DAW controller orientations and design

[endernoumea]

Everything have been ordered today except the frontpanel \o/

 

According the delay for shipment in New caledonia (1,5 month) and few vacations I should start my first solder by the end of April.

 

I want to thank everyone for your patience with all my "noob" questions and especially John E. Finster who was here from the first day :flowers:. Without people like you, newbies like me would never dare to start this kind of project.

 

Be sure i will have a lot more questions once i will have these 2500 little pieces on my desk :frantics:

 

A friend of me got a 3D printer and is very enthousiast with this project too. We will try to create some nice buttons by waiting the material.

 

I'll keep you in touch.

 

Kind Regards,

 

Ender 

[endernoumea]

Everything is finally arrived in New Caledonia :)

 

After a first night i declare the project officialy opened :D

 

[Duggle]

The real fun begins!

 

On the subject of power, USB from computer or power adaptor will power a surprisingly high number of MBHP modules (not sure about MF though).

 

My NG project is powered from an external 5V source plugged into J2. (For use with standalone midi equipment) If you do this please take care of polarity as there is no protection!

A 5V supply through the USB Socket is a safer solution of course. (I have a shottkey diode in series with J2, gives polarity protection but also a small voltage drop to solve an obscure problem with driving 24 LCD modules!)  

[endernoumea]

Day 3 : First board first issue.

 

 

Heya everybody ! Everything goes smoothly but i had an issue that 'im not able to solve alone.

 

On one board, the 5th upper led ring is not working properly. I will try to describe here the symptoms:

- when the NG start the first led which is alway light on (lower left one) shine much more than other.

- when i move up others encoders I can see a ghost of them on this 5th ring.

- the last led never light.

 

I've checked all the soldering... try to replace all the IC but nothing seems to work. 

Any idea ? 

 

Thank in advance,

Ender

[Duggle]

 

On one board, the 5th upper led ring is not working properly. I will try to describe here the symptoms:

- when the NG start the first led which is alway light on (lower left one) shine much more than other.

- when i move up others encoders I can see a ghost of them on this 5th ring.

- the last led never light.

Have you changed the LED that never lights? (sometimes they get damaged soldering etc, I had about 2 per 1000 or so on my 4 Fairlightiii PCBs).

Have you swapped the PCBs in the SRIO chain? (need to eliminate signal integrity problems and isolate faulty hardware.)

It's good to have a settup.ngc that works with just one PCB, and have just one PCB connected to do this.

 

PS I like you colour scheme with the LEDs!

[endernoumea]

Hi !

I've replaced the led, change the ICs and nothing... i made some tests and it's linked to the board itself. I think that i have a solder not good somewhere... i need to recheck everything again.

 

To vary the excitement i've worked on the lcd circuit. After some headache to find how to enter again in bootloader mode (noob i know now...) i've now declare my 4x5 lcd pattern. I need to solder the rest of the adapters "8<->16" and prepare the RGB matrix to color the blacklight according the function i use.

 

[Duggle]

Hi,

so you have isolated the Encoder boards and have only that is faulty?

 

Those adaptors are really nice! Recently I soldered 24(!) 16 way single-in-line header sockets to a long bus. It took many many hours. The good news is that at least 24 LCDs does work! (I had to adapt the interface to work though, let me know if you plan on connecting a large number and I'll explain what I did).

 

The RGB matrix works great. If it was possible to illuminate the LCD Backlights with them, I think that's the way to go. I made diffusers from "opal" white acrylic, so that the area adjacent to the encoder and LCD shines different colours depending on the parameter. This approach has been very successful though.   

[endernoumea]

yes only this encoder is showing these errors. The ghost from the others encoders from its board and the last led never shining.

 

The RGB matrix is what I will try to achieve as next step indeed. if you count the number of pin on the LCD you'll understand that I have 3 differents entries on each to setup the blacklight color.

 

Do you have uploaded somewhere a picture of this "opal" diffuser ?

[TK.]

yes only this encoder is showing these errors. The ghost from the others encoders from its board and the last led never shining.

 

Maybe the selection lines are set to the wrong polarity,

Are you using ULN based sink drivers?

And could you please show me your DOUT_MATRIX configuration?

 

Best Regards, Thorsten.

[Duggle]

If I understand correctly 3 of 4 PCBs work perfectly, and 1 has ghosting plus a LED that never lights, right?

 

I have some very minor ghosting in my own PCBs. I suspect it is because of a relatively slow switching of ULN drivers. The theory is that the drivers take a tiny fraction of a second too long to turn off, in which case the next column pattern (which is present on the previous row) becomes faintly visible. I think the timing overlap is very small, but sufficient to be seen with bright LEDs.

 

As a test you could set inverted=0 in the matrix setup and bridge the appropriate driver pins. (To both diver IC's as the change to the DOUT_MATRIX will affect the whole PCB).

I'm assuming you have sockets and can easily remove the ULNs to do this.

 

In my situation I ignore the ghosting as it will be invisible once I fit the translucent acrylic front plates.

 

I'll upload a picture of the diffusers.

 

BTW, the LCD's with integrated RGB backlights are a very cool idea!

 

[endernoumea]

Maybe the selection lines are set to the wrong polarity,

Are you using ULN based sink drivers?

And could you please show me your DOUT_MATRIX configuration?

 

Yes i'm using ULN2803 on these board. I tried to replace them with no effect.

 

The dout matrix is here.

 

If I understand correctly 3 of 4 PCBs work perfectly, and 1 has ghosting plus a LED that never lights, right?

Yes. I do not have any ghosting effect on the others encoders or board. Just on this encoder ring.

Each time i move up an encoder i can "see" the ghost changing as well. If i light them all the ghost is glowing nearly as strong as the others.

