Construction Diary -- Part 1, The LED matrix/display

Original build of a televisor by a complete novice.

Moderators: Dave Moll, Andrew Davie, Steve Anderson

Postby Andrew Davie » Tue Apr 03, 2007 1:26 am

gary wrote:Andrew, have you looked at your 'clean' 12V unconnected to the LEDs with the 'scope? If not then it is possible that this supply is not so clean after all so have a look at it.


I had done this, but having just re-checked, I can confirm that it is a nearly dead-flat 12V measured on the scope, and 11.81V via multimeter.

gary wrote:As a sanity check, if you have a couple of 9V batteries handy you could connect them in series to provide a guaranteed clean DC supply to the LEDs.


I will do this tomorrow.

gary wrote:At first glance the level of the unwanted signal seems to be too high to be just 'hum'. Its almost as though the rectifier is halfwave rather than fullwave although if this were true you would still expect it to be much smoother due to the smoothing capacitor.

It is important now to identify the signal at the emitter of the transistor, I am pretty sure that the intention of the circuit to not have any syncs at this point (truncated by the .7V drop across the transistor base-emitter). If they are there then the DC level of the input to the transistor base is not right (this could be just a matter of setting the brightness).


I had wondered, too, if I should be seeing synch signals, and now I think I understand how they're being 'dropped', thanks. I have done a whole new suite of pictures, each with their own comments, below. I have been more careful to identify the readings this time.

gary wrote:The images of your scope seem to indicate that it is out of focus or maybe that is just an artifact of the snapshot itself.


The scope is producing quite crisp images. Good enough, anyway. I am having trouble holding the probes on the correct pins while at the same time taking the pictures with the correct focus :) The signal doesn't synch, either -- so I am using the knob to reduce the scan width to try and get the image to remain stable oncreen. Usually when I am photographing it, it is moving horizontally, slowly.
Attachments
IMG_7486.JPG
Signal seen at emitter with LED matrix connected. There apear to be two components here -- a large sine wave at 50Hz, I'd say... and the NBTV signal superimposed.
IMG_7486.JPG (25.52 KiB) Viewed 17084 times
IMG_7485.JPG
Same is seen at OP#2pin6, base, emitter, collector (measured with black on ground, and red to appropriate pin). 0.5V/div
IMG_7485.JPG (22.53 KiB) Viewed 17084 times
IMG_7484.JPG
op#2, pin#2. Here the signal is sitting at 2V or so with a 0.7V synch.
IMG_7484.JPG (23.65 KiB) Viewed 17084 times
IMG_7483.JPG
Measured at OP#1, pin#6. This one has me stumped. I could not get a NBTV-looking signal, just this spike.
IMG_7483.JPG (23.51 KiB) Viewed 17084 times
IMG_7482.JPG
at OP#1, pin#2. 0.5V/div. signal sitting on 2V with a 1.4v synch.
IMG_7482.JPG (24.86 KiB) Viewed 17084 times
IMG_7480.JPG
video in to NBTV LED driver board. 2V/div.
IMG_7480.JPG (24.84 KiB) Viewed 17083 times
User avatar
Andrew Davie
"Gomez!", "Oh Morticia."
 
Posts: 1590
Joined: Wed Jan 24, 2007 4:42 pm
Location: Queensland, Australia

Postby Klaas Robers » Tue Apr 03, 2007 4:38 am

Andrew,

now you have an oscilloscope you can debug the circuits the way it should be done, from the input to the output.

First don't bother about the LED-driving transistor. Desolder it or at least desolder the base wire from the PCB. If you can follow the signals from the input to the base-connection then the remaining point is the transistor. If the base-signals are Ok, the transistor should work, or not, but then we know that almost everything is working fine.

I am going to give you some tests to do. We will check the different stages one by one. First I have to print the circuit diagrams of the sync-board and the LED-driver board for myself. Then I can see what to do for you.
User avatar
Klaas Robers
"Gomez!", "Oh Morticia."
 
Posts: 1656
Joined: Wed Jan 24, 2007 8:42 pm
Location: Valkenswaard, the Netherlands

Postby Klaas Robers » Tue Apr 03, 2007 8:01 am

Your oscilloscope pictures look well, but lets have a closer look.

Select the NBTVA CD number 1, tracks 20, 21, 22. This is the grey bars and this is a suitable signal to look at. You will see sync pulses negative and 8 different levels during the line, black to white. A saw tooth would be better, but is not on the CD.

