NBTV Scope Evolution

[Steve Anderson]

A bit later...but not that much...

Using this Frame Displacement Modulation for SSTV on a P7 CRT as you suggested is something I had not considered, and is a valid idea. As the Copthorne McDonald standard was a 1:1 aspect ratio it wouldn't matter if you did it vertically or horizontally. In the light of this when I return to the 3" CRT SSTV monitor I will certainly consider it - it does away with all that mucking around with the grid circuit and providing the high-voltage isolation required.

Good suggestion Harry.

Steve A.

[Harry Dalek]

Harry, I've spent a little time studying Jeremy's design and I've sent him an e-mail asking if he used an additional negative (probably -12V) supply. If he didn't I'm starting to wonder how it functions. I'll await his reply and given some work-bench space maybe breadboard it.

The concept is definitely sound, I'm trying to simplify the last stage somewhat.

Steve A.

Yes that would be good to know if he used a negative supply ..... if so any one copying would perhaps fail without that knowledge...

If you do breadboard the circuit its sure the way to find out ,i should of copied it in hindsight ,my thinking was it looked to me like most of it was just another way of doing PWM...still not sure about how the PWM square wave is increased in amplitude 3 X perhaps the sawtooth is decreased with the width control then getting 2 scans adjusting the height control get a correct aspect ratio speculating.

Only had a little time today looking at the waveforms on my shed crt scope i still don't see correct PWM the Triangle wave looks a worse and the pulse width of the not so good looking square wave does not seem to changing width just shaking on video ,could be a mistake some where or bad ic .

I will check its still doing Z axis modulation then its down to perhaps the oscillators wave shape ? or the lm311/
\

[Steve Anderson]

Well I've not had a reply from Jeremy as yet...we'll see. I have had a go at coming up with an alternate last stage for this and it's no simpler than Jeremy's, so it's probably best to stick with something that has been demonstrated to work.

Steve A.

[Harry Dalek]

Well I've not had a reply from Jeremy as yet...we'll see. I have had a go at coming up with an alternate last stage for this and it's no simpler than Jeremy's, so it's probably best to stick with something that has been demonstrated to work.

Steve A.

Yes i think so to, i think i could go back to what i was doing but it sounds best to see how jeremys works first ...have the schematic parts so may as well ...

I have started on constructing it only up to the first opamp and black level control via that npn transistor.

using strip board this strip board so first for me .

Perhaps jeremy might let us post the schematic ?

[Steve Anderson]

I've had a reply from Jeremy, the pertinent part follows...

I'm pleased the shift mod principle is still of interest and I'm always interested in people's use of it. The LM348 is indeed described on National's spec sheet as 'a true quad 741'. (The quad on my circuit could perfectly well be replaced by separate 741s if more convenient).

No - I hadn't used a negative line. My integrator uses the fact that the 741 behaves as if its non-inverting input rests internally at 2 volts above its negative pin. This means that earthing the charging resistor drives the output positive.

I suppose we are used these days to much more modern op amps capable of swinging close to either rail. But that feature of the 741 series is a useful way of often avoiding the rather tedious negative supply line!


Best regards

Jeremy

So no negative supply required and it should work as advertised...

I'll ask Jeremy if we can post the circuit or the whole article, but I get the feeling he would prefer that we didn't. It could be said that this may affect new memberships and/or orders of the back-issue CD. The whole of the newsletter is copyrighted anyway.

Steve A.

[Harry Dalek]

Thanks Steve for posting Jeremys reply
No problems i think we all have it to review and rest will have to join Ok .
i knew looking into the quad opamp it was like 4 741's and could be my mistake in the last circuit as i didn't use one ..i didn't know why sounds like the old 741 still has life a useful life ..i had 2 348s so i am using one in this ,
Very good no negative supply that makes life a little easier on this one .
Building more of it at the moment ..

Update ....working on the sync seperator,noticed Jeremy saying it needs a germanium PNP transistor good reason to use one i think

[Steve Anderson]

Getting Germanium transistors can't be that easy these days, but a source could be early transistor radios from the late 50s/early 60s. I haven't used any in many decades.

Seeing as you're using stripboard a useful tool is a 'spot-face cutter' for breaking the tracks. Picture attached. Mine is a standard 3.5mm drill bit with a plastic handle molded on. Others seem to use a fluted design, but I guess they work equally well. They're quite cheap even though I have to get them from the UK, not seen them here.

They can be sharpened on a standard grinder...usually every 2-3 years. Probably more frequently if using the fibreglass-based stripboard versions, which I don't.

Steve A.

[Harry Dalek]

Getting Germanium transistors can't be that easy these days, but a source could be early transistor radios from the late 50s/early 60s. I haven't used any in many decades.

