Mini version of Scophony - its do-able

Forum for discussion of narrow-bandwidth mechanical television

Moderators: Dave Moll, Andrew Davie, Steve Anderson

Re: Mini version of Scophony - its do-able

Postby Gregory » Tue Jan 26, 2021 9:02 am

Harry,
I have looked through your post on scope modulation, you got very good results, especially the syncing on both line and frame scans.
Although Steve is right, most of this can be done using just a single pic microcontroller, my thoughts are also to go for a simpler way by using similar circuits to the Klass sync and NBTV circuits, and add to this the additional logic electronics to be able to drive the stepper motors. When I complete this project, successfully I hope, it would be interesting to do a pic version of this.

The first part which I am currently building, the sync separator is very similar to the NBTV televisor and Klass circuits as previously shown. As for the missing pulse restorer, I have now re-designed this as shown below, to use 4098 dual monostable instead of the 4001 gates, as this may be more accurate.

missing pulse restorer 3.jpg
missing pulse restorer 3.jpg (28.81 KiB) Viewed 1568 times

To simplify things for different line resolutions the first monostable makes the sync pulse width standard at 500 usecs for all resolutions (32 lines 400 hz, 60 lines 750 hz, 120 lines (1500 hz) Etc at 12.5 frames. Although this pulse width is much higher than the standard sync pulse width, the sync pulse here is only required to drive the stepper motor and even at 1500 hz it will operate.

The second monostable provides a delay of exactly one line scan interval with selection of line/frequency resolution by manual switching . The final monostable again provides a pulse with 500 usecs width from the delayed output of the second monostable.

By mixing the 500 sync pulse from the output of monostable 1 with the delayed output of monostable 3 by using an OR circuit consisting of the 2 diodes and resistor, the missing pulse is replaced, without any phase change.
Last edited by Gregory on Sun Feb 21, 2021 2:15 am, edited 2 times in total.
Gregory
Misunderstood Genius
 
Posts: 80
Joined: Wed Feb 10, 2016 5:06 pm
Location: Nicosia, CYPRUS

Re: Mini version of Scophony - its do-able

Postby Harry Dalek » Tue Jan 26, 2021 6:55 pm

I am glad you are still working on your own idea more circuits on the subject the better i say ,do post up part values when you got it working to your needs ,always helps others .
Mines a Klass circuit from a while back as well i was very happy with it any case know you can get this working one way or another.
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: 4780
Joined: Fri Sep 26, 2008 4:58 pm
Location: Australia

Re: Mini version of Scophony - its do-able

Postby Gregory » Thu Jan 28, 2021 3:18 am

Finally finished the sync separator and even better, it worked. Scope signals show a stable sync at 400hz and 750hz
The potentiometer on the breadboard doesn't seem to make much difference, so finally may I may replace this with a fixed resistor at the end. The rotary switch on the front panel is for later to switch between different line/frequency resolutions.

Some mistakes I made on the schematic, on the power pinouts of the ca3140 IC so I will revise the schematic uploaded so its correct.

Next job, when time allows, is to build and test the missing pulse restorer.

build 2 electronics .jpg
build 2 electronics .jpg (86.12 KiB) Viewed 1831 times

Below block schematic of all the electronics I think will be needed to drive the line and frame stepper motors synchronized to the input video signal, but not necessarily in phase.

block1.jpg
block1.jpg (14.08 KiB) Viewed 1831 times
Gregory
Misunderstood Genius
 
Posts: 80
Joined: Wed Feb 10, 2016 5:06 pm
Location: Nicosia, CYPRUS

Re: Mini version of Scophony - its do-able

Postby Harry Dalek » Sat Jan 30, 2021 8:47 am

What are you using for the multipliers a 4046 ?
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: 4780
Joined: Fri Sep 26, 2008 4:58 pm
Location: Australia

Re: Mini version of Scophony - its do-able

Postby Gregory » Mon Feb 01, 2021 8:06 am

Harry, yes I will be using 4046 together with a ripple counter for the frequency multipliers. Unfortunately my mirror/motor combination for the line scan is not ideal as the multiplier factor is non integer, and will lead to slight complications, But there's no choice available, polygon mirrors are hard to find. Perhaps a new post on ways to construct a custom polygon mirror would be useful.

