#662

A short story about how awesome the TXo is.

I was working on my piece for the upcoming LCRP 2017 2.0 this morning and I needed more modulation. So I fired up the TXo and created a script that random numbers to create a slow slewed CV output using the OSC operators. It was perfect! Almost like a faux slewed sample and hold from a noise source.

Such a simple use of it, but it completely saved my patch :slight_smile:

5 Likes
#663

Wondering if it would be possible to add a slew time to TO.OSC.WAVE?

1 Like
#664

@emenel great to hear it is working for you! Love the feedback!!

@GoneCaving - I’ll put it on my to-do list for the firmware. We’re pretty CPU limited at this point, but I might be able to squeeze it in.

@Leverkusen - I’ve been looking into that gnarly aliasing that you reported. I’ve not been able to track down the root cause yet. Still trying to figure out if there is something I can do. It is puzzling as I haven’t found the culprit. Was pretty confident that it was firmware - but I am less so now. Hmmmm.

Still in a work whirlwind and also crawling out from under the massive life debt that I accumulated while hand-building all of the TELEX. Might be a while before I’ll feel like I’ve exhausted my investigation options.

:slight_smile:

#665

Quit teasing me with my lack of toys. :wink:

1 Like
#666

Great to know that you are still after it - on the other hand it is not about live and death and I hope you get the aftermaths of dedicating so much of your live to designing these modules balanced out soon…
:ear_of_rice::sun_with_face::shamrock:

1 Like
#667

For those curious and following the process, I posted a micro postmortem of things I learned along the way over in the Eurorack Development Advice thread. Find it (along with some outstanding advice from others) here:

3 Likes
#669

Nope. I asked for the same thing.

1 Like
#670

by slewing do you mean have it change the waveform gradually when you assign a new value to it? you could do something similar by updating it from a really fast metro script.

2 Likes
#672

This might be a stupid question but this thread is really long and I couldn’t find the info anywhere : are these modules available for sale anywhere or is it strictly DIY at this point ?

There’s a Teletype in my future and having at least a Txi with it would be quite useful I think…

1 Like
#673

It’s never been stricly DIY, as @bpcmusic made them all.
I would contact him to know. Good luck !

#674

yeah it’d be cool if there was a built in slew similar to other parameters, but not sure if it can handle that… i did get good results using a fast metro script reading from ansible levels and using it to update wave shape.

#675

Not at the moment. It shouldn’t be super hard to fit in (though we are pretty much out of CPU headroom - see below). I’ve put it on the top of my to-do list. :slight_smile:

That’s the rub. With 4 oscillators, envelopes, pitch slew, and waveform morphing all going at the same time, we are well beyond the capacity of the little Teensy. I’ve done a lot of optimization to get us this far by doing a lot of ugly fixed-point implementations that perform orders of magnitude faster than their floating point counterparts. There might be some things that can be done around processing rate for different things - though my past experiments left me less than enthusiastic.

I’ve sold out of my run and my backup reserve units. I have a small number I’ve promised out to some folks using reserve components and PCBs - but I’ve been focusing on some life catch-up things and summer travel and haven’t gotten to them yet. Might be a few leftover if the folks I’ve promised them to change their mind.

I’ve also mostly costed out what doing a manufactured production run of the Telex would be. I’ve figured most of the details out with a manufacturer/assembler, just need to see if there is enough demand that it would make sense pulling the trigger.

7 Likes
#677

Both TX modules have been my second and third SMD project after the Music Thing Simple EQ (which I can recommend for first trying it out - it’s named for a reason and a great module too for a delay feedback path!) Both are working fine and I am very happy that I dared to do it. So don’t be shy…just do the impossible…
:muscle::sunglasses::+1:

https://www.youtube.com/watch?v=7YkbgvRMpW0

2 Likes
#678

Haha, I used to do quite a lot of circuit bending ages ago and have some soldering background (though I’ve never been that good at it) but I have a baby at home and thus not a lot of free time now. DIY modules are definitely cool but it’s not something I can afford time-wise these days. :pensive:

Thanks for the answers, I contacted bpcmusic just in case a module becomes available.

1 Like
#679

need a little bit help with troubleshooting my TelexO.
The main functionality is ok, but what really confusing me are the behavior of the LEDs.
If I use 10k cureent limiting resistors, the LEDs are too dark (means not really visible). If I use 100Ohm Resistors the brightness is ok, but if more then 3 LEDs on at the same time, the fuse set off.

Any ideas?

#680

100 ohm is much too low!

Assuming a 5V LED power supply, that’s about 33mA per LED (5V - 1.7Vled / 100 = .033A)–probably the max current those LEDs can handle.

10k is too high: .33mA, which is practically nothing even for very high brightness parts.

I’d try something in the 510 ohm to 1k range (6.5mA-3.3mA).

#681

Hi jnoble

I was confused too by the 10k resistors, but thats written in the BOM.

Regarding the 100 Ohm, I think it’s not to less, because the Teensy Vout is 3.3V.
I use yellow LEDs with If 18mA and Vf 2.0 Volt.
=> (3,3V - 2,0) / 0,018 = 72 Ohm

For sure, with low current LEDs, the Resistor is way too low. But in any case 10k doesn’t make sense

Maybe @bpcmusic can explain why he choose this value.

#682

The LEDs that I used had very high Luminous Intensity (MCD) (Yellow at 2-3k; White at 14-16k) for the build. This allowed me to keep the overall power draw for the module down when all LEDs were lit.

The parts that I used are here:

I started with 10k resistors and, with these LEDs, the brightness matched the TT quite closely when at full! (I did, however, ultimately implement a logarithmic lookup table in firmware so that the brightness scaled better over the range of values - but that is another story.)

To be honest, I was just getting my head around all of these details at this point in the design process. @Galapagoose was amazingly kind and helpful … he nearly carried me through my darkest moments of confusion and ignorance. I went back through our PM thread from the time and here were his recommendations from back then (hope you don’t mind my sharing):

Thing is, the mA current rating isn’t a ‘recommended value’ but rather just a test case at which they measure the mcd rating. 10mcd LED @20mA is the same brightness as 100mcd @2mA, but you draw one tenth the power. This is important to help keep noise low in the power system.

The LEDs you linked are 2000mcd which is super bright, so I think you’d be safe starting around 10k for the resistor. If you’re looking for a reference point, I’d aim to have the LEDs be at least close to the brightness of TT just for continuity sake.

Also - think about the situation where all 8 LEDs are on full brightness at the same time. You don’t want 160mA (@3v3) being drawn just to indicate some levels.

These values are very easy to change after the circuit is built. The tradeoff is very simply between brightness and power draw.

Hope this helps!!!

1 Like
#683

Thanks for the clarification

#684

What’s your minimum run and retail price at that run size, if you don’t mind my asking?

1 Like