NBTV Scope Evolution

[M3DVQ]

There's a photo of the display on the nbtva website on this page: http://www.nbtv.wyenet.co.uk/c06jago.htm

[Harry Dalek]

Thanks for tor link Alistair for the link to the photo prety nice results .\

Also thank you Graham for the newsletter number to track down i will read up on this result shown in Alistairs link photo.

Testing the sawtooth for a raster today i ended up changing a few things to clean it up and rid the gap space between the waveforms... changed the line sync sawtooth oscillator syncing connection from pin 2 of the 555 to pin 5 ...with pin 2 it was syncing but there it was also causing some problems with the shape of the sawtooth .

Something i forgot about was the scanning directions this is more than likly off ,took a short video hard to film a raster correctly with my little camera .

I will have a good read of the news letter may be i can look into that form of displaying video .

I'll try video display on the other scope has z and works but not from the X Y external signals i think ,perhaps i have to dispaly it side ways and use the scopes timebase unless i work it out .

[Harry Dalek]

Got some images with the z scope input modulation .this was just direct from pin 6 of my 5534 opamp enough to see some thing on this scope ,the video is inverted and reversed ..

Not great but a start ,least its enough to show some results ...

[Klaas Robers]

On the website of NBTV.org you may find a photograph of this in action:

http://www.nbtv.wyenet.co.uk/c06jago.htm
Scroll somewhat down.

On the horizontal input a frame scan is placed, with superimposed a fast square wave (I guess 200 kHz) which is duty cycle modulated with the video. Then you get two pictures next to each other, one in negative and one in positive. The disadvantage is that, if you do it straight forward, you have no gamma correction. With direct Z-modulation this comes in automatically.

And remind that the NBTVA-frome has an aspect ratio of 2 : 3 in portrait orientation. Then the lines tend to get closer. Then you may defocus somewhat.....

[Steve Anderson]

I remember Jeremy's item in the newsletter, it was quite some years ago. Indirectly it was the trigger for my use of PWM beam modulation even though there's no direct connection between the two, how that happened I don't know.

It uses a PWM waveform to shift the beam left (or right) to create both a negative and a positive image. It is quite a clever idea.

There's no reason why it couldn't be done in the vertical direction with one image above the other, but due to the aspect ratio of NBTV less shift is required if done horizontally.

You could view this as an extreme example of spot-wobble using a varying duty square wave! The duty-cycle of the square-wave is modulated by the video, just as in the case of the PWM grid modulation. The difference primarily is that the square-wave is added to the horizontal ramp instead of the grid circuit which if you're using a commercially made 'scope is easier.

One thing of note: You can't use PWM modulation on more modern 'scopes like mine. Because it has a LCD flat-panel display there is no way to modulate it, there is no Z or 'blanking' input - there of course is no grid!. So here you would have to use Jeremy's method. But you could use a flat-panel PC monitor with an up-converter to VGA. There are some advantages with a digital 'scope however- you can turn the scale/graticule off and you can choose the colour of the display.

Steve A.

[Harry Dalek]

Thanks Klaas and Steve for your input here

I read in past posts Steve had mentioned Jeremy's idea and looking into it very clever !

On the horizontal input a frame scan is placed, with superimposed a fast square wave (I guess 200 kHz) which is duty cycle modulated with the video. Then you get two pictures next to each other, one in negative and one in positive. The disadvantage is that, if you do it straight forward, you have no gamma correction. With direct Z-modulation this comes in automatically.

Graham had mentioned 35khz and reading the newsletter Jeremy says he picked this frequency as low as possible so that i suppose the pwm was invisble to the eye on a scope screen .

The results on Jeremy's scope screen from the club site look excellent better than i would of expected ,the split screen is confusing but again reading the article i think the split is more caused by the amplitude of the 35khz square wave to the frame scans sawtooth in the link again jeremy's description says its 3 times the amplitude .

This makes me wonder it the amplitude of the carrier frequency and the saw tooth was say the same would both left and right just be over lapped ?

And remind that the NBTVA-frome has an aspect ratio of 2 : 3 in portrait orientation. Then the lines tend to get closer. Then you may defocus somewhat.....

Yes Klass i didn't play around with the settings to much Yep very gappy in the lines i will try and improve things for next time ,i was just happy some results for a start reversed image direction and inverted video wise does it get any better

I remember Jeremy's item in the newsletter, it was quite some years ago. Indirectly it was the trigger for my use of PWM beam modulation even though there's no direct connection between the two, how that happened I don't know.

