A new 90-line Mirror Screw

[Panrock]

I have now scanned the complete final-version Manual as .pdf's and have added to it a .png file of Karen's NBTVA Newsletter article, for convenience. I shall now try to send the whole caboodle in an email to Steve A.

Steve O

[Steve Anderson]

It looks like all have arrived OK, I'll choose a later, quieter time to download them. The 'net get's throttled here during the business day, by 8pm I usually get my paid-for bandwidth...banana republics, I love 'em. But TV is unaffected as it carries ads, can't have those interrupted can we? All our TV, radio, pnone (landline, mobile Wi-Fi and 'net) is carried on a single 10G fibre in/out of the house. Bottlenecking is common and maxes out the NIC frequently. 100G NICs are still too expensive, plus my PC would then become the bottleneck, IF it supports them, most unlikely, and it's only two years old.

Steve A.

When I can view then I will send some feedback, whatevr form that may take.

The next day....I'll try again later today, I simply forgot to download them yesterday evening....

[Steve Anderson]

That didn't work out. Simply 'cos of thunderstorms interrupting the power supply. We've got a 7kVA generator here, but with the general houshold load there's little left for luxuries like TV, Internet and so on. (A lot of fridges, freezers and air-con.). More rumblings now, probably much the same again this evening/tonight....It's already p***ing with rain at 4:17PM...(my clock is a bit out)....

Steve A.

Just checked. this message uploaded OK....it is sometimes a bit hit-and-miss....

[Panrock]

A routine update. The order for the 90 slats, in aluminium with polished edges, was placed 10 days ago. It will soon be time to chase it.

Steve O

[Panrock]

I needn't have worried. The slats were already ready.

Here is the pile of them, roughly assembled into a block. The central holes for the shaft are nice and accurate. However it looks like further flatting and polishing of the reflective surfaces will be needed.

Steve O

[Panrock]

The basic layout and structure of the future televisor begin to emerge.

Everything is evolving out of the bits and pieces I have to hand. There is little 'planning' involved. I have now obtained a three-stage kit to do the slat polishing.

I am already finding it a satisfying project - it keeps me focused on building something. My aim is to have it finished and working by the next NBTVA Convention in Spring '25.

Steve O

[Panrock]

Continuing the monologue...

I have now finished the polishing of the slat edges. These are still clamped together as a block. I found the use of 0000 wire wool and metal polish helped the process along.

Today I have sprayed the reverse side and end face of the block with etch primer, then with matt black. Later, when the block is split up into its 90 component slats, this means reflections from the corresponding slat edges shouldn't interfere with the picture. We want only one (long) edge on each slat to reflect the light and be 'active'.

I have also prepared the shaft on which the screw will be assembled and made a base panel, and a panel housing the upper ball race. Tomorrow I expect to make ready to get started on assembling the actual mirror screw. First, I'm going to have to make an accurate wedge-shaped positioning piece. I will use this to set up each slat angle to 360/90 or 4 degrees, as I build up the screw, slat by slat. I haven't tried using this technique before. More about this in due course.

Steve O

PS. Steve A, I hope you're all right. Did you get those files in the end?

[Dmitrij]

Hello. The method of adjusting the screw using a template is described in the Radio Front magazine, issue 4, 1937. But it is considered preliminary even for 30 plates. 2 lines drawn on a long-term machine with a difference of 4 degrees will give a more accurate result.

[Panrock]

Hi Dmitrij,

Well, all good advice, but I want to first try this method to see what can be achieved. The template will engage directly, matching the end of each slat.

Whatever the method used, in practice the final timing correction at pixel resolution for a high line count will have to be done by electronic means. Peter Yanczer's uncorrected 120-line screw gave a fuzzy picture. I wonder how good the TeKaDe 180-line screw was?

Another approach has been to make each slat angle individually fine-adjustable. I believe Albert/Viewmaster has done this. He may read this.

