Large number of flickering/ghost inputs on Master Cards

[iwik]

Jack,
Well at least if anyone knows about J3 then o/c should.
I'm curious, the cable that you have been plugging into j3
on all 4 of your m/c to get the flickering, is that the same one for
all of your testing. If it is then try another termination pcb/cable combo
into each of J3 on all m/c.
Good to try another USB EXP as hopefully that rules out cards.
Les

[fordgt40]

Jack

When you need the card, just PM me with your address

Regards

David

[Boeing 747 Flyer]

Hello lads,

@Les: Each IDC cable that leads to each set of inputs is different, so I have been plugging a different IDC cable into J3 on each card every time.

@David: I have sent you a PM, thank you very much!

Jack

[Boeing 747 Flyer]

After reading through all of my cockpit scripts today, I have made another interesting discovery.

All of the inputs that were flickering in FSX were inputs with switch numbers that lie in the J3 range for each master card; in other words, all J3 inputs. None of the flickering inputs were found on the J4 banks.

I am going to take a look at the circuit diagrams for the master card and visually inspect the circuit tracks near and around J3 on all my cards, see if there's something special about J3 that makes it liable to the flickering whilst J4 is ok.

I certainly feel we've made a breakthrough though; the discovery that the flickering is localised to J3 only must mean something!

[iwik]

Jack,
I agree that localizing it to J3 is strange. I find it hard to believe that all 4 cards bought at different times
could have a fault around J3 input. The only way this could happen was if originally the cable that was plugged into
J3 was faulty somehow and partially destroyed that chip and so subsequently plugging into the remaining cards
destroyed these one as well. But it definitely wont hurt to have a good look at the boards for poor joints.
Did you make these yourself or brought built up one's?.
One thing you could look at is to put your scope on the +5v supply pin 20 of U10, this is the supply, how noisy is it with ALL ON commanded?.
Les
P.S A long shot but you could try replacing U10, don't take one from another m/c.
Chips like U10 that are exposed directly to the outside can quite easily be partially
destroyed.

[Boeing 747 Flyer]

Hi Les,

The problems must have been occurring on J3 even before I started swapping all the cables around for testing; it is only now, having gone back through all my scripts that I realise that all of the problem inputs are on J3. So I honestly think it's not a cable related issue, especially since J3 is not outright broken; the inputs still work when a switch is flicked on/off, they just flicker (alot!).

All of my masters/USBExpansions are pre-bought from Opencockpits.

I'll get back to you regarding pin 20 of U10.

On another note Les/David, could you please tell me what happens if, once ALL ON has been commanded and all LEDs are lit, what happens if you click ALL ON again. Any flickering on your part? Similarly, what happens with all LEDs/Outputs OFF and you click ALL OFF?

Cheers,

Jack

[Boeing 747 Flyer]

Hi Les,

I have done some work on U10 today. The oscilloscope was connected to the 5V supply pins of U10. I did not connect anything to the master card apart from the USB expansion card (which was connected to my PC) and a 5V supply; in other words, the setup was as basic/bare as possible. I'll also mention in this case that the 5V supply was from a very high quality precision bench source, not any old 5V supply, very smooth and stable.

The timebase in all of the following shots is 5 seconds per square, with 1 square representing 1 volt. Before testing, I connected the 5V supply to the oscilloscope to test the smoothness of the supply, this was the trace:



After connecting the 5V supply to the card, and then connecting the oscilloscope to U10 with the setup plugged in via USB, the trace shows some interference (note ALL ON not yet activated):



There are regions of no interference, and then regions of interference, the gaps between which are irregular.





I then swapped master cards, and repeated under the exact same conditions. Again, initially no interference, and then interference. The sudden rise in voltage is when I turned the 5V supply on:



The interference randomly gets more severe:



And then very severe:



I believe that the type of interference shown in the last shot could be perhaps what is causing the flickering. The fact that the voltage drops from 5V to 0V could obviously be interpreted as a switch press by SIOC. Note that I have not even commanded "ALL ON" in any of these shots.

What confuses me is why this is happening. Both cards that were used are quite clean and have no visible damage upon them. If this sort of interference is happening with nothing attached to the master card, one can only imagine what happens when all of the displays etc are connected.

Let me know what you think.

Jack

[iwik]

Jack,
To get the last pic is the setup as follows:
Usb Exp/ 1 MasterCard/5v supply only and after say 2minutes?.
When does this voltage drop to zero?.
Are you plugging the USB Exp card directly into the M/C and no cable?.
How about a pic of your setup for the above last scope pic.?
Les

[Boeing 747 Flyer]

Hi Les,

Yes that's correct. There really is no "when" to when this occurs, it is completely random. I did not initiate an ALL ON or ALL OFF when the large spikes occurred, I literally just left the setup completely untouched. It will be absolutely fine and then will suddenly and randomly spike.

[iwik]

Maybe you should wait till David sends you the USB expansion bd. If these two board are sitting on the bench with no input then it should not randomly generate rubbish like this. If all M/C are like this then one has to eliminate the Exp
bd at this point. By continuing on you may cause yourself more grief.
I will try and get a chance to look at my setup with a scope and no input and see what happens.
Les
I have had some more thoughts, focusing on USB EXP bd. Try swapping ic1 with ic2 and ic8 and ic9. They seem to drive MasterCard 1.
I checked my Combo with inputs disconnected I had about 500mv on noise on U10 pin
20 and when inputs connected and ALL ON about 1v.
What I saw was what I considered normal.
With M/C disconnected I had some random pulses but these were only a volt.
Let us know how you get on.