[ORDER] Euclidean Circles - LEDs with rhythm

From circuitbending to homebrew stompboxes & synths, keep the DIY spirit alive!

Moderators: lisa, luketeaford, Kent, Joe.

Post Reply
User avatar
av500
Super Deluxe Wiggler
Posts: 1705
Joined: Mon Feb 02, 2015 8:11 am
Contact:

Post by av500 » Fri Mar 22, 2019 2:47 am

I fear you still have some dead LEDs. the chips themselves seem to work OK as they pass the control data through the whole chain, but indidividual RGB LEDs seem to be dead or damaged.

murtsi
1-Post Wiggler
Posts: 1
Joined: Sat Apr 13, 2019 2:38 pm

Post by murtsi » Sat Jun 01, 2019 7:21 am

Silver wrote:Anyone know where else we can find the encoders? looks like sparkfun no longer has them in stock.
This perhaps: COM-15141, https://www.sparkfun.com/products/15141

-murtsi

User avatar
av500
Super Deluxe Wiggler
Posts: 1705
Joined: Mon Feb 02, 2015 8:11 am
Contact:

Post by av500 » Wed Jun 26, 2019 12:59 pm

murtsi wrote:
Silver wrote:Anyone know where else we can find the encoders? looks like sparkfun no longer has them in stock.
This perhaps: COM-15141, https://www.sparkfun.com/products/15141

-murtsi
Yes, the part number changed, it's COM-15141 now.

User avatar
av500
Super Deluxe Wiggler
Posts: 1705
Joined: Mon Feb 02, 2015 8:11 am
Contact:

Post by av500 » Tue Aug 27, 2019 3:56 am

forgotten 100uF/25V thru hole cap added to Mouser cart

Arneb
Wiggling with Experience
Posts: 364
Joined: Thu Jul 04, 2019 4:50 pm

Post by Arneb » Thu Sep 19, 2019 1:31 pm

Is it possible to hack the firmware? Non-DIYer here so far, but I do coding for a living and EC is one of those things which have made me think awfully hard about how an already best-in-its-class module could be even better.

User avatar
av500
Super Deluxe Wiggler
Posts: 1705
Joined: Mon Feb 02, 2015 8:11 am
Contact:

Post by av500 » Fri Sep 20, 2019 1:33 am

Arneb wrote:Is it possible to hack the firmware? Non-DIYer here so far, but I do coding for a living and EC is one of those things which have made me think awfully hard about how an already best-in-its-class module could be even better.
it is not open source, but nothing would stop you from putting your own firmware...

Arneb
Wiggling with Experience
Posts: 364
Joined: Thu Jul 04, 2019 4:50 pm

Post by Arneb » Fri Sep 20, 2019 7:39 am

I see :waah: doing it from scratch will probably be a bridge too far, I'll see what I can do. Thanks anyways!

User avatar
Silver
Common Wiggler
Posts: 212
Joined: Mon May 02, 2016 5:02 pm
Location: New York

Post by Silver » Fri Sep 20, 2019 2:05 pm

av500 any plans for an envelope mode???

Arneb
Wiggling with Experience
Posts: 364
Joined: Thu Jul 04, 2019 4:50 pm

Post by Arneb » Fri Sep 20, 2019 2:09 pm

Uh... how would you design an envelope generator with three encoders with 16 discrete values?

User avatar
Silver
Common Wiggler
Posts: 212
Joined: Mon May 02, 2016 5:02 pm
Location: New York

Post by Silver » Fri Sep 20, 2019 3:38 pm

I was envisioning something like the Pamela's New Workout.

For an LFO mode, you could have one encoder select the waveshape of the envelope or LFO and then use the discrete values to select the pulse width or skew of the wave.

For envelope mode, each output trigger of a circle would trigger the envelope. Each of the circles could be set to envelope mode, in which case you would split each circle into quadrants or half, each of which quandrant/half sets a part of a crude AD/ADSR. or alternatively have only one circle triggering the envelopes on one channel and use the other circles to set the ADSR values.

