Audio/CV Signal Display : Graphical Oscilloscope with Midibox

[Psykhaze]

Hello everybody,
These are some design Questions!
I was wondering about assembling some midibox modules to achieve a little graphical ocilloscope to watch my audio/CV like in this vid : https://www.youtube.com/watch?v=utr23hp2nRM (like on the 4 OLED screens) 

- I think a great part the work is already accomplished if TK already realized that. Am I Wrong?
- What is the exact voltage range of the Core32 DACs ?. Edit : From The Ucapps Page STM32F4 Core module : 0-3.3V for native ADC  From AINSER Module Page : 0-5V
- Audio Inputs Should be ok for Core32 DACs/AINSER? What about bipolar voltage?And if i have "high" voltages like for example +15/-15v , is an OpAmp setup with dividing gain of 3 ok to reduce voltage range to -5v/+5V?
- What is the biggest resolution and screen size that MIOS32 already supports? (i was expecting a 5cmx5cm wide 256px*256px monochrome display or something like that)
- Is the combination of Core32+AINSER8 is modules enough to support 8 analog inputs (audio/CV)? 

I also thought Few DIN/DOUT Serial Registers may allow to have a little Control Surface to switch between inputs and make few changes about display calibration.

Please correct me if i am going in a dead end way =) Thanks in advance for your advices!
Greets,
Jerome
 

Edited May 19, 2016 by Psykhaze

[Hawkeye]

Hi Jerome,

maybe unrelated, but probably the cheapest (money and timewise) solution:  I recently added a DSO138 from china to an old sampler - total costs about 17$ including shipping, with a nice display (320x240px TFT). It works really well out of the box (see my latest video :-) - and no coding is necessary.

Now, on topic, if you want to use MIDIbox :-): TK. has created a waveform output display on the small OLEDs for MIDIbox CV - these look super-awesome! The output part therefore is readily available, you'd just need to "sample" from an ADC. Regarding displays, just have a look into the MIOS supported graphical displays sections... The cutest displays are the small OLEDs, but then there is support for bigger graphical LCDs, too.

Many greets,

Peter

[latigid on]

This is for MBCV v2, which I'm working on with TK. (when he has time).

What's happening is the digital waveforms are visualised on the scopes in parallel with the analogue output. 

Audio inputs will need amplification, but modular levels need attenuation/limiting and offsetting. You can't feed over- or negative voltages into the Core IO pin. (3.0V nominal peak voltage btw.)

This type of ADC is probably only useful for slow-ish waveforms. The accuracy (jitter) is reportedly better for AINSER, but at a cost of slower transfer rates as the digitised signal runs over SPI if I recall properly. I'm wondering if a proper ADC is a better solution, we can maybe look at some of Olivier's (Mutable Instruments) designs. 

So at the moment, ADC for the Core is a concept rather than a proven solution. 

 

Similar things are being developed elsewhere though

https://www.muffwiggler.com/forum/viewtopic.php?t=157649

 

 

 

 

[Psykhaze]

Hi guys !
@Hawkeye This DSO138 kit is really great for the price! Just it only has 1 input..
But should do the trick in many situations , i will think about ordering  one at least for electronic troubleshooting . 

@latigid on First, zeroScope sounds nice ! But did not found the schematics :(
I've been looking a bit on mouser about ADC at a reasonnable price and , even if not PDIP , found this one : MCP3428 (it's a quad input one with correct definition and trough I2C should be quite quick)
Was thinking about normalizing input levels trough an OpAmp setup with adujstable gain (from dividing to multiplying).
To offset the signal , mixing(adding) it with an OpAmp to a DC signal sounds ok? 

Thanks again for your advices !
Jerome

[latigid on]

I have a concept and PCBs for scaling CVs

 

 

The circuit isn't exactly the same as it uses a rail to rail op amp instead of clamping diodes and the offset is a precision +5V source. I have it built up but I think it needs troubleshooting. Maybe I can have a go today.

 

For your other points: I don;t think the zeroscope is not quite released yet, and I suppose it will be in the form of PCBs.

Trouble with I2C chips is that these data lines are normally reserved for MIDI outputs, but it could work for a custom application. In other words, it would be great to have common modular blocks that are usable across several applications (like MBCV).

For mounting OLEDs, you might be interested in this:

 

 

 

[lis0r]

It'd be a useful addition for midibox synths to have a trigger input available, so the display can be synced with the fundamental, rather than at an arbitrary zero crossing. The extreme stability and predictability of the miniscope on the Korg Minilogue makes it so much more useful than some of the other 'scopes I've played with.

[Psykhaze]

@latigid on Hey, this is a quick out-of-the-box design but sounds cool for CV scaling. Thanks for sharing ! Keep me tuned about your tests.An amplifying-alike design would have to be made to adjust Audio inputs.

I will try to design something with 8 inputs (4*Audio+4*CV) / 2 ADC (ADC Audio + ADC CV). I thought about I2C because i did not see why a MIDI I/O should be useful for this app .
The point about using midibox in here is the big display size option / using a custom multi-inputs ADC / Building a Control surface on a known platform wich could suit.

@lis0r Using multi input may allow to be able to sync with fundamentals i think

Once Again,thanks for your interest !
Greets,
Jerome

[Psykhaze]

Hello !
I am thinking about buying a double-trace oscilloscope kit DSO138-like to extend my Single Trace Vacuum tube Oscilloscope.
Will be useful for a lot of debugging ,including this 3inputs+1sync midibox scope viewer :D

@Hawkeye May you advice me?

@latigid on Maybe you have something to advice from your secret's bag?

Edited May 31, 2016 by Psykhaze

[Hawkeye]

Yo man, there are some Youtube reviews/tests of the DSO 138 vs other digital oscilloscopes. The consensus is: it is unbeatable for its price, but if you intend to do "real work" with it (and not just display audio waveforms), there might be a few better (but more expensive) alternatives. If you only want something for the audio range, it will be fine, though!

If you want to display left/right waveforms of a stereo output, you could always put two displays next to each other and wire up two DSO 138 boards - they are not really huge and it would be still cheap.

Many greets and have fun!
Peter

[Psykhaze]

Thanks @Hawkeye . This time it is not only for audio signals, it is to extends my electronics instruments and do real work with it. So if it needs the money to be correct, it's ok to me.
May you to advice me a double trace oscilloscope kit like the DSO 138 in size , at the best quality/price rate you think? 

To make my 2*3 inputs display (synced),i will go on my idea doing this with midibox and my MCP3428 :p

But i need correct instruments for troubleshooting my audio signals and some basic digital signals first.

Edit : For real work oscilloscope , the DSO203 is going to be my next tool !

Edited June 7, 2016 by Psykhaze

[rolfdegen]

My miniscope circuit with xmega128 MCU. I sample with 12Bit/40KHz one cannel.

Input signal is clipping with IC16a. For best noise reduction i make  ADC in Differential Input Mode. The comperator input is for trigger function. For better Ref Voltage i need an LM336 with 2.5 Volt.

 

Scope Input in my DIY Synthesizer

 

Youtube: 

 

Edited June 7, 2016 by rolfdegen

[Psykhaze]

Thanks a lot for sharing this Rolf, this is really nice from you !
I download the scheme and give it a deeper analytic review .
By the way this Xmega board sounds to be a nice tool, i'm wondering about how hard it should be to port the scope thing code to MBHP_Core32
In the end this is going to lead to a MB-Scope  project :D

regards,
Jerome

[Psykhaze]

created a user project for this : 
http://wiki.midibox.org/doku.php?id=mb-scope

[Phatline]

any progress on something like this?