I wonder if hes still doing NBTV Steve? Hes started off with an AM version 50khz this time and it looks to me like the AM modulated 50khz went to both the H and V in that 3 transistor circuit, can you post these things up ? if not the original a hand drawn copy ?

It uses a PWM waveform to shift the beam left (or right) to create both a negative and a positive image. It is quite a clever idea.

I wonder if he expected the results of what you mentioned ,i like the results ,he did also say you could ignor the negative image which i think you just incease and center the postive image so the other is out of scan viewing .

There's no reason why it couldn't be done in the vertical direction with one image above the other, but due to the aspect ratio of NBTV less shift is required if done horizontally.

A Tekade System !

You could view this as an extreme example of spot-wobble using a varying duty square wave! The duty-cycle of the square-wave is modulated by the video, just as in the case of the PWM grid modulation. The difference primarily is that the square-wave is added to the horizontal ramp instead of the grid circuit which if you're using a commercially made 'scope is easier.

I have talked to steve about this but was the system ever named seems a bit above spot wobble ? PWM spot wobble ?

The overall idea of it seems i hope within my skills low and inferior as mine are i would not mind giving it a go or at least seeing how far i can get with it ...I have tried mixing a low frequency pulse with the frames Sawtooth what i had handy it split the raster but half and half not two full scans arrr its a start I will do it bit better next with something around that frequency and have to look into the amplitude differences of the carrier and sawtooth.

[Harry Dalek]

Having a try at the PWM spot wobble i looked into changing my circuit a little on the 4060 to produce 35khz Triangle waveform for the PWM, already had the opamp to mix with oscillator to the LM311 comparator ,same sort of idea in the circuit below apart from i am using a 4060 oscillator and a 5534 opamp ..

Not sure what i should of expected as in a waveform mixing the PWM with the frame sawtooth but getting this result ..

I have not copied jeremy's circuit just the idea to try and get the same result.

[Steve Anderson]

I wonder if hes still doing NBTV Steve? ....can you post these things up ? if not the original a hand drawn copy ?

Yes, he is. I still get the occasional e-mail from Jeremy though not as often as in the past. He's still the editor of the newsletter as far as I know.

...I have no news r.e. the newsletter by the way, I'm awaiting my first copy in .pdf format. I need to be patient I guess.

'cos I use AutoCad so much and have done since the mid-80s it's quicker for me to use that rather than hand-draw something. It's the same with writing, I'm much faster on a keyboard than with a pen - a plus-point is that others can read it!

Steve A.

[Harry Dalek]

Thanks Steve never know who comes and gos doing this hobby i suppose its really the silent majority !

I hooked every thing up today for another go just playing around with the PWM mixed to the frame ...i didn't get 2 split images on this go more just wobbled lines as in a pooer 32 line version of those statue of liberty pictures posted a while back .

Really very similar results to just using Am from the opamp direct via a cap mixed with the frame sawtooth but the pwm way looked a touch better ...

I still didnt get the split frame effect pretty poor results at first go at it .

sorry about the video crappy camera do a better one next time

[Steve Anderson]

After the 'shift oscillator' there are two NOR gates, at the output of the second one you should be getting an approx. 35kHz square-wave, the duty-cycle, not the frequency, should vary with the video signal.

At the op-amp end of the 'Frame Scan Output' 1k resistor you should see a 12.5Hz negative-going ramp (timebase). There will probably be some interaction between the signals, why the mixing resistors (1k) are such a low value I don't know.

I realise that you're not using an exact copy of Jeremy's circuit, it may help to post a schematic of whatever you have built.

Steve A.

[Harry Dalek]

After the 'shift oscillator' there are two NOR gates, at the output of the second one you should be getting an approx. 35kHz square-wave, the duty-cycle, not the frequency, should vary with the video signal.

Hi Steve i posted my schematic below as theres always more than one way to do the same thing that was my thinking here ...Jeremy's circuit is another way of doing PWM ? if i got it right you mix a high frequency for the bandwidth of nbtv around the 30 or 40 khz triangle or sawtooth in a comparator in one of its inputs the other you input the video then the pulse width of the high frequency changes width but frequency is the same ....
I have mixed a Triangle wave with the video in the LM311 the result adding it to the frame sawtooth looks more Am to me i was expecting ramp line of the sawtooth to be a series of broken lines perhaps to fast to see ...any case head scratching ...

