and delivery.

I went to PepBoys and got a filter made in the USA with a lifetime warranty for $15.16 including tax. I installed it in the car park in less than a minute.

So I called my local Honder Dealership and talked to the service manager to see what he thought about this. Here are his comments and what I think he meant.

Service Manager: “$89.99. It’s the going rate.”
I think he means: “Everyone else is going to rip you off, so why can’t we?”

Service Manager: “It takes our technicians 15-20 minutes”
I think he means: “You have no clue how much effort it is, so I’ll lie to you”

Service Manager: “It’s Cost Averaging - we charge one price for everyone and some Hondas take a lot of effort to replace the filter”
I think he means: “You have a newer vehicle so you have to pay more to subsidize people with older vehicles”

Service Manager: “One price makes it easier for our customers”
I think he means: “Most of our customers don’t to know, and one price makes it easier for us to remember the price”

Service Manager: “I called the other Honda dealership and they charge $125″
I think he means: “We don’t rip you off as badly as the others so keep coming here”

Service Manager: “Sorry I took so long to call you back to appease you… errr… give you an answer I should say…”
I think he means: “All I do all day is appease complaining customers. Oops, did I say that out aloud?”

Yes, he really did say that. I don’t have a good enough imagination to make this up…

I then called the other Honder dealership. Sure enough they charge $125. When I asked why I was told “It’s labor intensive. There is some assembly required after the filter is changed”. Yeah, right… Snap on the small clip.

So, check around your local dealerships and see if they do “cost averaging”.

Bookmark to:

For Christmas 2008 I made coaster sets as gifts. I took pictures at the time but only now am I getting around to uploading them. I hope they provide inspiration for someone else. If the subject matter for the coasters seems a bit odd or random, it’s because most of them were designs intended specifically for the people they were given to.

Each set contains four coasters and a holder and involved about seven hours of work. The wood is 1/4″ poplar. The inlay is Premo! polymer clay. The items were finished with Varathane. Cut using 1/8″ spiral upcut endmills and 0.0571″ endmills. Designs were made in Solidedge 2D and the toolpaths were generated in CamBam Plus. On the bottom are thin discs of cork, glued on with Titebond II.

Bookmark to:

I have finally completed my dust box. Although this box is designed for the FireballCNC V90, it can fit any CNC machine the same size or smaller. Some features:

• Removable sides, top and front
• Cavities for optional sound dampening
• Reinforced floor
• Electronics storage with plenty of room for expansion
• Window on the front
• Made with easy to find materials

Some pictures are below. This will allow me to use my CNC machine in the house with people and animals around without worrying about snapped bits, sawdust, etc.

Bookmark to:

I am making my configuration files available to help anyone trying to get their Fireball V90 machine working with EMC2. The configuration is for inches, 1/4 microstepping and the HobbyCNC Pro board.

To use these files you will first need to perform the latency test and work out the base period for your PC. Open the Stepconf wizard and choose to update this configuration. Put the latency test results into the Stepconf wizard. Then you must double-check the pin configurations for your board if you are not using the same one I am. These changes are easy to make. I recommend 1/4 microstepping as it gives smooth operation without much reduction in power and speed.

Note that Stepconf will tell you that the INI file has changed since the configuration was generated. I changed the DEFAULT_VELOCITY to 0.9 so the default jogging speed is a bit faster.

Download Fireball V90 EMC2 Configuration Files.

Update: Now with Stepconf generated config files.

Bookmark to:

The Acer Aspire One is a pretty nice netbook - for a reasonable price you get 1Gb RAM, 160Gb hard drive, 1024 x 600 screen and 5 1/2 hours of battery life. However it comes with Windows XP or Linpus, which is a customized Linux distribution from Acer. Neither of these options appealed to me. In particular the Aspire One user forums seemed to have a lot of postings on limitations with Linpus.

Enter Ubuntu Netbook Remix (UNR), which is a special version of Ubuntu 8.04 for netbooks. Essentially it is the same as Ubuntu 8.04, but with some additions that make it easier to use on a small screen with limited height. Because netbooks don’t have optical drives it presents a challege to get a new operating system installed. Below is the process I found to work for me. I used Ubuntu on a desktop PC for all the steps.

