taking apart a camera

First tests to build a battery-grip for my GH1 camera by modifying a Chinese battery Grip for the Canon 450D. (Nearly the same size.)
However some Chinese thought it would be funny to *reverse the meaning of red and black wires*.
So I ended up frying my expensive camera.

...Took me ages and multiple failed attempts but I was able to find the fried sub-micron diode.
Tried to source a replacement diode.
...it turned out that a regular SMD- diode is still 5 times bigger then the one they used and does not fit mechanically.
Placed a 0,8mm wire over this 0.2mm diode. (Sorry, I can't cut shorter wires without a microscope.)

for (int i=0; i<7; i++) {
Put everything back together, black screen.
Took it apart, found a missing connector.
(dozens of them, all sub-millimeter mylar types.)
It switches on but I can't shoot pictures. Remote-trigger works.
Took it apart.....
Camera is working again.

It really hurts to touch the high-voltage part that powers the flash.


Good News and Bad News

I've some good news and some bad news....

the good news is:
My new threaded rods to upgrade my Repman arrived.
(And some 15mm aluminum pipes to replace the heavy
steel-pipes in my shoulder-rig.)

The bad news:
My long awaited ABS from china also arrived.
The surface is marred, it's brittle and there
are 0.5mm air bubbles inside the filament.
I have no hope that one can use this for 3d printing at all.
And I got the minimum of 20Kg of this junk.

The company is:

Qingdao TSD Plastic Co.,Ltd.
Add.:No.48, Ruijin Road, Qingdao, China.
Website: www.qdtsd.net
MSN: hongyong_zhang@live.cn
Free Line: (86)400-688-0532
Office Line:(86)532-81936899
Mobile No.:(86)158-54279111

I found a wiki-page about this to compare noted on different companies.

3d printed camera belt clip

Like I planned before I jsut designed myself a belt-clip for a camera.

I was impressed with my B-Grip. However it would not accept my Manfrotto quick-release plates that I have on all cameras and tripods,... .
So I designed this one, based on the existing camera-plate of my 3d printed shoulder rig.
I guess it will be hard to print due to it's size and the large overhang.
I hope to include a pause G-Code in the overhang to insert the actual belt as a support.
That should make it 3d printable.

We'll have to see.

This time I not only did a colored rendering and technical drawing with dimensions and a 3d PDF with the movable model and the drawings but also included the thingiverse-URL to the part into the PDF.

re-uploaded Micscreen-video

With the cleanups of intro, outro and the new broadcast-logo
I at least re-uploaded the latest video last night to include these changes.


Going on digital TV

It so happens, that during the CCCamp 2011
I may broadcast some of my videos to the Camp TV.
Thats a real, terrestrial, digital TV-station that can be received with a normal diital TV set.
(Analog TV has been switched off already.)

I took that as an opportunity to clean up my intro, create a separate outro for non-youtube (you can't have a clickable "subscribe"-button on TV) and create a small logo for the upper right corner.

I also just recorded the introductions to 3 smaller [HOWTO] -videos that I may or may not find the time to record before the camp.
(Extremely busy right now.)






Mirrored Zoom H4n Belt Clip

Last night I noticed a flaw with my Zoom H4n belt clip....you can't wear it on the other side without the microphones facing backwards or the speaker being obstructed.

So I created a mirrored version, so I can wear my camera on the right and sound on the left.
(I'm used to carrying my camera on the right side.)

Video: [HOWTO] Building a microphone screen

And another video is done.
The one about how to build a microphone screen.
As you can hear in the video, it improves sound quality dramatically.
I'm slowly getting the the level of quality that I want to see in my YouTube channel.
Still a lot left to learn and to practice but I'm making progress.

Video: [REPORT] Gullasch Programmier Nacht

It is done.
My report about GPN11, what it is, what to expect there, what it looks like to be there is uploaded and public.


