midiphy SEQ v4+

[latigid on]

45 minutes ago, ilmenator said:

Peter, nice work! Also good to see a mini-TPD,

To differentiate from Peter's TPD (or your PCBs), the one built into the SEQ is called the "activity" matrix. This means we have a JA board (jog wheel + activity). When he has time, Peter intends to add some extra features to display. Not sure if they'll necessarily be compatible with the TPD. 

 

45 minutes ago, ilmenator said:

albeit this seems to have only a single color?

Well spotted! The colour choices on the smaller ones are quite limited, although it goes with the "clean" look of the UI.

 

45 minutes ago, ilmenator said:

Are the UI schematics publicly available or do we need to reverse-engineer them based on the v4+ config file?

I'll probably put the matrix schem up at some point. For now, I'm happy to supply it to those who have purchased the PCBs (just Peter, Adrian H, TK. and lazy beta testers :P ) or to those who need troubleshooting advice.

[latigid on]

With regards to naming, I don't want to take the v4+ moniker away from Wilba ones. Although it's true that the version I've designed (with input from Adrian/TK./Peter/Bruno) does represent UI improvements and thus fits with the "plus" suffix as put forward by TK., any F4 Core version will run the V4+ firmware. 

TK. created the hardware config "antilog," which is the name I use for PCB projects. You could also call it the latigid version, the andy version, whatever :).

[ilmenator]

13 minutes ago, latigid on said:

I'll probably put the matrix schem up at some point. For now, I'm happy to supply it to those who have purchased the PCBs (just Peter, Adrian H, TK. and lazy beta testers :P ) or to those who need troubleshooting advice.

That's not so nice, what a pity. So I'd need to buy a PCB from you in order to receive the schematics, although I don't even want to use your board?

[latigid on]

It's not a priority at the moment, sorry.

[Antichambre]

2 hours ago, ilmenator said:

Are the UI schematics publicly available or do we need to reverse-engineer them based on the v4+ config file? I'm thinking about my own homebrew version using Marquardt switches, I don't like those christmas tree ones you are using :-).

29 minutes ago, ilmenator said:

That's not so nice, what a pity. So I'd need to buy a PCB from you in order to receive the schematics, although I don't even want to use your board?

I suppose you will use the Marquardt switches just to replace the Mattias ones(I supoose they are the "christmas tree ones" you don't like, Humm). Then you will need the PCB set from Andy and Peter, you need the Jog menu and Encoders boards, you just have to replace the LeMec boards by your own, so no trouble... if you really want to do that...
I you don't want to use any Andy's Boards then you can go from scratch, with the pinout of J8/9 and a blank Config file, no?
 

Edited September 17, 2018 by Antichambre

[ilmenator]

2 minutes ago, Antichambre said:

I you don't want to use any Andy's Boards then you can go from scratch, with the pinout of J8/9 and a blank Config file, no?

Sure, that is always an option. However, having two different config files doesn't really make sense with the underlying functionality being exactly the same. But then, that approach also leaves a lot of freedom for changes, so that could possibly be a good solution as well.

52 minutes ago, latigid on said:

It's not a priority at the moment, sorry.

I don't think you'd need to be afraid of schematics in the public - on the contrary, it would probably help you a lot once you have to do troubleshooting and support on your customers' builds. This community has been awesome in the past when it came to solving build problems. However, I do respect your priorities.

I'm looking forward to seeing the v4+ taking off!

[Antichambre]

2 minutes ago, ilmenator said:

having two different config files doesn't really make sense with the underlying functionality being exactly the same

As you said, if you've got the config file then you've got the position of the elements over the SPI chain, like schematic you just have read it, not(more) a big deal. So if you want to respect the V4+ config file with your own board, that's possible too, without waiting for Andy to get time to perfect and release his schematics ;)

Anyway...

2 minutes ago, ilmenator said:

I'm looking forward to seeing the v4+ taking off!

 

Edited September 17, 2018 by Antichambre

[ilmenator]

Just now, Antichambre said:

If you've got the config file then you've got the position of the elements over the SPI chain. So if you want to respect the V4+ config file with your own board, that's possible too, without waiting for Andy to get time to perfect and release his schematics ;)

I'm aware of that already, but that is a reverse-engineering process that I'd rather not spend my time on.

If the schematic was clean enough to create the board file with, then that is clean enough. I understand the term "priority" a little differently, but hey... It's good to see progress with our beloved SEQ!