At the output of the CD-player this should be 1.4 volts from bottom sync to the highest line of white. To sync the oscilloscope turn the trigger level control as low as possible, while synchronisation is still obtained. This should give you a stable trace on the screen.

At the output of the contrast control there should be a signal about 0.5 volt peak to peak. If not adjust control pot.

At the output marked video output of the sync-board should be again a 1.4 volt video signal. Adjust the contrast pot to make it as good as possible 1.4 volt peak to peak. And yes, this signal will be shifted to about 6 volt positive as a whole. Opamps only work between 0 and the supply voltage, now 12V. So 6V is well in between these limits. We will look at the sync board later. I heve seen on your oscillograms that this works, but it should be adjusted well now.

Over to the LED-board:
At the video-in pin the 1.4 volt video should still be there. The pin "clamp select" should be connected to ground (0V).

At pin 2 of the first opamp and pin 3 of the second opamp you should find the same amplitude of video back, so 1.4 volt peak to peak, however shifted in level. The level (called DC-level) should be adjustable by changing the setting of the brightness pot (sun). Adjust the brightness pot to a bottom sync on second opamp - pin 3 = 1.6 volt. If still Ok:
the bottom sync = 1.6 volts
the lowest line (black) = 2 volts,
the highest line (white) = 3 volts.

Eventually adjust contrast pot and brightness pot fine to obtain this.

Now the output of the second opamp = base of driver transistor should show sync pulses:
bottom at 0 volts,
lowest line (black) = 0,7 V,
highest line (white) = 2,7 V.

If you are this far the video processing is working fine.

Select track 35 - 37. This is the test pattern. It contains a staircase signal. Adjust the scope trigger level lowest possible for a stable signal.
User avatar
Klaas Robers
"Gomez!", "Oh Morticia."
 
Posts: 1656
Joined: Wed Jan 24, 2007 8:42 pm
Location: Valkenswaard, the Netherlands

Postby Klaas Robers » Tue Apr 03, 2007 8:19 am

And then the transistor.

Reconnect the base. Now connect the collector (LED-NEON-pin) to your own made stabilised 12V, without LED's in between! Run again the tracks 35 - 37 and check the base voltage with the oscilloscope. It should be unchanged.
Now check the emitter voltage (the 150 ohm resistor). There should be the same signal, however shifted 0,7 volts down. So there are no sync pulses any more!! The staircases will be visible still, withe their lowest stair down to earth.....

Caution, the transistor might run warm to hot. Don't do it too long. A "heatsink" some aluminium might be bolted to the transistor. There is the hole for.

If this works fine the LEDs. Connect the LEDs from the LED-NEON-pin to the unstabilised +17V.!!!! Not to the 12V. This is because the transistor needs some voltage left for operation. The LED-array that you made works on 12V without a transistor (3 x 4 volts), but with the transistor you need more volts. From the 17 volts going down by 12 volts still 5 volts is left. This is enough. The voltage on the collector should never go lower than 4 volts.

When this works it should be that when you stop the CD-player the LEDs go out, on tracks 2-4 (black) they should glow very dim (adjustable with the brightness pot), on tracks 5-7 (white) they should be very bright. Try the different tracks!!!

But they can give even more light. But first this should work.......
User avatar
Klaas Robers
"Gomez!", "Oh Morticia."
 
Posts: 1656
Joined: Wed Jan 24, 2007 8:42 pm
Location: Valkenswaard, the Netherlands

Postby Klaas Robers » Tue Apr 03, 2007 8:24 am

Oh Andrew,

don't bother about the spike. This is Ok. It is the signal that recharges the 1,8 uF capacitor at the peak of the sync. This chargeing keeps the voltage of the sync pulses to the voltage that was adjusted by the brighness pot. I saw that and I knew that the circuit worked at least to this point.

The circuitry around the first opamp is rather cluttered and difficult to explain. But when it works.......
User avatar
Klaas Robers
"Gomez!", "Oh Morticia."
 
Posts: 1656
Joined: Wed Jan 24, 2007 8:42 pm
Location: Valkenswaard, the Netherlands

Postby Andrew Davie » Tue Apr 03, 2007 8:14 pm

Klaas Robers wrote:To sync the oscilloscope turn the trigger level control as low as possible, while synchronisation is still obtained. This should give you a stable trace on the screen.


