Mechanical video recorder

Centralised area for notes about construction projects. Each project has its own sub-forum. In the sub-forums will be topics relevant to parts of that project (e.g., there might be a topic on motor control). If you start a construction diary, just post in the forum with your project name as the topic, and a moderator will create a sub-forum for your project.

Moderators: Dave Moll, Andrew Davie, Steve Anderson

Re: Mechanical video recorder

Postby Klaas Robers » Sat Nov 30, 2019 11:10 pm

There have been radio amateurs, in the time that SSTV was popular, that used a separate 1200 Hz oscillator, which was keyed into the SSTV FM signal when a sync pulse was given. Then of course, the FM subcarrier was discontinued and you saw after demodulation a positive or a negative glidge at the beginning of a sync and in most cases also at the end of the sync. This was visible at the right edge and at the left edge of the picture. But if for sync detection a separate tuned circuit at 1200 Hz was used, it gave a more stable sync. I never did this because, if you wanted to recover the sync directly from the demodulated video signal, it gave also a jittering sync due to the glidges.

On the other hand, if you insert a baseband sync pulse in the video signal and send that through the VCO / FM-modulator, The phase of the 1200 Hz is different for each sync pulse and may give also some jitter on the sync. Yes, sync detection on an SSTV-signal is not that easy. I think that an electronic flywheel could have solved this problem, but I have never seen that implemented.
User avatar
Klaas Robers
"Gomez!", "Oh Morticia."
 
Posts: 1656
Joined: Wed Jan 24, 2007 8:42 pm
Location: Valkenswaard, the Netherlands

Re: Mechanical video recorder

Postby Harry Dalek » Sun Dec 01, 2019 9:59 am

Klaas Robers wrote:There have been radio amateurs, in the time that SSTV was popular, that used a separate 1200 Hz oscillator, which was keyed into the SSTV FM signal when a sync pulse was given. Then of course, the FM subcarrier was discontinued and you saw after demodulation a positive or a negative glidge at the beginning of a sync and in most cases also at the end of the sync. This was visible at the right edge and at the left edge of the picture. But if for sync detection a separate tuned circuit at 1200 Hz was used, it gave a more stable sync. I never did this because, if you wanted to recover the sync directly from the demodulated video signal, it gave also a jittering sync due to the glidges.

On the other hand, if you insert a baseband sync pulse in the video signal and send that through the VCO / FM-modulator, The phase of the 1200 Hz is different for each sync pulse and may give also some jitter on the sync. Yes, sync detection on an SSTV-signal is not that easy. I think that an electronic flywheel could have solved this problem, but I have never seen that implemented.


Hi Klass
Bit of head scratching as to reprocessing the video ,inserting a new 1200hz sync start is what a thinking of doing next the slight time delay framing might be out but i would expect from using an external clock using a 4066 to switch on the start of the baseband video to the vco as in Steve's circuit would think the resulting video line would be lined up better for framing .
There would still be slewing drift perhaps from the platter speed...
All speculation but interesting .
I did check the mechanical sync pulse's frequency playing the recording.. it can go from 14 hz to 17hz most of the time it looks to swing from 15 to 16 hz does account for the sinewave look to the frame .
Its interesting others in the past inserted or tried to correct back syncing , i have never heard of electronic flywheel idea !
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: 5364
Joined: Fri Sep 26, 2008 4:58 pm
Location: Australia

