#182

Assembly of the 1.1 prototype boards went well. Only found a couple of small silkscreening issues. Still - I think I’ll do one more run before considering it “done”.

Firmware hacking has gone very well. I’ve been optimizing things to squeeze more out of the processor. I was able to implement audio out for all four CV outputs with a simple oscillator with selectable waveform (sine, triangle, saw, square with variable width, and noise/sample and hold), several methods to control frequency (integer voltage number, note number or frequency), a sync method, and a few rectification modes. It is a balancing act of processing and not meant to replace a “real” oscillator - but so far I’ve had a lot of fun playing with it.

Here is a quick demo I put together the other day. The two connected TXo modules are the only sound source.

If you are curious as to what is going on: I’ve a four chord progression stored in a pattern on the Teletype. The right TXo is playing all four parts from each CV output. The left TXo is doubling the top and bottom notes with outputs 7 and 8 (its 3 and 4). Middle notes are sine waves, outer notes are square and saw. The first output of the second TXo (output 5) is oscillating at an LFO rate and slowly opening and closing a filter that is operating on all six mixed voices. Aside from a filter and mixer, the only other modules in the loop are a VCA (triggered by the first TXo trigger output) and reverb unit. The parameter knob is controlling the movement through the progression in four different modes.

Bonus points if you guess the music that I’m covering. :wink:

Cheers!

b

8 Likes
#183

Exciting …!
Sounds a bit like one of the first Ligeti’s ‘musica riccercata’ to me

#184

Not bad…it’s the Tristan chord from Wagner’s Tristan und Isolde mutated forwards and backwards. You can hear the sequence near the end (though not following the rhythmic suspensions).

:slight_smile:

b

1 Like
#185

UPDATE

Fixed the silkscreen goof and have another pressing in the queue at OSH. Reflow toaster should be ready later today to bake the boards (I’ll need to practice). Will be ordering some bulk pots for the TXi later today to go with my big bag of knobs.

Continue to tweak and optimize the TXo firmware. I implemented some AD envelope generators and threw together this little cover of Reich’s Piano Phase to show how they can interact with oscillator mode.

Patch Notes:

TXo 1 (CV 1-4):

  • four oscillators with different waveforms (sine, triangle, saw, and square) for each voice
  • single 12-step pattern with each oscillator offset by a couple steps and octaves for each voice
  • envelope mode active with short attack and decay (100ms) for each voice

TXo 2 (CV 5-8):

  • three slewwing cv values (0-10V)
  • been enjoying spreading the slews over an hour (defined in minutes using the new command SLEWM) - but for the demo set it in seconds to stay under instagram’s limit (SLEWS) :wink:

General:

  • Voice 1 is unfiltered; Voices 2-4 are each going through a different filter
  • All signal generation is from the TX modules
  • Slewing voltages are driving the filter cutoff and a VCA mixer
  • Everything gets a little reverb
9 Likes
#186

So exciting!..
so this is very close!

1 Like
#187

Great news !
Should i understand that the oscillators will be available for the TT as well ?

#188

Not as far as I can tell
(could this code could be folded into TT firmware easily?)

I’m inclined to leave it seperate and have the osc/env powers unique to these expanders

then tt, tx, and ansible each have their own lane to operate in

1 Like
#189

TT functionality has not changed; all new operators and capabilities require the expander hardware.

  • Oscillators and -10V support require the hardware.
  • Envelope generators and longer slew times (in seconds and minutes) could be coded for the TT, but it would be bespoke for the module and have implications in the core firmware itself - an area that I have tried to stay out of. I have no plans to do this at this point.

Now that the TX is so playable - I need to get back to finishing it instead of jamming!

Realized I didn’t share the frequency slew on the oscillator with you (portamento):

Reflow oven arrived, btw. Now I need to learn how to use it.

b

6 Likes
#190

rad

happy microwaving

1 Like
#191

SMD went well with the reflow oven. I added another pair of output modules to the prototype fleet. Here is a video of a single TT driving 20 TR outputs and 20 CV outputs. Pretty nuts!

b

7 Likes
#192

How many HPs is it ? 4 ?

#193

Each expander is 4HP. The TXo has 4 trigger outputs and 4 CV outputs for that width. The TXi has 4 potentiometers (PARAM) and 4 CV inputs (IN).

Theoretically you can have eight of each device type on the bus at one time using jumpers to set each device’s address. I’ve only had two inputs and four outputs going at one time. Given the practicalities of the TT, 4x TXo feels like a good peak of usefulness for the modules - with the sweet spot at 2.

