Page 9 of 11

Re: Slow Scan Television Experiment

PostPosted: Thu Dec 11, 2014 11:22 pm
by Harry Dalek
Heres what i think is going on with a little research as far as the light sensor gos ,on Alan shorts DIY SSTV and NBTV cameras late last century ,he used the ocp 71
which i always wondered how it could be used side on as well as normal top i think he used it side on
He Swapped to the TiL78 looks like the type we can get these days with out the lens it sounds like it works better as a point of light detector .
I will out of interest try this on my sensor worth a shot as i need to improve the detail it sees .

Re: Slow Scan Television Experiment

PostPosted: Fri Dec 12, 2014 1:12 pm
by Harry Dalek
Well i gave a try to grinding down the lens on one of my light sensors but it wasn't really necessarily as my light sensor has 3 parts the light sensor is in a type of epoxy resin very close to the light sensor non lens idea, it has a light proof casing around that then the lens on that which i found out is not resin but glass .

The glass could be ground down but as it heated the case which it became unstuck from the rest of the light sensor main casing so didn't have to polish any thing yippy :mrgreen: .

I gave this a try and it was worse as in focusing :? .

I then tried some thing i had done before over the light sensors lens but it didn't work well that was using the lens and casing from a broken laser with the laser diode removed you just have a lens in a focusable if thats a word casing .

It drops the light level the sensor picks up and i have just placed it over what remains of the light sensor with some blu tack to hold it and stop light around it but it did the trick on focus .

Since it works from a quick try perhaps it can be improved on a little ,as you see in the last photo the superimposed faces of the block can be made out much easier.
Well now i know what the problem was here, nothing like trying out and idea information is good but to see for your self what works and what doesn't .

Re: Slow Scan Television Experiment

PostPosted: Sat Dec 13, 2014 7:29 pm
by Harry Dalek
Went back to testing light sensors this afternoon looking again at the Old MEL 12 and comparing it to the 3DU5C with out the lens it clicked that the MEL 12 does not have a lens either just an epoxy covering.

So i gave the MEL12 one last shot with a laser lens as with the other sensor it did work but i found it less sensitive than 3DU5C ,but any case it showed me another reason why things were not working out on the NBTV camera so that was good .

I also gave a reflection a test run it was very dark but i managed to get an image via a mirror still not practical but all that was in natural room light.

Now the focusing problem has been worked out the next problem is to see if i can image some thing that isn't a two tone image.

Re: Slow Scan Television Experiment

PostPosted: Sun Dec 14, 2014 7:34 pm
by Harry Dalek
Back to it again today and looked into adjusting the speed of the stepper motor and see if i could improve picture detail this way as i know my current speed is wrong but i am limited to the top speed of this stepper motor unless i change the pulley size on the stepper.

So any way my experiments today were to improve picture detail at current speed 8 sec then adjust to 12 sec Scan and see what happens there .

At 8 sec setting i see 2 faces of the block scan the other 2 are really superimposed over other 2 faces and so i have left the 3 faces blank i think i need to get twice the speed to get one large picture or as in a cylinder just use all the area at current speed or use all 4 faces half speed my head hurts :roll: .

Well trying 8 sec i got some better detail results ,i then swapped to 12 seconds and adjusted the speed it shows all 4 faces of the block but picture detail is good enough on one of the faces to show some thing other than a 2 tone picture.

The experiments again were all done in room light which does fluctuate a bit with time of day and cloud cover so i had to adjust my head amp a bit for near on each photo , i get it right others little dark some times syncs off .

All still hand scanned down the photo as it rotates,getting the quality a little better each try as i learn .

Counting the non interlaced lines in the 8 second pictures i am getting 60 lines so half way there :wink:

Thinking about the mirrors and why its so hard to do this in reverse if i stick with it by the time i end this topic i hope to know ,at the moment i am thinking either my mirrors are to wide or i need a mask ,i could be wrong i will test away till i run out of ideas getting this thing running at the right speed for maximum quality for a picture as i am trying to do now will help .

Re: Slow Scan Television Experiment

PostPosted: Mon Dec 15, 2014 12:19 am
by Steve Anderson
Harry, with each posting things are getting better, keep at it! It's a shame about the lack of display software for the Copthorne McDonald standard though.

But what you have done here surely can eventually be carried over to NBTV.

Steve A.