Re: Mechanical video recorder

Postby Harry Dalek » Sun Dec 01, 2019 4:10 pm

Looked into the switching and VCO circuit today so far the VCO works Testing ...the 4066 mounted but not connected up yet apart from power supply .
If any thing it should line up the raster lines even if with a record platter speed change's the framing this is expected to go off ...should give a better image any case lining up the raster lines straight .
Just using a 4060 for the 15 or 16 hz clock for the 4066 switching a 15hz crystal clock had been made before still have it on another board come in useful here later for testing
Not sure if i need any thing after the VCO should just be FM sstv again with the syncing corrected to the new clock but not to the line start thats why i am thinking framing might be a touch out perhaps with a platter speed adjustment it might be fine enough for 7 or 8 sec's image ,could not think it would be any worse than uncorrected video works now .
Circuit is a slight change to Steve's design posted a post of 2 back working on 5 volts here...
DSCN9847-crop.JPG
DSCN9847-crop.JPG (204.25 KiB) Viewed 11230 times
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: 5364
Joined: Fri Sep 26, 2008 4:58 pm
Location: Australia

Re: Mechanical video recorder

Postby Klaas Robers » Sun Dec 01, 2019 7:55 pm

What you do is inserting a new sync into the base band video signal. That is better than inserting a 1200 Hz burst into the FM subcarrier SSTV signal. In the way you do it, the SSTV FM-signal is made new in your 4046, so glidges in the FM signal are not occuring.

I still do not understand why you are doing this. When you insert new syncs that are not synchronous with the 1200 Hz syncs in the original SSTV signal, your video content (picture) will be unstable in the horizontal direction. In case the new sync pulses are synchronous with the original 1200 Hz SSTV-sync, I see no reason why you should reinsert them.....
User avatar
Klaas Robers
"Gomez!", "Oh Morticia."
 
Posts: 1656
Joined: Wed Jan 24, 2007 8:42 pm
Location: Valkenswaard, the Netherlands

Re: Mechanical video recorder

Postby Steve Anderson » Sun Dec 01, 2019 8:57 pm

I must admit that I don't understand quite why a non-synchronous (in terms of sub-carrier) sync pulse needed to be inserted. Surely the modulators/VCOs of the period could cover a 2:1 ratio? Simplification possibly? No idea!

Now the discontinuities introduced by this method will introduce glitches in the demodulated signal, probably equally on each edge of the sync pulse on average, though maybe not on every single line.

The SSTV stuff I've done previously hasn't really displayed much in the way of sync jitter, I guess it is there but there are other artifacts and distortions I'd rather eliminate or reduce first.

Getting that sub-carrier to blend each end of a line into and out of of the sync 'burst' of 1200Hz is no easy challenge!

Steve A.
User avatar
Steve Anderson
"Fester! Don't do that to 'Thing'"
 
Posts: 5360
Joined: Fri Mar 30, 2007 10:54 pm
Location: Bangkok, Thailand

Re: Mechanical video recorder

Postby Harry Dalek » Mon Dec 02, 2019 7:46 pm

Klaas Robers wrote:What you do is inserting a new sync into the base band video signal. That is better than inserting a 1200 Hz burst into the FM subcarrier SSTV signal. In the way you do it, the SSTV FM-signal is made new in your 4046, so glidges in the FM signal are not occuring.


OH i see Klaas ! i had FM in my head i forgot about that would be possible to add a pulse to the baseband ...i think because the original pulse would be there still i was thinking of a way to replace it not because its bad sync pulse but because its changing frequency with slight changes in the platter speed?

I wanted a stepper motor of the platter rotation as speed control but its that noisy vibrations and all it would of been picked up in the recordings so i dumped that idea for the quiet dc geared motor .

I still do not understand why you are doing this.


Well if motor speed control would keep my platter speed correct so my sync pulse on play back stayed dead on 16 hz ,i would be happy as is as the image would not be wavy a nice squared up image lines all matching one after the other .
At the moment the recorded sync pulse is swinging from 14 to 17 hz thus the sine wav look to the image .

I was thinking perhaps another way might be add a correct 16 hz sync pulse to the video if possible to straighten up my play back image ,i was trying to replace the recorded sync with a made version via my clock and pulse width circuit .

When you insert new syncs that are not synchronous with the 1200 Hz syncs in the original SSTV signal, your video content (picture) will be unstable in the horizontal direction. In case the new sync pulses are synchronous with the original 1200 Hz SSTV-sync, I see no reason why you should reinsert them.....