I am not sure Steve but are all the tin can types Germanium .i was thinking something like a BC177 to 179 could be used to but i was not sure if they were Germanium so i made sure what i used was 2Sb22 i know thats a Germanium...Jeremy saying use something like a OC71 thats a bit rare even for me but i have a few old tin can transistors/.....Yes all scrapped from old radios or ampifliers . I recall using them years ago and they were easy to distroy well in my hands they were !

Seeing as you're using stripboard a useful tool is a 'spot-face cutter' for breaking the tracks. Picture attached. Mine is a standard 3.5mm drill bit with a plastic handle molded on. Others seem to use a fluted design, but I guess they work equally well. They're quite cheap even though I have to get them from the UK, not seen them here.

I had one laying around seemed the right size i was thinking this circuit might fit on ,the one and only i ever had ! so i was thinking there were 2 ways to brake the tracks one a stanley knife which i didn't like the idea of in case i got cut ! other a drill so a low tech way copying your tool ...i didn't know they made those.I do double check with my meter when i brake a track for no conductivity its easy for forget using this strip board ....I need to test soon make sure i have not made a mistake sync sep is finished .

They can be sharpened on a standard grinder...usually every 2-3 years. Probably more frequently if using the fibreglass-based stripboard versions, which I don't.

Well since this drill has that bit at the end it makes this job of hand drilling easy enough ...but i do like your tool !

[Steve Anderson]

Earlier Silicon transistors were also in 'tin cans' though generally smaller, e.g. the BC107/8/9 series. I just looked up some prices of Germanium transistors in the UK...horrific! There surely is a better way than using Germanium in this day and age. Schottky diodes? (1N60 and the like).

If the European type number starts with an 'A' (AC126) it's a 99% chance it's a Germanium device, if it starts with a 'B' (BF173) it's Silicon.

It sort of followed on from the European valve/tube designation system...really sorta though. A resemblance, but that's all.

The US and Asian semiconductor numbering system is totally haphazard it seems, if there's a logic to it it defies me!

As for destroying Germanium devices, I recall the early days of CMOS and MOSFETs...it was as if you only had to look at them and they died!

Steve A.

[Harry Dalek]

Yes Steve they would be pretty rare now ,may be ebay might be cheaper but more than likly second hand that someones kept and corroding away for 40 50 years in a shed .

I like the old boxes they came in like valves ,i wonder if that box was just for one transistor !


OK i fired up he circuit that i have made so far no forgotten tracks on the strip board shorting any thing out ,so that was good ...

Sort of worked ( there was nice sync pulse but video or noise between)looking at my little scope then i noticed on the schematic the frame sync selector the trimmer pot should be a 4.7meg i put in a 4.7k WHOOPS need to put my glasses on !

Now 4.7meg trimpots are as rare as oc71 transistors in my neck of the woods so i am going to have to be creative here use a 1meg and play around with some high meg fixed resistors ..bit of a bugger even my resistor wheel only gos up to 1meg ,i have resistors up to 10meg so i will have to do a bit of trial and error, at the moment i have a 1meg trimmer and a 1.5 meg as a quick replacement for the 4.7k mistake! its a lot cleaner just a sync pulse but the now 1 meg trimmer still out of range as it has no effect on the pulse if its to adjust the width or any thing .

So when i have some more time tomorrow i will try some other higher resistors see if i can get some adjustment happening ...

I tried to be descriptive on the photo scope shots where in the schematic points they were taken ..forgive the little scope only time for card table work today .

[Steve Anderson]

Early transistors were packaged individually, I used to have some in the 60s. They were so expensive they justified individual boxes. But as we all know the price per device today is nothing, what you're paying for is mainly distribution, packaging (still), profit and inevitably taxes. In the late 50s a single transistor was about a sixth of the average blue-collar workers weekly salary!

Now you can literally buy billions for a few dollars - an 8GB USB flash drive for example.

Agreed, my supplier here only goes up to 2M in cermet trimmers, and only 500k in carbon. Even then, the 2M cermet trimmers are only available in either 15 or 25 turn varieties, no single-turn.

You might try reducing the trimmer to 470/500k, the resistor connected to the wiper to 15k, and increasing the cap to 33nF...should work, but not 100% certain. Or pro-rata for other values,

A wide range like this is required due to the variability of the logic-threshold of the 4011, it's very loosely defined, it can be anywhere between 20 and 80% of the supply voltage. This is generally typical of the whole 4000-series logic devices.

