New NBTV camera project

[AncientBrit]

Hi Gary,

Not packed 5bit, just bog standard 8 bit with only 5 or 6 bits actually used, the other bits left at zero.

I could start another thread and publish some of my processing software here such as diagonal interpolation, movie editor and colour MAC but they will all be based on bitmaps as opposed to WAV files which is yet another standard for members to embrace.

And of course source code is ancient VB6!

Not sure if this forum will allow .exe files to be uploaded

Cheers,

Graham

[gary]

Ok, well I can decode those easily enough - but yeah, why not post your software? It's always another option and if it allows people like Harry and Troy to see some of your camera output I am sure it will be a great incentive to them. I wish I had kept all my old stuff on tape but - oh well...

[Harry Dalek]

Ok, well I can decode those easily enough - but yeah, why not post your software? It's always another option and if it allows people like Harry and Troy to see some of your camera output I am sure it will be a great incentive to them. I wish I had kept all my old stuff on tape but - oh well...

Yes i for one am interested

Yes also to the strive for excellence seeing what grahams mechanical cameras can do .

For me perhaps one day but seeing all this stuff is a big incentive.

The forum is the modern day tape diary what better place to keep your accomplishments.

[AncientBrit]

Okay Harry,

I'll dig out some of the old software and associated picture files and post them.

I might even take a few pix of the disc camera...

Cheers,

Graham

[Harry Dalek]

Okay Harry,

I'll dig out some of the old software and associated picture files and post them.

I might even take a few pix of the disc camera...

Cheers,

Graham

Thanks Graham i think a lot of us will be interested not just me so if anything you have an Australian audience !

What ever you can do will be great i am sure Troy will be interested since he is working on one at the moment .
On your electronic design i have said once before for me i need something not so much to copy but an example at least to work with ..makes me wonder what other hidden NBTV treasures you have made over the years .

[aussie_bloke]

G'day all.

Thanks for all shared knowledge in response to my camera/monitor project, I am learning a great deal as I read!

With outdoor cameras, whilst my camera is not exactly battery operated portable, I could stick a hood over it all and work it outside the garage.

In the meantime I am still yet to get the image pickup sensitive enough to even film natural sunlight which is what this post is going to be mainly about.

I have since added a cardboard hood to the condenser lens/phototransistor sensor unit to keep stray light out and have added some black tape to cover the exposed area of the less domed condenser lens so only light from the holes in the disc hit the phototransistor. And of course I have been doing more testing and have added another 4 stages of amplification to the head amp in hope to make the image pickup more sensitive to lower light.

Unfortunately with image sensitivity I have hit a stumbling block.

Initially the head amp consisted of 3 stages of amplification, first stage was this circuit http://www.hawestv.com/mtv_cam/mtv_cama ... ansAmp.png (transistor Q2 is BC548), then I added 2 snaptogether stages PNP then NPN using these circuits http://www.hawestv.com/mtv_cam/mtv_camart/addAnAmp.png (transistors of choice were BC547s and BC557s). Now with 3 stage amplification, it was sensitive enough to pick up bright light from my torch and from the 150W incandescent lamp shining directly into the lens and reflected metallic objects like the shaft of a screwdriver.

But I could not get the phototransistor to work with just shining the 150W lamps onto general objects, still not bright enough for it. So I decided to add another 4 more stages of amplification once again using those snaptogether stages http://www.hawestv.com/mtv_cam/mtv_camart/addAnAmp.png and using BC547 and BC557 transistors and tested it out again. Unfortunately that did not increase the sensitivity of the image pickup from the phototransistor and of course with this amount of amplification I was getting unwanted signal noise going to the LED driver circuit causing it to intermittently flash, so I rolled the amplification back a couple of stages which fixed that issue, so my head amp now consists of 5 stages. But with 5 stages the image pickup is no better in sensitivity than that of 3 stage amplification.

So it now makes me wonder, are phototransistors poor image sensors or do they vary greatly in sensitivity depending on what one you buy? The phototransistor I am using is a
L-51P3C bought from Jaycar http://www.jaycar.com.au/productView.asp?ID=ZD1950 (see attached pdf for its details). When I did some basic testing with this phototransistor I found that for the oscilloscope to even display a signal on it I literally had to shine the torch directly over it and it seems to have a hard on almost "digital" response to light as in light or no light instead of the desired "analog" response to light with varying levels to create a picture of varying light levels.