Well my want is a better played back image ,the recording process causes drag friction as it emboss's the record with the sstv video so errors in speed happen which i am seeing in playback ...the band width of the video is there just a fluctuation seen in the sync pulses frequency.
Its very hard what i am doing and understand all the problems just seeing if there's another way around the problem .
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: 5364
Joined: Fri Sep 26, 2008 4:58 pm
Location: Australia

Re: Mechanical video recorder

Postby Harry Dalek » Mon Dec 02, 2019 8:03 pm

Steve Anderson wrote:I must admit that I don't understand quite why a non-synchronous (in terms of sub-carrier) sync pulse needed to be inserted. Surely the modulators/VCOs of the period could cover a 2:1 ratio? Simplification possibly? No idea!


A lot of this explained in the reply to klaas ,i am just seeing if there is a way to correct the wavy play back image ...problem is keeping a correct 16hz sync pulse and its swinging on back causing that wavy image ..

If it stayed at 16 hz i would have a great image so the sync pulse changing frequency is the problem ,i thought it might a way to just replace it with a 16hz clock timing the 1200hz subcarrier to start adding the base band video to the VCO remaking the SSTV with new sync arrrr but sounds like i may be barking up the wrong tree ?

Now the discontinuities introduced by this method will introduce glitches in the demodulated signal, probably equally on each edge of the sync pulse on average, though maybe not on every single line.


I was expecting the line start to be off a touch so like a sync bar seen to one side with perhaps a little bit of video to one side of the bar and most on the other until i adjusted the record platter speed to to move the image to one side of the screen ...speculating what i think would be the result ?

The SSTV stuff I've done previously hasn't really displayed much in the way of sync jitter, I guess it is there but there are other artifacts and distortions I'd rather eliminate or reduce first.


Mechanical SSTV sure has it :wink:

Getting that sub-carrier to blend each end of a line into and out of of the sync 'burst' of 1200Hz is no easy challenge!

Steve A.


Klass's idea adding a sync pulse to the baseband sounds good but i would still have the old sync there may i would need a filter first ...?
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: 5364
Joined: Fri Sep 26, 2008 4:58 pm
Location: Australia

Re: Mechanical video recorder

Postby Klaas Robers » Mon Dec 02, 2019 8:42 pm

If your disc speed during recording is varying, it is very difficult to recover the original picture. Inserting a new sync during play back is of no use, as the video itself is varying already. So you will Always end up with a wavy picure.

The only solution is: see that the disc rotation speed during recording (writing the disc) is very constant. The only solutions for that is a rather powerfull synchronous motor and a large and heavy turntable.

The synchronous motor is to keep the speed constant. A synchronous motor is in fact a stepper motor, that is running on sine wave voltages in stead of square wave voltages. That ensures that it rotates continuously and not stepwise.

The heavy turntable is to smoothen the rotation. A synchronous motor runs a steady number of revolutions per second, but if the driving sine waves are having some distortion (in most cases they have) there is some fast irregularity on the speed. Think of a slight step by step behaviour, added to the continuous rotation. When you recorded a pure tone, this gives a fast (50 Hz or higher) frequency modulation on the tone, something that you can hear at play back. Recorders for grammophone records always were large and heavy. I have seen them in operation in about 1980.

When you want to check your recording - play back system, record a pure sine wave, e.g. 1 kHz and listen at it during play back. I do have a .wav file of such a 1 kHz tone.

When this is Ok, record a slow sweep tone and look at play back if the amplitude is constant. I have a .wav file with such a sweep from 2 Hz to 20 kHz. If you like to limit yourself to SSTV, a sweep of 300 Hz to 3000 Hz would be sufficient. I do not have a file for that.
User avatar
Klaas Robers
"Gomez!", "Oh Morticia."
 