Longer videos on YouTube

Herzlichen Glückwunsch! Dein Konto unterstützt nun Uploads von Videos, die länger als 15 Minuten sind. Klicke auf die Schaltfläche "Video hochladen" und wähle ein Video aus.

= For whatever reason I just got a notification in YouTube, that I now can upload videos longer then 15 minutes.
From the help-pages I thought that thus required an old directors-account (that you can no longer get) or YouTube-partner status (that you needs lots and lots of followers to get).

BTW, the GPN11 -video is finished and a [HOWTO] -video on how to build my mic-screen just needs the introduction to be recorded.


new extruder design - prototype 5

Only 2 weeks until CCCamp and my Thing-o-Matic is still out of order...
I managed to get my new extruder design up to prototype IV.
The print-quality of my Repman suddenly dropped dramatically with the update from firmware 2.0.8 to 4.1.1 .
Reason still unknown. Maybe they changed something in the handling of reversals.
The photo below is the same Skeinforge 16mm/2 profile I used to do really high quality prints the day before.
(After the quick 48mm/s prototype. That was ugly but served it's purpose.)

I changed some dimensions, turned the magnetic encoder board around, got more pressure on the filament, ...
I even got the axis for the magnetic encoder to be turned by the filament already.

There is not enough time left this weekend to also print the new pressure-arm and possibly even test-drive this extruder. :(
So that leaves me with 2 weekends to finish this extruder, get it to print, pack the printer for CCCamp and get some last camera parts printed.

Firmware upgrade to Repman

I haven't looked at the firmware-version of my Repman in a loooong time.
It was still running 2.0.8 while the current firmware is 4.1.1 .
They only listed the changes since 3.x and even that takes multiple pages
(with one line per improvement).
I'm not sure I like the high travel speeds when moving from homing-location
to warmup-location. That could be dangerous. But during a print that does
come in handy.


Found the right sound for my Intro

So far the HOWTO-intro has been silent because I just could not find the right click-noise I wanted.
...I just found it.
The new emergency-stop for my Thing-o-Matic has just the right two sounds.


CCCamp planning has started

CCCamp 2011 is drawing near and I'm starting to plan the last details of my stay there on this wiki page.

I plan to:
  • do 3d printing (of cause)
  • get more experience in interviewing people by doing so for [REPORT]-style videos for my YouTube channel. Where they can present their project.
  • Provide a place where people can plug in their recorder/laptop to record podcasts

<<Wiki page

building a micscreen from Basotect pyramid foam

A few days ago I got myself a nice T.Bone SC600 large disphragm microphone.
That one should be just perfect for recording voice-overs.
But since my room is not a sound studio, I was looking around for a good mic-screen.
(Something to absorb sound in the back and on the sides of the microphone.)

However these things are expensive in any reasonable size.
And they are not very portable.

So I decided to make my own.
My first thought was to make a half-circle of felt and glue some 5cm cussioning on.
But last night I found a source where I could get me 1x0.5m of Basotect pyramid foam.
(The real stuff [tm])
So now I'm preparing a howto-video about building yourself a (hopefully) really good mic-screen with the proper materials for cheap.
...and to build it in a way, that can be quickly disassembled and reassembled and
folded flat for transportation.

The mic-screen is finished and I made a video about it.
(Will be on my YouTube channel



new extruder design

I managed to get the last required meassurements for a first prototype of my own extruder design.

Not printed yet and certainly not tested yet.

The magnetic encoder can be mounted above the drive pulley. The head of an M3 capped bolt inside a 3mm ball bearing (6mm outer diameter) is driven by the filament streaming by.
The encoder-board is held by 3 M3 bolts driven directly into the extruder-block and cutting their own thread.
I think this will hold securely but this remains to be tested.
The magnet is glued to the bottom of that bolt, sticking out of the extruder, and rotates below the board.
In the design it may look like this but in fact this M3 bolts does not get in the way of the M5 spring-loaded fastener bolt. (I hope.)

