LCD Black Row

[Rics]

Two days ago i've received the smashtv core32 pcb and parts, i've builded it and then i've connected it via usb to my mac laptop.

I've then opened mios studio, uploaded the BLM app and connected to the DIN/DOUT, J8/J9, socket a small BLM module to do some tests.

Got all working fine.

Then i had an old 2*20 characters display, used with a PIC Core, this display was one of these that smashtv sells some years ago.

So, i've changed the pins order like requested on the new core32, i've then checked all the connections with a tester to be sure and only then i've fired up via usb my core32.

the result was a black row on the led, like the following picture:

After this, i've tried to do a Mios upload, i've downloaded the file on this page, the mios studio rebooted the core32 and then, on firing up, the result was the same, a black row.

No system messages, nor booting or initializing messages. As soon i connect the core32 to the usb, that black row appear.

Now i don't know if i have to code and upload something more, i thought that like in the pic core, also if no app were loaded, a welcome message from the mios were printed on the lcd.

PS

I've already did a search over the forum, but sincerely i haven't found something useful to solve my problem.

[JRock]

Backwards?

[Rics]

Well, it's just the picture that's upside down!

Btw here it is the datasheet:

Link

[julienvoirin]

upload an app on your core32.

on the pic your lcd isn"t initialised, like there was no app

[julienvoirin]

btw, in core 32 there is no more mios to upload before the app as it is included inside the code whle compile

there is only bootload, then upload of compiled app. tr the one of the sequencer v4

[smashtv]

Hi Rics!

The 'all black row' thing happens any time the display has initialized but has not received valid data..... It could be as simple as a dodgy connection on any of the data or control lines, a pair of data lines swapped, etc.

You can get the same behavior from most character displays connected to a Core8 without it's PIC in the socket, meaning most will give the black row with only power and contrast bias connected.

Let us know what you find, hopefully it will give up more hints about the cause.

Best regards

Tim

[Rics]

Ok, today i i got some ribbon cable and a pair of connectors and i've did the connections like the ones found here:

I've compiled and uploaded the MidiBox SEQ V4.0 and then did a power cycle via usb.

The LCD still remained as before, black row.

What i noticed is that the P2 contrast trimmer do his work, in fact i see the black row fading in and fading out.

But the P1 Luminance trimmer isn't working, to be more precise, as i fire up via usb the module, the LCD isn't backlighted, also when i modify the P1 Luminance trimmer no backlight is going on.

I'll do another ribbon cable to reconnect the lcd to the pic core to see if it is working fine.

Later i'll post infos about this last test.

Thank you all for the support!

[Rics]

Update:

I've reconnected the LCD to the pic core and is working fine.

Then i've reconnected it to the core32 and it isn't working, black row as ever.

[Rics]

2nd update:

I've checked another time the connections, and the lcd wasn't illuminated because the pins 15 and 16, respectively B+ and B- were swapped.

I went with closed eyes about the 1:1 pins scheme with the core32.

So i took the occasion to recheck all the pins, and seems that were all right except the two B+ and B- immediately fixed.

By the way the black row still remain and as before, the P1 trimmer luminosity don't do anything, i've modified the P1 value but the illumination (now working after the B+ & B- swap) remain the same, no variations with the trimmer variation.

P2 trimmer, for contrast, work fine.

Stille remain that on firing up, no system messages were show, only that black row.

At the moment i've the BLM app uploaded.

I've tried also with the tutorial#20, but i have the same behaviour.

I'm 99% (i leave 1% to murphy's law) sure that the connection now are correct...without apps uploaded, a minimum of initializing data at the startup must come on the display? All the core32 apps support clcd displays?

Last question, could be something related to the R33 resistor array?

[smashtv]

So i took the occasion to recheck all the pins, and seems that were all right except the two B+ and B- immediately fixed.

hehe I noticed that, wrote a post about it, went back to check pinout one last time and before I hit send you posted about it. SmashTV lag level up! :)

the P1 trimmer luminosity don't do anything, i've modified the P1 value but the illumination (now working after the B+ & B- swap) remain the same, no variations with the trimmer variation.