Reading he says a high frequency square wave is mixed with the video can you do PWM with a Square wave i thought you had to use a triangle or sawtooth to do this ..

At the op-amp end of the 'Frame Scan Output' 1k resistor you should see a 12.5Hz negative-going ramp (timebase). There will probably be some interaction between the signals, why the mixing resistors (1k) are such a low value I don't know.

I had to ac capacitor couple the PWM to the frame saw tooth dc connection tried resistors but nope.

I realise that you're not using an exact copy of Jeremy's circuit, it may help to post a schematic of whatever you have built.

Steve A.

Yes as below pretty much it at the moment i only left out A sync indicator and a few leds ...the 12 hz is free running at the moment .

[Steve Anderson]

The first possible problem area may be the use of a NE5534 which isn't a single-supply op-amp. I say maybe because it might just operate as intended. If the video waveform at its output looks OK then perhaps you've gotten away with it - note: the waveform will be inverted (deliberate). The AN1358 is a single-supply dual op-amp, so that's fine. Jeremy used a LM348 quad single-supply op-amp.

The use of the 4060 is an error. The 35kHz waveform needs to be either a sawtooth or a triangle. Use a 555 to generate this and feed the triangle waveform into pin 3 of the 311. The triangle waveform with a +12V supply will be approx. 4V p-p, from +4V to +8V. A 555 used convetionally as an astable doesn't generate a linear triangle but for this application it's OK - but could be better.

The video part of the signal (311, pin 2) should be in the same range as the triangle waveform, the syncs will be higher than +8V but that doesn't matter, we only need the picture information here. You may need to increase the value of the 22k feedback resistor to get the gain you need (pins 2-6 of the 5534), possibly increase the value of the 15k resistor is series with the pot to make setting the output DC level easier.

The output of the 311 is now the desired PWM variable-duty square-wave. Disconnect the 25uF cap to view this on a 'scope. The posting of Ron Jones's laser-link above explains the principle quite well. What you are trying to achieve is shown in the sketch below.

Lastly, the circuit downstream of the 311 isn't going to work well, if at all. I'll have a think on this and post an update a little later,

Steve A.

Later...I just realised that the LM348 isn't single supply, I got it confused with the LM324 which IS single-supply. Either that was a typo by Jeremy or he used a negative supply not shown in the circuit.

[Harry Dalek]

[/quote]

The first possible problem area may be the use of a NE5534 which isn't a single-supply op-amp. I say maybe because it might just operate as intended. If the video waveform at its output looks OK then perhaps you've gotten away with it - note: the waveform will be inverted (deliberate). The AN1358 is a single-supply dual op-amp, so that's fine. Jeremy used a LM348 quad single-supply op-amp.

The 5534 part i sort of replaced the 741 used in the laser circuit i post a few posts ago if you scoll up to it ,i can easy swap it with some thing else pin connections on the used area are same on a lot of opamps .......comes out inverted when i tried the z axis so want to be a positive image any way ..i will look into this ,i have been looking at and testing it at the moment.

The use of the 4060 is an error. The 35kHz waveform needs to be either a sawtooth or a triangle. Use a 555 to generate this and feed the triangle waveform into pin 3 of the 311. The triangle waveform with a +12V supply will be approx. 4V p-p, from +4V to +8V. A 555 used convetionally as an astable doesn't generate a linear triangle but for this application it's OK - but could be better.

I added the circuit at the start for an idea but as i changed things i didn't need it ,instead of replacing it with the 555 Triangle wave oscillator again in that laser circuit past post i made a high frequency square wave oscillator out of the 4060 then added a cap across the square wave which made it a Triangle wave with out changing that frequency .....IT still sounds like i made mistake as you say as if i need to run it at 12 volts and get a 4 volts p-p waveform out ... this is a 5 volt 74hc4060 version of the ic so unless i find a 5 to 15 volt version i need to swap this ..OK i will adjust to a 555 .

The video part of the signal (311, pin 2) should be in the same range as the triangle waveform, the syncs will be higher than +8V but that doesn't matter, we only need the picture information here. You may need to increase the value of the 22k feedback resistor to get the gain you need (pins 2-6 of the 5534), possibly increase the value of the 15k resistor is series with the pot to make setting the output DC level easier.