The mounting at the bottom is only for the Thing-o-Matic in this early prototype.
The later version will also allow other hot-ends and I'll see about an adapter-plate to mount 2 of these extruders (one mirrored) in the limited space of a Thing-o-Matic.
2 of the M3 bolts use trapped nuts, the other 2 are inverted and need to face downward instead of up. (not enough space below the pressure-arm.)

Update: I did a very rough prototype at 48mm/s for the first tests.

parametric lens gear set

After my 3d printed geared follow focus it was inevitable to one day publish this:
A set of lens gears.
Fully parametric design.
Each lens-gear labeled with it's size already engraved into the surface.


compact middle_piece for 3d printed shoulder rig

I'm currently printing parts for a new shoulder rig on the Thing-o-Matic.
Thus I noticed that thee middle_piece (an often reused clamp in the design)
is not as easy to print on the Thing-o-Matic as it is on the Repman.

Thus I designed a variation that is smaller and easier to print.

Speed against warping

I just can't get a special part printed on my RepMan right now.
One edge is always curling up until the print fails.
...this time I'll try to print it much faster.
The hope is, that it does not curl up fast enough to cause any problems.

...let's see if this can be a viable strategy.

preliminary result:
Printing the first layer at 8mm/s very, very tightly pressed into the build-surface and the following ones at 48mm/s looks promising so far.
It has progressed higher then any previous print, minimal warping so far and of cause it's MUCH faster then 16mm/s.

From earlier experiments I already know the top-speed to be 87mm/s limited by the power supply to the extruder-heater and very close to the structural limits of the table my RepMan sits on.
(Have to consider sand-bags and slow deceleration programmed into the G-code to move faster.)


It worked once and the second one is currently printing and seems to work just fine too.

Upgrading the RepMan

How that I have done so much on the Thing-o-Matic (and will do much more in the future),
it's time to care for my RepMan too.

My current heated bed is only 2mm thick aluminum and bending ever so slightly.
So I just got myself a 300 x 300 x 6mm AlMg3 Aluminum plate to replace it.
I plan to use my milling-head to make it absolutely, perfectly flat.
(That should solve a lot of issues in using all of that amazing 270x300mm
build area.)

I also ordered precise M8 threaded steel-rods to replace the original ones.
These ones should be much more straight and reduce my wobbling.
(wobbling=repeated XY-shifts due to Z-axis not being straight.)

While at it, I also ordered 15mm aluminum rods for my camera rig,
to replace the heavy 15mm steel rods I used to test is.
There was just no 15mm aluminum to be found locally at that time.

The new headed bed will be temperature-controlled by attaching a spare thermistor
to it and a 12V->230V relay to attach it as a third extruder to the control board.
Thus temperature is controlled like for an extruder but without putting load on
the common 12V heater-line in the controller board that would be missing to heat
extruder #0 and extruder #1 .
(Yes, I have the dual-extruder upgrade and a spare extruder right here, waiting to be installed.)


heavy Printfest

Wow...that was a night.

After work both printers where (for the first time) running simultaneously.
I got the new H4n belt-clip done and the newer RepMan-to-ToM toolhead adapter.
Failed on some new H4n-spider parts and some middle-pieces for the shoulder rig.
Well...you can't get everything working in one night.


Belt clip for Zoom H4n field recorder - Mark II

Finally got my Thing-o-Matic fixed enough to get the belt clip printed.
I'll design a "Mark II" version on the train later today
that make it more practical.
I'm not satisfied with the way I can reach the 1/4" bolt and the volume-buttons yet and want to get the center of gravity lower by attaching the belt higher.
This will give better access to the 1/4" -bolt to quickly release the H4n audio recorder.

Google+ added to links-bar

...I just added a link to my Google+ -postings to the "Links"-widget in this blog.
I'm trying to have more and finer updates to what I'm doing there and reserve the blog for postings with more "substance" then mere progress-updates.

I'll probably do a lot more small Google+ -postings with the Android phone in the future.


