MUFF WIGGLER Forum Index
 FAQ & Terms of UseFAQ & Terms Of Use   Wiggler RadioMW Radio   Muff Wiggler TwitterTwitter   Support the site @ PatreonPatreon 
 SearchSearch   RegisterSign up   Log inLog in 
WIGGLING 'LITE' IN GUEST MODE

SSSR Labs SM010 Matrixarchate
MUFF WIGGLER Forum Index -> Eurorack Modules Goto page 1, 2, 3  Next [all]
Author SSSR Labs SM010 Matrixarchate
rpocc
Hi pals!

I just had released a new module called SM010 Matrixarchate. It is a sequenceable crosspoint switch matrix with user programs, you can read more info here:
https://sssrlabs.com/en/store/sm010/



A couple of videos:
https://www.youtube.com/watch?v=0CT25qEIg0U
https://www.instagram.com/p/BXA34MVgCpB/

This module has ISP interface for easy update and the I2C interface for remote control, but the control device itself isn't yet designed even as the prototype, although I'm going to start working on it very soon and this device will be not only for Matrixarchate but for the whole series of modules.

Also, I've already had requests to implement control by Monome with Teletype interface and Macro Machines Storage Strip. I do not have this equipment, but I'm totally open to any suggestions.

If I had an understanding of the interface requested for support, I can release different versions of the firmware.

What suggestions could you give to this module? I understand, there is not much information: the manual isn't yet written, I did not make a video with explanations, but I'm working on it.

The short list of currently implemented features:
* Passive, bi-directional computer-controlled 8x16 crosspoint matrix switch, supports +/-6v. (Based on CMOS switch array)
* Can store up to 60 programs (or up to 30 presets with custom graphic dot art per program.
* Particular set of programs can be picked for sequencing.
* 4 sequencing directions (forward, backward, pendulum, random)
* Can overwrite a program with another one.
* Can temporarily change a program and revert it to the last saved state.
* Adjustable brightness
* A special "lights-off" mode for improved s/n ratio and reduced current draw
* Quick program recall mode
* Two styles of program numbering (1, 2...60 and 1A, 1B...4D) with optimised quick-recall key combinations for each mode.

Owner's manual
Southfork
Looks monstrous!
peripatitis
Very interesting, especially with Teletype tools driving it..
desolationjones
This looks incredible for patch automation! And great value. So now at the same MSRP ($330) we now have the WMD Sequential Switch Matrix and the SSSR Matrixarchate.

SSM:
    + Buff mults for each row
    + Summing for each column
    + DAC and comparator outputs
    + CV control
    + Trigger/gate expander
    - 4 inputs, 4 outputs (dedicated in/out)

Matrixarchate:
    + 16 x 8 matrix (!!!)
    +/- Passive connections, bidirectional (feedback patching, anyone?)
Paranormal Patroler
Cool. Saw it on your site and then MG. Congrats on a lovely design, it looks so playable!
adh82
This looks awesome!!!
mbartkow
One question, regarding the passive nature of the switch matrix. Assuming I would like to route one signal to several destinations while having it mixed with some different signals (a different one for each of the destinations) - is it possible?

For example - signal 1 mixed with signal 2 goes to A, and signal 1 mixed with signal 3 goes to B. Am I right to suspect both A and B will receive a mixture of 1, 2, and 3?
rpocc
mbartkow wrote:
One question, regarding the passive nature of the switch matrix. Assuming I would like to route one signal to several destinations while having it mixed with some different signals (a different one for each of the destinations) - is it possible?

For example - signal 1 mixed with signal 2 goes to A, and signal 1 mixed with signal 3 goes to B. Am I right to suspect both A and B will receive a mixture of 1, 2, and 3?


This is exactly what I'm going to discuss.

Yes, your suspicion is correct, all three signals will be shortened together. This is a nature of matrices.

The passive matrices do not allow summing of signals, only weighted averaging if your signals have similar or better, the same output impedance. In this case, the passive matrix can distribute one source to many destinations perfectly, as a usual passive multiple, and can mix signals together in the EMS manner. It's not OK for V/Oct connections, but totally OK for modulations.

