Narrow-Bandwidth Television Association

AncientBrit wrote: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

Hi Graham

Yes i just made this one for SSTV where the time it takes to fly back is not important ,i am starting to want to free my hands from hand scanning so see how this gos .
I just mounted it i hope all the motor rattling doesn't cause another problem doesn't seem to bad when bolted down .
If i didn't have so much trouble with mirrors and optics i would stick with another mirror but one rotating mirror for a camera is hard enough till i work it out i am not pushing my luck .
I think it would brake at higher speeds or jam up .
Heres a look it mounted and running.

Well since i last posted the below animation is already out of date , the basic idea i wanted to do scanning wise works ruffly .

I did find now some noise on the video this is a metal case that might be the cause any case another problem to work out .

What i noticed and wasn't happy with was the vertical movement causing a little vibration on the light sensor this could be because the movement frame was only bolted down on one side,its now turned around and bolted better to the case .

I also wanted to test different mirror sizes and so something needs to be adjustable here .

Also since i have a lens for the light now i just want the light via that so some sort of cover for now till the lid gos on should help.

Since i swapped it over to this case i need to test every thing from the start such as block scanning just to make sure every thing is working so it needs to be able to do that as well with out hand scanning this time .

I will also want to test masks just to see if a larger mirror could still work as it seems to on Alan Shorts camera which has me beat so far as i see no masks on hes designs, but i do see it in cased but he says it does still work lid off , i do find the surrounding light does effect the sensor for me .

I was testing the optics where the focus would be ,i had to add another lens and the focus is now where the vertical movement is ,i think i can even use the hole where the old dvd laser used to be so i will give that a go .
I tried to take some pic's to show where the focus is by projecting my front window to some paper where the movement area is ,so its lens to thin mirror to screen where the sensor will be ..
Problem or just live with it is the image is now upside down
So i hope i have the optics right as far as working with the sensor ,i am not to bothered if the image is flipped ,i suppose reversing the vertical scan movement would get it the right way around scanning wise ,not sure about if the image would be need to be reversed have not checked that yet any case thats some thing to worry about later need to get it working as is .

Thought i would post as i see there a few people viewing
At the moment i have redesigned the vertical movement with smaller gears and a different stepper motor to get a better speed range .
I am also going to change the 555 timer that controls the vertical movement with a 4060 the 555 is to volt sensitive controlling the line mirror stepper effects the other stepper motor because that 555 changes frequency with a slight voltage change its not getting its own power supply
I have not had much time over x mas holidays to work much on it but will do and keep trying .
I have had a think about the results i have got so far at the different speeds,full screen pictures at 30 line half at 60 and half that at 120 line, its like the image being scanned needs to be larger with more scanning lines this was a result i was not expecting .
You do get a increase in image size with the sensor close to the image or mirror and decrease with small distance away but thats a small factor to the line number change.
I will be back at it in a day or 2 .

One last change for the last few hours of the year .

I am going to try the head amp sensor this way seem to get less vibration ,i can only see this on the fly back where i don't care .

its connected to the vertical movement by the L bracket to a magnet so its movable back and forward when needed .

The optics part is still hard part for me so see how this gos

Harry, I don't care if this apparatus "works" or not - it is a work of technological art if I have ever seen one - it needs to go on display somewhere! Well done!

gary wrote:Harry, I don't care if this apparatus "works" or not - it is a work of technological art if I have ever seen one - it needs to go on display somewhere! Well done!

Happy new year Gary

I am pretty sure it will at least block scan only difference being no hand needed for the vertical this time,i haven't tested for noise the last motor was giving me trouble with rf . i have been naughty and not encased the head amp so might be one thing i can try if i have trouble again.

I am pleased you like the look of it there is some thing magical about mixing electronics with motors gear pulleys but i would like it to work at the last stage and be a direct light camera ,i will try my best a while back i had bit of success with out a lens poor results as it was .

I started this as a test to see why i was getting no further on the nbtv one its answered a few questions i had ,you were sure right when you told me once cameras a hard !

Just going to start with tests i know that worked before i cased it that helps me on the optical side ...the focusing with a lens and the reflecting mirror i will have to see keep shifting things around changing mirror sizes and such till it works or i run out of ideas .

I call this post look no hands
I had my gadget going good enough to do a few tests on the vertical scan mechanism .
I did learn some thing today that the size of the image distance and such determine the speed needed not so much 8 second scan you sort of have match the image size so it has to run a bit faster or the image is to long ,lucky i put in a speed adjustment on my panel for this movement.

I screen grabbed a a few tries the light levels were dreadful today fluctuating very quickly .

This test was just the head amp on the vertical movement scanning my early block drawing just using light via the front lens ,having to start all over again is a pain .

You can see the scan start and end is a bit short i will look into that ,least the vibration level is low enough not to cause problems.

Getting a bit of interference in the video had the Tv on so it might just be that , once again need to improve results .

Looking into some problems with the design looking at the line scan ,problem being only be able to show quarter of the line scan at 120 lines when using a 4 sided surface i am using .

To explain....
If i do a block scan and i can wrap a long image around the 4 sides of the lego block i would get to use full length of the line scan ..same as using a round drum and wrapping an image around the drum like a fax machine ,one rotation one line.

At the moment quarter of a rotation is the image other 3 quarters of the line wasted .

Now the problem with a mirror i can't do this, i can adjust the speed but i can only increase the image size by decreasing the number of lines this is a major problem .