Also I wonder a solar cell from a calculator or a general purpose light dependent resistor have better light sensitivity and could they easily be adapted to the head amp without modification? And hearing that dome sensors seem to be the image pickup sensor of choice for NBTV camera makers, what are typical dome sensor model numbers and where does one buy a dome sensor from as I don't seem to see them commercially available on Jaycar nor even on Ebay? On both Jaycar and Ebay search the results are security cameras housed in dark tinted domes.

[gary]

Troy, in my best Mythbusters accent - "well there's your problem" - the photosensor is infrared! All the visible light spectrum will be severely attenuated.

LDRs and solar cells have both been used in the past (as have phototransistors but not IR versions) with varying degrees of success. In particular I have had good success with solar cells but that was back in the era of monolithic solar cells - these days they tend to be broken up into cells which cause problems if your image flass over 2 or more adjacent cells - an area for experiment however. The other problems one can have with them is response time and visible spectrum limitations.

Troy the dome cells can be bought from the club shop through Vic Brown - you would need to be a member however - something I would very much recommend anyway. As far as I am aware there is no other known source.

Troy, in the meantime, I have a spare dome cell I can send to you - PM me with your address and I'll post it to you.

PS: With your current rig you might want to try and replicate Baird's "night vision" experiments - break out your old radiator heater as a source! - LOL (or use an IR LED light source).

BTW - I am not sure I have already mentioned this but you would also be well advised to record the output of your head amp via a sound card on your PC/laptop - that way it is easier for you and others here to analyse the quality of the signal.

[Harry Dalek]

Well it could well be the circuit i never tried the one you used but looking at it ......all the ones i have tried use an amount of feed back to the first transistor they are all fixed in the circuits but i put a trimmer pot in as i found for me it got the circuit to work ...also have a variable power supply via the old lm317...that helped to gave a bit of adjustment to the thing .

I was using a torch via the nipkow and that worked to a point ..i had better luck with the solar cells...but i think once you get the dome sensor off Gary you might have better luck..Gary could just about chuck it off that mountain might end up in your back yard !
Use the club dome sensor circuit more than likely the best ..

[aussie_bloke]

Troy, in my best Mythbusters accent - "well there's your problem" - the photosensor is infrared! All the visible light spectrum will be severely attenuated.

Haha silly me I only just read that today when I checked the Jaycar online catalog, I had a feeling that might be the case after reading, I don't recall thinking about the infrared factor when buying it.

LDRs and solar cells have both been used in the past (as have phototransistors but not IR versions) with varying degrees of success. In particular I have had good success with solar cells but that was back in the era of monolithic solar cells - these days they tend to be broken up into cells which cause problems if your image flass over 2 or more adjacent cells - an area for experiment however. The other problems one can have with them is response time and visible spectrum limitations.

Cool, I have recently got a solar cell from a trashed calculator so that might do, also have some LDRs too. Will have to experiment with them.

Troy the dome cells can be bought from the club shop through Vic Brown - you would need to be a member however - something I would very much recommend anyway. As far as I am aware there is no other known source.

I have just paid for membership via Paypal a few weeks ago so I am a member now as far as I know unless I have to wait for some code number to be mailed to me like was the case with http://www.radiomuseum.org/ .

Troy, in the meantime, I have a spare dome cell I can send to you - PM me with your address and I'll post it to you.

Thank you I will PM you my address shortly, how much do you want for it?

PS: With your current rig you might want to try and replicate Baird's "night vision" experiments - break out your old radiator heater as a source! - LOL (or use an IR LED light source).

That's a really cool idea, I have some infrared LEDs so I will bunch them together and shine them down the lens and see if the phototransistor will respond and display a picture!

BTW - I am not sure I have already mentioned this but you would also be well advised to record the output of your head amp via a sound card on your PC/laptop - that way it is easier for you and others here to analyse the quality of the signal.

I have just tried that last night when testing my camera with your NBTV Big Picture program, here is a sample when I position the 150W lamp in front of the lens and moving it a bit here and there. I think the disc might not be spinning fast enough as I'm not getting intelligible display on your program nor am able to sync it for an intelligible display of a light.