Belt clip for Zoom H4n field recorder - 2

I managed to get my mobile printer (Thing-o-Matic) more or less printing again.
The first test-print did not work out well in terms of usability but enabled be to correct some measurements and make this work much better on printers with warping-issues.
At the moment a tension lock is printing because every now and then the tension-bolt on the MK6-extruder shakes loose.


I just got the spam about Autodesk 123D -Beta5 being out.
They list IGES export and STL import+export.
I try not to get my hopes that high but it just may work as an STL to IGES converter.
Will have to try if that works ASAP!

Result: 123D can work with meshes and solids in the same project but not export a mesh as a solid.
This could be nice to build assemblies of solids with part that I only have as meshes.

Inkscape gears plugin

I just learned that inkscape has a gears-plugin.
And inkscape can export 2D-DXK.... finally gears in Alibre Design!
They look very simple but at least it works!

Pause mode

The ToM / ReplicatorG needs a better pause mode, that raises the head like the Repman does....
This is what you get, after pausing a print that threatened to go well into the early morning hours:
Published with Blogger-droid v1.7.2


ToM filament problem

During GPN11 I had problems after switching to a new roll of ABS.
The answers in the Makerbot-Operators Google-Group said that the PTFE-tube may have been to small for the thicker ABS I tried to use.

So now I detached the MK6 stepper-motors and then the hot-end and checked.
My filament was stuck fast inside the PTFE and would only come out after 220°C and lots of force was applied.
The PTFE-tube of the Thing-o-Matic is (now) 3.2mm on the upper side but only 2.99mm on the tapered business-side while my new filament is 3.06mm.

I drilled my PTFE to 3.2mm (3.0 and 3.2 are standard drill sizes while I could not find 3.1mm).
In a first 30second test it seems to extrude fine.
The filament-thickness into thick air is 0.60-0.63mm .
Time for a test-print....

On the plus side, I did manage to solve the 129°C temperature-reading issue with the Automated Build Platform.
The connector was plugged into the only socket that would accept it all the tine.
It's left 3 pin-holes should have been connected to 3 free-standing pins in the middle if the board instead.

My plunger was too tight, the drive-pulley was no longer centered because the cork-spacer between extruder and stepper was getting squashed over the first few month of operation and thus the stepper-shaft deeper in the extruder then when I mounted it the first time.
There are still issues but I can work on finding a reliable set of parameters and then calibrating skeinforge for the filament-sizes and hole-diameters resulting from these parameters.

DIY shift lens

I just ordered myself a Canon-FD to Micro 4/3 adapter from HongKong for my second camera.
...and another one to destroy it, add some 3d printed parts and hopefully make it into a universal Shift-lens adapter that turns all Canon FD -glass into manual Micro 4/3 shift lenses.

It will probably fail terribly at stage 1 "destroying the adapter" but it just meight work out...

Arcol hot-ends

I just found out that Arcol.hu makes some great hot-ends with 0.35mm nozzles that work for RepMan, Thing-o-Matic and Mendel. This review calls it an "overengeneered thing of beauty" and from the photos that looks just about right :).
I'm tempted to get one of their's for my dual-head upgrade to the RepMan.
They also have a universam mount for the Thing-o-Matic that looks much more sturdy then the acrylics that Makerbot provides.

ToM back in business

"My printer is over the ocean, my printer is over the sea, my printer is over the ocean, oh bring back my printer to me..."

The 2 PCBs of my Thing-o-Matic that I sent in for RMA returned.
Makerbot could not reproduce the issue and a new cable didn't help a bit.
The temperature-sensor of my Automated Build Platform (the belt) still shows 129°C at ayny temperature.
This means that an automated build never detects that it has cooled down and ejects the part to start printing the next one.

I reassembled everything very, very late at night, packed the mobile TOM v2 again and am taking it with me for the week.
As I found on GPN 11 the PTFE-tube to be too thin for my next batch of filament, I checked that.
The filament is 3.04-3.06mm perfectly round (not oval).
I could not find any 3.1mm drill, so I took 3.0 and 3.2mm with me to make that tube wider.
I guess I should create a collection of such tubes in 2.8-3.2mm and label them.
Every roll of filament will have a slightly different diameter after all.