:roll: One more "mysterious" knob on my oscilloscope explained! And how much easier it is to see the signal waveform when synched! I was previously obtaining synch magically by reducing the trace width which sometimes 'worked', sometimes didn't.
User avatar
Andrew Davie
"Gomez!", "Oh Morticia."
 
Posts: 1590
Joined: Wed Jan 24, 2007 4:42 pm
Location: Queensland, Australia

Re: The Art of Electronics.

Postby Andrew Davie » Tue Apr 03, 2007 8:16 pm

Steve Anderson wrote:Here's a link to the book...
http://www.artofelectronics.com/
Steve A.


I have now purchased a copy of this book through eBay (from India). Cost was US$29 including postage. Will advise if/when it arrives.
User avatar
Andrew Davie
"Gomez!", "Oh Morticia."
 
Posts: 1590
Joined: Wed Jan 24, 2007 4:42 pm
Location: Queensland, Australia

Postby Klaas Robers » Tue Apr 03, 2007 9:59 pm

I feard already that you didn't know that control. It works together with a flip switch marked + and -. It works like this:

When the input signal (selectable with an other switch) crosses a certain voltage (adjustable with "trigger level") in the positive (switch on +) or negative (switch on -) direction, the horizontal scan starts.

So when you turn the trigger level control you will see the start of the scan moving up and down. Adjust the horizontal position to a position such that you can see the start of the scan. When the trigger level is half way the sync pulse the display is most stable. Then the video part is not affecting the triggering.

When no triggering occurs, e.g. with no input signal, the scan goes to "automatic scanning" or it just stops scanning at all. This is selectable by another switch that reads "auto" and "trig" or similar words.

Yes an oscilloscope requires that you get a well experienced operator before you get what you want. This oscilloscope features also a delayed time base. It starts working when you pull and rotate the time base rotary switch. Forget it for the time being.
User avatar
Klaas Robers
"Gomez!", "Oh Morticia."
 
Posts: 1656
Joined: Wed Jan 24, 2007 8:42 pm
Location: Valkenswaard, the Netherlands

Postby gary » Tue Apr 03, 2007 11:02 pm

gary
 

Postby Andrew Davie » Tue Apr 03, 2007 11:16 pm

Klaas Robers wrote:Now the output of the second opamp = base of driver transistor should show sync pulses:
bottom at 0 volts,
lowest line (black) = 0,7 V,
highest line (white) = 2,7 V.

If you are this far the video processing is working fine.


Thank you Klaas for this marvellously detailed walkthrough :)

I have everything OK up until the base of the driver transistor (=pin 6 of OP#2). I did the measurements with the transistor in place; I did not want to remove it if nothing was wrong. Something WAS wrong, so I removed it, and did the measurements again.

Aside from a 0.1V difference (1.3V instead of 1.4) at pin 2 of OP1 and pin3 of OP2, when there was 1.4V at the video in on the LED board (ie: there was a 0.1V drop between input and those OP pins), everything at the beginning looked OK.

...up until the output of OP#2, as I said. Instead of the lowest line being at 0.7V, it is at 1.2V, and the highest white line is at 3.1V. So, the signal is being amplified 'too much'?

One other oddity. The 8 'lines' representing each vertical 'colour' are not totally horizontal on the 'scope as I expect. They slope gradually upwards, from left to right. This would mean, I'm sure, that the image will be brighter on the top than on the bottom.
User avatar
Andrew Davie
"Gomez!", "Oh Morticia."
 
Posts: 1590
Joined: Wed Jan 24, 2007 4:42 pm
Location: Queensland, Australia

Postby Klaas Robers » Wed Apr 04, 2007 8:00 am

The 1,3 volt in stead of 1,4 volt at the opamps inputs can be corrected by a tiny bit more input gain, oh sorry, contrast setting, until you get the 1,4 volt bottom sync to top white. However, 10% deviation is not at all a problem......., remember?

The amplification of opamp 2 is correct, 2x, because 1 volt variation at the input (2 volt to 3 volt) gives 2 volt (ah, 1,9 volt) difference at the output. The amplification is correct, however there is some DC offset (the voltages are all somewhat higher).
When you adjust the brightness pot slightly you will get the lowest line at 0,7 volt and the highest at 2,7 volt. May be that the resistor values are not exactly calculated by me, I am going to redo that. But the brightness setting is doing it too.