Just brainstorming... but the envelope and LFO modes on Pamela's New Workout add a tremendous amount of functionality.

Arneb
Wiggling with Experience
Posts: 364
Joined: Thu Jul 04, 2019 4:50 pm

Post by Arneb » Sat Sep 21, 2019 6:37 am

Frankly, trying to beat Pamela at her own game is not what modular is about.

Specifically, how would you go about setting an LFO frequency or an envelope attack length? Unless I'm misunderstanding something really badly EC doesn't make any assumptions about incoming clocks, it just responds to triggers. It has no concept of "time" at all. Which is a good thing in my book because you can feed EC a swung clock or do breaks via reset input and it won't flinch. Or even trigger it at audio rates and get a bunch of weird FM suboscillators. But it also means that you can't do something like "rise until next step" because EC is completely clueless about when the next step is going to be.

And that's not even going into the additional menu-diving in a module that's plain not supposed to be Pamela's or T_U in this respect, or into technical limitations of processing power or flash size (because I know nothing about those).

User avatar
Silver
Common Wiggler
Posts: 212
Joined: Mon May 02, 2016 5:02 pm
Location: New York

Post by Silver » Sat Sep 21, 2019 7:01 am

I think adding functionality is never a bad thing. If you don’t want to use the function, don’t. It would be a mode, that could be selected, much like the other modes that are already there. I think that was the whole point of v2, no? (Was also the whole point of Pamela’s NEW workout vs the original)

Also, as it is I am often triggering an envelope with at least one of the channel outputs from EC, so why not try to bring that function internally.
If you think creatively about the visual interface of the led rings, I think there are a lot of options that don’t have to be too menu diving heavy.

EC also has an internally driven clock option, so you could certainly set a rate for the LFO mode.

For an envelope you don’t need to trigger each step. You just need to set a rise/fall time for each step after it’s been triggered. I described one solution for the selection of the time for each segment of a AD or ADSR above.
Based on the EC and his other creations, I’m sure av500 could come up with an even more elegant solution.

Personally, it would save me the need for a separate envelope module, which is highly appealing in a small rack.

pld
Ultra Wiggler
Posts: 894
Joined: Thu Mar 05, 2015 5:15 am
Location: Germany

Post by pld » Sat Sep 21, 2019 8:26 am

"Challenge accepted!" is fun and all, but a perhaps more fundamental question is "how would you design an envelope generator given all outputs are logic only"?

User avatar
Silver
Common Wiggler
Posts: 212
Joined: Mon May 02, 2016 5:02 pm
Location: New York

Post by Silver » Sat Sep 21, 2019 3:20 pm

Oh, can the outputs only be set as logic? I assumed the microprocessor outs were just voltage outputs? Or is there something else in the hardware architecture that would only allow for logic outs?

pld
Ultra Wiggler
Posts: 894
Joined: Thu Mar 05, 2015 5:15 am
Location: Germany

Post by pld » Sun Sep 22, 2019 1:50 am

It'd be really nice if the outputs were all that flexible, but the processor is digital logic. A variable voltage is analog. Somehow you have to get from one domain to the other, which isn't free so most pins will default to logic outputs only. Similarly on inputs.
I don't think those nice small & cheap F0s even have a DAC at all, and even larger chips may have only 2 channels. So you can either go to PWM (which has tradeoffs) or you need an external converter (which costs extra). Either way there's a bunch more "stuff" than just a simple inverter/buffer (see EC schematic) and you kind of have to decide up front what you want.

User avatar
Silver
Common Wiggler
Posts: 212
Joined: Mon May 02, 2016 5:02 pm
Location: New York

Post by Silver » Sun Sep 22, 2019 7:22 am

I see. Bummer. That totally makes sense though. Thanks for the quick lesson!