Re: Slow Scan Television Experiment

PostPosted: Mon Dec 15, 2014 1:32 am
by Steve Anderson
I'm still working on the small Euro-P7 SSTV display device (the DP7-5), but in looking through my old notes I found reference to (I1LCF) Prof. Franko Fanti's SSTV P7 monitor. It used a 3FP7 CRT which I also have an example of. See below.

The professors' design appears to be all tube from the scanned photo below. It's from Dave Ingrams' "The Complete Handbook of Slow-Scan TV", Tab Books, April 1977, ISBN 0-8306-6859-4. It's pure Copthorne McDonald SSTV, not a mention of VIS codes at all.

Compared to the DP7-5 I also have it's about twice the length and as an average requires 3kV to operate it. It has the same screen size, the same spot size and the deflection factors aren't far apart. But it was designed for high altitude radar applications. Hence the wide spacing on the pins to prevent arc-over at low atmospheric pressures. Being airborne it is ruggedised which resulted in the almost 2" diameter neck. And with some 3kV probably much brighter than a DG7-5 to cope with above cloud sunlight, both direct and reflected off clouds below. (Believe me, that is bright!).

But in the simple application I have in mind I think the smaller tube should do fine, I had almost forgotten I had this 3FP7.

I have unfortunately been unable to find any further information on the Professors' design.

Steve A.

Re: Slow Scan Television Experiment

PostPosted: Mon Dec 15, 2014 3:13 pm
by Harry Dalek
Steve Anderson wrote:Harry, with each posting things are getting better, keep at it! It's a shame about the lack of display software for the Copthorne McDonald standard though.


Thanks Steve ,i think i can do better , i know what is happening ,focus is no longer a problem and its just a matter of getting the right speed to double the quality of the 8 second pictures ,i will try and cheat and get a bigger pulley wheel .

Once i get that right i will try reflection scanning again .

Yes very little i think JVFax the dos version does do 8 sec but i am not sure if its the original system or a robot ,its the one i used way back when .

It is a shame no one has kept the original system as is when doing SSTV software ,it like no one will ever use that again so why bother .


But what you have done here surely can eventually be carried over to NBTV.

Steve A.


Yes the head amp and sensor optics should work for the other camera its fine for room light, the mirror scanning needs to be worked out as i am finding out its not just a reverse of a laser raster idea ,its hard even to just view a scan as i am doing off a stationary mirror ,i could just do a FSS but i heart is in working out a direct light camera with mirrors ,it must be possible others have done it.

Re: Slow Scan Television Experiment

PostPosted: Mon Dec 15, 2014 4:55 pm
by Harry Dalek
Hi Steve
I don't have or have found any thing on 3fp7 apart from this circuit see mention on last page.
Theres a little in the below booklet on other p7 types but your going to have to google translate hungarian to english.
I too have the Dave Ingrams book :wink:
i have this valve design i don't remember if i posted it before ,may come in useful think i got a page scan wrong number whoops.

Re: Slow Scan Television Experiment

PostPosted: Tue Dec 16, 2014 8:39 pm
by Harry Dalek
I tried a larger pulley wheel on the stepper and it knocked up the speed of my block rotation and it is now giving me similar results to the 12 sec go the other day ,i sort of ran out of light today for more testing but got these results slightly larger to the 12 sec test .

i suppose the only way to improve image quality now is have only the one picture on the screen than the 4 sides of the block spread out across the frame.

The easy solution is one full photo around a drum at this speed for a rotating image scan .

Camera wise i am back to mirrors , some thing that comes to mind that a rotating image such as this photo on a block and a image reflecting off a rotating mirror there are differences ,the mirror does see more area before and beyond the image which could just mean i need a larger image to scan or mirror dimensions need to be smaller , in any case it seems logical that to get the same result as the block image turns the mirror must reflect identically as it turns the same way .
Only thing i have found when trying to scan a block scan image the reflection off a stationary mirror is very dark light loss, which i suppose there must be some loss via the mirrors glass ? but loss there is ,i have been using natural light which probably shows this loss up more than using a lamp would but that is yet to be tested .

Re: Slow Scan Television Experiment

PostPosted: Tue Dec 16, 2014 9:18 pm
by Steve Anderson
I'll be starting a new thread on my SSTV demodulator/monitor build as I don't wish to clog up Harry's thread any more here...though he says he doesn't mind...

There may be a few spin-offs that could be adapted to NBTV as a result.

I'll call it "Retro SSTV anyone?"

Steve A.