[gary]

I have just paid for membership via Paypal a few weeks ago so I am a member now as far as I know unless I have to wait for some code number to be mailed to me like was the case with http://www.radiomuseum.org/ .

Then you should be able to purchase now. Contact Vic Brown and place an order (I'll still send you one because it will be quicker than waiting for shipment to the UK but it's a good idea to have spares - isn't it Harry? ).

BTW the first thing you should buy is the past newsletters CD - there is an absolute wealth of information there that will answer any question you have about NBTV.

Thank you I will PM you my address shortly, how much do you want for it?

Oh that's OK - just keep posting your build log and videos - as most of my NBTV stuff is still packed away since my move a year ago I am living vicariously via this forum (at least in an NBTV sense )

I have just tried that last night when testing my camera with your NBTV Big Picture program, here is a sample when I position the 150W lamp in front of the lens and moving it a bit here and there. I think the disc might not be spinning fast enough as I'm not getting intelligible display on your program nor am able to sync it for an intelligible display of a light.

Ok, I have attached the best picture I could extract out of that wave file Troy, you do seem to be able to make out a light moving around.

There are a couple of things you can do to make it much easier to analyse the picture:

1) Record in 48kHz/32 kHz/24kHz = anything that is an integer multiple of the line rate - in fact ANYTHING but 44.1kHz - that's a nightmare.

2) If you can, arrange to have a pulse signal on the right channel of a stereo recording - this is easily set up by using a IR led/phototransistor pair (you have the detector! )that picks up a reflective "patch" stuck on the disk thus producing a pulse that is related to the speed of the disk. This can also be a hole in the disk with the led placed one side and the photodetector on the other.

There is an encoder pdf on this forum site which is ideal for this purpose and can also be used to generate the sync pulses that can be mixed into the video signal if you ever want to go down that path - just one thing, if you use the encoder print it out on a laser printer NOT an inkjet as Harry has discovered, and Graham confirmed, it produces black segments that is opaque to IR and therefore reflects it back into the sensor.

But even one pulse per revolution is better than nothing.

Using this information it is easy for me to do timebase correction.

[aussie_bloke]

G'day Gary. Thanks for extracting that test recording into something recognizable, I was shining my 150W lamp onto the scan area and the lamp has a distinct hexagonal honeycomb pattern on the glass and the pattern became somewhat slightly recognizable particularly near the end of the video so I am happy about that alone!

Now I have made me an infrared torch using 4 infrared LEDs and have tested it out with my camera and to my joy, I got a spot of light on the monitor!!! So my camera does work in the infrared spectrum. I had to of course get the infrared torch on focus and aimed dead straight through the lens to the disc for the phototransistor to pick it up which was difficult considering the obvious we can't see infrared . So I pointed my sensitive Panasonic RX3 camcorder to the camera's monitor and looked at the TV for guidance and at the same time I have made a telerecording of my camera in action displaying infrared light and light from my torch and 150W lamp.

I will post pics and video in my next post so stay tuned!

[gary]

hmmm, it would be interesting if you lit, say, a match or candle in front of it to see how sensitive it really is. Both should be good sources of IR...

Also, if you made a test card A4 size with big chunky letters NBTV or something from a *laser* printer and shone your IR source on it - on the basis that the reflective ir sensor that I use works with a printed out encoder, you may see the letters - something to play around with anyway.

[Harry Dalek]

Good results really better than my first gos .
The photo trany you got from jaycar is something i did buy but didn't use because of it being IR..
If you still want to try photo trany idea on the cheap i am not sure they are any different but watching secret life of machines the fax ep they cut the top of a can transistor then its light sensitive ..might have to stick a clear plastic over it stop oxidation over time ....they used a power transistor you can see the you tube video of it.

Yes Gary is right good to have spares of these things the dome sensor can be damaged applying voltage to it by mistake as i found out

And listen to our Aussie Sensei and you will prevail

[Harry Dalek]

Hi troy not sure how sensitive to normal light this is yet but heres a cheap way to make a photo trany also not sure how long it would last without a cover but interesting .

http://rimstar.org/renewnrg/make_solar_ ... 2n3055.htm

[M3DVQ]

also not sure how long it would last without a cover

Not long at all...

That's what's so good about using the junctions of an EPROM. The quartz window keeps all that nasty oxygen out