Page 5 of 19

Re: NBTV/SSTV-625 Up Converter MkIII

PostPosted: Sat Aug 26, 2017 8:10 pm
by Steve Anderson
As a post-script to the simulation mentioned above, below is the frequency response of both versions - simulated of course.

The red plot is of the original 5V version, the green the 3.3V modified version. As you can see they're very much the same except the gain. They're slightly different around the 'knee', this is probably due to specifying E12 resistors and 5% capacitors.

Steve A.

Just to ease any confusion, when I refer to the '5V version' I mean that it was to be used on the original MkI up-converter which used 5V for the logic. The '3.3V version' means it's for use with the MkIII up-converter which uses a 3.3V supply for the logic circuitry. The demodulator itself uses +/-12V in both versions.

NBTV/SSTV-625 Up Converter MkIII

PostPosted: Sat Aug 26, 2017 11:29 pm
by acl
All comments accepted Steve. I wonder if someone in the 'team' is willing to design a PCB for the demodulator so we can optimise the performance and repeatability of this unit as well as grounding arrangement. I'm still working out how to drive this new 'scope so I can capture waveform on USB stick.

Regards Chris Lewis

Re: NBTV/SSTV-625 Up Converter MkIII

PostPosted: Sun Aug 27, 2017 12:04 am
by Andrew Davie
acl wrote:All comments accepted Steve. I wonder if someone in the 'team' is willing to design a PCB for the demodulator so we can optimise the performance and repeatability of this unit as well as grounding arrangement. I'm still working out how to drive this new 'scope so I can capture waveform on USB stick.

Regards Chris Lewis


I could maybe do the PCB using Eagle, if Steve doesn't want to. I couldn't do the "design" but I would be happy to convert a given circuit into a board layout, as directed.

Re: NBTV/SSTV-625 Up Converter MkIII

PostPosted: Sun Aug 27, 2017 2:43 am
by Steve Anderson
I'm in the process of learning Eagle. I hope to be able to produce a PCB in due course. With at least 5 interested so far it's worth the effort, not only for this but future projects. Thanks for the offer Andrew, maybe I'll run the first (or more) outputs I generate past you to see if I've made any major drop-offs'.

I've also made contact with a forum in the US who deal in all things electronic and one of their themes is SSTV of the CopMac standard. So watch this space. Though so far I haven't seen any reference to NBTV. Perhaps it needs someone to kick it off...or at least to draw attention in our direction...

Steve A.

Re: NBTV/SSTV-625 Up Converter MkIII

PostPosted: Sun Aug 27, 2017 9:46 pm
by acl
Hi all,

As last resort I had to read the 'scopes manual and reading between the Chinese text lines worked out what I was doing wrong.I am impressed with this £200 scope, it's got lots of features I am not going to use .Anyway, here is a screen dump of the output from my demodulator for your comments. I am going to try some 100nF discs across each of the devices directly on the IC pins. Have a great weekend lads.

acl00.jpg
acl00.jpg (192.45 KiB) Viewed 16370 times


Chris Lewis

[Edit by Andrew Davie - file converted to JPG for forum inline viewing]

Re: NBTV/SSTV-625 Up Converter MkIII

PostPosted: Mon Aug 28, 2017 7:20 am
by Steve Anderson
Chris, that's looking much better. I expect some ripple in the bottom of the syncs which is no problem. Depending on your layout and the inherent capacitance between stripboard tracks we may need to clean up the high-speed outputs of the full-wave rectifiers. I'll look into this in the morning, just got back from having a good Indian curry with some friends - as one does when in the UK!

Steve A.

NBTV/SSTV-625 Up Converter MkIII

PostPosted: Mon Aug 28, 2017 4:39 pm
by acl
Hi Steve,

Just to avoid confusion. The waveform is as original. I am still to put these capacitors on once they arrive. Our local Mapin has only two in stock. Yes you cant beat good old British cuisine.

Regards Chris Lewis

Re: NBTV/SSTV-625 Up Converter MkIII

PostPosted: Mon Aug 28, 2017 6:00 pm
by Steve Anderson
A squint at the demod circuit - the waveforms at pins 7 & 8 of IC202 are large and full of harmonics. The lead-length/track-length around here should be kept to a minimum. If you look at the photo a few posts above IC202 is the lower of the two 14-pin packages. The diodes are close-in to the chip and the track minimum length and near the board edge with nothing significant around.

I'm not saying 'this is it', but it may be a factor.

This is a WIP project (Work In Progress) so these little (and not so little) wrinkles will occur as things progress.

Steve A.

Re: NBTV/SSTV-625 Up Converter MkIII

PostPosted: Mon Aug 28, 2017 6:51 pm
by Klaas Robers
Although the picture is nice and crisp now, I still don't like the noisy character of the traces. It can be that it is a matter of where you connected the ground of your "oscilloscope", but it can be too that longer strips pick up HF noise from your environment. That is the reason that I dislike the stripboard and use "dot board" and as short as possible, Teflon isolated, wires for experimental circuits.

These circuits should be built as small as possible and with as small as possible "antenna area" to the connections.

Re: NBTV/SSTV-625 Up Converter MkIII

PostPosted: Mon Aug 28, 2017 7:53 pm
by Steve Anderson
Just to reiterate, this demodulator is an updated version of the Robot 70 version, it's almost an exact copy with the exception of the full-wave rectifiers and the slightly different output filter.

I chose this as it was a benchmark or reference at the time and copied by many. Once the rest of the up-converter is well under way or completed I intend to return to it and hopefully produce a somewhat better performance version. This is why I suggested building it as a separate unit on it's own so it can be replaced if desired. Even so it does perform quite well given that it's a 40+ year old design.

I haven't come across any analogue demodulators which have any significant improvement over this one, they might be 'out there' but I'm not aware of them.

Steve A.

Re: NBTV/SSTV-625 Up Converter MkIII

PostPosted: Mon Aug 28, 2017 11:43 pm
by Harry Dalek
I see the good work chris has put into he's board i was worried about the strip board design due to the length of the resistor wires but fingers crossed the decoupling caps help .
Some times the signal looks better on a analog scope well i find on my crappy little digital scope not that the stair case looks that bad any way little bit noise .
I always check the resistance of all my ground connections some times the solder connections cause a a few ohms always get 1 or 2 ohms due to the wire when you get 3 to 5 ohm i start resoldering those connections
Talk of the custom circuit board is it best to wait Steve ? i was going to build the demodulator on matrix board .

Re: NBTV/SSTV-625 Up Converter MkIII

PostPosted: Tue Aug 29, 2017 1:53 am
by Steve Anderson
There are advantages and disadvantages with using either, matrix/dot-board or stripboard/Veroboard. I prefer stripboard as it's quite easy to create fat earth and supply rails. But it's drawback is increased capacitance between tracks. Six of one, half a dozen of the other.

If you look at the demodulator photo a few posts back you'll see there are five strips between the two 14-pin chips, three are used for ground with links at each end and in the middle. Interspersed are the +12V and -12V rails. Throw in the disc-ceramic capacitors and the two electrolytics and you have a firm clean supply and ground.

It does take a bit more effort to produce a layout on stripboard but to me the results are generally better. To each there own. Use whatever you feel comfortable with, there is no reason why one is better than the other.

The ultimate is the colander board. Where the component side has a ground plane with clearance around the through holes and either stripboard or matrix board on the other. Not often seen and expensive.

A 'proper PCB solves all those issues of course.

Steve A.

NBTV/SSTV-625 Up Converter MkIII

PostPosted: Tue Aug 29, 2017 4:19 am
by acl
Hi all,

I have noted all your comments and will try a few things later this week. I am not disappointed with my results as I am learning a lot from this experience. As this is a long term project we will see how I can improve the performance. I have moved the scope earth directly to the board and it made no noticeable difference. I will try the 100nf capacitors directly across the I c sockets and either solder 22swg tinned copper wire along the matrix board power busses or solder coat the tracks to reduce track impedance and possibly 'slap' more electrolytic capacitors across the supply rails. I will also review track lengths. One thing I always do when using strip board is to cut unused section of tracks. I still believe a dedicated PCB with suitable ground planes will be the answer.

When I was in my 'real job' I remember going on a short EMC course at York university and using a 'near field' probe on a spectrum analyser to compare the EMC emissions from various construction methods. What stuck in my mind was the huge difference between the 'wire wrap' version and a PCB version of the same circuit. I understand on several PCB CAD packages there is an optimisation auto routing function for minimising RF emissions and susceptibility. I remember the lecturer saying "if it's getting out then its easy to let in"

Regards from a Sunny UK Bank Holiday Monday

NBTV/SSTV-625 Up Converter MkIII

PostPosted: Fri Sep 01, 2017 6:57 am
by acl
Hi all,

Tried fitting capacitors across plus and minus 12 volts to ground on all the devices. 10nFs instead of the 100nF I ordered from eBay and they had minimal effect to the noise on the output signal. Attached are the before and after traces.

Regards Chris

Re: NBTV/SSTV-625 Up Converter MkIII

PostPosted: Fri Sep 01, 2017 9:35 am
by Harry Dalek
acl wrote:Hi all,

Tried fitting capacitors across plus and minus 12 volts to ground on all the devices. 10nFs instead of the 100nF I ordered from eBay and they had minimal effect to the noise on the output signal. Attached are the before and after traces.

Regards Chris


That is frustrating ...sometimes lowing the supply a touch might fix it and show up then there's a problem somewhere else , but then you still have to work out why its doing it on the dual 12 volt supply ...did you check how close or far off each resistor is to wanted values and capacitors even those are marked one value and if tested with a meter can be off ...any pickup from a connection soldered to close to another connection.
Try it on another power supply ..even battery would be better can't get cleaner .
Maybe also looking on your scope at different opamp inputs outputs might how up something as in the last opamp input output waveform would be a good one to check against each other first
Apart from that i am not sure ...