3.1 Synthex board problem (1 row of solid white boxes)

[der Papst]

I just solved my LCD-Problem:

When i startet my Lab-PSU, the backgroundlight worked but nothing else.
The problem was the connector, you have to press it extremely tight together (i used a pipe-forceps, then it worked )
It's tight enough when you can't see the knifes in the connector...

[zoggynog]

Thanks for all the advice so far.

I tightened down the end clamp on the LCD ribbon again, just to make sure.

I visually rechecked the back of my board for any cross connections due to bad soldering/scraps of metal connecting two pins.

I re-ohm'ed (since connecting the new LCD) my board from the actual pins on the Atmega to the pads on the LCD. All connections were correct according to the schematic provided in Synthex's manual.

In addition, I checked all voltages on the LCD pads, they were as follows:

1. 0 vdc
2. 5 vdc
3. 0.5 vdc
4. 5 vdc
5. 0 vdc
6. 0 vdc
7. 0 vdc
8. 0 vdc
9. 0 vdc
10. 0 vdc
11. 0 vdc
12. 0 vdc
13. 0 vdc
14. 0 vdc
15. 3.75 vdc
16. 0 vdc

I'm still in the same boat. I only see one lit row of solid boxes on my LCD. I can adjust the contrast with both the keyboard (blindly) and by touching pad 1 to pad 3 on the LCD.

I may be able to get a hold of an oscilloscope in the next few days, if it would be of any use (check the crystal). I could also test the capacitance of any caps, I do have a capacitance meter available.

thoughts?

[dmitri]

Thanks for all the advice so far.

I tightened down the end clamp on the LCD ribbon again, just to make sure.

I visually rechecked the back of my board for any cross connections due to bad soldering/scraps of metal connecting two pins.

I re-ohm'ed (since connecting the new LCD) my board from the actual pins on the Atmega to the pads on the LCD. All connections were correct according to the schematic provided in Synthex's manual.

In addition, I checked all voltages on the LCD pads, they were as follows:

1. 0 vdc
2. 5 vdc
3. 0.5 vdc
4. 5 vdc
5. 0 vdc
6. 0 vdc
7. 0 vdc
8. 0 vdc
9. 0 vdc
10. 0 vdc
11. 0 vdc
12. 0 vdc
13. 0 vdc
14. 0 vdc
15. 3.75 vdc
16. 0 vdc

I'm still in the same boat. I only see one lit row of solid boxes on my LCD. I can adjust the contrast with both the keyboard (blindly) and by touching pad 1 to pad 3 on the LCD.

I may be able to get a hold of an oscilloscope in the next few days, if it would be of any use (check the crystal). I could also test the capacitance of any caps, I do have a capacitance meter available.

If you can blindly adjust contrast then Atmega is running fine, no need to check the crystal.

thoughts?

There is no magic here. Since you claim that all the connections between Atmega and the LCD are fine and, as we established, that Atmega is running fine then either the LCD is fault or one (or some) of the Atmega's pins going to the LCD are faulty.

[samdrum]

I have a Synthex 2.8 board and am having exactly the same problem / pulling my hair out. But as always there is a solution for everything and am sure that with some persistence and a little bit of help we will be able to work it out

So story is: Megadrum works in MCT (device detected as usb sound device and MCT detect firmware version correctly).
LCD backlight works but shows boxes, just like the above post.



In Dimiti's LCD problem solver the following causes are mentioned for this problem (see http://www.megadrum.info/forums/viewtopic.php?f=3&t=492):

• Atmega is faulty. Replace Atmega.
• Atmega is not programmed. Program Atmega.
• A faulty Atmega crystal. Replace the crystal.
• Incorrect capacitors connected to the crystal. Replace capacitors.
• Atmega is working properly, e.g. tested with MIDI activity, but data communication with LCD is broken. Check connectivity between pins 4-14 of the LCD and the board right to the pins of Atmega and that there are no breaks/short circuits on those data lines. If you're sure that the connections are ok, than your LCD is faulty.