 

And its last led (last on right) is never lighting on whaterver I do.

 

I double check all my solders and can't find where it comes from.

Do you think i can use a cutter to isolate this encoder's led ring and patch it from the IC directly ?

[Duggle]

Do you think i can use a cutter to isolate this encoder's led ring and patch it from the IC directly ?

Please don't do this. The PCB is correct, you can't improve it. If there is an etching fault, we can find this later using non destructive methods.

 

The config you linked to is for 4 PCBs.

This one here is for 1 PCB.

I'd like to know that you've tested each PCB by itself (wired with a short SRIO cable to the core, nothing connected to the "out" connector on the PCB) with this config.

 

I.e. test each PCB separately and report back, if you could.  

[endernoumea]

I tried the file you attach here with same result.

3 boards are 100% ok.

The last one have the same issue.

[Duggle]

Just to be clear, you connected them one at a time, by themselves, right?

[Duggle]

Now, assuming the above is correct: (single cable, single faulty PCB connected)

 

First of all we should sort out the LED which is not lighting. 

Have you downloaded the schematic?

If so are you able to identify the LED (which is not lighting) on the circuit?

If you are not able to read the schematic, then please you post a picture clearly identifying which LED does not light. 

 

The procedure will be to energise the LED by setting the DOUT pins with statements in an NGR section.

Then it will be easy to read the voltages with a voltmeter, as they will be constant (rather than pulsing when as a matrix).

 

I'll take you through the process step by step if you want, but the first stage is to identify the faulty LED...

[endernoumea]

Now, assuming the above is correct: (single cable, single faulty PCB connected)

 

Yes it is.

 

We are talking about the LED 143 :)

 

The procedure will be to energise the LED by setting the DOUT pins with statements in an NGR section.

Then it will be easy to read the voltages with a voltmeter, as they will be constant (rather than pulsing when as a matrix).

 

I'm not sure to know yet how to do that but that sounds interesting :)

[Duggle]

OK, if I understand correctly, if one starts with blank NGC (no config),  but in default.ngr has:

if ^section == 0
  # turn on the first 3 LEDs connected to the DOUT module:
  set LED:1 127
  set LED:2 127
  set LED:3 127
endif

When you click "save" or power up the core, the statements in section 0 get executed.

Now with a blank config there is no matrix, but there is DOUT lines that are addressable as LED:x  (where x is the DOUT pin counting from Shift register1: D7..D0, SR2: D7..D0, etc).

So what I would ask you to do next is to identify for the LED, the row and column ( terms of SRx:Dy) of the anode (+ve, column driver either SR3 or SR4) and cathode (-ve, row driver either SR1 or SR2)

Then count the bits to work out how to address the anode and cathode. Then put the following in your default.ngr:

 if ^section == 0
#set the annode column
 set LED:X 127
#set the cathode row
 set LED:Y 127
endif

Of course replace X and Y with real numbers, test a known LED that works first, then energise the one that doesn't.

 

 

[endernoumea]

I finally found the issue... to resume, this board was the first i made and i'm not proud of it. Probably my iron was too hot too... so the anode under the led 144 was burned and causing short cut.

I isolate it and the board is working in standalone.

 

It has generated a new issue. Now when i plug this board after the 3 others I have random change of values on the others board. I guess that I have another "burned" track around the RIOS chain. I'm gonna check that.

 

Hopefully I have a spare board on its way just in case...

[Duggle]

Please swap the boards in the chain.

You may find that a very similar problem is always on the 4th PCB connected.

If so let us know as there is an easy fix for this (signal integrity issue).

[endernoumea]

I tried all the combinations and it seems the effect is reduced when it's the 4th of the chain.

 

For you to know:

the "bad" board alone : ok

the "bad" board + any other board : ok

the "bad" board + 2 others boards : first gliches appearing

4 boards : party time

[Duggle]

Ok, I think I have the general idea.

Try this:

[endernoumea]

I tried this and it drastically improve the result.

I still have some glitche but they are rare.

 

FYI: i did not have 100ohm resistor at my disposition so i use 2 x 220ohm in parrallele. I will be able to try with real 100ohm in a week.

[Duggle]

The main signal that is an issue is the SCLK.

The idea behind this is to match the resistor value with the impedance of the transmission line. The actual line impedance is a little hard to determine (an estimate can be made using an inductance/capacitance meter) so you could experiment tweaking (try changing up and down) the value of the resistor on the SCLK line. At a certain point the serial I/O will cease functioning completely and at the other extreme you'll see the spurious results you've seen without any termination resistor(s). Somewhere in between things will work.  I just checked mine, and it is 100R. If 110R is not perfect, then try 3x 220R in parallel(~=75R) or  2x220R (in parallel) in parallel with 440R (in series, 220+220) .   

[endernoumea]

Tada !!!!

 

Everyhting works ! Finally :)

[endernoumea]

I just setup the dout Matrix to control the RGB screens. I take some time to understand how it works but now it's ok and it's amazing to watch all the colors available.

 

I'm now facing a new issue. I think it's related to the "non-permanent" current going thru the matrix. I mean that my LCD are not glowing as expected. (on the picture the LCD glowing OK but they are directly connected to a power source and not to the matrix)

 

I remove all the resistor from the dout and it's kinda better but not enought.

 

I tried two circuits, on with ULN 2803 and one with 74HC541 which improve the result but not enought again.

 

I'm am now wondering if I can use a capacitor to smooth the current drop but I'm not sure on the way to do it.

 

If this kind of trick cannot works i guess I will have to use DOUT module without matrix but this will make the RGB control much more complicated.

 

Any advises ?

 

Thank you in advance,

Ender

Edited August 10, 2013 by endernoumea