hell schreiber

Hi Zsolt,
I think INOX is a mark of anticorro steel.

BR from Ivan OK!SIP

Ivan wrote:Hi Zsolt,
the FeCl3 solution can be regenerated using electrolysis. The recommended materials are anticorro steel for cathode and carbon for anode. Copper collects on the cathode and the solution increases its power while turning FeCl2 to FeCl3.
Etching:         2 FeCl3 + Cu -> 2 FeCl2 + CuCl2
Regeneration: 2 FeCl2 + CuCl2 -> 2 FeCl3 + Cu

BR from Ivan OK1SIP

Thanks for the info . I definitely give it a try ! 
I wonder if anticorro steel is what we call here  INOX  .

Hi Zsolt,
the FeCl3 solution can be regenerated using electrolysis. The recommended materials are anticorro steel for cathode and carbon for anode. Copper collects on the cathode and the solution increases its power while turning FeCl2 to FeCl3.
Etching:         2 FeCl3 + Cu -> 2 FeCl2 + CuCl2
Regeneration: 2 FeCl2 + CuCl2 -> 2 FeCl3 + Cu

BR from Ivan OK1SIP

Admin wrote:Thank you for the information.

I thought you were using a pen, but now I see you are using PLA as the resist. Very clever.
And when you create the file you also remove any raft or brim.

Did you have any PLA adhesion problems? This is one of the biggest problems I have, and I feel sure that my PLA will have difficulty sticking to copper without Pritt-stick. I never had any problems with one particular brand of filament, when I borrowed someone else's machine. I wonder if the colour makes any difference?

But I would really like to see a pic of a finished board, if it is not too much trouble for you. It would be really appreciated.

BR Harry - SM0VPO

Hi, 
 I'm not using PLA to print the board . I use a non expensive 0.4 mm permanent marker .
I meant that i use the same printer for making things from plastic and the same printer to plot PCB board . 
Now i'm thinking to move a step further because unfortunately my FeCl solution is saturated with cooper so the board needs some scratching after. I can buy new Fecl solution but i think it's not enough concentrated, or we get third order products (we had a very strong chemistry industry before starting with imports ).  So i plan to use milling . I need to get a biax small enough and with enough rpm to engrave the cooper board.
Probably i need to build an other machine since any mistake with that kind of tool would harm the heated bed.
Regarding the PLA i observed the same thing, color does give different adhesion behavior that can be compensated with +- 5'c adjustment of bed temperature.
I also experienced that the thermistor from these printers go broke quick . For this o plan to replace them in the far future with industrial PT100 sensor .

I believe to be able to print directly on the cooper board you need to give it quite a good polish . Or change the plastic from PLA to something more soft .

Thank you for the information.

I thought you were using a pen, but now I see you are using PLA as the resist. Very clever.
And when you create the file you also remove any raft or brim.

Did you have any PLA adhesion problems? This is one of the biggest problems I have, and I feel sure that my PLA will have difficulty sticking to copper without Pritt-stick. I never had any problems with one particular brand of filament, when I borrowed someone else's machine. I wonder if the colour makes any difference?

But I would really like to see a pic of a finished board, if it is not too much trouble for you. It would be really appreciated.

BR Harry - SM0VPO

Admin wrote:That's nice. How did you get the pen to lift? Or do you use custom SW?

Don't forget a heatsink on the transistor.

BR Harry

Hi,
i use the same SW with some modifications .In CuraEngine settings i have checked the option Enable retraction and in the Retraction distance field i entered 2 mm. Also i modified the startup G code from:
; Default start code
G28 ; Home extruder
G1 Z15 F{Z_TRAVEL_SPEED}
M107 ; Turn off fan
G90 ; Absolute positioning
M82 ; Extruder in absolute mode
;G1 Z1.5
{IF_BED}M190 S{BED}
; Activate all used extruder
{IF_EXT0}M104 T0 S{TEMP0}
G92 E0 ; Reset extruder position
; Wait for all used extruders to reach temperature
{IF_EXT0}M109 T0 S{TEMP0}

to his:

