New programmable switcher/looper with LCD touch screen

[vagos21]

Hello everyone, and thank you for accepting me here 

Total noob on the forum, so forgive me if i break the rules from time to time 
I've been working on this for a couple of months and i think it's time to show it around, get some feedback, do some troubleshooting etc. It's a pedalboard switcher, a bit different than what we usually find on the market. I wouldn't say it's got brand new features, but the whole thing is very easy to use

First things first, a BIG thank you to bryan for sharing so much info and marshall arts for the guidance with this project so far. It's built around the arduino mega, and features:

• Separate boxes for controller and looper, this means you can have it all set up on your pedalboard somewhere far away, and only use the controller with the foodswitches in front of you.
• 10 loops, the last 2 of which have trails/spillover option (thank you guys)
• midi control with 7 different parameters stored per patch. that includes program change, CC and other stuff
• 4 amp switches, which can also be used for tap tempo input for pedals that support it
• expression pedal support right next to the controller. It can control any midi parameter real-time
• switchable pre-gain on guitar input, post-gain on guitar output, i intend to give it a +/-9dB range
• 100 presets stored in banks of 10
• tap tempo stomp (green led blinking)
• bank up/down stomps
• patch/stomp mode stomp (when active, you can directly control any of the loops via the 10 stomps. for improvising situations)
• Color LCD resistive touchscreen. Yes i wanted to have a capacitive touch, but could not find something that fit the project. It works just fine though!
• The whole setup/programming is done on the screen. This makes it very easy to understand and fast to configure, even if you've never seen it before!

The relays box is under heavy construction/design, the prototype of the controller is already here and working (attachments)

So what do you think about it guys? Any features worth adding?

[rockola]

Looks gorgeous, but what does "HVYGBJI" mean?

[vagos21]

Looks gorgeous, but what does "HVYGBJI" mean?

hehe it's just random gibberish i wrote for the program name
I will post the menu screens with all the settings later so we can see how deep this thing can go 

[gordo]

Very cool. I like having bpm shown on screen. I don't have any midi capable delays but would be cool to do presets and my Nova delay is annoying at best to preset.

[vagos21]

Some eye candy from the design:

[vagos21]

And here is my spillover problem, breadboarded.

I'm sure there's something else wrong with my setup, not the schematic itself. I have been swapping caps and op amps to see if i spot anything bad, but i get the same behavior: from bypass to fx on, no pop. From fx on to spillover, huge pop. From spillover to bypass, huge pop. Measuring from ground to output jack, when spillover is on, i get about 1V DC. I have everything powered from an isolated supply, so i have separate supply for the delay, the audio signal of the looper, and the relays/arduino, just to make sure the pops don't come from the relays themselves. I followed Marshall Arts fantastic looper schematic for loop 7, exactly as it is, the only difference is i replaced all 100K resistors with 150K (that's what i had in hand), and the Vb resistors of 10K replaced by 18K. My op amps under test are all TL072. I know the breadboard pics won't help, it's a tiny mess...

I also tested another mixer a couple of days ago, but still get that annoying DC which gives the huge pop... i also attach a hand drawn schematic of that one.

[Willybomb]

The BOTY competition hasn't even opened yet and we have the winner...

[vagos21]

The BOTY competition hasn't even opened yet and we have the winner...

As promised, here are a few of the main screens/menus on the controller. I finally got to embed pics in my posts, yay..! 

Hope i didn't make your eyes bleed! 

[Marshall Arts]

Nice. I am very much interested in the technical details. Schematic, Display, code :-)

[vagos21]

The code is of dual nature. the screen is a Nextion 4.3", which happily does all the heavy weight lifting, running its own code for the menus buttons etc, and reports stuff to the arduino via serial port. The arduino on its turn sends its own data to the screen. it's simple, but it's like writing the same thing twice. I'd love to share schematics and some sample code, heck might even make a kit or tutorial
But i have to get over the spillover problem first so i can complete the relays box. And i'm also in the research for volume control via digital potentiometers. The MCP41010 (10K) has some basic circuit in the datasheet and that's where i'll begin. I also note that, even if i do the 10ms mute while changing patches, i think i'll need to do the same on the inputs of the spillover fx (9+10), since the pop will be repeated inside the delay, should there be one at its input.

[Marshall Arts]

Summarizing some points for your debugging endeavor concerning the pop:

I see no reason in the hand drawn schematic of the mixer above for a DC on the output. For the spillover circuit used in Eoo3 I provide a simplified view of the relevant part of loop 7:


It's hard to image, where the DC on the output should come from in bypass mode (that's the mode the schematic shows): C27 should block all DC coming from the circuit, if R52 is connected to GND, there is no chance for a potential difference on the OUT.

In FX mode, Return is directly connected to the OUT Jack, so if the FX itself has DC in it's ouput, that would be the source (unlikely, but easy to measure). The IN signal is fed to the FX in AS WELL as towards C28. It is directly fed to the Output via C30, where it is mixed with the return signal (this is where phase issues might occur with the direct signal part of the fx return, but I did not care, as I never had any issues).

On the input, the send is grounded in bypass mode. Switching to FX mode should not create any popping. The output switch has pulldowns on both sides, this should not create any popping either.

If I fully understand my approach (it's been a good year now...), the direct signal in fx mode coming through C30 should not cause any issues. I think I tested that as well, but to be 100% sure, you could test the following:

Pull R60 (any trails would disappear)
Switch from bypass to FX mode
If the pop is gone the phase inverted (IC1A!) direct signal does not go well with the non-inverted signal from the return path (RET7)

Can you try that?

[Marshall Arts]

P.S. This is an impressive work, if you get all of this to work. Having designed Eoo3, I know about the effort. Having a display and other digital components in such a unit can induce some serious HF noise in the audio path, and fixing that is a PITA. I had issues with the 7-segment display driver... my goodness!

If I should ever design a 3rd version of this, I would try to add a WIFI Chip (ESP-wroom) and make the unit configurable via smart phone. Other features I failed with were configurable fx order and configurable displays/label on the fx buttons.

For the digital potentiometer: It does not work for continous volume control, it's just not discrete enough. It's ok for setting a static volume, though. You will have to mute the output for the volume change time, though, because it creates some nasty artefacts during the change... They are available in 50k and 100k (recommended)as well.

[gordo]

Damn, I'm looking forward to getting back home to the bench now.

[vagos21]

Yes i've done some research on digital pots and i don't intend to use them to change the volume continuously, only once when the patch changes. Continuous changes of the MCP digipots creates zip noise, i've tested it. There are special digipots that detect zero crossing, but i think my choice is more than enough for this application.

Also, here's some fresh new pics of the controller and relays units, side by side, and how  they are connected via serial cable:




I will research the schematic you proposed again tomorrow afternoon when i get back from work. Yes it is quite an amount of work and man-hours invested in this! Weeks, if not months...

As for the LCD, i don't know about any EMI from it yet, i'll find out soon when i connect the guts of these 2 units But i'm counting a bit on the 6 foot long cable that might get rid of any possible digital noises on the way... yes i've been through hell with all kinds of digital noises too! Imagine this: making a digital tremolo with patches, with an LED and LDR. Even after changing the PWM frequency above the audible range, the LED emits something that the LDR picks up as noise, it was totally WTF? Unexplainable noise, optical fibres also carry it with them... it's like part of the optical wave. Something i'd like to develop further one day when i have more time  Long story short, i had to apply a sallen-key low pass filter to the LED, the only way to make it go away...

P.S. Are you using KiCad for the schematics?

[Marshall Arts]

I am using eagle for schematics :-)