Firstly note that I tried and failed to get a USB hard drive to work. I partitioned and formatted it every way I could and couldn’t get it to work. I managed to the the UNR installer to boot, but it always failed with “sdb: unknown partition table”. The process only seems to work with USB flash drives. I used a Sandisk Cruzer Micro 4Gb.

1. Download UNR 1.0.1 image from:

http://oem-images.canonical.com/unr/unr-1.0.1.img

2. Plug in USB drive and identify device location. Mine was /dev/sdb mounted to /media/disk.

3. At command prompt:

sudo dd if=./unr-1.0.1.img of=/dev/sdb bs=1024

4. fdisk -l will then show:

Disk /dev/sda: 250.0 GB, 250000000000 bytes
255 heads, 63 sectors/track, 30394 cylinders
Units = cylinders of 16065 * 512 = 8225280 bytes
Disk identifier: 0x88000000

Device Boot      Start         End      Blocks   Id  System
/dev/sda1               1           8       64228+  de  Dell Utility
/dev/sda2               9        1314    10485760    7  HPFS/NTFS
Partition 2 does not end on cylinder boundary.
/dev/sda3   *        1315        7769    51849787+   7  HPFS/NTFS
/dev/sda4            7770       30394   181735312+   5  Extended
/dev/sda5            7770       23109   123218518+   7  HPFS/NTFS
/dev/sda6           23110       29908    54612936   83  Linux
/dev/sda7           29909       30394     3903763+  82  Linux swap / Solaris

Disk /dev/sdb: 4016 MB, 4016045568 bytes
124 heads, 62 sectors/track, 1020 cylinders
Units = cylinders of 7688 * 512 = 3936256 bytes
Disk identifier: 0x8ef631df

This doesn't look like a partition table
Probably you selected the wrong device.

Device Boot      Start         End      Blocks   Id  System
/dev/sdb1   ?      274784      529564   979374166   66  Unknown
Partition 1 has different physical/logical beginnings (non-Linux?):
phys=(734, 123, 14) logical=(274783, 70, 21)
Partition 1 has different physical/logical endings:
phys=(120, 143, 6) logical=(529563, 65, 22)
Partition 1 does not end on cylinder boundary.
/dev/sdb2   ?      448668      961719  1972168331    7  HPFS/NTFS
Partition 2 has different physical/logical beginnings (non-Linux?):
phys=(187, 180, 14) logical=(448667, 16, 52)
Partition 2 has different physical/logical endings:
phys=(784, 0, 13) logical=(403059, 76, 1)
Partition 2 does not end on cylinder boundary.
/dev/sdb3   ?      426615      680707   976730017   7d  Unknown
Partition 3 has different physical/logical beginnings (non-Linux?):
phys=(252, 59, 46) logical=(426614, 84, 39)
Partition 3 has different physical/logical endings:
phys=(139, 118, 4) logical=(122048, 22, 28)
Partition 3 does not end on cylinder boundary.
/dev/sdb4   ?       36178       37261     4161536   6f  Unknown
Partition 4 has different physical/logical beginnings (non-Linux?):
phys=(370, 101, 50) logical=(36177, 96, 57)
Partition 4 has different physical/logical endings:
phys=(10, 114, 13) logical=(37260, 47, 62)
Partition 4 does not end on cylinder boundary.

Partition table entries are not in disk order

Don’t worry about the errors - they don’t seem to be important.

5. Unmount the USB drive by right-clicking on the icon on the desktop.

6. Plug USB drive into Aspire one and boot. At Acer screen press F12 to get to the boot menu.

7. On the boot menu the USB drive will appear twice:

USB Floppy Drive
USB CDROM

This because there is a second hard-coded, read only partition that appears as a CDROM drive. Choose the floppy drive option. Press Enter at the prompt and UNR will now install. Follow the on-screen instructions.

8. At this point you probably want to change the name of the computer. It seems to default to the name of the first user. Go to Administration -> Network, click on “Unlock” and enter your password.

9. Click on the General tab and enter a new name in the Hostname field.

10. Reboot. Clicking on “Quit…” didn’t seem to work. Holding down the power button for a second or so brought up the shutdown options screen. Note that after the reboot the “Quit…” option started working, so I guess the reboot is necessary.