G28     ; Home extruder
G1 Z3 ; go up 3mm
G90 ; Absolute positioning
M82 ; Extruder in absolute mode

In the above sequence i just took out the heating . This could be done also in Repetier software by checking Dry run option
So now i have two slicing configurations saved. One for melting plastic and one for plotting with a permanent marker. 
And ... i added a small offset to the Z axis so that the nozzle won't hit the PCB. This would have been to tedious to do by software so i just added a 5 mm bolt on the Z axis limiter  which i can take off when i print PLA

That's nice. How did you get the pen to lift? Or do you use custom SW?

Don't forget a heatsink on the transistor.

BR Harry

I managed to draw the circuit for the suggested controller ... with a 3d printer 

Ivan wrote:Hi Zsolt,
I found a nice circuit here.

BR from Ivan OK1SIP

thanks i'll give it a try

Hi Zsolt,
I found a nice circuit here.

BR from Ivan OK1SIP

Ivan wrote:Hi Zsolt,
a DC brushed motor regulator with speed feedback can be built with cca 3 transistors. It makes use of the voltage generated by the running motor, which acts against the power source voltage. Such regulator can keep the motor speed stable even at changing mechanical load. Using an icremental sensor and Arduino is a real OVERKILL in this case.
A stepper motor with simple logic keeps its speed perfectly, too.

IMHO = in my humble opinion

BR from Ivan OK1SIP

hi ,
i did try out a basic configuration , i put a bigger tranzistor in emitter follower mode to drive the motor than i used a second tranzistor to adjust the bias of the base of the first one with the emf voltage of the motor. It does work , just that it seams somewhat late in reaction. After a while heat comes in action and all biases move away ..

Thanks to the encouragement i get here,  i took a little time to align up the printer head and the nipp roll/counter nipproll on a support. 
Next i need an an unwind and rewind . (the paper can be printed on the other side too )

Hi Zsolt,
a DC brushed motor regulator with speed feedback can be built with cca 3 transistors. It makes use of the voltage generated by the running motor, which acts against the power source voltage. Such regulator can keep the motor speed stable even at changing mechanical load. Using an icremental sensor and Arduino is a real OVERKILL in this case.
A stepper motor with simple logic keeps its speed perfectly, too.

IMHO = in my humble opinion

BR from Ivan OK1SIP

Admin wrote:Hi Zsolt,
I think it may be better to make a constant voltage source.
If a motor encounters a change in mechanical load then it will try to draw more current to overcome the load.
If you current is restricted then the motor speed will change, until the load becomes constant.

On the other hand, if you have a huge gear reduction then the motor will be buffered from small load changes, but constant voltage will still be better for controlling the motor speed, IMHO.

BR Harry

Hi,
indeed torque is mostly related to current (..) so the actual torque will be limited . I taught something and i wrote something else because i'm working on a rewind tensioning control for a label sorter....  
Somewhere i used to have circuits of sony walkman (for even more younger papanasos than me those are portable things to play casette tapes See wikipedia  ) . If i don't find them most probably i put an Arduino to work since i already have incremental encoders on the motors. Than a PID will do the job to keep speed around setpoint. 
Still, i remember some drawing with only 2 transistors or so and maybe some diode, and they where silicium transistors  ...  Some walkman's also had a little board on the back of the motor with a little cip which i don't remember ..
I have seen a casette player in a car 8 years ago 
What is IMHO ?

Hi Zsolt,
I think it may be better to make a constant voltage source.
If a motor encounters a change in mechanical load then it will try to draw more current to overcome the load.
If you current is restricted then the motor speed will change, until the load becomes constant.

On the other hand, if you have a huge gear reduction then the motor will be buffered from small load changes, but constant voltage will still be better for controlling the motor speed, IMHO.