This is a point in the project identical similar to my mechanical sstv monitor where i came to the conclusion i had got the scanning wrong as much as i didn't like changing the design it did eventually point me to the right design and this what i had to decide today .

So next time i post a picture this part of it will be very different ,the line mirror will be the frame mirror back using the large 4 sided mirror which may even be to small for my updated idea but i will try it for now as i could still use it and test block scanning before trying this mirror eventually.

As much as i hate it have to disconnect the geared vertical movement as it will no longer be needed least i tried it out and it worked .

I will be using one of the few laser scanner mirrors i have, bit of luck i can use one as is just with a clock a power supply both there already so nothing there to make apart from soldering some wires.

In a way similar to the start idea but also opposite when looking at my block scanning ,this time the block will rotate slowly stationary light sensor and fast rotating line mirror, so a sort of opposite like scan idea of the glow drum monitor .

My decision to try this isn't taken lightly, the optics work out better and i would rather know than never know .

I have no idea how well this next attempt will go but it is some thing i wanted to test and try .

If any thing i now know how to focus to my lens, i have focus on the light sensor, i know my electronics work and the top thing i know is doing this with mirrors isn't easy .

The design idea this time is with the large 4 sided mirror is now the frame mirror the vertical so size here is better , a line mirror in this design would make it so much easier in getting an image to use all the line scan just a matter of getting the distance right i would rather a smaller polygon or small square mirror so i can get the distance to not over shoot either side of the frame mirror sides........ the length of each polygon mirror is better smaller as you do have to get the mirrors close as it is to each other my main mistake in the first design ,once again optics !

What i have been working on today is having an adjustable focus or best i can do with what i have ,so the whole mirror assembly is on a rail ,i can focus from about 30 cm to any thing past that i can see the trees across the road fine out my window projecting on that white block .

I was at first wanting to use the square laser mirror motor i posted a picture of a while back from one the scrapped laser printers but it was dead ,one of the 2 polygon mirrors tested worked fine ,they just need 12 volts and a high frequency clock pulse and spin they do .

I will be adjusting this position of this when i start to mount the light sensor head amp .

I can still do a block scan with this so this good for a image testing.

Oh well fingers crossed see how it gos this year.

Update report on the design change .

Just been working out how to use the polygon mirror motor with the light sensor on its own for now i am happy with mirror results to sensor .

Theres a problem with motor speed it drifts much more than a stepper so i think this one would need mechanical feed back from the mirror so really no better than if i used a Normal DC motor ,with the extra sides 6 to the 4 sided one the speed needed is slower so i expect this is also the problem...running it to slow

I will try a spare laser mirror on a stepper tomorrow that might be better to test this .

I am back to experimenting .
I have taken a step back and removed the second axis scanning for now and remounted the original line mirror .
Reason looking into a mask, but i wasn't sure where the mask had to mounted and testing today it looks me it has to be at the line mirrors reflection focus any where else its like a pin hole camera and just projects the image at a lower light level by the looks of it ,just wanting a line of light it this seems right .
So what i will do next time is make a mountable mask and have the light sensor behind it .
I think i have not taken into account the correct focus from the second axis scanning mirror to the sensor .
I will try again this year !

I have to report the mask idea for me is another dead end ,perhaps fine for a solar cell large surface area detector but did nothing more than drop the light level for my sensor .

Having tried many ways to scan see my crude drawing....first focus to mirror surface and scan close to this area ,second idea reflect off the mirror and the focus area is away from the mirror larger image scan this area,third idea focus image to a screen and or mirror this image is reflected to the rotating mirror and scanned there .

I have been using a slide projector using its reflected light off a screen and testing reflected images but really no luck so far ,the difference to a rotating image to a stationary image scanning is like chalk and cheese ,even at the same scanning distance the results are not comparable.

Just a quick retest of the focus and image results of a rotating image theres nothing wrong here .

I still have a few more ideas to try let you know how they go.

To reiterate on the last posting and since the experiments today are on one of the scanning ideas i will show here best i can what the problem is on a mirror scanning.
What i am seeing with the eye is hard to video the focus area is a bit easier to take a photo with this camera anyway,i will see if my Nikon camera works better next time i try.

I have a projector with a drawn slide of that cross thats projected to a white paper screen then picked up by my large lens and focused to the mirror area .

As the mirror rotates at the line frequency unlike a picture wrapped around a drum or block type scan the mirror does 2 things that i can see it scans the image which i can sort of see flashing as the mirror rotates but you also see the image of the cross as is as shown in the images just as a mirror reflection at the same time.

I think this means the light sensor is seeing both and i suspect the mirror reflection i can see is cancelling the scan part out .

This is an effect you don't have with a picture rotating on a drum or block .

Just an effect i have noticed and a bit of head scratching .

I am plodding on with the mirror camera idea and still head scratching results with mirrors.
I masked off the other 3 faces of the block ,i was seeing why when using a mirror the scan does not use all of the area of that face of the block with the mirror left as is , you just get that thin light strip .i drew on the other faces to see how the focus was ,bad lighting today sorry.

At best with line speed dropped showing 2 mirror faces and lighting ok pointing at my window i get this last picture result not sure i had the focus correct there ,its a bit wider but i am not sure why it matters that the image is not using up all of the area of the mirror face a drawing to mirror result shows one works one doesn't as in face of the block usage,i would of expected even if i couldn't get an image just a full square of light.

I think i would get this at the lowest quality picture 32 lines ,the last picture is at 64 lines looks to be half a face usage .