Posts: 1656
Joined: Wed Jan 24, 2007 8:42 pm
Location: Valkenswaard, the Netherlands

Re: Mechanical video recorder

Postby Harry Dalek » Mon Dec 02, 2019 10:04 pm

Thanks for the advice there Klaas i am listening blink blink !

I did use a stepper motor when starting this but step wise but and it was noisy to noisy to record any thing vibrated the platter ! ,i didn't think of using sine wave to run it .

The platter i am using is heavy it is steel plate from an old beta vcr used as a flywheel i think .

I will have think about this and reboot my thoughts on the progress .

Yes SSTV for now till i get this right using the 8 sec SSTV to learn how to record and play back video right ...at the moment badly :roll:
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: 5364
Joined: Fri Sep 26, 2008 4:58 pm
Location: Australia

Re: Mechanical video recorder

Postby Steve Anderson » Tue Dec 03, 2019 2:54 pm

Harry, I'm not sure if this is possible in this set-up, but for what it's worth here's a suggestion...

Break the project down into two parts, first the electronics, second the mechanical.

Build up the record and playback electronics and feed the output of the record section into the input of the playback section - via a resistive attenuator probably. See what you get. This should have a flat frequency response over whatever bandwidth you're interested in - and hopefully no distortion too. Measure maximum signal levels where they start introducing distortion, usually clipping of the top and bottom of a sine-wave.

The next issue is equalisation. Depending on the type of record/playback 'head' you're using will determine what sort of equalisation is required. Usually equalisation is required in both the record and playback processes. Both for analogue discs and analogue magnetic tape.

Disc recording is a bit of a 'black art' with few left now who know how to do it properly...

Steve A.

Klaas is right, the record-cutting lathes I saw had motors bigger than a large (commercial) washing machine. (Decca, New Malden, UK, 1977. Though I never worked for Decca Records, I was employed by Decca Radar, the two divisions were co-located).
User avatar
Steve Anderson
"Fester! Don't do that to 'Thing'"
 
Posts: 5360
Joined: Fri Mar 30, 2007 10:54 pm
Location: Bangkok, Thailand

Re: Mechanical video recorder

Postby Harry Dalek » Tue Dec 03, 2019 5:31 pm

Thanks for your Advice too Steve i will look into resistive attenuator ..i sort of looked into this for the ceramic play back head .

I am not sure i have any problem on the recording or play back side apart from syncing distortion of the image ,but i will look into what you have mentioned .

Yes record Lathe's can be large and i understand the need for torque for the recording side every action has an equal and opposite reaction cutting a grove will slow down a motor if its to weak for the job playing hell for the recording ,here i am not cutting a record i am embossing a record which is different i am scratching a grove in the soft plastic of a cd or dvd .
The weight of the head will result in the depth of the scratch ,i seem to have got this right as i can play back every time i have tried now at first i could only do this 2 times for play back .
Designing the record head was a lot of trial and error not so much the thing doing the embossing needle stepper motor but working out it has to ride the undulations of the cd or dvd if not 100% flat ..without this design to the lathe arm the lathe needle will just dig into the cd dvd and you will know it when it happens!
Black art indeed ! when i stated i had no idea what the difference was cutting and embossing .
I've having been mulling it over on the platter motor and Klaas's suggestion does sound very good way to fix things so i will look into this ,its something i have tried before but never crossed my mind for this .
Last edited by Harry Dalek on Tue Dec 03, 2019 8:54 pm, edited 2 times in total.
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: 5364
Joined: Fri Sep 26, 2008 4:58 pm
Location: Australia

Re: Mechanical video recorder

Postby Steve Anderson » Tue Dec 03, 2019 6:39 pm

OK Harry, good thinking and good luck! It's an area I have no experience in...Albert is probably our resident expert...

Steve A.
User avatar
Steve Anderson
"Fester! Don't do that to 'Thing'"
 
Posts: 5360
Joined: Fri Mar 30, 2007 10:54 pm
Location: Bangkok, Thailand

Re: Mechanical video recorder

Postby Harry Dalek » Tue Dec 03, 2019 9:37 pm

Steve Anderson wrote:OK Harry, good thinking and good luck! It's an area I have no experience in...Albert is probably our resident expert...

Steve A.


Thanks Steve You and klaas set me on the right direction once again , see if i can improve on the results .
Yes Albert is our Baird !
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: 5364
Joined: Fri Sep 26, 2008 4:58 pm
Location: Australia

Re: Mechanical video recorder

Postby Klaas Robers » Fri Dec 06, 2019 10:25 pm

Harry,
it is more complex than you might think at first sight. Some recording heads give a side way displacement proportional to the voltage that you apply, others give a displacement SPEED proportional to that voltage. And to make things even more complicated: some play-back cartriges give a voltage proportional to the displacement of the "needle" (crystal and ceramic), while others give a voltage proportional to the displacement speed (magnetic dynamic). So just inserting a resistor voltage divider between the recorder amplifier output and the play-back amplifier input, will not guarantee that you get a flat characteristic.

Besides that sound, music, has different amplitudes for low, mid, high tones. So a flat recording characteristic might not be the optimal solution. In the 78 rpm era almost each disc producer (label) had his own optimal recording characteristic, and the better play-back amplifiers had about 10 different compensation networks to get an optimal reproduction. They all had the name of the record label that you were playing and you should switch it by hand. But I have read that in quite some cases a play-back compensation network of a different Label sounded better than the very label that you were playing. Besides that: video is not music, so the optimal compensation, or equalisation, will surely be different. Good Luck!

This all came to an end with the introduction of vinyl Long Playing 33 1/3 rpm and 45 rpm records. Then the RIAA committee standardized an optimal equalization. And since about 1950 this became the standard. In ceramic and crystal play-back cartridges this is more or less mechanically built in, so a "straight" amplifier will do. But for the better MD (Magnetic Dynamic) elements quite some equalization is needed. Amplifiers from 1950 to 2000 would have a "phono" input with this RIAA characteristic.

If you would like to record NBTV-signals, you should dive deeply into this matters. Then the characteristic should be flat from 2 Hz to 10 kHz. But for the SSTV audio-FM signals it is less important, because if there are amplitude changes during play-back the limiting input amplifier will compensate automatically for that. Your "disc channel" should be only more or less flat for frequencies from 300 to 3000 Hz (telephone bandwidth). It worked even over an amateur radio channel with quite some noise, distortion and interference in the "aether". (You know, there is no "aether", but we still talk about it as if it is existing.) But speed variations, never occuring in the aether, are forbidden. That will be immediately visible.
User avatar
Klaas Robers
"Gomez!", "Oh Morticia."
 
Posts: 1656
Joined: Wed Jan 24, 2007 8:42 pm
Location: Valkenswaard, the Netherlands

Re: Mechanical video recorder

Postby Harry Dalek » Sat Dec 07, 2019 4:39 pm

Klaas Robers wrote:Harry,
it is more complex than you might think at first sight. Some recording heads give a side way displacement proportional to the voltage that you apply, others give a displacement SPEED proportional to that voltage. And to make things even more complicated: some play-back cartriges give a voltage proportional to the displacement of the "needle" (crystal and ceramic), while others give a voltage proportional to the displacement speed (magnetic dynamic). So just inserting a resistor voltage divider between the recorder amplifier output and the play-back amplifier input, will not guarantee that you get a flat characteristic.


Oh Yes Klaas the idea of record recording sound doe's seem simple but it is its own science and still experimental by many ,so i do understand there's a lot of factor's to just the record and play back heads what ever design they are as you mentioned.

I have been thinking of reversing a play back head as a recording head but think you would need something up around 50 or so watts to make that crystal vibrate ,i was told on the labels of some old records the record head was mentioned if ceramic and 70 watts powering it .