BR Harry

HI,
i have 2n3055  ..
What if i make a simple constant current circuit to supply the motor?
like this http://blog.novaeletronica.com.br/en/circuito-de-fonte-simples-de-corrente-constante-e-limitador-de-corrente/

Hi,
the first letter means the material resp. the width of the prohibited energy gap:
A - germanium
B - silicon
C - galiumarsenide.

The very first trasistors were the types OC72 etc. It should read as "zero si" i.e. a triode with zero
heater voltage. IMHO they all were germanium ones.

The first letter for parts from former Soviet Union and Czechoslovakia:
G - germanium
K - silicon ("kremnij", "kremik")

AD162 is surely a germanium power transistor.

BR from Ivan OK1SIP

hi,
i reprinted the gear , i found out the cause of the noise it was a small error at the shaft part I also made it wider than the small metallic one on the motor. I greased it a little (thous useless with plastic on metal) Now it runs like it should . 
The gear was drawn in Autocad 3D , i also published it on Tinkercad in the net (it's similar to thingiverse...)   There are all my printings about this thing .
 It's easy only when you already have one gear than you just need to find the transmission ratio to the other shaft ... 
I want to build some circuits for the motors now .
On the schematics provided by members in a post below  i found some simple circuits , the thing is that those tranzistors are obsolete for me . AD162 ? is that a germanium transistor ? Younger papanasos live only with silicium now days 

I think it also looks rather well made.

I tried to mess about with gears on my 3D printer. For large gears where the teeth are greater than about 1.5mm it is perfect, but for smaller gears I found that rubber ring or belt-drive is more durable: less concentrated stress.

I tried to make a gearbox for an R/C buggy, according to www.thingiverse.com but the 540 motor stripped the gear threads after 30 minutes. I re-made the gears about 3x the size and they were much more durable (but BIG). Original size and rubber-ring/traction worked better.

Your project looks very good and I can see a lot of work and great attention to details - hex-bolts, recessed, etc. What modelling program did you use?

BR Harry - SM0VPO

Very nice!

Ivan OK1SIP

meanwhile from the nozzle of my printer

I found on the net that the elicoid should have periphery speed 20 times greater than paper advance ...
My mechanics are not so good , so it's kind of noisy .

The gear i printed has a module of 0.5 . I think that's the lowest i can get with my amateur printer. The motor i fond has an encoder . Should i get a suitable circuit for it ? Or just a simple circuit without speed control loop?

That's what I thought, but not exactly secret.
If I use my EPROM project to cycle through small memory chunks, one could generate several waterfall text messages with a series of 8 push-buttons:

CQ DE SM0VPO
CQ NO CONTEST
73 DE SM0VPO

The possibilities are endless. But I still like the idea of the MP3 player. You can get them for just $2.

BR Harry

Thank! That is a great trick!
Secret messages in a MP3 file, what a joke!

Hi Ruud,
I simply use Adobe Audition and select spectrum view.
In that I can highlight a section of colour spectrum, Ctrl-X and it becomes black.

As I said, it took me about 5 minutes to crop out that SM0VPO.COM text in there.

At the moment i am sunbathing, but thinking about a method of generating 5 selected tones,

5 - 1040
4 - 1030
3 - 1020
2 - 1010
1 - 1000

If I send the tone sequence:
3
2+4
1+5
1+5
1+2+3+4+5
1+5


Stuff them into the MIC socket of my HF rig, using HRD PSK31, then everyone should see the letter "A" on the waterfall display.
__X
_X_X
X___X
X___X
XXXXX
X___X

It will remain there as it scrolls up the screen. That way I can send CQ SM0VPO CQ for people to actually read.