The rising lines is strange. I would expect them to drop somewhat, but not to rise as they do. Is this also visible at the input signal? When you follow the signal, is there a point where this starts to happen?
User avatar
Klaas Robers
"Gomez!", "Oh Morticia."
 
Posts: 1656
Joined: Wed Jan 24, 2007 8:42 pm
Location: Valkenswaard, the Netherlands

A re-start?

Postby Steve Anderson » Wed Apr 04, 2007 9:18 pm

Gents,

Could we re-start this thread? Sifting back through some eight pages and trying to find out what you are doing is not easy.

i.e. What are you trying to do? What circuits are you using? A snapshot of where you are and I'll put in my tuppence for what it's worth.

Steve A.
User avatar
Steve Anderson
"Fester! Don't do that to 'Thing'"
 
Posts: 5361
Joined: Fri Mar 30, 2007 10:54 pm
Location: Bangkok, Thailand

Postby Andrew Davie » Thu Apr 05, 2007 2:09 am

Oh, so close! I tried connecting the LED circuit powered from the 17V supply -- no go. Nothing at all. This seemed strange to me, especially as I saw a signal all the way to the transistor (which I have re-inserted). So much inserting and removing of stuff, and resoldering, that the board looks a bit like it's been through the wars.

Anyway, after the LED didn't light up, I got the bright idea (pun intended) of connecting directly to the collector of the transistor. And... voila. My lights are flashing nicely, and stop when I stop the CD player. And flash differently with different tracks. In other words, it's working! It turns out that the trace from the collector to the pin for connecting to the LED circuit is damaged/broken. Probably from removing/inserting the transistor twice.

Anyway, I know now that the fundamentals are now all working, and I can proceed to the next stage, which will be learning how to control the motor, and spinning the Nipkow disc at the correct speed.

Wahoo! Many thanks to everyone who's helped so far.

I will address some of the issues related to the problems in the signal (sloping upwards lines that should be horizontal, 'phantom' negative voltage signal bits) at some later point. For now, time to move on.
User avatar
Andrew Davie
"Gomez!", "Oh Morticia."
 
Posts: 1590
Joined: Wed Jan 24, 2007 4:42 pm
Location: Queensland, Australia

Re: A re-start?

Postby Andrew Davie » Thu Apr 05, 2007 2:13 am

Steve Anderson wrote:Gents,

Could we re-start this thread? Sifting back through some eight pages and trying to find out what you are doing is not easy.

i.e. What are you trying to do? What circuits are you using? A snapshot of where you are and I'll put in my tuppence for what it's worth.

Steve A.


Hi Steve

I understand where you're coming from with this request, but this whole thread is my journey through the construction process. It belongs together, and is intended for someone to read from beginning to end (if they wish) so they can see the pitfalls and solutions as they happened. If each problem was split into separate threads, the forums would be cluttered with this 'beginner stuff' that most will not be interested in, and any cohesion in the understanding of the development of this particular monitor would be lost.

On the other hand, it's clear that you have a strong understanding of electronics and I'm hoping not to turn you away from assisting :)
User avatar
Andrew Davie
"Gomez!", "Oh Morticia."
 
Posts: 1590
Joined: Wed Jan 24, 2007 4:42 pm
Location: Queensland, Australia

Postby Klaas Robers » Thu Apr 05, 2007 5:28 am

Andrew, I recalculated the resistors around the "second opamp". Things will work more conforming the voltages that I gave when you change the resistor 39k into 33k. Then you will be able to give the brightness pot a swing to a somewhat higher setting. The clamp will also work somewhat better.

Have you checked the voltage on the collector of the transistor? You will see a kind of video signal, but "upside down" (inverted) The voltage should nowhere be lower than say 5 volts.

The top brighness (tracks white) will be lower than what is possible with these LEDs. The max current is 40 mA devided over 8 parallel series. If you disconnect 4 series you will see the brightness of 6 LEDs burning on 20 mA each. It is easier to make longer strings and increase the unstabilised voltage than increase the current.
User avatar
Klaas Robers
"Gomez!", "Oh Morticia."
 
Posts: 1656
Joined: Wed Jan 24, 2007 8:42 pm
Location: Valkenswaard, the Netherlands

PreviousNext

Return to Andrew Davie's Televisor #1 (2007)

Who is online

Users browsing this forum: No registered users and 0 guests