I noticed with the 120-line screw that its slat-angle evenness and mechanical balance improved with use. I never understood the mechanism behind this...

Steve O

[Steve Anderson]

I guess in Karens' version of electronic 'slat trimming' might have been done by some form of elelectronic delay line, each individual slat having its own delay parameter. I've not yet thought this through fully, but may only require a delay line the equivalent of a couple of lines. maybe/probably less. times 3 for RGB.

If need be I'll wake up the grey matter and 'have a go'...but a maybe simpler arrangement, Karens' version could be 'adjusted'.

Steve A.

[Viewmaster]

Hi Dmitrij,
Another approach has been to make each slat angle individually fine-adjustable. I believe Albert/Viewmaster has done this. He may read this.
Steve O

Yes, but my screw was only 32 lines.
I made 32 tiny U clamps to lock on each slat. Each clamp had an adjustment
screw #hat 'pushed' on the next slat.

It was a bit hairy to run at speed as any loose clamp
might decide to leave its slat and explore the room

It would be difficult to do this with more lines as there
would be less room for the clamps on each slat, there
being less and less inter slat angle and so clamp space available as the lines increased.

[Panrock]

Thanks for the info, chaps.

Here are the various bits and pieces as things stand at the moment. After I've got it basically working and before a suitable housing or cabinet is made, I'll have it all apart again to paint or change the parts and smarten up the general appearance.

Steve

[Panrock]

Today I've made a start mounting the very first slats. At once it strikes me how difficult it will be to set the angles critically. But then I remember how I did this on the 120-line screw... I used a laser torch and set the reflection 'markers' two or three at a time by ensuring they were separated by a set distance as seen on the opposite wall of the workshop. This makes a sensitive and accurate method, equivalent to using a protractor about 10 metres across! The standard distance on the wall between the laser beams should be easy to calculate and I must remember it will be rendered as twice the slat angles, or 8 degrees... giving a doubling in positioning sensitivity.

Later, it should be simple to run a check on the whole screw, when a procession of moving red dots will be seen following each other along the wall, as the screw is slowly rotated. All of these should be at the same standard distance from each other. Any outliers can be identified and corrected at this point.

I have now ordered a line-projecting laser torch from Ebay. The line will be shone onto the screw perpendicular to the slats, resulting in a reflected spot on the opposite wall for each slat.

Steve O

[Panrock]

While I await the arrival of the laser, I'm spending the time fitting the various parts needed for sync'ing the WC-01 to the future Mirror Screw. All that's needed is a frame pulse.

So far, a small disc marked with a reflecting strip on its black underside has been fitted to the shaft, and a socket (at lower right) - to plug straight into the existing control unit - I'll be re-using as much as possible of the 120-line electronics, originally designed by Steve A.

One rare piece he didn't do is the 555 monostable (attached) which takes the raw pick-up from the sync disc and transforms it to a respectable-looking pulse.

Steve O

9888.jpg (131.22 KiB) Viewed 832 times

pulse_proc.png

(6.93 KiB) Not downloaded yet

[Steve Anderson]

One rare piece he didn't do is the 555 monostable (attached) which takes the raw pick-up from the sync disc and transforms it to a respectable-looking pulse. Steve O

Maybe I assumed it had already been done? But if I recall correct I was thinking of using a Hall-effect device at the time, a magnetic detector insensitive to stray light. But then again it may have been upset by any errant magnetic field from the motor. But still a suggestion for the future...

Steve A.

https://en.wikipedia.org/wiki/Hall_effect_sensor

Although I was unaware of the following at the time, a possible alternative...

https://en.wikipedia.org/wiki/MEMS_magn ... eld_sensor

There are many Hall-effect devices 'out there' to choose from, here's a typical example..some have a linear output, some are logic-level, the following has a latching logic output...depends on what you need...

https://uk.farnell.com/allegro-microsys ... dp/1791385