Re: Slow Scan Television Experiment

PostPosted: Wed Dec 17, 2014 5:33 pm
by Harry Dalek
Steve Anderson wrote:I'll be starting a new thread on my SSTV demodulator/monitor build as I don't wish to clog up Harry's thread any more here...though he says he doesn't mind...

There may be a few spin-offs that could be adapted to NBTV as a result.

I'll call it "Retro SSTV anyone?"

Steve A.


Thats no problem Steve mostly just us two any way but i will be watching how the monitor gos ,more SSTV on the off topic is good one day we might have a header SSTV !

I am going to now try different ideas on reflection the only way to do this is one at a time trial and error ,might hit it first go but we will see or not ! .

First go will be the narrowing of the mirror for the horizontal ,my reasoning is the wide mirror is perhaps or i know it is seeing to wide an area i want to narrow the seeing angle so light from say a picture as in the photo is all thats scanned ,keeping the hight for the vertical i will see if this narrow mirror idea works or not ,its a quick prototype so i used some little mirrors i had self sticking and just the right size for the smallest lego blocks i have found ,i think i would still like about half this width but i will see how this gos if its worth making a better one .

Re: Slow Scan Television Experiment

PostPosted: Wed Dec 17, 2014 7:31 pm
by Harry Dalek
I have tested the new mirror idea and it sort of works ,i am back to rudimentary out of focus pictures Yippy ! but the narrow mirror for horizontal does what i was hoping for scan a picture without over shooting the scan area .

These tests were at 12 sec Slow scan as i was starting to get results and didn't want muck around with settings .

This is the first time i have used 4 mirrors first time and got viewable results and i could see the reflected image and say yes thats what i am trying to film and again all in room light only.

I once again tried this felix picture gave me the best results bad as they are .

You may notice one or two of the pictures where the mirror join ,theres also a slight wobble causing those lines ,i am surprised i got any thing .

Its also another answer for me on the nbtv camera on the mirror side of things .

Another day to sit back and think ,In a way its sort of a camera now a very bad one but its a start.

Re: Slow Scan Television Experiment

PostPosted: Thu Dec 18, 2014 7:11 pm
by Harry Dalek
Since i had a bit of encouragement with the experiments and once again its getting to the point where i need an upgrade to house the bits to see if i can improve results so i took a brake and am converting a case to house everything for further experiments its still a mess but worked out the hard part today.

I am going to copy in a way with my own twist Allan shorts camera in the way of large lens ok for close ups :wink: and direct to sensor frame scanning do away with a second mirror for now .

Looking at hes first design he didn't like this system as he had problems with full framing he went more with the design here from CQ tv magazine, I like the original design the vertical movement can be worked out ,its easier having its own motor to do its movement.when i was working on the glow drum sstv i worked out a system using an old dvd laser drive movement apart from the time to reset it would more than likely be fine for this but i am still thinking about best way to do the vertical moving of the sensor .

On to where i got the 2 sided mirror .....

I found 2 laser printers on the side of the road as you do ,one was a colour tower laser printer the other a cheapy , the colour one i should say was ! amazing as i scrapped it soon as i got home ...4 lasers and 2 polygon motor mirrors and 4 mirrors just the right side i want ...2 sided will do for a test. :wink:

Re: Slow Scan Television Experiment

PostPosted: Fri Dec 19, 2014 2:14 pm
by Harry Dalek
Today i have been working on a vertical movement idea for moving the light sensor and or head amp .

To do the Alan Short pivot idea the sensor would move in an arc with my testing i found it needs to be moved at the same distance or you have focus problems either end ,so i just wanted an up down movement with some sort of reset .

Using an old DVD laser movement 2 gears from the scrapped laser printer a spring and a stepper motor i should be able to move the sensor down in a vertical scan and at the end of the scan the point where the gear cogs have been removed the spring pulls the sensor movement back to the start position .

Only problem i have not checked if the stepper motor will cooperate and run in the right direction the spring can be reversed to the other side and i will have to turn it around .

Its large so i will see how it mounts i think it can still be out of the viewing angle from a placement test.

UPDATE same day same work well the stepper gos the right way so thats good ,worked out and placed the boards , every thing is really shifted over to this case now only thing it a bit noisy in this steel case .

Re: Slow Scan Television Experiment

PostPosted: Fri Dec 19, 2014 6:38 pm
by AncientBrit
That's neat Harry.

I like the mechanical flyback, very simple but effective for slow scan.

I doubt if it would scale from slow scan to NBTV frequencies though, pity.

Kind regards,

Graham