I can generate a small MP3 file in Audition :-)

BR Harry

This is really amazing!
What was the software you used for making the 'Get back' demo Harry?

i know i know ... some of my teachers (actually most of them where not teachers from the beginning) worked at the local calculator machines institute ,  i heard a lot of old stories with electromechanic machines that suppose to calculate ....
Anyway this thing with turn your pictures to sound is amazing 
As engineer in electric power systems i had to do a lot with furier analysis , we had a data acquisition board attached to a computer . Big consumers generate a lot of harmonics and we had to see how could we improve the quality of electric energy .... or at least find out what is the biggest harmonic generator .
I'm glad im not dealing with fourier and this things anymore.

zsolt wrote:all this reminds me of furier transform ... 

Yes you typically use som sort of Fourier transform to make a spectrogram/frequency domain/power spectrum - and back to time domain.

Please see and read about this more than 100 year old design of an old analogue mechanical Fourier transform machine:

17 Nov 2014, Pre-digital computer 'cranks out' Fourier Transforms. Boffins get a handle on pre-digital computer, restore it to working order:
http://www.theregister.co.uk/2014/11/17/predigital_computer_cranks_out_fourier_transforms/

Albert Michelsons harmonic analyzer:
http://www.engineerguy.com/fourier/pdfs/albert-michelsons-harmonic-analyzer.pdf

Video-series:

(1/4) Intro/History: Introducing a 100-year-old mechanical computer:
https://www.youtube.com/watch?v=NAsM30MAHLg

(2/4) Synthesis: A machine that uses gears, springs and levers to add sines and cosines:
https://www.youtube.com/watch?v=8KmVDxkia_w

(3/4) Analysis: Explaining Fourier analysis with a machine:
https://www.youtube.com/watch?v=6dW6VYXp9HM

(4/4) Operation: The details of setting up the Harmonic Analyzer
https://www.youtube.com/watch?v=jfH-NbsmvD4

all this reminds me of furier transform ... 

Hi Glenn,
Not at all sure what it should be called, but I have seen it on may files published on YouTube.com.

If anyone else is wondering what we are writing about, here is a screen-shot from my spectrum view:



It took me about 5 minutes as I was not particularly accurate with the scale or size of the characters. But it shows just how easy it is to do this sort of stuff.

But the point I was trying to make is that if this was a PSK waterfall display, then something very similar can be easily produced,
in fact, with the correct software for the encoder, a PSK waterfall is the only decoder actually needed. No synch problems and no need to have dual lines of text in case they slip in time.

BR Harry - SM0VPO

Admin wrote:Hi again Glenn,
That was really interesting watching those links, and others similar.

Take a look at a little piece of music I have edited:

http://www.sm0vpo.altervista.org/get-back.mp3

I put in there SM0VPO.COM along the entire length of the file, without any noticeable difference.

BR Harry - SM0VPO

Ahh - you have removed the soundpattern SM0VPO.COM between 7-15kHz :-)

What should it be called; watermarked or imagemarked sound?

I use mac and found this program that can show it (both for Windows and Mac):
http://photosounder.com/download.php

Hi again Glenn,
That was really interesting watching those links, and others similar.

Take a look at a little piece of music I have edited:

http://www.sm0vpo.altervista.org/get-back.mp3

I put in there SM0VPO.COM along the entire length of the file, without any noticeable difference.

BR Harry - SM0VPO

A new way to exchange QSL cards?:

Encoding pictures in frequency waterfall display:
https://www.youtube.com/watch?v=ugdgYl9aLXY

KQ4ZH Sending waterfall pics:
https://www.youtube.com/watch?v=VS7MHgjJAag

Some Odd Things on the Radio:
https://www.youtube.com/watch?v=OffrnGB4RpY

-

How did they do it?:

Converting images to IQ streams that make images appear in waterfall plots:
https://github.com/polygon/spectrum_painter

August 22, 2015, Spectrum Painting on 2.4 GHz:
http://hackaday.com/2015/08/22/spectrum-painting-on-2-4-ghz/
Quote: "...
Give a software-defined radio (SDR) platform to a few thousand geeks, and it’s pretty predictable what will happen: hackers gotta hack. We’re only surprised that it’s happening so soon. Spectrum Painter is one of the first cool hacks to come out of the rad1o badge given out at the CCCamp 2015. It makes it dead-simple to send images in Hellschreiber mode on a few different SDR hardware platforms.
..."