The known issues of this matrix is it's low input voltage range, no buffering and no overvoltage protection. Again, it's OK for EMS-style systems unless you're using 10v envelopes, but may be considered too primitive for advanced users.

Understanding that I have developed three unreleased projects of upgrade/plugin/extensions for this matrix, each has its own pros and cons. I would like to discuss them. I really need your opinion on this because starting any production cycle requires a lot of money and I can't just produce 3 more projects with no understanding how much each of them is actually needed.

All prices are estimations, cannot guarantee anything, but I'm trained to estimate budgets. All options increase the module depth by 13mm.

Option 1: Diode protection plugin board
Simply, a board with a voltage reference circuit and clipping diodes in parallel to all terminals. Total 48 diodes.
Cost: $20-$25/kit, few bucks more assembled.
Pros: Adds overvoltage protection to use this module with 10v sources and crappy manual gates. The bidirectional connection is still possible.

Option 2: A simple summing plugin board
1/2 gain stage at each of the 16 inputs, summing stage with 2x gain at each of the 8 outputs. 6 OpAmps, LOT of resistors. Can fit one board.
Cost: $50-60/kit, twice more assembled.
Pros: Adds perfect summing at low cost, plug and play; allows keeping all programs.
Cons: The signal distribution no longer makes sense. Only one-way connection.

Option 3: A summing option with allowed distribution
The alternative solution that requires not 1, but as much as 8 matrix switch chips. along with all contents of the option 2. Makes this module a perfect, but quite pricey.
Cost: Max. $200-250/kit, another $300-$350 for an upgrade. It hardly depends on prices for the used chip.

Pros: Correct summing, correct distribution, precision channels as a free bonus.
Cons: Most probably, impossible to keep the logic board, so it's rather an upgrade than a plugin, but can be an option with the lowered total cost of the complete set.
blinosynth
this module is insane
option 3 w00t w00t w00t <3
thank you SSSR guys
fishdog
any thoughts about a diy kit version?
desolationjones
rpocc help me understand the need for option 1: is there danger of damage to the Matrixarchate from voltages in the +/-10V range? If so then I feel option 1 should be standard on all units. Does it handle +8V okay currently?

Option 3 sounds like it would put the module into a $500 price bracket. I think that is still quite competitive for the feature set.
RussiaZero23
rpocc wrote:
mbartkow wrote:
One question, regarding the passive nature of the switch matrix. Assuming I would like to route one signal to several destinations while having it mixed with some different signals (a different one for each of the destinations) - is it possible?

For example - signal 1 mixed with signal 2 goes to A, and signal 1 mixed with signal 3 goes to B. Am I right to suspect both A and B will receive a mixture of 1, 2, and 3?


This is exactly what I'm going to discuss.

Yes, your suspicion is correct, all three signals will be shortened together. This is a nature of matrices.

The passive matrices do not allow summing of signals, only weighted averaging if your signals have similar or better, the same output impedance. In this case, the passive matrix can distribute one source to many destinations perfectly, as a usual passive multiple, and can mix signals together in the EMS manner. It's not OK for V/Oct connections, but totally OK for modulations.

The known issues of this matrix is it's low input voltage range, no buffering and no overvoltage protection. Again, it's OK for EMS-style systems unless you're using 10v envelopes, but may be considered too primitive for advanced users.

Understanding that I have developed three unreleased projects of upgrade/plugin/extensions for this matrix, each has its own pros and cons. I would like to discuss them. I really need your opinion on this because starting any production cycle requires a lot of money and I can't just produce 3 more projects with no understanding how much each of them is actually needed.

All prices are estimations, cannot guarantee anything, but I'm trained to estimate budgets. All options increase the module depth by 13mm.

Option 1: Diode protection plugin board
Simply, a board with a voltage reference circuit and clipping diodes in parallel to all terminals. Total 48 diodes.
Cost: $20-$25/kit, few bucks more assembled.
Pros: Adds overvoltage protection to use this module with 10v sources and crappy manual gates. The bidirectional connection is still possible.

Option 2: A simple summing plugin board
1/2 gain stage at each of the 16 inputs, summing stage with 2x gain at each of the 8 outputs. 6 OpAmps, LOT of resistors. Can fit one board.
Cost: $50-60/kit, twice more assembled.
Pros: Adds perfect summing at low cost, plug and play; allows keeping all programs.
Cons: The signal distribution no longer makes sense. Only one-way connection.

Option 3: A summing option with allowed distribution
The alternative solution that requires not 1, but as much as 8 matrix switch chips. along with all contents of the option 2. Makes this module a perfect, but quite pricey.
Cost: Max. $200-250/kit, another $300-$350 for an upgrade. It hardly depends on prices for the used chip.

Pros: Correct summing, correct distribution, precision channels as a free bonus.
Cons: Most probably, impossible to keep the logic board, so it's rather an upgrade than a plugin, but can be an option with the lowered total cost of the complete set.


Very Hard to pick just one. I would say Option 1 is a must. In many systems I can find many envelops that are 8 Volts + in my own system I have 16 envelops that are 8volts or greater (10 volts max). Also there are some Random Modules that out put more than 8Volts as well. SSF URA is one and Qbit discontinued model is another of the top of my head.

Option 2 does not seem worth it for the compromises.

Option 3 is the best but what would be the total cost for the end user then?
$700 for 16 I/O bi directional 8 I/O that can accept 10 volts and do summing + Precision 1 Volt per Octave + Distribution. That seems Worth it to me.

As the module stands right now, I could not use it at all for most of my signals are out of its range.

Maybe a DIY version of Option 3 where you buy the whole module as a DIY project to keep the cost down and allow advance users (or just users with 8 Volt + modules) to enjoy the Fast re patching Matrix control.
dodecabilly
I'd say - make the module the best it could be, and price it accordingly
rpocc
I have a very good news!

Fact 1:
According to the datasheet, the matrix switch chip have total maximum voltage swing of 16 Volts, so it's OK to run it from +10 / -6 V. So, at least you can consider that it SUPPORTS up to +10v, at least I can replace 6v regulators with 10v.

Fact 2:
I've just tried to replace one of the positive regulators I use with the matrix chip, with a simple 300R resistor, limiting 12v current to about 30 mA and in fact, right now I'm running this chip of +12/-6v supply and it works perfectly with Intellijel Dual ADSR. I even can't see any trace of distortion, clipping, or whatsoever, comparing the input and output signal with my DSO.
I will keep this thing to work all night and will see if there is any visible degrade.

And by the way, there are actually clipping diodes inside the chip, so no external voltage can really make damage to the chip unless it's really high-current.

I think it's possible because the chip is designed for switching video signals, so absolute maximum ratings are calculated for intended use. Maybe, CMOS transistors inside may overheat from passing 75 Ohm signal through it when powered from too high voltage, or slew rate may degrade and this chip will not correctly pass 4 MHz composite signal powered of 18v p-p, but this is definitely not our case.

Summarizing the above, I think, the voltage range issue is solved and option 1 is not really needed.

I appreciate every comment left in this topic. It helps me to figure out how to seriously improve the Matrixarchate even when it's officially released.
rpocc
I had tested extended voltage range for more than 50 hours, so I confirm that all modules will be rated for -6...+10 Volts. It doesn't affect the price or anything and all modules that are scheduled for shipping on 20 November, will have this feature.

Also, I have decided to do the following things:

1. This module remains as is since it features passive connections, that may be unique for some applications (I like how it works already with my modules)

2. The summing/ buffered distribution upgrade board will be the right next product to develop.

3. This module will support Teletype.
Trebbers
Forgive me if I am misunderstanding, but as is, a <-6v input would still have the potential to damage the module? Not uncommon to have envelope generators with inverted outputs that would give you -8v .

Understand there's extra cost involved, but inputs should be safe for the entire range of voltages possible in a Eurorack system. No matter how well you spell out the danger/limitations to end users, some won't understand until they've run into a problem.
gruebleengourd
Trebbers wrote:
Forgive me if I am misunderstanding, but as is, a <-6v input would still have the potential to damage the module? Not uncommon to have envelope generators with inverted outputs that would give you -8v .