User avatar
strathTerrace
Learning to Wiggle
Posts: 15
Joined: Fri Nov 30, 2018 10:46 am
Location: Sacramento, CA USA

Post by strathTerrace » Mon Oct 07, 2019 7:29 pm

Hello All,

I just completed a Euclidean Circles V2 DIY. The encoders are lighted and it seems to be functioning except that not a single LED in the circles is lighted. I can send signals from all six channels to modules and get rhythms out and tap tempo is working. I've checked and rechecked, and all of the LEDs are oriented correctly, seem to be fully soldered in place, and all seem to be getting voltage across them.

Otherwise all components look like they are properly oriented and have the correct values.

It seems like there is a very obvious solution, but I can't find it. Any ideas?

User avatar
av500
Super Deluxe Wiggler
Posts: 1705
Joined: Mon Feb 02, 2015 8:11 am
Contact:

Post by av500 » Tue Oct 08, 2019 2:34 am

strathTerrace wrote:Hello All,

I just completed a Euclidean Circles V2 DIY. The encoders are lighted and it seems to be functioning except that not a single LED in the circles is lighted. I can send signals from all six channels to modules and get rhythms out and tap tempo is working. I've checked and rechecked, and all of the LEDs are oriented correctly, seem to be fully soldered in place, and all seem to be getting voltage across them.

Otherwise all components look like they are properly oriented and have the correct values.

It seems like there is a very obvious solution, but I can't find it. Any ideas?
can you post a detailed pic of the main board and the front panel? with the encoders working it means the main board is mostly OK

User avatar
strathTerrace
Learning to Wiggle
Posts: 15
Joined: Fri Nov 30, 2018 10:46 am
Location: Sacramento, CA USA

Post by strathTerrace » Tue Oct 08, 2019 9:59 am

Here are some pix. Thanks

Image
Image

User avatar
strathTerrace
Learning to Wiggle
Posts: 15
Joined: Fri Nov 30, 2018 10:46 am
Location: Sacramento, CA USA

Post by strathTerrace » Tue Oct 08, 2019 10:02 am

Whoops, too small. Trying again. Thanks

Image

Image

User avatar
av500
Super Deluxe Wiggler
Posts: 1705
Joined: Mon Feb 02, 2015 8:11 am
Contact:

Post by av500 » Tue Oct 08, 2019 10:16 am

so all 3 encoders work and change colors when you press them? that means that the LED control signal is OK on the main board

User avatar
av500
Super Deluxe Wiggler
Posts: 1705
Joined: Mon Feb 02, 2015 8:11 am
Contact:

Post by av500 » Tue Oct 08, 2019 10:36 am

from the last WS2811 its only a short trace to the front panel and I assume the 5V are also OK since the encoder LEDs work.

Image

User avatar
strathTerrace
Learning to Wiggle
Posts: 15
Joined: Fri Nov 30, 2018 10:46 am
Location: Sacramento, CA USA

Post by strathTerrace » Tue Oct 08, 2019 3:09 pm

Yes, all three encoders seem to work and change colors when I press them. Also functionality assigned to encoder two and one appear to work as specified.

User avatar
strathTerrace
Learning to Wiggle
Posts: 15
Joined: Fri Nov 30, 2018 10:46 am
Location: Sacramento, CA USA

Post by strathTerrace » Tue Oct 08, 2019 3:17 pm

I will test the voltage across the pins that you show in the schematic above. I did use the test points on the back of the board earlier and noticed that the Five V tp read about 4.7 V. The 3.3 V test point read 3.3 V. I have not checked the 5 V output from my power supply. When I get home from work later this evening I will double check that

User avatar
av500
Super Deluxe Wiggler
Posts: 1705
Joined: Mon Feb 02, 2015 8:11 am
Contact:

Post by av500 » Tue Oct 08, 2019 3:22 pm

4.7V sounds about right, there is a slight drop across the 10R resistor (3x 33R)

did your check the soldering of the SMT header on the back of the panel?

Post Reply

Return to “Music Tech DIY”