The frequency multiplication calculations work out:

LINE SCAN MULTIPLIER
Stepper motor has 768 micro-steps/rev (48 steps on 1/16 micro-step)
Using a 15 sided mirror, 1 line scan =768/15 = 51.2 pulses per line scan

Multiplication factor = 51.2
Divider required = 256 + additional divider of 5 on input to stepper drive to achieve 51.2 MF

FRAME SCAN MULTIPLIER
Stepper motor has 3200 micro-steps/rev (200 steps on 1/16 micro-step)
Using an 8 sided mirror, 1 line scan =3200/8 = 400 pulses per line scan

Multiplication factor = 400
Divider required = 400
Gregory
Misunderstood Genius
 
Posts: 80
Joined: Wed Feb 10, 2016 5:06 pm
Location: Nicosia, CYPRUS

Re: Mini version of Scophony - its do-able

Postby Harry Dalek » Mon Feb 01, 2021 4:53 pm

Gregory wrote:Harry, yes I will be using 4046 together with a ripple counter for the frequency multipliers. Unfortunately my mirror/motor combination for the line scan is not ideal as the multiplier factor is non integer, and will lead to slight complications, But there's no choice available, polygon mirrors are hard to find. Perhaps a new post on ways to construct a custom polygon mirror would be useful.


Well easy to construct badly ! to make something to match a laser printer polygon octagon or what ever your 12 sided mirror is called is very hard ,more so the motor mounting correctly true ...this wobbles your mirror it will work but you get a wobble in the raster ...need something like a video head or HD motor no wobble with them .
There's a bit on the old newsletters via Allan Short years back use to make them but from what i saw be nothing like what you could do with your CNC machine these days...you made the frame drum that's pretty good !
What mirror number of sides are you looking at thinking about making .
https://www.taswegian.com/NBTV/forum/vi ... f=1&t=2607

The above 18 sided mirror is good what was not good was it was mounted on that motor shaft no matter what you do here it will wobble ..better again use a vcr head HD motor even if you don't use the motor a pulley and you can use what ever to drive it.
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: 4780
Joined: Fri Sep 26, 2008 4:58 pm
Location: Australia

Re: Mini version of Scophony - its do-able

Postby Gregory » Wed Feb 03, 2021 9:24 am

Well easy to construct badly ! to make something to match a laser printer polygon octagon or what ever your 12 sided mirror is called is very hard ,more so the motor mounting correctly true ...this wobbles your mirror it will work but you get a wobble in the raster ...need something like a video head or HD motor no wobble with them .
There's a bit on the old newsletters via Allan Short years back use to make them but from what i saw be nothing like what you could do with your CNC machine these days...you made the frame drum that's pretty good !
What mirror number of sides are you looking at thinking about making .

16 sided mirror would be good, as nearly all stepper drives with micro stepping have divide by 16 step, so the frequency multiplier factor would always be an integer.

Frame mirror.jpg
Frame mirror.jpg (58.97 KiB) Viewed 1815 times

On the frame mirror I constructed (above) a few years back I used acrylic mirror sheet, which was optically quite accurate, although the mirror surface was on the back, but this didn't seem to noticeably affect the reflected beam. each mirror face was adjustable by 4 small screws. I think a better way of doing this maybe to have some sort of jig to hold the mirror faces accurately and then glue the mirror mount to the mirrors.
Gregory
Misunderstood Genius
 
Posts: 80
Joined: Wed Feb 10, 2016 5:06 pm
Location: Nicosia, CYPRUS

Re: Mini version of Scophony - its do-able

Postby Harry Dalek » Wed Feb 03, 2021 8:09 pm

Gregory wrote: I think a better way of doing this maybe to have some sort of jig to hold the mirror faces accurately and then glue the mirror mount to the mirrors.


An accurate Temp mount say 16 sides but do not glue back at all they will all be off no matter how little you use all faces will reflect different ! i learnt that mistake the hard way got to fix it flat to the mount your screw mount idea is good ,if you can stay away from glue its the way to go .