An OFDM Spectrum Painter for GNU Radio:
https://github.com/drmpeg/gr-paint

-

He explains waterfall displays with yellow rubber ducks :-)

Waterfall Displays Explained: Ask Dave #21:
https://www.youtube.com/watch?v=91XtLZNIyK8

-

Unknown audio waveform containing hidden images:
https://www.youtube.com/watch?v=FnzIpAAzP3w

SonicPhoto - What do pictures sound like?:
https://www.youtube.com/watch?v=8BGr1u-Oz8M

Hi ,
using a simple 2x 16 character lcd was first spark in my mind.. its interesting that also the characters in that lcd are 5 x 7 pixel . It would be to easy    all that one has to do is to scan the lcd pixel by pixel in a loop and lit up those ones that are beeng scanned when tone is present ...and shift the display . And maybe add some data storage to hold up larger text 
The electro-mechanic  thing is more of a thrill . It's simple as walking . I'm specially interested in the printing . If this works it will be very interesting. The most fun part for me is that  i'm not  telling absolutely nobody what it is,  so everyone looks amazed trying to figure out what the hack is that   . So fun to see  very curious people not understanding things

Good grief!! I had heard that there was something along these lines, but that was back in 1968 when i was under training in the military. Never saw anything, no photos, pictures, or anything. All we had was a particularly well informed, civilian radio school teacher (Mr Horry (or possibly Horrie)).

The mode itself look fantastically simple, would need little or no actual decoding to make a software encoder/decoder. I can see that the original concept needed a mechanical arm, and some form of logic decision, but on screen there can be a direct display of noise in sections. It should be possible to see characters forming in the noise when no ON/OFF tone is present.

TX/RX? What about a simple HF rig, with SSB TX and an audio tone generator + keyboard (TX) and amplifier (RX) plus HACD (Human Adaptive Context Decoder)?

If you could use a rolling LCD display then perhaps the decoder could be decoded using a simple 16F624 driving two text lines.

This also reminds me of my old Creed 7R/P teleprinter that printed paper tape with the Baudot code. The whole caravan workshop used to rock when the "bell" decoder started champing its way through RTTY.

BR Harry

Hi,
you should keep the ability to fine tune the helicoid speed (with respect to the sender speed) while keeping the paper speed constant. The easiest way is using two motors. Another way is to construct a continuously variable gear.

BR from Ivan OK1SIP

hi ,
i will use the reley  driver circuit. My approach is somewhat different , probably awkward for some older foxes ... 
i will put the ink container down , it seems most normal to me . In every DIY model on the net you see the impression roll on the top . Probably because they don't want to turn the paper back again.
Also i want to use a single motor and gear transmission . For the start i use a pulley and 2 motors until i get the relation between "web advance" speed and "impression roll" speed . This could be calculated , but i don't like mechanics so i try the natural way ... Anyway when i have the geared transmission i can still reprint the helicoid with another pitch
This is what i printed so far , ink container ,fountain roll , helicoid and a pulley . The spikes on the pulley are for speed measurement .

Not much yet , i still need the traction roll and nip roll ,and of course the electromagnet . Then i can test things out 
I downloaded a program which can send and receive and uses only the sound card of the computer   . Called IZ8BLY Hellscreiber

Hi,
lots of inspiration:
https://www.nonstopsystems.com/radio/hellschreiber-homebuilt.htm#My-Hell-printer

BR from Ivan OK1SIP

Hi Zsolt,
maybe the strength is not an issue. The printing arrangement must be very lightweight to minimize inertia. Consider using a polarized relay as the actuator, so that the printing helix can be both pushed down and lifted actively.

