NBTV without any moving parts. THE NIPTRIX

[Viewmaster]

Very nice Albert. It's thanks to your influence I have refrained from throwing away the drinking straw that came with a recent visit to McDonalds. It 'just might' come in useful some day.

<pedant_alert> You have got me wondering why you have 'only' 1532 LEDs to wire up when I thought it should be 32x48=1536. </pedant_alert>

Have you got a missing LED or sync pulse in there somewhere ... or have I got a screw loose?

Steve O

PS. As for my line-o-light, I've now settled down to doing 10 LEDs a day... it's all I can stomach.

1536 it is. O deary me, I have to solder up an extra four LEDs....
...on top of which I now find that I have 48 PNP trannies wired the wrong way
My wife thinks I'm nuts...I do too

[gary]

My wife thinks I'm nuts...I do too

The fact that you are a member of this forum bears witness to that...

[Viewmaster]

My wife thinks I'm nuts...I do too

The fact that you are a member of this forum bears witness to that...

Yes Gary, and I have just realised that my method of switching the Niptrix is badly flawed.

Grant Dixon used PNP trannies because the 4515s he used for 48 row switching have LOW outputs whereas my 4017s all have HIGH outputs.

Incorrectly I followed, using PNPs toooooo!

SO, I need to cut out all my 2N3096 PNP trannies and replace them with
NPN 2N3904s.

So Gary, I am nuts and an ignorant one at that.

[Panrock]

SO, I need to cut out all my 2N3096 PNP trannies and replace them with NPN 2N3904s.

That's a swine. Bad luck. When putting it right, if you work at my ten-a-day rate then perhaps it won't seem so bad... but it still soon will be done.

So Gary, I am nuts and an ignorant one at that.

The man who never made a mistake never made anything.

All the best,

Steve

[Viewmaster]

SO, I need to cut out all my 2N3096 PNP trannies and replace them with NPN 2N3904s.

That's a swine. Bad luck. When putting it right, if you work at my ten-a-day rate then perhaps it won't seem so bad... but it still soon will be done.

So Gary, I am nuts and an ignorant one at that.

The man who never made a mistake never made anything.

All the best,

Steve

..........and the man who makes too many mistakes never makes anything either.

Not too bad a job to do, I shall chop off all the 48 x PNPs at their tops leaving maximum leads intact. Then just solder in the new NPNs.

I must write out 100 lines,"Make sure you get them the right way round, you nut."

[Panrock]

I must write out 100 lines,"Make sure you get them the right way round, you nut."

What with my 'loose screw' and your 'nut', it would seem we are well matched here!

Steve O

[gary]

Dear me Albert, that's rotten luck! You couldn't sneak an inverter in there could you? I suppose that would be no less work than replacing the transistors. The 4515s aren't pin compatible either so you can't just switch them in...

The bad news is you now have a lot more work to do.

The good news is you are now a very well practised solderer(?)... you should do it in no time.

A little slap on the wrist though - didn't you say somewhere you were going to test as you went along?

[Viewmaster]

As you say Gary, putting in an inverter would mean as much work as changing them. It's quite an easy job because I don't intend to actually remove the PNP soldered joints. Just leave their leads sticking up and solder new NPNs in place.

Re testing. I tested out the 32 column NPNs and all seemed OK, so I was confident that the 48 rows would be OK too.
In fact they are except the o/p are all on and just flick off and on again at every clock position.

It was just TWO words in Grant Dixon's write up that I ignored. The LOW and the HIGH outputs.
I shall never get the hang of transistors. Black devils

PNP.NPN. Grrrrr......
Give me good old KT66s every time where the electrons always travel in the same direction, the grid always controls the flow and it's ON when its on and OFF when its off. And it has that friendly red hot glow

[Viewmaster]

48 short legged 2N3096 PNP trannies for sale.
Only used once.
Buy early for Xmas and beat the rush.
Would make a great stocking filler.

[Viewmaster]

The good news is you are now a very well practised solderer(?)... you should do it in no time.
?

All done now.
Finished soldering in the last of the new 48 x NPN little blighters this morning, just after the postman delivered a few more.

Except for two base R's from a 4017 going to the same Tranny and a low o/p from one pin all the clocking seems OK.

The master reset using OR gates seems to work too so I do hope that the missing frame pulse inserted will do the trick at the start of every frame, maybe with a slight adjustable timing shift to get the first frame pixel to be clocked exactly when it should to ensure a correct picture

A little slap on the wrist though - didn't you say somewhere you were going to test as you went along?

Well Gary, even my wife too, told me to test after I had soldered in but 8 new NPNs.

Of course, until such time that I have a full column of 48 x LEDs installed will I be able to see if all the clocking is correct from bottom up to the top of the display, without any in the wrong order sequence.

[gary]

Well Gary, even my wife too, told me to test after I had soldered in but 8 new NPNs.

Behind every great man...

Of course, until such time that I have a full column of 48 x LEDs installed .

Dear me, I had led* myself into thinking you had done the LEDs already.

BTW you seem to have a very fast post there...

*(pun intended)

[Viewmaster]

Dear me, I had led* myself into thinking you had done the LEDs already.

The LED matrix will work providing all joints are OK. So I tackle the questionable part first.
It's the sync/clocking circuits where trouble may lay, hence building these first.

Bear in mind Gary, that the 4017 cascade method of switching many LEDs has never been used for NBTV before so I may be taking a big risk in using this method, which I slightly adapted to reset at the start of NBTV frame.

Not having the knowledge to design my own complicated cct from scratch, I just hooked a cct off the net, after reading about it from many sites (including 4017 manufacturers) as a cross check .

I had the mouse in one hand and fingers crossed in the other as I still don't understand how it works really. Maybe the odd glitch here and there will be my downfall, in which case I shall have to revert to the tried and tested Grant Dixon cct.

Added.........re fast post.
My PC battery is down so I am resetting the PC clock at every boot up. Maybe I set it a bit wrong at times.

[Viewmaster]

Well, here is the completed 4017 cascade switching board, all tested and passed inspection.
All the 1536 LED matrix and sync/timing etc still to be done though.

If this board doesn't work I may send it to the Tate Modern Gallery in London.
I shall call it, "Transistorised Ribbon Fountain."

[gary]

A thing of beauty Albert, but I now feel justified in my own procrastination in embarking on this kind of project - it used to be cost but now it is complication (and old age).

It should work Albert, but as you say the odd glitch/timing issue/etc could be a problem if you don't have access to a logic analyser or at least a CRO (preferably the storage type).

Any way step by step and you'll have your self a great Christmas present! Best of luck.

[Viewmaster]

A thing of beauty Albert, but I now feel justified in my own procrastination in embarking on this kind of project - it used to be cost but now it is complication (and old age).
.

Old age? Excuses, excuses.

Bet I'm well senior to you, Gary, so age is no excuse to tackle anything........except Mount Everest maybe!

Complications just keeps you mentally active in old age. So get that soldering iron out, and as Wilfred Pickles use to say on the wireless,
" Come and 'ave a go."