I suppose you could redo the frame drum to fit what you have made as is ?
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: 4780
Joined: Fri Sep 26, 2008 4:58 pm
Location: Australia

Re: Mini version of Scophony - its do-able

Postby Gregory » Thu Feb 04, 2021 9:59 am

Harry Dalek wrote:An accurate Temp mount say 16 sides but do not glue back at all they will all be off no matter how little you use all faces will reflect different ! i learnt that mistake the hard way got to fix it flat to the mount your screw mount idea is good ,if you can stay away from glue its the way to go .

I suppose you could redo the frame drum to fit what you have made as is ?

On second thoughts, yes I think your right, best to keep with the screw mount idea, The frame mirror is OK for the moment and quite accurate as I had this already built, but it would be good to modify & improve this further, so it could be used for the line scan mirror which requires a much higher RPM. This may be of interest for other members wanting to use mirror drums for scanning, as polygon mirrors are very hard to get.

For 60 lines / 750hz

15 segment mirror = 3000 rpm
18 segment mirror = 2500 rpm
24 segment mirror = 1875 rpm

Both 7.5 and 1.8 degree motors with low resistance windings, can reach 3000 RPM reliably under light load conditions.

Getting back to the electronics I have started building the missing pulse restorer & hopefully should have this working the following week.
Gregory
Misunderstood Genius
 
Posts: 80
Joined: Wed Feb 10, 2016 5:06 pm
Location: Nicosia, CYPRUS

Re: Mini version of Scophony - its do-able

Postby Harry Dalek » Sun Feb 07, 2021 8:25 am

Gregory wrote:On second thoughts, yes I think your right, best to keep with the screw mount idea, The frame mirror is OK for the moment and quite accurate as I had this already built, but it would be good to modify & improve this further, so it could be used for the line scan mirror which requires a much higher RPM. This may be of interest for other members wanting to use mirror drums for scanning, as polygon mirrors are very hard to get.


Yes i agree your frame drum is fine only improvements i could be picky about is smaller screws holding the mirrors on but as is is not going to effect any thing its fine beautiful better than any thing i could make .

A home made line mirror is harder for sure but making some thing a touch flatter and more mirror faces with the idea similar to the frame mirror your construction ideas it should be possible always take into account the wobble problem very thing has to be rotating true ...thing looks great as is .


For 60 lines / 750hz

15 segment mirror = 3000 rpm
18 segment mirror = 2500 rpm
24 segment mirror = 1875 rpm

Both 7.5 and 1.8 degree motors with low resistance windings, can reach 3000 RPM reliably under light load conditions.


Should be able to do do 30 and 48 l,60 or 64 ine as well ,free NBTV you can custom make the video line frame rate if you want to go out side standards .

Getting back to the electronics I have started building the missing pulse restorer & hopefully should have this working the following week.


Let us know how you go fingers crossed not that you need it . :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: 4780
Joined: Fri Sep 26, 2008 4:58 pm
Location: Australia

Re: Mini version of Scophony - its do-able

Postby Gregory » Mon Feb 08, 2021 5:01 am

re-done
Last edited by Gregory on Mon Feb 08, 2021 6:29 am, edited 1 time in total.
Gregory
Misunderstood Genius
 
Posts: 80
Joined: Wed Feb 10, 2016 5:06 pm
Location: Nicosia, CYPRUS

Re: Mini version of Scophony - its do-able

Postby Gregory » Mon Feb 08, 2021 5:14 am

Harry Dalek wrote:Let us know how you go fingers crossed not that you need it

No, I think I'll need all the finger crossing I can get.
I have done a few calculations on different mirror/frequency multiplier combinations as below:

FREQUENCY MULTIPLIER divider requirements for 48 step motor

Micro-steps per rev = 48 x 16 =768 (for divide by 16 micro steps)

For 8 sided mirror, Divider = 96:1
For 16 sided mirror, Divider = 48:1
For 24 sided mirror, Divider = 32:1

For the 15 sided mirror I have for the line scan it requires a frequency multiplication of 51.2, which is non integer. This will require a higher frequency multiplier of 256:1 and then separate frequency dividing on the pulse input of the stepper motor of 1:5 to obtain the 51.2 (256/5) frequency multiplication required.