OK I will just see what results i get with a few different opamps ,i would need to swap the 22k for them anyway i'll connect me resistor wheel in there and test.

The output of the 311 is now the desired PWM variable-duty square-wave. Disconnect the 25uF cap to view this on a 'scope. The posting of Ron Jones's laser-link above explains the principle quite well. What you are trying to achieve is shown in the sketch below.

That was a 25NF OK i have made this circuit a few times but sort of always did it with the 555 oscillator part so sounds like i have gone wrong here ,i will check the results when i do the changes .

Lastly, the circuit downstream of the 311 isn't going to work well, if at all. I'll have a think on this and post an update a little later,

Well so far the pwm sort of works with out that 25nf cap the image is really just a wobbled line with that cap mixing with the sawtooth my nbtv letters in the wav file i was displaying came out as solid letters no lines in the letter parts but even so best i could do with this so far any thing more complex to display didn't come out well at all .
The beam splitting is the harder part for me .

Steve A.

Later...I just realised that the LM348 isn't single supply, I got it confused with the LM324 which IS single-supply. Either that was a typo by Jeremy or he used a negative supply not shown in the circuit.

Well i never knew the 5534 was a dual supply opamp ..yes a mistake in the ic could change every thing mmmmm

[Steve Anderson]

Yep, the 5534 is primarily an audio chip for low-noise applications, turntable pre-amps, microphones etc.. In most applications it was used with dual supplies, where used single-supply the inputs are close to the centre of the supply voltage, e.g. +12V if the supply is +24V. The inputs stop behaving if they get within 2-3V of either supply, or in this case ground/0V. The same applies to 741s, LM348s and the like, I don't know of a single single-supply op-amp i.e. not a dual or a quad, they must be around though I guess. The LM358 is a dual version of the LM324 quad single-supply op-amp.

In the laser circuit above the 741 inputs are close to half the supply voltage and the signal is small, a microphone output.

The major concern I have with the cap on the output of the 4060 is the the internal current-limiting circuit is being used to provide a current source to charge/discharge the cap. It is meant for protection of the device and under normal circumstances should never be activated, it also increases heat dissipation and power supply current. Yes, change it to a 555.

I've still to look at the last part, I got dragged away by something else...

Steve A.

[Harry Dalek]

Yep, the 5534 is primarily an audio chip for low-noise applications, turntable pre-amps, microphones etc.. In most applications it was used with dual supplies, where used single-supply the inputs are close to the centre of the supply voltage, e.g. +12V if the supply is +24V. The inputs stop behaving if they get within 2-3V of either supply, or in this case ground/0V. The same applies to 741s, LM348s and the like, I don't know of a single single-supply op-amp i.e. not a dual or a quad, they must be around though I guess. The LM358 is a dual version of the LM324 quad single-supply op-amp.

I have really only ever used it with a single supply in my head amps, it has a wide bandwidth and yes low noise ,ok i think i have LM324 its a catch 22 a rebuild or dual supply i will think about this ,so you think those circuits using a single supply for their opamp should really be a dual supply ?
OK if so i will take this on board as something to do from now on.

In the laser circuit above the 741 inputs are close to half the supply voltage and the signal is small, a microphone output.

So this should have been a dual supply ...this was the first 741 circuit i had seen with a single supply i always thought the only way to use a 741 was with a dual power supply .

I tried btw replacing the 5534 with a few different opamps it has a ca3140e this one and the Tl071 seemed about the same the 741 didn't work with the 22k feed back resistor ,has a 100k in the laser circuit so might be that .

The major concern I have with the cap on the output of the 4060 is the the internal current-limiting circuit is being used to provide a current source to charge/discharge the cap. It is meant for protection of the device and under normal circumstances should never be activated, it also increases heat dissipation and power supply current. Yes, change it to a 555.

I tried a 0.5 to 20 volt version of the 4060 ic and have to say that its output was not as good perhaps half the 5 volt version and frequency went down to 27 khz tops ...theres a lot of difference when it comes to different versions of the same ic ...i found this in transistors as well made by different companies .
Ok my test on the other 4060 and your advice .....a 555 it is .

I've still to look at the last part, I got dragged away by something else...

Steve A.

[/quote]

NO worries Time to make an oscillator now.