Edited September 17, 2018 by ilmenator

[Antichambre]

Just now, ilmenator said:

I'm aware of that already, but that is a reverse-engineering process that I'd rather not spend my time on.

If the schematic was clean enough to create the board file with, then that is clean enough. I understand the term "priority" a little differently, but hey...

I did not understand for a moment that you wanted the original files, I thought rather a pdf format. Sorry

Maybe I'm wrong... but for me retro-engineering is analyzing a hardware and extract the data from it, not recreating something by reading a schematic or a text file, cause in both cases the data are already available.

Best
Bruno
 

Edited September 17, 2018 by Antichambre

[eptheca]

It seems that my suggestion for a clarification about the name was not that stupid, because you all explain the SEQ V4+ a little different.

How can the V4+ be the new UI/frontpanel when I am already running it with a Wilba frontpanel?

The way I see it, V4+ is the firmware version that only runs on a STM32F4 core. It can be used with a Wilba frontpanel and a latigid on frontpanel just with a different HW file. Some new functions will work on both frontpanels, like it already does with CC layers for drum tracks, and some new functions will probably only work or work best with the latigid frontpanel.

Therefor it makes sense to me to be clear about the name.

From the changelog:

MIDIbox SEQ V4+ is a special firmware variant for the STM32F4 core.

 

[Hawkeye]

@ilmenator no worries, we are planning on releasing the schematics, we are the good guys! :) Also be assured, that In this case "priority" really is that, we have to still validate the new UI boards, adjust configuration for left hand / right hand jog models, create the second/final soldering/build video tutorial, slightly modifiy the firmware for the new activity matrices, and have a dozen other little tidbits on our todo lists. It is just a lot of work, and the release of the schems is not being on the top of the list. As we have normal day jobs and families, these are all time-consuming "after-work" or night activities, deeply cutting into the hours normally allocated for sleep, atm :).

@eptheca yes, technically you are automatically running a "v4+" with any stm32f4. It is caused by a code preprocessor "define" looking for the processor variant. This is for simplicity, so that only a single firmware .hex is needed for any stm32f4. Nevertheless, you won't be able to use the new v4+ UI features until you enable the "antilog panel" hwcfg section to benefit from its features.

We have no problem whatsoever to call the v4+ officially "midiphy SEQ v4+", if it helps to differentiate things. For me, the only important point is that "v4+" is in the name, because that defines the hardware capabilities of the unit, that the software can rely on.

Best regards, Peter

[ilmenator]

34 minutes ago, Hawkeye said:

we are the good guys! :)

Thanks Peter, I never really doubted this - I can see how much work was put into this project. I'm glad to see the documentation coming up, eventually and as your time allows!

9 hours ago, Antichambre said:

I did not understand for a moment that you wanted the original files, I thought rather a pdf format.

I do not care so much about the format, and at least with the tools that I use, a PDF file is only a mouse click away...

43 minutes ago, Hawkeye said:

we have to still validate the new UI boards,

...but of course verification and validation is essential before putting out documentation.

Thanks, and keep up the good work!

[Smithy]

14 hours ago, latigid on said:

happy to supply it to those who have purchased the PCBs (just Peter, Adrian H, TK. and lazy beta testers :P ) or to those who need troubleshooting advice.

Hey some of us lazy beta testers were delayed due to components missing in the previous BOMs, just saying! ;)

Also I'd rather wait to have my hands on the case so I'll know exactly how long to cut the ribbon cables or cables that connect the boards together.

 

 

[Antichambre]

35 minutes ago, Smithy said:

Hey some of us lazy beta testers were delayed due to components missing in the previous BOMs, just saying! ;)

lol, I'm a lazy beta tester too, I totally assume that  I preferred waiting for you Guys, instead of find some difficult part by myself and make mistake e.g. rgb leds
And finally now I prefer waiting for the new UI boards set, the one which matches with the definitive enclosure.

Have a good day everybody!

[latigid on]

 

Lefty version from Adrian. The Apem caps on the JA board are too tall, they should all be the same height. The blue will be lighter in real life.

[Antichambre]

1 hour ago, latigid on said:

The blue will be lighter in real life.

I like this blue too... Is it the picture? If it's not a the picture and not a color accident in the dosage I'm interested in this color code.
 

1 hour ago, latigid on said:

The Apem caps on the JA board are too tall

Finally what is the best height for the Apem? How much is a 15mm button flush at rest?