As much as the digital Tek scope I bought a few years ago is a huge leap forward compared the 20-year old analogue CRT version it replaced, there are times when the analogue version is better suited. So once in while it gets dusted off and fired up. Tek have a range of 'scopes that address this difference, it's mainly to do with the display to human eye interface. It's not easy to explain but there are times where you don't want to see fast rising or falling edges. The downside is that the Tek DPO range (Digital Phosphor Oscilloscope) are 4-5 times the price of the standard range of digital 'scopes of otherwise similar specifications...well, that was the case last time I checked.

So there's nothing wrong with using the smaller digital devices, just sometimes the good old analogue versions are better suited. In other words you need to be aware of the limitations of both.

Steve A.

[Harry Dalek]

Hi Steve yes i suppose a different types of scope for different jobs ,i just use this little one for inside work .

Agreed, my supplier here only goes up to 2M in cermet trimmers, and only 500k in carbon. Even then, the 2M cermet trimmers are only available in either 15 or 25 turn varieties, no single-turn.

Yes not much luck getting any thing higher than 1 meg here ,i have an old 2meg thats the highest i have ever seen in variable resistors.

You might try reducing the trimmer to 470/500k, the resistor connected to the wiper to 15k, and increasing the cap to 33nF...should work, but not 100% certain. Or pro-rata for other values,

Thanks for that ,i will give it a go i tried a 3.9meg fixed and a 1meg trimmer see rushed drawing sorry ! still does not adjust that well ,the pulse width looks wider ,a little untidy see the effect at the out put of the 4011 picture , but the pnp transistors collector the pulse is wider and i am not sure why the trimmer effects the pulse here unless a slight voltage change does as i see adjust the amplitude of the pulse .

I will try your idea Steve see what i get .....well update here changed the parts it didn't really work random waveform i lost the 400hz pulse from the sync seperator and no real adjustment ...again i am a bit worried as to why the 4011 frame sync selector part of the circuit effects the PNP transistor so much ..i might disconnect the 4011 from the pnp as it should just work there as a sync seperator .

[Steve Anderson]

Something I just thought of...and the photos above seem to confirm this...with these high resistance values the scope input, even if it's 10M will possibly load the waveform such that it can't reach the logic threshold of the 4011. It's easy to get caught out by this, I have several times. In most circuits you can ignore the scope loading, but here, maybe not.

You can get 'active probes' for 'scopes which have input resistances of gig-ohms, 1000M, but they're not cheap, often limited to a maximum input of 50V and have a limited bandwidth, say 10MHz...unless your pockets are very deep.

Steve A.

[Harry Dalek]

Something I just thought of...and the photos above seem to confirm this...with these high resistance values the scope input, even if it's 10M will possibly load the waveform such that it can't reach the logic threshold of the 4011. It's easy to get caught out by this, I have several times. In most circuits you can ignore the scope loading, but here, maybe not.

You can get 'active probes' for 'scopes which have input resistances of gig-ohms, 1000M, but they're not cheap, often limited to a maximum input of 50V and have a limited bandwidth, say 10MHz...unless your pockets are very deep.

Steve A.

OH this is a problem...Steve i don't think my shed scopes would be any better ,i did notice if this is the case the led did drop intensity a bit when i was testing on pin 11 so that would make sense .

This stage has got me stuck a bit ,i feel the sync seperator pnp part of the circuit is more important to get right ...this was fine when i used 1m and 1.5m fixed resistor on the frame sync selector part ,had no time today to work on the circuit but i will take the 4011 out of the circuit and see .....the sync seperator should not be effected by the next stage of the circuit.

Perhaps there might have been a difference in the 4011 Jeremy used to mine bit like the 4060 ic i used a while back may be it needs some resistance or to be ac coupled between the sync seperator and the frame selector circuit ..

The circuit is some what explained but not enough for electronics for dummies like me ..on the frame selector ,i am not great at the digital ics i suppose it is explanatory the 4.7meg trimmer adjustment just delays detection period....i will check this also on my dual trace scope if theres any difference to what gos in to what comes out ...

If i have time tomorrow i will look at some of these points .

[Steve Anderson]

For the moment leave the 4011 in circuit but disconnect the diode on the output which feeds the capacitor. What you should see on that output will be sync pulses roughly coincident with the input video, possibly narrower though, at the usual 400Hz. The edges should be nice and clean and fast - well as fast as 4000 stuff will go.

Reconnect the diode and what you want on the output of the third 4011 gate is a positive pulse at 12.5Hz after fiddling with the 4M7 value of the pot...or my suggested values of earlier. Too low a value of the pot and you'll get 31 pulses per frame, too high, nothing. If using the higher resistance values do not put a 'scope on the capacitor. With my suggested values you should be OK to do this, disconnecting the 'scope may still require a slight adjustment though.

Steve A.