BR from Ivan

hi, now i m stuck with the receiver ..i printed some parts based on what i saw on the net .Now i got to the rellay (in direct translation from hungarian that part translates to "signal catcher ").
I think to solve this in 2 ways . One i could savage a 24Vdc rellay and work around the mobile armature ... or i could build one from scratch . I know that it has to be a very fast one , the impedance should match the circuit driving it , histeresis should not be at all  ... Here i really  can't estimate the value that should be around. I know that the circuit driving it is like a radio . Radios usually drive speakers . A speaker is about 4 ohm ... should i build a coil about 50 ohm ? 
I think it will take sufficient current to have enough force ...  I have doubts about this , i think this is a critical part . I will dig out some old manuals about electromagnets . Anyway i will create a weak and adjustable  tension zone in front of the printing screw so that little force is needed

Hi Zsolt,
so you consider a hybrid system? Why not.
The most spread code uses a 5x7 dot matrix for serialization of each character. It would be reasonable to employ 5 bytes to store it. Considering cca 50 different characters, all bitmaps get into 250 bytes - no problem for Arduino!

BR from Ivan OK1SIP

hi ,
just for the sake of try out things i printed some 3d parts similar to the receiver parts i see online... the transmitter could be programmed on an arduino i think ...

Hi,
the original mecanical devices are quite rare now, although lots of Feldhellschreibers were made in Germany during WWII for Wehrmacht etc. The mode is still alive in the ham community, see e.g. http://www.blackcatsystems.com/ipad/iPad_iPhone_Hellschreiber.html
When I was a boy and read about those mechanical devices, I dealt with how to make one. The printer is quite simple. The hard think is to make the encoding wheels or drum driven by a keyboard. I abandoned the idea at last.

BR from Ivan OK1SIP

What it means is, that people do not need to learn morse - just use hellschreiber via a keyboard :-)

Hellschreiber:
https://en.wikipedia.org/wiki/Hellschreiber
Quote: "...
Improvements that came as a result of software implementation:
* Depicting the received signal as shades of gray instead of monochrome, thereby making it much easier to read weak signals.
* Changing to a different font. Here is one mode that is truly international and independent of character sets: any thing that can be depicted as markings within a 7 pixels high grid, can be transmitted over the air.
..."

WHAT IT IS - AND WHAT IT IS NOT:
https://www.nonstopsystems.com/radio/hellschreiber-function-operation.htm#top-of-page
Quote: "...
Note that this clever printer is stupid! It prints all sufficiently strong signals that it receives – be it the intended pixels, noise or interference. It has no notion of fonts, bits, pixels, encoding, speed, or synchronization. It does no decoding or interpretation, other than deciding "signal (not) present". Signal distortion may add unintended pixels, and may cause omission of intended pixels. This only makes the printed text harder to read. But unlike teleprinters, a Hell-printer can never print a wrong character! The human reader, with its impressive pattern-recognition capability, can read messages that are well "down in the noise":
...
can be used with a very simple CW transceiver. No other digi-mode can do this!
..."

It should even be possible to send pictures and receive them robustly, after insertions of line-sync-starts (7 lines at a time) and frame-sync-starts ("blacker than black") like analog B/W TV :-)

hi ,
i am absolutely amazed ..this man was a real engineer 
Its not difficult cu guess what id like to build now, thous i did not end a single project i started lately     
  wonder if i could successfully build a receiver how can i test it since i believe no one has a transmitter anymore

zsolt wrote:Hi,
you will not believe what i found on the internet . https://www.youtube.com/watch?v=4Aebc8Ur5Gw
Can anyone tell things about this ?

Hellschreiber: WHAT IT IS - AND WHAT IT IS NOT:
https://www.nonstopsystems.com/radio/hellschreiber-function-operation.htm#top-of-page
Quote: "...
Well, it is a telegram system in between fax and teleprinter.
..."

Electronic circuitry & schematics:
https://www.nonstopsystems.com/radio/hellschreiber-circuitry.htm#top-of-page

Hi,
you will not believe what i found on the internet . https://www.youtube.com/watch?v=4Aebc8Ur5Gw
Can anyone tell things about this ?