Besides that sound, music, has different amplitudes for low, mid, high tones. So a flat recording characteristic might not be the optimal solution.


An equalizer ?

In the 78 rpm era almost each disc producer (label) had his own optimal recording characteristic, and the better play-back amplifiers had about 10 different compensation networks to get an optimal reproduction. They all had the name of the record label that you were playing and you should switch it by hand. But I have read that in quite some cases a play-back compensation network of a different Label sounded better than the very label that you were playing. Besides that: video is not music, so the optimal compensation, or equalisation, will surely be different. Good Luck!


I don't think for video i would go below 78 rpm i noticed the difference between a 45 and 78 rpm and it was much better

Yes that does sound complex what you say on compensation if you want the best results ,i am happy with any result or any improvement on what i have done

This all came to an end with the introduction of vinyl Long Playing 33 1/3 rpm and 45 rpm records. Then the RIAA committee standardized an optimal equalization. And since about 1950 this became the standard. In ceramic and crystal play-back cartridges this is more or less mechanically built in, so a "straight" amplifier will do. But for the better MD (Magnetic Dynamic) elements quite some equalization is needed. Amplifiers from 1950 to 2000 would have a "phono" input with this RIAA characteristic.


Well for record side i have only ever tried the the re purposed tiny stepper motor using half the winding's ,the play back is a ceramic modern head ..to a low noise amplifier..
I see they designed these things for simpler electronics even before i was born .

If you would like to record NBTV-signals, you should dive deeply into this matters. Then the characteristic should be flat from 2 Hz to 10 kHz.

In time for sure but i know i have some way to go ...that advice above will be kept in mind ..once i get around to doing sweep tests for that ...what i have at the moment is a touch short for 10hz i can get up to 8 khz i don't think this head is good enough for NBTV but fine for SSTV.

But for the SSTV audio-FM signals it is less important, because if there are amplitude changes during play-back the limiting input amplifier will compensate automatically for that. Your "disc channel" should be only more or less flat for frequencies from 300 to 3000 Hz (telephone bandwidth). It worked even over an amateur radio channel with quite some noise, distortion and interference in the "aether". (You know, there is no "aether", but we still talk about it as if it is existing.) But speed variations, never occuring in the aether, are forbidden. That will be immediately visible.


Well aether may not exist space time does and there's a lot of noise in space time :wink: I think there is FM NBTV sounds like if would be better for recording things ...but i still have the thought to use am modulate NBTV at 10 khz once i get a record head that can do 10khz .

On the dreaded Platter speed control i started to look into stepper motors but trying the circuit below using sine waves its unfortunately just as noisy plus vibrations similar to using it the traditional way it also lacks torque which may be fixed using and stereo amplifier to each coil and a oscillator for speed adjustment so just on the noise and vibration side its not good for a platter it would end up playing up with the recording .
bipolar (1).jpg
bipolar (1).jpg (6.95 KiB) Viewed 11103 times


Looking in my junk box of dc motor stepper motors i came across a microwave oven platter motor which happens to be a synchronous motor but running a low 5 rpm similar to the photo below ,this would have to be geared or pulley design up to give around the 78 rpm around meaning does not have to be record player standard if play back is only from this machine .
I like it testing the motor it is pretty quiet little vibration and has more torque than needed here many times over ,i have another i need to look at but so far this seems promising .
I have not looked at also standard record player motors but they would be synchronous also ?
220px-Synchronmot.jpg
220px-Synchronmot.jpg (31.19 KiB) Viewed 11103 times


Interesting pull apart for one of these motors

youtu.be/WT-rWaPFcQE
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: 5364
Joined: Fri Sep 26, 2008 4:58 pm
Location: Australia

PreviousNext

Return to Construction Diaries

Who is online

Users browsing this forum: No registered users and 3 guests

cron