Now, since the device is working properly in MCT, I assume ATMega is operating/programmed correctly and it is not a capacitor problem (right?). So this leaves us at the connection between ATMega and the LCD. I measured the voltages and these are the results:
01 - 0
02 - 4.5
03 - 0.48
04 - 0
05 - 0
06 - 0
07 - 0
08 - 0
09 - 0
10 - 0
11 - 0
12 - 0
13 - 4.61
14 - 0.49
15 - 4.64
16 - 0

I got the following info from the schematics:
gnd are LCD pins 16, 10, 9, 8, 7, 5, 1. So these should be zero, which they are -> ok.
vcc are LCS pins 15 and 2 -> These should be around 4.5, which they are -> ok.
Connected to ATMega are the remaining pins:
03 - 0.48 (connected to pin 14 with 1k resistor)
04 - 0
06 - 0
11 - 0
12 - 0
13 - 4.61
14 - 0.49 (connected to pin 03 with 1k resistor)

While I don't know what values these pins should have, it seems strange that pins 4, 6, 11 and 12 are zero. My guess is that this is due to my sloppy soldering job (the connector to the LCD is pressed in right; I can see the pins). This is what the connection looks like from below:


My questions to you guys are:

• Do you see an direct soldering conflict here that might cause the problem?
• What should the voltages be of the mentioned pins? I.e. Is it possible that some of the (connected to ATMega) pins are 0?

I will continue to debug when I have some spare time again (starting with measuring straight from the ATMega and comparing with the above results). Can't wait to do some serious megadrummin'

Thanks guys!

[Firelord]

Did you check the main board/LCD connections for shorts (the device shold be powered off during this)? Did you clean your board after soldering? The image depicting soldering joints doesn't look too good, I'm afraid.

[dmitri]

Can you blindly adjust LCD contrast?
It almost pointless to measure voltage with a tester on Atmega's pins when Atmega is running and it is pulling those pins up and down. Rather you should test these pins for shorts between themselves/to ground/to Vcc while the board is powered off.

[samdrum]

That makes perfect sense gentlemen. Thanks for the advice. Yesterday I did a quick check (using a ohm meter) on the pins that are connected to ATMega and they are all connected. I will got through all pins of the LED Connector using synthex PCB drawing. There should be a short somewhere Can take some time though (vacation). I will let you know what the results are.

[zoggynog]

I've actually been meaning to make my giant "SUCCESS" post for a week or so now.

I was able to resolve my problem, and my Megadrum fully works!

The solution for me was odd, apparently I had an issue with the connector on the LCD ribbon. I say it was "odd" because I was able to test accurate connectivity from the pins on the Atmega...all the way to the solider pads on the LCD display.

Out of desperation, I attempted to break apart my existing 16 pin connector and reapply it, but that was (of course) a bad idea. Eventually, I went down to my local Fry's Electronics and bought a couple of extra connectors. Applying the new connector resolved my issue completely.

I bet I'm not the only one who has had this problem. Although I didn't do this myself, I would highly recommend buying a prefabbed cable to folks who are having similar problems. I'm sure most of us are already doing that with the 32/24 pin connections. Plenty of spare UDMA-33 IDE cable/ Floppy Disk cables laying around.

This site:

http://www.provantage.com/supermicro-cbl-0049l~7SUPM0WM.htm

seems to carry the 16 pin ribbon cable /w connector, albeit at a high premium.

Thanks very much to everyone (especially Dimitri/Synthex) who helped me out. This has turned out to be a great project. Been playing with the BFD2.0 demo...I'm having a blast...even with a cruddy rockband cymbal pad as my only method of input at the moment.

[samdrum]

Good to hear that Zoggynog. Guess what: my problem was in the connector as well!

These connectors are so flimsy that I broke already 3 of them. Particularly since you need to use A LOT of force to force the knives through the cable. Apart from that, it is easy to put the cable in with a slight angle, with shorts in the cable as the result. So my advice is to check you connector first when you experience problems with your LCD. And indeed, preferably buy a prefab cable to prevent this whole problem.

I'm quite happy that the LCD is working now: Dimitri's name shines when I boot up now. Respect I still have to overcome two problems though:

• Without the piezo connector plugged in, I can see the first VU on the LCD at 100%. So probably there is a short there. The other VUs are at 0%, which is right.
• I have wired the buttons wrong. When I plugged them in, the menu goes very quickly through all the items. I based the wiring on Synthex button drawing, but strangely enough the buttons have four pins instead of 2 pins, so I had to improvise with my limmited knowledge of electronics. This is probably a no brainer for you guys. What I did is connect each connector to each other via one pin of each and then hooked that circle to the ground. ThenI attached the various pins to one pin of each putton. See picture. Is this right?

Please not that the numbers in the image are arbitrary; they do not match with the pin numbers.

[ignotus]

There should be (at least mine did) a slight groove running down the centre of the bottom side of the button: each side of this groove (2 pins) corresponds to one "side" of the switch; it's as if there were only two pins anyway.

Hope this helps