Forums :: NEW! Web Resources :: Features :: Photo Gallery :: Vintage Radio Shows :: Archives
Support This Site: Contributors :: Advertise


It is currently Nov Tue 24, 2020 2:48 am


All times are UTC [ DST ]





Post New Topic Post Reply  [ 111 posts ]  Go to page 1, 2, 3, 4, 5, 6  Next
Author Message
 Post subject: Digitizing Hickok Cardmatic Test Card data
PostPosted: Jan Wed 29, 2014 3:25 pm 
Member

Joined: Dec Thu 06, 2012 11:47 am
Posts: 115
Location: Portland, Oregon
I just made a small utility program for converting Cardmatic test card into pin mapping data.
Program screen shot as attached.

I just started this project yesterday, functionality is limited.
Any one would like to give it a try? Also let me know what you think, we can put in all the function we want, if that is do-able.

Please leave me message for email address, so that I can send you the program for testing.

Jason Huang


Attachments:
PinMapping003.JPG
PinMapping003.JPG [ 121.76 KiB | Viewed 4975 times ]
PinMapping002.JPG
PinMapping002.JPG [ 89.73 KiB | Viewed 4975 times ]
PinMapping001.JPG
PinMapping001.JPG [ 95 KiB | Viewed 4975 times ]
Top
 Profile  
 
 Post subject: Re: Digitizing Hickok Cardmatic Test Card data
PostPosted: Jan Wed 29, 2014 4:46 pm 
Member

Joined: Jan Thu 01, 1970 1:00 am
Posts: 25381
Location: Pocasset, Cape Cod, MA
I assume you know about these military manuals:
Attachment:
94716.jpg
94716.jpg [ 18.36 KiB | Viewed 4965 times ]
Attachment:
94716 p1.jpg
94716 p1.jpg [ 108.4 KiB | Viewed 4965 times ]

It's 50 pages and I can't get very excited about xeroxing the whole thing but it ought to be available somewhere. There's a commercial version of it with 121 pages:
Attachment:
2490-427.jpg
2490-427.jpg [ 15.58 KiB | Viewed 4961 times ]

There are also other manuals showing the hole locations for individual cards; I only have a partial copy of pages 31-66:
Attachment:
TO 33AA21-9-1.jpg
TO 33AA21-9-1.jpg [ 151.13 KiB | Viewed 4965 times ]
TOs are Air Force Technical Orders.


Top
 Profile  
 
 Post subject: Re: Digitizing Hickok Cardmatic Test Card data
PostPosted: Jan Wed 29, 2014 9:22 pm 
Member

Joined: Dec Thu 06, 2012 11:47 am
Posts: 115
Location: Portland, Oregon
Alan, what I intend to do is to convert scanned image into binary mapping data. This program is not that smart yet, you have to click on the lower left "hole" (A-17) and drag mouse all the way to upper right (L-1), then it generates grid, calculate and judge each intersection being a "hole" nor not.

My next step is to eliminate this TWO-CORNER locating, let program do the image processing/judging itself.

With this pin mapping ( 17 * 11 = 187 Boolean value matrix ), thousands of records is just a small table. Computer program can easily re-create a card image, clear and perfect. Maybe there is a kind of printer that can print the "hole" or "non-hole" on blank card, you can then punch it.

(Or there is a kind of computer controlled punching machine...well, this has gone too far, just my day dreaming, too expensive....)

The whole point is an easy way to

1. append/edit/validate card information, we can even put this on a cloud server so that we all can get access to it.

2. easy to find the record you want, if you need to "punch" a card yourself

3. people can exchange and share their database easily. We can create our own database and merge database from others and expand the whole set. Well, if there is this kind of database already, then ignore my propose, we can just get it with or without some cost.

4. with the pin-mapping, program will be able to parse the meaning of these cards, filament voltage, Ef, Ec..Gm or Ib and so on, according to the description in the technical manual and other information available online. This would provide you another way of looking up the "Tube Test Condition Table".

5. computer aided design, you can even design your own special test card with this kind of data. I don't really know about tube or electronic circuit, I can not say much about this.

6. other application...people can always figure out different application or improvement on this kind of database, once the database is available.

7. as for commercial value, sorry, I did not think of that. This is my hobby, not my job. I am willing to share the program and database (not now, program is still under construction, database is to be defined). It is to build a database and a set of application programs. I can not do this all alone. I can not scan and process thousands of cards all by myself. It takes us to do it together. And I do this just for fun.


Top
 Profile  
 
 Post subject: Re: Digitizing Hickok Cardmatic Test Card data
PostPosted: Jan Wed 29, 2014 11:17 pm 
Member
User avatar

Joined: May Fri 29, 2009 4:35 am
Posts: 2174
Location: Chicago, IL USA
Yes, I'm interested and have been thinking about doing something similar for a while now. However, my main interest is the less common Hickok 121 rather than the 123 which is already pretty well documented. I have the complete basic set plus about 75 additional cards.


Top
 Profile  
 
 Post subject: Re: Digitizing Hickok Cardmatic Test Card data
PostPosted: Jan Wed 29, 2014 11:39 pm 
Member

Joined: Dec Thu 06, 2012 11:47 am
Posts: 115
Location: Portland, Oregon
This program is a prototype. It works for Hickok card format, 11 * 17. it works with 121, 123, too
Currently it just processes the image and generate a 11 * 17 Boolean matrix, I am still thinking what the output format it will be and what other contain is to put in.

To build a database and its application program, we have to figure out what we want to do with these records.

1. Tester Model Type
2. Tube #/ID
3. for the technical meaning, Ef, Ec and so on, it depends on what kind of information we can get. I have only read part of the spec for KS-15874. At least, for the part that I have read, it should be able to parse the values from the pin mapping. I guess it applies to other model of the same style. As for 123 or 121, I have no information in hand.

I am still working on this image processing. I have downloaded some low-resolution card image and I am testing it. Lower resolution >> Less Information >> Higher Noice >> you know........


Top
 Profile  
 
 Post subject: Re: Digitizing Hickok Cardmatic Test Card data
PostPosted: Jan Wed 29, 2014 11:55 pm 
Member

Joined: Mar Thu 12, 2009 2:20 am
Posts: 1521
Location: Dayton, OH
Funny thing .. I've been kicking around a similar database for "non-card based" tube testers, and a little web front end, that's tablet/phone friendly.

Pick your tube tester
Type in the tube number

Viola, your tester settings.

But my availability of time is kinda non-existent. Coming up with a reasonable database structure to handle multiple tester types, with all their varying controls, isn't coming easily. (Never mind scanning/OCRing a roll chart is going to be kind "teh suck")

Would be kind of nice to have an online database of it all though. (Such as for the EICO 666 where alot of the data is (so I hear) plain wrong).

David


Top
 Profile  
 
 Post subject: Re: Digitizing Hickok Cardmatic Test Card data
PostPosted: Jan Fri 31, 2014 12:29 pm 
Member

Joined: Dec Thu 06, 2012 11:47 am
Posts: 115
Location: Portland, Oregon
Current Status:


Attachments:
2014-01-31_032900.JPG
2014-01-31_032900.JPG [ 171.5 KiB | Viewed 4870 times ]
Top
 Profile  
 
 Post subject: Re: Digitizing Hickok Cardmatic Test Card data
PostPosted: Jan Fri 31, 2014 8:34 pm 
Member

Joined: Dec Thu 06, 2012 11:47 am
Posts: 115
Location: Portland, Oregon
Contrast adjustment for semi-transparent plastic cards or low contrast ones
and manual Pin Mapping Editing function, if only couple holes were missed or mis-judged.

It can handle slightly tilted image, if one horizontal/vertical line can go through all the holes in one row/column.
If they are too far off, it will be a problem to form the grid mesh.

Next thing is in what form it is to be stored?

How about put it in SQL Express with :

1. comma separated text import/export function with record checksum? So that people can exchange their data via text file safely.

2. scanned image can be put into SQL, but that would take quite some space.

3. comparison and validation, for example:duplicated records should be compared and warning should be raised if Pin Mappings are different.


Attachments:
2014-01-31.png
2014-01-31.png [ 238.07 KiB | Viewed 4824 times ]
Top
 Profile  
 
 Post subject: Re: Digitizing Hickok Cardmatic Test Card data
PostPosted: Feb Sat 01, 2014 3:36 am 
Member

Joined: Dec Thu 06, 2012 11:47 am
Posts: 115
Location: Portland, Oregon
For now, it can save Pin Mapping to a CSV file.
I have made a PDF file for program description, but it is kind of big, 4.27MB.
I can not upload it here, if you are interested, drop me a message with your email.
Have fun!!


Top
 Profile  
 
 Post subject: Re: Digitizing Hickok Cardmatic Test Card data
PostPosted: Feb Fri 07, 2014 2:52 am 
Member

Joined: Jan Thu 01, 1970 1:00 am
Posts: 575
Location: Monte Vista, CO. USA
Just a FYI, most of what you are trying to accomplish has already been done by my Computerized Cardmatic software. It has 3,924 cards in it, and the software provides on-screen data showing the test conditions for each test. Here's a sample screen shot:

Image

I am currently working on a way to make this data available on the web. My plan is to make all the cards available as a download for a small fee since in my case this isn't just a hobby :).

-Phil Frakes


Top
 Profile  
 
 Post subject: Re: Digitizing Hickok Cardmatic Test Card data
PostPosted: Feb Fri 07, 2014 6:22 am 
Member

Joined: Dec Thu 06, 2012 11:47 am
Posts: 115
Location: Portland, Oregon
Phil, that is a nice software!!

You know, I am kind of "lazy", when I read the instruction manual, describing the meaning of pins, I just think that, hey! if I have the data of pins then I can parse the testing condition, pin assignment and so on........
And use a scanned image would be not too much work to do to build up a library.

Well, kind of fun thing to do.


Top
 Profile  
 
 Post subject: Re: Digitizing Hickok Cardmatic Test Card data
PostPosted: Dec Sun 09, 2018 6:07 am 
New Member

Joined: Dec Thu 06, 2018 6:36 pm
Posts: 1
Hello, I have a Cardmatic, I have some cards but I'm missing a lot, could someone tell where I could find manuals showing the hole location?
thank you


Top
 Profile  
 
 Post subject: Re: Digitizing Hickok Cardmatic Test Card data
PostPosted: Apr Wed 10, 2019 2:32 pm 
Member
User avatar

Joined: Mar Sun 31, 2019 3:44 am
Posts: 126
Location: Alberta,Canada
this is interesting . Phil did you get this done

I am currently working on a way to make this data available on the web. My plan is to make all the cards available as a download for a small fee since in my case this isn't just a hobby :).

-Phil Frakes


Top
 Profile  
 
 Post subject: Re: Digitizing Hickok Cardmatic Test Card data
PostPosted: Apr Wed 10, 2019 4:27 pm 
Member

Joined: Jan Thu 01, 1970 1:00 am
Posts: 809
Location: Opelika, AL
if there is a way to output the data to generate an autocad or adobe illustrator vector file, it is a relatively trivial job to use a laser cutter to punch the cards, and even laser etch the info such as tube type onto the card.

I'm only familiar with the Universal Laser machines that we have at work, but any line weight in Autocad greater than zero (and line weight greater than (.07 points/.001") makes it do raster cutting instead of vector, so it will engrave/etch a picture instead of lines. Various colors are set up to tell the machine what power level and speed it should use, which allows different depth of engraving, along with full through cutting.

The only issue is with cutting tolerances, since there is no kerf compensation in the driver for the laser cutter. Part of that is because kerf width changes as the laser ages, and based on cleanliness of optics and perfection of focus. Also kerf width can be variable depending on natural variations in density of the material being cut. In principle, if a .125" hole is needed, and a .125" hole is drawn, it will always be slightly larger rather than slightly smaller. The same goes for the size of the finished card.

If someone can figure out how to get your database to automatically generate a usable file, I'd be happy to make a few sample cards as proof of concept. I periodically have to test the machines we have as part of routine maintenance that I do, and I might as well cut something useful rather than random squares and circles.

_________________
filling in the gaps in my education...


Top
 Profile  
 
 Post subject: Re: Digitizing Hickok Cardmatic Test Card data
PostPosted: Apr Wed 10, 2019 4:30 pm 
Member

Joined: Jan Thu 01, 1970 1:00 am
Posts: 809
Location: Opelika, AL
of course, I could draw the stuff manually, but it would be much easier/faster if there were a way to generate the basic info automatically! if a person were to want to make more than a couple of cards, it would get very tedious very quickly, and on top of that, there is the ever present risk of human error when doing such work.

_________________
filling in the gaps in my education...


Top
 Profile  
 
 Post subject: Re: Digitizing Hickok Cardmatic Test Card data
PostPosted: Apr Wed 10, 2019 5:24 pm 
Member
User avatar

Joined: Oct Sun 18, 2009 5:38 am
Posts: 3417
Location: Tyler, Texas 75707-4212
I suppose on the AutoCAD end, you guys know you can paste in an image file (jpg or several others)? If you have to scan several thousand cards it would indeed take a bit of time, but if several owners of card sets would divide it up it might not be too bad. You can also put several images on one sheet, so it may not be too bad. Once that it done, the mechanics of working out kerf width, etc., could be standardized and then exported to the laser cutter.

_________________
I'm right 97% of the time. Who cares about the other 4%!

-.-. --- .-.. -.. / ---.. ----- --...


Top
 Profile  
 
 Post subject: Re: Digitizing Hickok Cardmatic Test Card data
PostPosted: Apr Wed 10, 2019 5:44 pm 
Member

Joined: Jan Thu 01, 1970 1:00 am
Posts: 809
Location: Opelika, AL
Tony Wells wrote:
I suppose on the AutoCAD end, you guys know you can paste in an image file (jpg or several others)? If you have to scan several thousand cards it would indeed take a bit of time, but if several owners of card sets would divide it up it might not be too bad. You can also put several images on one sheet, so it may not be too bad. Once that it done, the mechanics of working out kerf width, etc., could be standardized and then exported to the laser cutter.


The laser cutters need vector lines to cut efficiently. I'm not sure if there is an easy way to convert a scanned image to a working vector file with Autocad. I've only experienced poor results when doing this with illustrator.

_________________
filling in the gaps in my education...


Top
 Profile  
 
 Post subject: Re: Digitizing Hickok Cardmatic Test Card data
PostPosted: Apr Wed 10, 2019 5:54 pm 
Member
User avatar

Joined: Oct Sun 18, 2009 5:38 am
Posts: 3417
Location: Tyler, Texas 75707-4212
Been a while, but I believe you have the intermediate step of converting to a .dwg, then exporting as a vector file.

_________________
I'm right 97% of the time. Who cares about the other 4%!

-.-. --- .-.. -.. / ---.. ----- --...


Top
 Profile  
 
 Post subject: Re: Digitizing Hickok Cardmatic Test Card data
PostPosted: Apr Wed 10, 2019 7:55 pm 
Member
User avatar

Joined: Mar Sun 31, 2019 3:44 am
Posts: 126
Location: Alberta,Canada
it sure would be nice to have a complete set on file that could be cut by simply taking the file to a laser cutter .
now to be able to have Hickok name and the tube number that is even a sweeter deal


Top
 Profile  
 
 Post subject: Re: Digitizing Hickok Cardmatic Test Card data
PostPosted: Apr Thu 11, 2019 2:03 pm 
Member

Joined: Nov Wed 30, 2016 7:35 pm
Posts: 8459
Location: Sunbury, Ohio 43074
"one of these days" I will get my entire set scanned and up on my museum website where the entire world can do whatever they want with it. Most likely they will be medium rez jpeg files, but I may provide links to external higher rez or pdf ... .still pondering. And by the time I get done pondering, I may be dead anyway .... but hopefully I'll get the cards posted first lol 8)

As for Autocad and automatic punchers, I doubt there's enough demand to make such a venture into an economically viable business, but who knows... maybe someone should make a brand new computerized interface to any cardmatic, or better yet design a brand new cardmatic using modern parts. Replace the card reader with a computer interface.

Hmm ... aren't there a few of these out there already ?? ;-)

_________________
Preserving the hist. of electronics, one boat anchor at a time! :)
https://www.bbtvtestequipment.com


Top
 Profile  
 
Post New Topic Post Reply  [ 111 posts ]  Go to page 1, 2, 3, 4, 5, 6  Next

All times are UTC [ DST ]


Who is online

Users browsing this forum: bob91343, John Limbach and 13 guests



Search for:
Jump to:  




































Privacy Policy :: Powered by phpBB