I think you missed this part of the post:

Quote:
nd by the way, there are actually clipping diodes inside the chip, so no external voltage can really make damage to the chip unless it's really high-current.



What will happen if you go below -6V is that the signal will clip, not that anything will be damaged.
rpocc
The owner's manual is ready! It's available on the product page, or by the following link:
http://www.sssrlabs.com/wp-content/uploads/2017/11/SM010_Manual_EN.pdf
TechForze
I was looking for the A-150-8 (Octa switch), but this SM010 is it's big brother..!
Am I right?
RussiaZero23
Any more videos showing the adding and taking away of row connections in real time? For example holding down the Vertical row 1 button then with the other hand quickly press in fast sequence Horizontal buttons A thru H. Showing how fast one can route 1 signal from input 1 to 8 outputs (A to H).

Also using the unite in a live performance type way. For example staying on patch 1 and just turning on and off parts in the grid to create a live dynamic song or changes in sound.
rpocc
TechForze wrote:
I was looking for the A-150-8 (Octa switch), but this SM010 is it's big brother..!
Am I right?

The Doepfer unit has 8 independent 1x2 "matrices", each with CV control having a threshold point for choosing between two choices. Might be interesting in use with a set of square LFOs.

SM010 is a single 8x16 matrix with digital-only control, so switching is either manual or a program-based.


RussiaZero23 wrote:
Any more videos showing the adding and taking away of row connections in real time? For example holding down the Vertical row 1 button then with the other hand quickly press in fast sequence Horizontal buttons A thru H. Showing how fast one can route 1 signal from input 1 to 8 outputs (A to H).

Also using the unite in a live performance type way. For example staying on patch 1 and just turning on and off parts in the grid to create a live dynamic song or changes in sound.

Nice proposals for a demo, thanks!
I'm going to shot it very soon.
A_Mc
I´m super interested in this module. it seems to get rid of some of my hustle when playing live. especialy the program change when triggered. super dope!

On modulargrid it is stated that it will be available as diy project to. though i can´t find any kits in your shop. i allways tend to build stuff myself so thats the prefered approach. Will there be any kits available in the future or was that a mistake?
rpocc
A_Mc wrote:
I´m super interested in this module. it seems to get rid of some of my hustle when playing live. especialy the program change when triggered. super dope!

On modulargrid it is stated that it will be available as diy project to. though i can´t find any kits in your shop. i allways tend to build stuff myself so thats the prefered approach. Will there be any kits available in the future or was that a mistake?


Yes, SM010 will be available as a DIY kit already this week with the price of $220/pcs.

By the way, there is firmware update v.1.1 released. (On the product's page) All further RETAIL units will be shipped already updated, but kits are already packed with the firmware v. 1.0, so I suggest to check if you have access to the ISP programmer or at least, to an Arduino board.
Daisuk
rpocc wrote:
A_Mc wrote:
I´m super interested in this module. it seems to get rid of some of my hustle when playing live. especialy the program change when triggered. super dope!

On modulargrid it is stated that it will be available as diy project to. though i can´t find any kits in your shop. i allways tend to build stuff myself so thats the prefered approach. Will there be any kits available in the future or was that a mistake?


Yes, SM010 will be available as a DIY kit already this week with the price of $220/pcs.

By the way, there is firmware update v.1.1 released. (On the product's page) All further RETAIL units will be shipped already updated, but kits are already packed with the firmware v. 1.0, so I suggest to check if you have access to the ISP programmer or at least, to an Arduino board.


Nice! Will the DIY build be with only through hole components, or will there be SMD as well?
rpocc
Daisuk wrote:

Nice! Will the DIY build be with only through hole components, or will there be SMD as well?


It has only one SMD component, the AMS1117-5.0. It's a pretty big thing, easy to solder.
MUFF WIGGLER Forum Index -> Eurorack Modules Goto page 1, 2, 3  Next [all]
Page 1 of 3
Powered by phpBB © phpBB Group