Cleaning Up Your USB Drive

1. Install GParted from Add/Remove…

2. Go to System -> Administration -> Partition Editor

3. Plug in the USB drive and select it from the list at the top right (for example /dev/sdb).

4. Right click on the drive on the desktop and unmount it.

5. Select the partition (there will only be one for some strange reason) and delete it. Click on Apply to confirm.

6. Create a new FAT32 primary partition. Click on Apply.

7. Remove and insert the USB drive to mount it.

Bookmark to:

There are a couple of features that I’ve come to the conclusion are essential, when generating g-code. Sure, there are a lot of features that are necessary, but these are features that perhaps a newcomer to CNC might overlook.

Tab Generation:

When cutting out a part from a piece of wood, there is a moment when the bit will cut through the last piece of wood holding the part to the rest of the wood. At this point the forces may cause the part to simply be forced out instead of precisely cut out. If the part is complex it is possible that this may happen even earlier in the last cutting pass, causing the wood to be snapped and the part to be damaged.

The solution is holding tabs. These are small pieces of wood that are left in place to hold the part to the rest of the wood. After cutting has finished the tabs can be broken, perhaps with a Dremel, then sanded away. This creates control over the process of removing the part, and will avoid any damage.

It is possible to manually design in the tabs when creating the part in a CAD program, but then some manipulation is required in the g-code generator to obtain the correct toolpaths. A quicker and easier solution is to use a g-code generate that can automatically generate the tabs. The usefulness of this should not be overlooked.

Lead In Moves:

Plunges, where the bit is moved vertically into the wood, put a lot of strain on the CNC machine and the spindle. For example on my CNC machine I can sometimes see the Z-axis assembly move slightly up and back when it encounters the resistance of the wood.

The solution is to use lead in moves. These are moves that gradually move the bit vertically into the wood at an angle or from the side. This reduces the stress on the mechanical components, which should last longer. As an example for my current projects I use a lead in move of seven degrees.

Bookmark to:

The best introduction I have found is from the Enco catalog. Fortunately you can view the catalog online. The introduction covers materials, types, number of flutes (including the trade offs), end cut type, shank type and surface treatments.

View the introduction at:

http://www.use-enco.com/CGI/INPDFF?PMPAGE=123

Note that this gives you page 123. Enco’s next release of the catalog may change the location of this introduction to another page. If you see this link is out of date please let me know the new page number.

Bookmark to:

Recently I wrote about the steps I took to reduce the cutting time of a particular part from 67 minutes down to 20 minutes. I have now purchased a couple of 1/8″ four flute endmills from Enco.

I can now cut the 1/8″ wide groove in the design, plus the outer profile with holding tabs much more quickly. In addition I can use the 1/8″ bit to perform a quick roughing pass on the pocket operation. Once complete I switch to the 0.0571″ endmill and make a finishing pass around the edges of the pocket, increasing the detail.

The new time, including the manual tool change, is about 10 minutes. Not bad. At the end of the cutting process the motors, stepper driver chips, transformer and dremel are all a lot cooler.

One small downside is that the 1/8″ bit make a lot of noise.

Bookmark to:

Previously I wrote about a DXF exporter for Inkscape. Bob has updated his Better DXF Output exporter for Inkscape 0.46. All the files are now in one place and it is under the GPL v2 license. This is great news for CNC users and I hope it leads to more use for this handy Inkscape extension!

Bookmark to:

I am working on a design that basically involves cutting out a 4″ diameter disk from 1/4″ thick wood, with some shapes pocketed into the surface of the disk. I have been using 10 inches per minute (IPM) to cut through the Poplar and I had no clue if this is average, fast or slow. What I do know is that it is painful to watch.

At 10 IPM cutting out the design took 67 minutes. That is probably split 60%/40% with the majority on the pocketing. Too long for my liking. I decided to try and reduce the cutting time. First I increased the speed to 30IPM, which is nearly the maximum that my PC can go. Next I reduced the depth of the pockets from 0.15″ to 0.10″. This saves an extra pass.

The new time to cut? 20 minutes. Less than 1/3 of the original time. But that’s no good if the result is a ragged mess. Tonight I found out - there are more burrs on the wood, but the vertical edges look just as good as the ones cut at 10 IPM. So I will likely continue to use 30 IPM in the future.

My next aim is to split the cutting into two parts. The pocketing with a 0.0571″ end mill and the rest with a 1/8″ end mill. This should reduce the time even further.

Bookmark to: