New NBTV camera project

Forum for discussion of narrow-bandwidth mechanical television

Moderators: Dave Moll, Andrew Davie, Steve Anderson

Postby aussie_bloke » Fri Oct 12, 2012 11:03 am

Thank you Gary for the templates and the program that does the 1.5 turn discs :) . Will definitely stick a masking in front of the disc to prevent light spillage.
aussie_bloke
Just nod and pretend you understand me
 
Posts: 219
Joined: Sat Apr 28, 2012 7:48 pm
Location: NSW, Australia

Postby Klaas Robers » Sat Oct 13, 2012 5:58 am

A lens from a slide projector can make an image of the size of a slide. Realize that for the Nipkov camera the disc is the slide, so the hole pattern should match the size of a slide or slightly smaller.

To have an easy start, you should have saved the distances of the slide projector before you scapped it. Then rebuild it with the Nipkov disc at the position of the slide and the photo transistor at the position of the filament of the projection lamp. Then you need only small adjustments for an optimal result.

It is easier and better to steal a good layout than to design a bad one yourself.
User avatar
Klaas Robers
"Gomez!", "Oh Morticia."
 
Posts: 1656
Joined: Wed Jan 24, 2007 8:42 pm
Location: Valkenswaard, the Netherlands

Postby aussie_bloke » Sat Oct 13, 2012 10:24 am

Klaas Robers wrote:A lens from a slide projector can make an image of the size of a slide. Realize that for the Nipkov camera the disc is the slide, so the hole pattern should match the size of a slide or slightly smaller.

To have an easy start, you should have saved the distances of the slide projector before you scapped it. Then rebuild it with the Nipkov disc at the position of the slide and the photo transistor at the position of the filament of the projection lamp. Then you need only small adjustments for an optimal result.

It is easier and better to steal a good layout than to design a bad one yourself.


Thanks for the info. I have re-read chapter 20 on optics and recall that being mentioned in there. Good news I have not chucked away the projector not its parts I've pulled off, I still have the lens carriage so I can use it as reference with distances, or alternatively do you suggest I put the lenses back in the carriage and use it instead as I think it may possibly be a much more convenient idea?

Now there's something else I need a bit of clarity on, I may have read over it in the optics chapter but not sure, and that is the focal point from the taking lens. Now does the focal point of the image need to be at the surface of the Nipkow disc or at the surface of the first (less domed) condenser lens?

Looking at the size of picture can be made from a record Nipkow disc, I think I can an image size of roughly 20m by 15mm give or take some millimeters.

Now I have printed out the 1.5 turn Nipkow template, the one with the holes closer to the edge and printed it poster style and with precision cut the template with a ruler and pen knife to accurately join and tape the 4 pieces together, unfortunately the size of the template exceeds the size of the record by I guess 10 to 15mm. I wonder if that's due to an issue with inches and cm as I measured a LP record to be pretty well dead on 30cm and converting that to inches gives 11.8 inches not 12, and converting 12 inches to cm gives 30.5cm? Or could some records be out in diameter by a cm? Or could there be a mismatch in size of image to size on printout? Anyhow will compare the sizes of my records and measure up the template I printed and get back on the details.

In the meantime here's some pics of my lens assembly (more domed lens I admit is loose and tilted but I will correct that):
Attachments
lens7.jpg
lens7.jpg (67.88 KiB) Viewed 12386 times
lens8.jpg
lens8.jpg (68.01 KiB) Viewed 12386 times
lens9.jpg
lens9.jpg (74.43 KiB) Viewed 12386 times
nipkow.jpg
nipkow.jpg (46.14 KiB) Viewed 12386 times
aussie_bloke
Just nod and pretend you understand me
 
Posts: 219
Joined: Sat Apr 28, 2012 7:48 pm
Location: NSW, Australia

Postby gary » Sat Oct 13, 2012 11:16 am

Sorry Troy, my mistake. All my records are still packed away since my move so I couldn't measure one - I looked on wikipedia where it *seemed* they only mentioned them as 12". I just looked again and buried in thousands of lines of text there is:
"Since its inception in 1948, vinyl record standards for the United States follow the guidelines of the Recording Industry Association of America (RIAA).[43] The inch dimensions are nominal, not precise diameters. The actual dimension of a 12-inch record is 302 mm (11.89 in), for a 10-inch it is 250 mm (9.84 in), and for a 7-inch it is 175 mm (6.89 in)."

Because this format originated in the US I assumed the dimensions were the other way around, and you know what happens when you assume.

I still have to use imperial because Sketchup seems to always assume the drawing units are imperial and does a conversion - that may also be a problem with my DXF library something I'll have to check up on.

BTW I called in to my local photo shop and they can do an A2 print for $5 - if you have problems with accuracy that may be an option for you - I am sure your local photo shop will be similar. Personally I think joining them together will likely introduce errors but I suppose the cost is just an old record and time.

I attach a corrected PDF.
Attachments
cammon nipkow 11.89-.125.pdf
(26.08 KiB) Downloaded 621 times
gary
 

Postby aussie_bloke » Sat Oct 13, 2012 12:23 pm

Thanks for correcting the template for me, yep believe me I have tripped up many times from "assuming" with a lot of things in both technical and non technical situations. There are some print shops in the Wollongong area, I have been thinking on getting them printed up there and can spare a few dollars per print no problemo.

Now question out of interest, a while ago when I made my multi-standard televisor, I made the templates using my crude method by print screen from your program and overlaying them on top of each other in Corel Photopaint from screen dumps which gave me crude results, I would like to know, if your disc making program can easily be edited to allow for multiple spirals to be made for those who want to make multi-standard televisors? I suppose the easiest way to go about this would be to just simply print 3 templates and do them one after the other on the one record.
aussie_bloke
Just nod and pretend you understand me
 
Posts: 219
Joined: Sat Apr 28, 2012 7:48 pm
Location: NSW, Australia

Postby gary » Sat Oct 13, 2012 12:35 pm

Yes it should be fairly easy to do - I will look into it.

If anyone else has an "features" they think would be useful let me know.
gary
 

Postby aussie_bloke » Tue Oct 23, 2012 9:43 am

G'day all.

It has been a while since last updating you all on my NBTV camera/monitor project.

I have since constructed the camera/monitor up to a makeshift completion just to test it out to see if it would work.

I have constructed the framework to hold the disc, motor, lenses/sensor and LED bank and placed all the parts in their appropriate spots. I then arranged the taking lens to focus in close range sharply on the record Nipkow disc and adjusted the condenser lenses so the incoming dissected light hits the phototransistor. I also added a piece of cardboard to the right side of the disc to block light coming in from the side of the disc to the lenses.

With the template for the disc, it was still slightly bigger than the record but I was able to fudge it a bit but the edge hole is literally close to the edge of the disc.

Anyhow I set it all up to makeshift test level and I placed an ornament doll in close focal range and switched off the lights and shun a torch onto it and powered the set up. Unfortunately on the monitor end, nothing at all came up on display, not even a response to shining a torch directly onto the phototransistor.

So the problem looks to be between the phototransistor circuit and LED driver circuit. I hooked my oscilloscope to the output of the LED driver circuit, shine torch on phototransistor, no response, but hooking the oscilloscope to the output of the phototransistor sensor circuit and shinging the torch on it I got a signal response. So I believe amplification is needed between phototransistor output and LED driver.

After I fix that issue I will then have to contend with the arrangement of the optics. I am seriously thinking of putting the condenser lenses back into the slide projector lens cradle and sticking the phototransistor in place of the halogen globe.

Anyhow here below are pictures of my camera/monitor in makeshift state for testing, I will of course be refining it all once I have it "sort of working" as in getting signal response from varying light hitting phototransistor and obviously there will be an enclosure over the whole apparatus to keep the light out ;) :
Attachments
disc.jpg
disc.jpg (56.8 KiB) Viewed 12346 times
framework.jpg
framework.jpg (62.71 KiB) Viewed 12346 times
camera1.jpg
camera1.jpg (86.71 KiB) Viewed 12346 times
camera2.jpg
camera2.jpg (84.39 KiB) Viewed 12346 times
camera3.jpg
camera3.jpg (78.32 KiB) Viewed 12346 times
camera4.jpg
camera4.jpg (82.95 KiB) Viewed 12346 times
camera5.jpg
camera5.jpg (90.85 KiB) Viewed 12346 times
camera6.jpg
camera6.jpg (74.76 KiB) Viewed 12346 times
aussie_bloke
Just nod and pretend you understand me
 
Posts: 219
Joined: Sat Apr 28, 2012 7:48 pm
Location: NSW, Australia

Postby gary » Tue Oct 23, 2012 2:09 pm

aussie_bloke wrote:With the template for the disc, it was still slightly bigger than the record but I was able to fudge it a bit but the edge hole is literally close to the edge of the disc.


Hi Troy, your printer must be printing out over sized then as I have dug out my record collection and printed one off and it aligns perfectly with the first aperture 1/8" in from the edge, which is nearly 6 aperture widths wide.

You may need to do some calibration.

aussie_bloke wrote:So I believe amplification is needed between phototransistor output and LED driver.


Yes, this is normally called the "head" amplifier - so called because it needs to be physically near the detector itself, and is usually a high to very high gain amplifier - for this reason it needs to be enclosed inside a metal container for shielding purposes and also requires a very quiet decoupled power supply - the latter I usually make a battery in the development stages.

aussie_bloke wrote:After I fix that issue I will then have to contend with the arrangement of the optics. I am seriously thinking of putting the condenser lenses back into the slide projector lens cradle and sticking the phototransistor in place of the halogen globe.


Well that is a very well tried and proven solution, personally I would stick with what you have until you get some kind of a signal and then make a decision as to whether it is necessary - certainly many cameras have been made using something similar to the arrangement you have.
gary
 

Postby aussie_bloke » Tue Oct 23, 2012 3:10 pm

G'day Gary.

I actually got the template printed out at a print shop in Wollongong so their printers must be out of calibration, my printer from memory is pretty well accurate as I've printed CD sized discs which fit over them perfectly.

I will endeavor to make the head amplifier when I get back onto the project, will have to find a schematic of one that works best with phototransistors, will look at the NBTV handbook's head amp circuits and some other ones on other sites.

I will test the optics out and see if it all looks good, it will be a matter of locking the hemisphere condenser lens and phototransistor when they are positioned as they are a bit loose when sliding. I wonder if there's some margin for error if the lenses are slightly tilted or skewed?

Anyhow I will firstly build the head amp and test the response between the phototransistor and the LED bank then work on the optics.
aussie_bloke
Just nod and pretend you understand me
 
Posts: 219
Joined: Sat Apr 28, 2012 7:48 pm
Location: NSW, Australia

Postby gary » Tue Oct 23, 2012 3:20 pm

aussie_bloke wrote:G'day Gary.

I actually got the template printed out at a print shop in Wollongong so their printers must be out of calibration, my printer from memory is pretty well accurate as I've printed CD sized discs which fit over them perfectly.


Well it definitely must be - I'd ask for my money back...

Quite honestly I would take it back, I am sure they would be happy to re-print it with the required adjustment - after all it's not all that cheap so it should be reasonable accurate.

BTW have you actually measured it?

You can also check it yourself by printing out one segment and measuring it from the hub centre to the outside - I'll stake my reputation that it'll be correct (ok ok I know that isn't much ;-))

The problem you may have now is that it may be accurate in one axis and not in the other (i.e. elliptical) - or did you not use it at all?
gary
 

Postby aussie_bloke » Wed Oct 24, 2012 5:39 pm

G'day all.

I have recently decided to add amplification to the phototransistor sensor (see pic at bottom) and I got a much stronger signal response on oscilloscope when flashing the phototransistor with the torch. However when I connect the phototransistor/head amp circuit to my LED driver circuit I've been using and have found that the LED driver circuit seems to attenuate the signal coming from the phototransistor circuit when testing with the oscilloscope. The LED driver I used is in the pic just below.

What I would like to know what would be the best phototransistor head amp circuit to use and the best monitor LED driver circuit to connect with the head amp that will work good without any problems? Any schematics anyone can provide would be much appreciated.
Attachments
led_driver2.gif
led_driver2.gif (9.58 KiB) Viewed 12311 times
photoQ2transHiZ.png
photoQ2transHiZ.png (4.4 KiB) Viewed 12311 times
aussie_bloke
Just nod and pretend you understand me
 
Posts: 219
Joined: Sat Apr 28, 2012 7:48 pm
Location: NSW, Australia

Postby Harry Dalek » Wed Oct 24, 2012 5:51 pm

http://www.hawestv.com/mtv_cam/mtv_cam2.htm

Keep adding stages Troy as this fellow has done till it works as in instructions.
The electromagnetic spectrum has no theoretical limit at either end. If all the mass/energy in the Universe is considered a 'limit', then that would be the only real theoretical limit to the maximum frequency attainable.
User avatar
Harry Dalek
"Fester! Don't do that to 'Thing'"
 
Posts: 5364
Joined: Fri Sep 26, 2008 4:58 pm
Location: Australia

Postby aussie_bloke » Thu Oct 25, 2012 3:41 am

G'day all, I've got some fantastic news!!!:D

My NBTV camera/monitor is now working!!!:D

I fixed the image sensor to LED driver problem by adding more amplification stages as Harry suggested and after 2 stages of amplification I got a good response on the LED display when flashing the torch over the phototransistor.

Then I rigged up the camera/monitor to makeshift state and once again tested it and I shined my torch down the lens and on display was a crude looking circle!!!:D Seeing that alone really made me jump for joy!!!:D

I then attempted to shine a 150W spotlight onto objects in front of the lens however unfortunately the reflected light from the object hitting the disc was too low for the phototransistor to pick up, so I then shined the spotlight through a brick and 3 of the holes in it were nice and bright and well focused onto the Nipkow disc and looking at the monitor end I see three holes!!!:D

I know usually I have trouble photographing the display but this time I succeeded in photographing the display of the brick holes and here below is a photo of it along with a few other snaps of the setup:
Attachments
picture.jpg
picture.jpg (40.62 KiB) Viewed 12291 times
scan.jpg
scan.jpg (49.36 KiB) Viewed 12291 times
brick.jpg
brick.jpg (35.5 KiB) Viewed 12291 times
setup.jpg
setup.jpg (71.36 KiB) Viewed 12291 times
aussie_bloke
Just nod and pretend you understand me
 
Posts: 219
Joined: Sat Apr 28, 2012 7:48 pm
Location: NSW, Australia

Postby gary » Thu Oct 25, 2012 8:49 am

Well done Troy - step by step, that's the way.

You are probably now getting a feel for how much gain (and light!) you need to get a reflected light signal, and then you will become cognisant that such high gain brings up the noise floor as well as the signal hence the need for PS decoupling, shielding etc. All these can be over come however so stick with it!

(There's nothing to match the first glimpse of a picture generated by something you have constructed almost totally from scratch - no matter how basic - and each iterative improvement gives a similar buzz).
gary
 

Postby Harry Dalek » Thu Oct 25, 2012 9:39 am

Great news troy very good !
Now you know it works perhaps a idea would be reducing the light levels and adding another stage or 2 ...i would also play around with a variable voltage to the head amp and gain control just see what it works best at never know...
Using simple shapes is some thing gary told me about so thats a good start as Gary i think would say another camera is born
:wink:
The electromagnetic spectrum has no theoretical limit at either end. If all the mass/energy in the Universe is considered a 'limit', then that would be the only real theoretical limit to the maximum frequency attainable.
User avatar
Harry Dalek
"Fester! Don't do that to 'Thing'"
 
Posts: 5364
Joined: Fri Sep 26, 2008 4:58 pm
Location: Australia

PreviousNext

Return to Mechanical NBTV

Who is online

Users browsing this forum: No registered users and 35 guests