FREQUENCY MULTIPLIER divider requirements for 200 step motor

Micro-steps per rev = 200 x 16 =3200 (for divide by 16 micro steps)
Divider for PLL = 3200 :1

For 8 sided mirror, Divider = 400:1
For 16 sided mirror, Divider = 200:1
For 20 sided mirror, Divider = 160:1
For 25 sided mirror, Divider = 128:1
Last edited by Gregory on Sat Feb 20, 2021 5:57 pm, edited 1 time in total.
Gregory
Misunderstood Genius
 
Posts: 80
Joined: Wed Feb 10, 2016 5:06 pm
Location: Nicosia, CYPRUS

Re: Mini version of Scophony - its do-able

Postby Andrew Davie » Mon Feb 08, 2021 11:02 pm

In 3D printing, it is common to have the print surface (a plane) adjusted by 4 screws in the corners. This always bothered me; a plane is *defined* by three points, and it is about these that the plane pivots when they are adjusted. So, I get the same sort of vibe seeing 4-screws in a mirror adjustment. I understand why so, but still bugs the purist in me! Keep up the good work; am watching your construction with interest.
User avatar
Andrew Davie
"Gomez!", "Oh Morticia."
 
Posts: 1584
Joined: Wed Jan 24, 2007 4:42 pm
Location: Queensland, Australia

Re: Mini version of Scophony - its do-able

Postby Andrew Davie » Mon Feb 08, 2021 11:07 pm

Gregory wrote:
Harry Dalek wrote:Let us know how you go fingers crossed not that you need it

No, I think I'll need all the finger crossing I can get.
I have done a few calculations on different mirror/frequency multiplier combinations as below:

FREQUENCY MULTIPLIER divider requirements for 48 step motor

Micro-steps per rev = 48 x 16 =768 (for divide by 16 micro steps)

For 8 sided mirror, Divider = 96:1
For 16 sided mirror, Divider = 48:1
For 24 sided mirror, Divider = 32:1

For the 15 sided mirror I have for the line scan it requires a frequency multiplication of 51.2, which is non integer. This will require a higher frequency multiplier of 256:1 and then separate frequency dividing on the pulse input of the stepper motor of 1:5 to obtain the 51.2 (256/5) frequency multiplication required.

FREQUENCY MULTIPLIER divider requirements for 200 step motor

Micro-steps per rev = 200 x 16 =3200 (for divide by 16 micro steps)
Divider for PLL = 3200 :1

For 8 sided mirror, Divider = 400:1
For 16 sided mirror, Divider = 200:1
For 20 sided mirror, Divider = 160:1
For 25 sided mirror, Divider = 128:1


Regarding micro-stepping in steppers this, too, has been something I've run across in 3D printing. The discussion there was the claim that layer height (the vertical stepping) should be a multiple of 0.04 mm. The argument ties down to avoiding micro-stepping the motor because it is inaccurate. The fundamentals are that the stepper driver is (depending on the brand) somewhat inaccurate at holding a motor at a micro-step position, and this translates to inaccuracies in layer height. While true enough, my analysis back in the day was that the actual error was too small to make any perceptible difference.

But in any case, interesting to see the same sort of thing being discussed in such disparate endeavours. Just be aware there is some positional error in microstepping, and this can vary between motors/brands. I could probably dig up the analysis/evaluation if really needed.
User avatar
Andrew Davie
"Gomez!", "Oh Morticia."
 
Posts: 1584
Joined: Wed Jan 24, 2007 4:42 pm
Location: Queensland, Australia

Re: Mini version of Scophony - its do-able

Postby Andrew Davie » Mon Feb 08, 2021 11:10 pm

User avatar
Andrew Davie
"Gomez!", "Oh Morticia."
 
Posts: 1584
Joined: Wed Jan 24, 2007 4:42 pm
Location: Queensland, Australia

PreviousNext

Return to Mechanical NBTV

Who is online

Users browsing this forum: No registered users and 11 guests