The filament I ordered in China is delayed (what a surprise).
They told my that their extrusion-machine has broken down and needs repairs.
So I'm stuck with that expensive roll of black BitsFromBytes filament and some left over white filament from that 2x10Kg rolls I got back in the first days.

I do have a raw version of the first half of the GPN -video.
(first half has me talking about what GPN is and what to expect there.
Second half is just collected footage and music.)
Still need to select some Creative Commons music for the second half apply all the little details I learned to make it perfect.
(Not like that "mobile TOM v1"-video.)


3d printed Battery grip prototype

I'm currently designing a 3d printed battery grip for the Panasonic GH-1.
Due to the complexity of the model,
I'm currently starting out with a highly simplified design.
This prototype only contains 2 batteries, no button, no electronics, no display, no connectors.
Not even routes for the cable. Just holes to mount tiny, 1mm thick copper plates for the battery contacts.
It is opened and closed with 3 M5 sinkhole bolts.
I want to get this prototype right before refining it into a "proper" battery grip.


trying to upload a favicon here....


Panasonic GH2 hacked

I waited for ages for the GH2 to get hacked.
Staying with my GH1 until that happened.
Now it is! :)
Vitaly had to wait to get a firmware. Since no firmware-update for the GH2 was released yet, that was a problem.
It seems that the V1.0E -firmware for the GH2 was leaked now!


Me on Google Plus

BTW, you can find me on Google+ since last weekend.
(I'm pretty much replacing Google Buzz with Google Plus.)


extruder design

Being unsatisfied with the filament drive of the Thing-o-Matic
(way too little pulling force),
I'm in the early stages of designing my own extruder.
  • geared down stepper motor
  • spring-loaded lever to adjust the pressure
  • mount for a rotary-encoder that meassures the filament being actually pulled in
  • integrated compartment for a sponge to clean the filament
  • use herringbone-gears
  • mounting holes for a fan for the stepper motor
  • low center of gravity (reduced stress in transport and backlash)
  • allow ToM and RepMan hot-ende and allow both extruder mounts (I want to design this only once for both my printers)
  • 3 point mouting (use 3 contact points to mount the extruder to a Thing-O-Matic z-stage as opposed to the flawed arc-like 2-point design of the default MK6-steptruder. Doesn't matter on a Repman.)

Originally I wanted to use 2 drive-grears to grip and pull the filament from both sides instead of clamping it between a drive-gear and a 608-bearing but I found that this would require a more complicated gear because the only stationary point on the lever is it's axis and that is not the axis of the intended second drive gear.
So that would require 2 more gears in a very small space.

Don't hold your breath. This will take me a while to get right but I'm working on it.

Belt clip for Zoom H4n field recorder

Not having a hand to hold the Zoom H4n but having a working design of a spider-mount for it,
I just modified that design to make a nice belt-clip.
Thus I hope that on CCCamp I'll be able to have the Zoom H4n on my belt,
phantom-powering the headset microphone and have both hands free to do some interviewing.
(I plan to interview some people to showcase their projects.)


packing light

I'm traveling a lot.
...but I don't like to carry much weight.
...and there's never enough power sockets around. Especially in trains.


You know how it is with me and tools...can't hold myself back. ;)

So I got myself a new pocket-oscilloscope for 40eur from China.
Sadly I clicked the wrong link and got mine without a probe. Thus I attached 2 cables with plugs for standard multimeter-probes to it.
Works very fine for small audio tasks.
And frankly... that's all I need it for.
Just to look at some signal or some low frequency noise.
So it has it's place just like 5eur pocket-multimeters have their place.

What I found interesting on the weg was:
an article about attaching other probes (especially X10 ones) to it.

C't even has an article about the next generation of this one in the edition that arrived today.
A small, cheap and simple tool.