[Need Help with SPI] Euxoloti for axoloti (Eurorack Module)

modular
#142

The build document is at version 1. By which I mean I used it to build mine and since my core board is sitting in its eurorack case passing its tests i think the document is ready for more eyes! The core board build was pretty straightforward. There are a couple of things I would have done different:

  1. headers after IC's. This just makes it a bit easier to flip the core board with the control board holding the headers in place to solder them down.
  2. I built a small header rig to make sure my alignment was correct. I should have used the Axoloti to line them up instead of just a few more male/female headers. Again not a big deal, just would have made attaching it after the fact slightly easier.
  3. Cleaned my board before taking pictures :smile:

Euxoloti build manual v1

#143

do you really need 2 cut the midi part / i have some space left in my eurorack and dont find
much info making it work again if anyone can give some advice ... don't want to break it !!!
still waiting for pots and pushbuttons and other components / i have to be patience ....
i have a paia 9700 modular frack rack hoping this wont be a problem with the power ?
i just bought an eurorack so the power will come from that ...reading a lot about it but want a second opinion

#144

If you dont cut the midi part it wont fit in a eurorack in the way Paul has designed it. You could change the orientation of the board, which is possible, but means you should most likely design a new front panel.

The board still works without the midi and power part, and even though I haven't had to put it back together it was designed by @johannes to work this way, you just have to reconnect the circuits with a bit of wire.

#145

You'd need to mount the Axolotl the other way round, and the module would be way too wide, with the MIDI part as well.

Paul is designing an expander module to break out the MIDI in and out ports, the micro USB programming socket, and 2x CV outputs.

a|x

#146

Moved discussion about the MAX11300 adc/dac chip into a separate thread here: https://sebiik.github.io/community.axoloti.com.backup/t/max11300-for-cv-out-and-cv-out/1568

#147

Further to the questions I asked in the wrong place (doh) https://sebiik.github.io/community.axoloti.com.backup/t/euxoloti-hardware-version-confusion/1667 will the new pcbs still be open source?

#148

@edwinm
yes, I will put the files online at github, if everything works fine.

#149

hi everyone, as I posted in an other thread, the new proto boards have arrived…after assembling everything, beside button reading (done by a 74hc165) where I have forgotten an tristate buffer to block miso message while reading other ICs, worked…but now I got stocked. my adc don't read the right values and jumping. I also have a conflicts with different spi clock req. rate. Some source on the internet wrote that this a clear sign for spi error…

VDD is stable at 3,31V and adc channel 0-6 is stable at 1,647V. So dial position should be around 32.75…
I have tried three different pairs of mcp6004 op amps and mcp3208.
I also connected a brand new mcp3208 to axo and a breadboard and recorded axo<->adc spi communication by logic sniffer.
maximum analog<->digital ground resistance is not higher then 0,1Ω...

1. test round
ADC CH0 = VDD = Axoloti hex object: 0x87FF0000

SPI results ch0 = vdd signal monitor.png1893x906 23.4 KB

SPI results ch0 = VDD.png1920x1012 94.6 KB

  1. test round
    ADC CH0 = 1,647V = Axoloti hex object: 0x83D40000 or 0x83E40000

    SPI results ch0 = 1,647V signal monitor.png1893x906 23.3 KB
    SPI results ch0 = 1,647V.png1919x1012 79 KB

  2. test round
    ADC CH0 = GND = Axoloti hex object: 0x00800000

    SPI results ch0 = gnd signal monitor.png1893x906 25.5 KB
    SPI results ch0 = gnd.png1923x1014 83.4 KB

I got the same results with adc on breadboard and axoloti only.
Does anyone have any suggestions to solve this problem?
current prototype revision schmeatic

#150

update spi debug session.
OK, I think I know what's going wrong at SPI. it's a problem of clock cycle length and transmitting datas on MISO.
the main core + control pcb works

but if some other ICs are added to the spi bus, I got odd value from ADC.
It seems that the adc value are shifted by 1 bit while transmit session on MISO line.
What I have done since now:

  • checked VDD (axoloti, mcp3208, mcp6004) by oscilloscope and it's stable while powering from eurorack psu (Axo 5V = 4,98V and axo VDD is 3,31V). If euxo core pcb and axoloti is only powered by usb, axoloti monitor shows red voltage values (4,73V) for 5V line but now voltage drop outs or disconnections. VDD stays stable at 3,21V.

  • no shorting between communication lines or VDD and GND.