[latigid on]

Just now, Antichambre said:

I like this blue too... Is it the picture? If it's not a the picture and not a color accident in the dosage I'm interested in this color code.

Would need to talk to Adrian about different colours. It might be that the cases are ordered in low (e.g. <10) quantity, which could leave some room for customisation. Very likely we'll keep a stock of "standard" colours though. It's too much to keep too many options, especially considering we start with two possible variants (LH/RH).

 

Just now, Antichambre said:

Finally what is the best height for the Apem? How much is a 15mm button flush at rest?

I'll leave you to check the datasheets if you're interested. The switches have 1mm travel. 19.0mm caps are suggested.

https://www.apem.com/int/multimec-3f-287.html
http://www.produktinfo.conrad.com/datenblaetter/700000-724999/705191-da-01-en-MULTIMEC_KAPPE.pdf

 

[Antichambre]

I know you suggest the 3F 1S 19mm.
But on all the pictures it seems high, like 4 or 5mm over the front panel surface. I just wanted to see if the 5G 1LS 15mm(which can be easily sourced too) could fit.

[latigid on]

Sure, might even be better. I left the panel spacing up to Adrian and have never seen the case in person yet :)

Not sure if 5G fits? But there's 16mm:

https://www.mouser.com/ProductDetail/MEC-Switches/1S11-160?qs=XkGcM6gst%2F8pgAO1bVm7xw%3D%3D

 

Edited September 18, 2018 by latigid on

[Antichambre]

30 minutes ago, latigid on said:

Sure, might even be better. I left the panel spacing up to Adrian and have never seen the case in person yet :)

Oh! Without worrying too much about, I think this is an important aspect of your machine for finishing...

Something like the 1LS096-16.0 should be the best, I know it's a question of taste but I really like the frosted round lens.
+
But this one is not distributed in 16mm, no stock anywhere.  And I'm afraid the 15mm is too short.
The only solution is to order it from APEM, but they want minimum quantity. Should be included in a future version of the essential kit maybe?

Best
Bruno

[latigid on]

I haven't tested the 5E, indeed it might be too short. The LED mounting is also different and has no "groove" to fit inside. But hey, feel free to try if you like the aesthetics better.

[rbv2]

Luckily, I ended up here again. :D

That unit looks fantastic. when i think of the panel wiring on my seqv3 wich was my first diy project about ten years ago :) ... I'm in

is it possible to build it with 16x midi-ouput and 8x cv-gate or do we also have to use the expansion port for additional midi outputs if you want more than eight?

thank you!!!

in reverent expectation

[latigid on]

Thanks for the kind feedback! 

I'm not sure where you got 16 MIDI outs from? The max here would be 8 (MIDI8 + I2C). There is a possibility to use a second I2C with a firmware hack to gain 4 more, but for space and latency versions this wasn't included in the standard build. @Antichambre is developing a 16x16 MIDI expander but again it doesn't fit in here. Another thing I can think of is hosting a GM5x5x5 or the newer chip, but it's also not in the plan nor complete at the moment.

Of course you have 8 outputs with 16 channels each, so plenty to work with using the 16 sequencer tracks. What was your plan with 16 MIDI outs?

CV/gate would go through the DB-25 line driver. On the other end AOUT/DOUT modules are required (under development, some solutions already available).

[Antichambre]

19 minutes ago, rbv2 said:

is it possible to build it with 16x midi-ouput and 8x cv-gate or do we also have to use the expansion port for additional midi outputs if you want more than eight?

Like me!? you don't want to connect/disconnect any machine anymore?

Just now, latigid on said:

. @Antichambre is developing a 16x16 MIDI expander but again it doesn't fit in here.

I think an external expansion will be possible for the Seq V4+( as it is ), there's already too possibilities to connect it, the Line Driver and the MCAN... I will build this seq too, and try this expansion once I've got it, so go for this sequencer! 16 MIDI outputs(and more) will come after.

Best Bruno

 

Edited September 22, 2018 by Antichambre

[rbv2]

thanks for your reply.

right.. twelve was said in the video.
it's certainly a bit exaggerated with the 16 ports but i'm using expert sleepers at the moment with 16 midi out and thought of a simple switch box that allows me to determine for each synth whether it gets its signals from the computer or the hardware sequencer without moving a cable.

guess eight outputs will be enough anyway :P