Got another batch of boards from OSH and I’m going to test my assembly process again. I’ve got some metal panels on their way for evaluation. At that point, I’ll be able to start looking at making these for other folks. :slight_smile:

There are also two TT firmware things left to lick: i2c timeouts and reading i2c from an interrupt. Both can cause your TT to hang and need bounced. The i2c code from Atmel has an endless loop waiting for a reply…so it waits forever. Send a command to an address not on the bus and zzzzzzzzz…

The second issue is trickier as reads must be processed before continuing with the rest of the script. I’ve been studying both problems and will be looking into clearing them up sometime soon.

b

6 Likes
#194

Excellent news, progress is faster than I expected. \o/

#195

will the expanders fit in an isms case - 41mm?

#196

It should fit in an isms case - but not tons of extra room. I’ve been contemplating building some without the full two-sided headers - it doesn’t need them and would be way shallower, but it makes testing individual boards before assembly tricky and harder to debug any problem. I’m looking into making a little testing rig that allows me to pair known good boards with companion boards without having to solder them together. Pogo pins should arrive in a few days for this testing.

blob.jpg970×728 387 KB

b

2 Likes
#197

noted thanks for the info. To fit the isms i soldered the teensy direct to the board. Is it possible the 2 could go behind an 8hp panel instead of 2, 4 hp modules. I’m guessing people who buy them would buy both? or maybe not. Or would you consider making an 8hp panel as an option?

#198

I’ve only planning to do 4HP panels at this point. No real benefit with going for a single panel for multiple units other than saving a few screws. That said, I’ll be sharing all of the files and it would be easy to combine a few together for a custom panel.

As for configuration, I don’t think there is a one size fits all multi-panel that makes sense at this volume level. It really depends on how you use your TT. I find myself in a sweet-spot with a single input expander and a pair of output expanders. But, I’ve also made good use of 4 output expanders and no input expanders as a wild and crazy pseudo-random trigger + slewing cv monster.

b

#199

TX Update (Panels and General Progress)

Just got a test set of panels from Lasergist and switched out for knurled nuts on the jacks:

IMG_6824.JPG3024×1701 991 KB

PANEL REPORT:

  • everything fit perfectly
  • weight of the panel is fantastic - really make the modules feel done now that i’ve taken off the 3D printed ones
  • their coloring is a little more shiny and metallic colored than the teletype (LED lighting and my fleshy face is causing the champaign tint of the photo - to the eye they look more steely grey)
  • one panel (out of 15 total) had a slight mar on its face
  • upper right corner of each panel has some extra material and is a little less than square (where the final cut of the laser is i guess); need to get a better tool to clean it up if i go this direction
  • lettering isn’t exact with the TT - but still looks pretty good (same font)

their price is the lowest cost option i have sourced for these things that still gives a nice metal 2mm panel (flush with the TT). I’ve made some panels with Front Panel Express before - they look AMAZING, but cost $$$$$$$. unless i can find a similarly priced option for a small run (thinking approx. 120 divided) that gets a color more like the TT, i think this will be the route that I go for the panels.

GENERAL PROGRESS REPORT:

i’ve already started bulk purchasing some components (knobs, jacks, passives, etc.). still have a couple TT firmware gotchas to resolve and want to do a “production” build test where I try to make a bunch of boards at once using my manual pick-and-place tool and reflow oven to see how many i feel that i can commit to making.

but, in reality, this is more or less the finish line. with a lot of help from others (especially @Galapagoose back in the circuit design and layout phase) they are"done" in that: they work, they integrate with the TT, they cover all the basic TT functionality + add some new cool stuff, i’ve assembled a bunch of them with a 100% success rate and consistent function, firmware for the teensy is stable, and mods to the TT framework are pretty straightforward (save a last couple of things i want to see done).

once i’ve resolved the two TT issues (adding timeouts for calling non-existent expander ports and problems reading from the input expanders on the metronome interrupt), i think i’ll open up a new thread and start seeing who wants to get these babies. i’ll also post everything up to my github site so that those who are more adventurous can make their own.

cheers!

b

22 Likes
#200

What a great week to be in monome-land. arcs, ansibles and now this!

Cheers @bpcmusic. Beautiful work!

3 Likes
#201

Awesome work… Could you please also announce the availability of the expanders in this tread (in order for us to receive a notification email :slight_smile: …)?

1 Like