P2 trimmer, for contrast, work fine.

This is normal with these displays, it has on-board current limit resistor(s) for the backlight. Somewhere here on the forum there are some very old posts with more info.

Last question, could be something related to the R33 resistor array?

Absolutely.....backwards orientation, shorted or open R33 connections could definitely cause this. With an open connection @ R33 the related LCD control line would only see 3v logic levels. Likewise any problem with the 74HC595/IC2 will keep the display from working.

Back to the kit mill!

Best

Tim

[Rics]

Thank you for the reply Thorsten Tim, today i'm away from home (:brr:), so i've not the possibility to do a check.

This evening i will check that R33, now i've the doubt that i've swapped one network with another.

Aaah, the rush to test a blm module...:blush:

Edited December 5, 2010 by Rics

[nILS]

Thank you for the reply Thorsten

Do you happen to mean Tim?

[Rics]

You know our Tim "nameexchange" issue!

:tongue:

_{(Sincerely this mistake fitted well with the Tim/Mike name exchange, so i took advantage....but the real thing is i wrote that post when I had just woken up! :blush: )}

Edited December 5, 2010 by Rics

[Rics]

Ok, i swapped a pair of resistor network and now the lcd is working fine!

But a problem go away and another come up!

Now, if i connect my small blm module, that worked until before this resistor swap, a lot of random messages are sent to the leds of my blm test module (3 leds).

I've all the analog inputs clamped to ground, but this don't stop the random messages downhill.

Without the blm module connected to J8/J9 i've this sysex message:

[15026.730] f0 00 00 7e 4e 00 01 10 10 02 01 01 01 f7

Don't know what mean this. Then, after the connect of the blm module i have, between the random messages, these sysex messages:

[15267.324] f0 00 00 7e 32 00 0f 4d 49 4f 53 33 32 f7

[15267.327] f0 00 00 7e 32 00 0f 4d 42 48 50 5f 43 4f 52 45 5f 53 54 4d 33 32 f7

[15267.330] f0 00 00 7e 32 00 0f 53 54 4d 33 32 46 31 30 78 f7

[15267.330] 92 0a 00

[15267.333] f0 00 00 7e 32 00 0f 30 30 30 30 30 30 30 30 f7

[15267.336] f0 00 00 7e 32 00 0f 33 30 46 46 45 39 38 44 33 31 34 42 33 34 33 34 37 36 32 35 31 39 34 33 f7

[15267.339] f0 00 00 7e 32 00 0f 35 32 34 32 38 38 f7

[15267.340] 93 0f 00

[15267.342] f0 00 00 7e 32 00 0f 36 37 31 30 37 38 34 30 f7

[15267.345] f0 00 00 7e 32 00 0f 42 4c 4d 5f 53 43 41 4c 41 52 20 56 31 2e 30 f7

[15267.348] f0 00 00 7e 32 00 0f 28 43 29 20 32 30 31 30 20 54 2e 4b 6c 6f 73 65 f7

Don't know what means also this.

Keep in mind that at the moment i have R32 resistor network not placed, but also J19 isn't used.

J16 isn't used and also the display at the moment isn't used.

The only resistor network placed are the R31 and R2, the other aren't placed.

J5A-B-C are all ground clamped.

Concluding i can't find the source of these random messages, i've the doubt that can be at the blm module as is done over a veroboard with ribbon cable wires.

(And also because if i disconnect it from the core, no more random messages appears over the mios_studio)

What i don't understand is that before the resistor networks unsoldering and resoldering all worked fine.

[smashtv]

Hi Rics,

Keep in mind that at the moment i have R32 resistor network not placed, but also J19 isn't used.

J16 isn't used and also the display at the moment isn't used.

The only resistor network placed are the R31 and R2, the other aren't placed.

J5A-B-C are all ground clamped.