  • In last revision the adcs and op amps are powered by an external 3,3V vreg and Vref of each ADC (AVDD) is connected to VDD and its filtered by a 10Ω resistor. Pots are also connected to AVDD. I also disconnected ICs' VDD from ext. vreg and connected them to axo VDD vreg.

  • traceroute length MOSI = 101,7mm; MISO = 98,35mm; SCLK = 110,39mm; CS1(ADC 1/cv ins) = 65,57mm; CS2(ADC2/Pots) = 88,29mm
    eagle files

    euxo-core-1-1-pcb-layer-overview(web).jpg3000x1500 2.4 MB

  • measured resistance of several different points(VDD,AGND-GND, AXO GND-EURORACK, SPI IC-AXO etc.) and it's at max. 0,1Ω(tool has a tolerance of 0,1% :sweat_smile: )

  • checked spi message by bus pirate and logic sniffer at IC socket of both mcp3208 and directly at the breakout row next to the axoloti core header row. no problems, some longer cycles are recorded ( check post before) and axoloti stops to receive correct adc values.

  • Setup different ICs for SPI communication on a new breadboard …same but after a while axoloti shuts down like there is a shorting or something consumes more power then axo can deliver. breadboard ICs are powered by axo's VDD.

  • tried axoloti 1.0.6 and axoloti 1.0.10; and different Macs.

  • tested different SPI modes and clock speed. no problem with euxo core adcs at SPI only. same with connected control pcb.

  • tried 4 different type pairs of ICs.

  • tested previous euxo prototype and this one works out of the box with more ICs on SPI bus.

So does anybody know, what's the cause for the spi (clock) problem? I don't know how to fix this problem.

Edit:
here are axo patch and my script2 object code for reading pots and cv ins:
euxo cv&pot read
script2 object code

euxo pot-cv.png992x870 27.4 KB

SPI trouble (MISO/MOSI) clean signals
#151

What's the news on the hardware?

#152

Hi!

This is a very interesting project. It would be fantastic to use the Axoloti in a Eurorack system!

Is this project still active?
How are the already owners working with the Euxoloti? How is the experience?
Will it be posible to order an Euxoloti already mounted in the future?

Thanks!

#153

Hi Everyone,

Does anyone know what the status of Euxoliti is? I've seen the Euxoloti site still says new boards will be printed March 2017... but he hasn't posted here since last August, Hasn't tweeted since October, and hasn't commited anything on Github since February. Has anyone heard from Paul?

#154

hey , i build two versions of euxoloti and i can say Paul's a friend but the project needs a serious update there is still no correct v/octave
i wish more people get involved or maybe helping out the problem would be awesome !!!

#155

Okay, in that case maybe I'll have to break down and buy a few pcbs on OSH park... unless someone already has some lying around? :slight_smile:

#156

Last I heard, Paul had been having persistent internet-connection problems, and was therefore effectively offline, much of the time.

I imagine this has been hampering his ability to continue with the project (though, knowing Paul, in pretty sure he's been keeping busy with some interesting project or other).

Hopefully we will all hear from him again soon.

a|x

#157

Oh boy. I'd love to see this project come to fruition. I've been following it once in a while for a few years now. It would be the mostly friendly and community supported programmable module to date.

What can we do to get this going again?

#158

Same for me!
I would love an Axoloti Eurorack module!

#159

If anyone is interested i´d like to sell my build. It is the latest version and fully working except the gate8 expansion.
Sold as is (in need of completion) to the expierienced DIYer
I´d ship worldwide. just send me a pm if interrested

The Euxoloti is know sold!!!!!!!!!!!!!!!!!!!!!!!

#160

I went the lazy way myself. I'm making a Eurorack module out of mine with all the analog and digital control inputs wired to pots and switches. For CV control, I bought a doepfer a-192 voltage-to-midi interface module (though discontinued, Analog Haven still has some) which provides 16 CV ins. I will probably mount them both in the same panel. With that setup, although maybe not as elegant as a custom interface made for the Axoloti, you still have no worries of overloading the CV in on the Axoloti and have PLENTY of controls to play with if your patch needs it. I thought of dedicating some of the digital I/O lines into being gate outs with a simple transistor circuit, but that and the front panel (Front Panel Express) would be the only custom parts.

#161

I spoke to a number of people at Knobcon about this project. If none of them are going to bite, we're going to add a Eurorack solution for Axoloti so people can buy finished modules / kits.