Without the R30 and R32 networks in place, you have some open lines and without anything connected these 'float'. In the same way that floating lines on J5 cause jitter, letting the serial lines @ J16 and J19 float can have the same effect. With analog lines it's common to 'pull down' or connect/clamp them to ground, with digital lines accepted practice is to 'pull up' floating lines to logic hi (typically through a resistor to +V/Vd).

What i don't understand is that before the resistor networks unsoldering and resoldering all worked fine.

oof that's what the 5 pin SIL sockets in the kit are for!

Best regards

Tim

[smashtv]

Thank you for the reply Thorsten Tim, today i'm away from home (:brr:)

Dave,

I am so terrible with names I tried a mnemonic technique.... but stinky, whiney, and JoJo were not amused when I accidentally spoke the mnemonic instead of their given names. :)

[Rics]

Hi Rics,

Without the R30 and R32 networks in place, you have some open lines and without anything connected these 'float'. In the same way that floating lines on J5 cause jitter, letting the serial lines @ J16 and J19 float can have the same effect. With analog lines it's common to 'pull down' or connect/clamp them to ground, with digital lines accepted practice is to 'pull up' floating lines to logic hi (typically through a resistor to +V/Vd).

oof that's what the 5 pin SIL sockets in the kit are for!

Best regards

Tim

Here i'm with the fixing show!

Ok, i've replaced the J8/J9 network resistor with a new one, i've placed all the resistor networks where required, J15A-B-C clamped to ground and now the behaviour is returned as before!

Also with the lcd working now!

What i don't know how to do is the J16 and J19 ground clamping: i've to put only the serial SI and SO to VS? The SC, RC1 and RC2 pins?

What i don't get now are the startup messages that mios studio reports on the mios midi in:

I start the Core32 with a BLM test module connected.

This test module have one 74HC165 and two 74HC595 ic's, three buttons and three single color led's (i don't want to build an entire module for testing purpose).

The buttons and the leds common are all placed on J3 C0 via a bc337 from IC2 (74HC595) and respectively I0-I1-I2 from IC1 (74HC165) for the buttons + G0-G1-G2 from IC3 for the leds (74HC165).

I haven't used IC4 because i'm not using duo-leds.

That's what i've on the mios_studio monitor (i used pastebin because the messages list is too long)

Messages list

It starts with a long series of note on/note off messages on the first four channels, ch.0-ch.1-ch.2-ch.3 (0x90-0x91-0x92-0x93), with note values B-2 A-2 C-0 B-1 (07-06-0f-0e) and velocity values between 0 and 127 (00-7f).

After 1-2 minutes of these messages i've the mios_studio monitor showing in a regular cadence the sysex message

f0 00 00 7e 4e 00 01 10 10 02 01 01 01 f7

with no more the previous note on / note off messages.

The leds and the buttons work fine, also when the note on / note off messages are show in the mios_studio monitor.

Perhaps these messages are coming because the other BLM test module IC inputs and outputs aren't connected?

PS

I think that i'm going a bit off topic, perhaps i've to put this last post to my

Edited December 6, 2010 by Rics

[smashtv]

What i don't know how to do is the J16 and J19 ground clamping: i've to put only the serial SI and SO to VS? The SC, RC1 and RC2 pins?

Those are digital signals and on this design they need to be pulled up not grounded.

With analog lines it's common to 'pull down' or connect/clamp them to ground, with digital lines accepted practice is to 'pull up' floating lines to logic hi (typically through a resistor to +V/Vd).

More info here: Pull-up resistor

Best regards

Tim

[Rics]

Thank you as ever Tim!

Later i will try with the pull-up and i will do a report. :thumbsup:

[Rics]

Ok, i did some tests with pull-up resistors over the J16 and J19 jumpers and the jitter still remained.

So i'm now sure that this is coming from my small BLM test module...and more worst, i leaved some pins of the 74hc595 and 74hc165 opened! :excl:

[Rics]

Problem solved!

But as i think that this thread is going out of his main title and the lcd problem is solved, i will continue on my other thread where i discuss about the BLM module and the entire my project.

Jitter is gone, i've builded a complete BLM module, but new problems comes out and i will disccuss over them here: