Dreadbox Erebus DIY kit

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

Moderators: Kent, Joe., luketeaford, lisa

Post Reply
User avatar
bartleby
Ultra Wiggler
Posts: 812
Joined: Wed Oct 20, 2010 2:20 pm
Location: berlin

Post by bartleby » Sat May 19, 2018 8:36 am

meanwhile, i've noticed another thing i didn't expect:

the echo time cv input on the lil erebus seems to respond only to voltages around +/-1v.

by contrast, the echo cv input of the desktop erebusses (erebi?) responds to +/-2.5v (according to the manuals of erebus and erebus 2).

it's still a great little lo-fi delay. but controlling it with your pitch cv signal has its limits. better find a separate cv source for the echo time...

geklimper
Learning to Wiggle
Posts: 19
Joined: Thu Aug 11, 2016 4:41 am
Location: Wuerzburg

Post by geklimper » Sat May 19, 2018 9:44 am

JAO wrote:Anyone know what are the estimated current requirements? The MG entry is just using the specs from the optional power supply. Thanks.
I asked what is left from the 500 mA/150 mA.

The answer I got from Dreadbox:
"You have about 420mA at +12 and 70mA at -12. Notice that if you draw more than about 200mA from +12 , a heatsink might be needed."

So I assume it's 80 mA on +/-12V.

User avatar
JAO
Common Wiggler
Posts: 231
Joined: Fri Jan 06, 2017 10:03 am
Location: Ireland

Post by JAO » Sat May 19, 2018 10:11 am

geklimper wrote:The answer I got from Dreadbox:
"You have about 420mA at +12 and 70mA at -12. Notice that if you draw more than about 200mA from +12 , a heatsink might be needed."

So I assume it's 80 mA on +/-12V.
Thanks for that. :tu:

geklimper
Learning to Wiggle
Posts: 19
Joined: Thu Aug 11, 2016 4:41 am
Location: Wuerzburg

Post by geklimper » Sat May 19, 2018 10:11 am

The user manual is online now:
https://www.dreadbox-fx.com/lil-erebus/

I did the calibration with a guitar tuner, and a repeating two-note octave sequence (C3-C4). Turning scale 1 until I had an octave (that was not C). Repeat with scale 2. After that, set OSC2 to C with tune 2 and OSC1 to C with master tune. Both TUNE knob and OSC2 slider are in middle position and are not touched.

User avatar
bartleby
Ultra Wiggler
Posts: 812
Joined: Wed Oct 20, 2010 2:20 pm
Location: berlin

Post by bartleby » Sat May 19, 2018 10:42 am

geklimper wrote:The user manual is online now:
https://www.dreadbox-fx.com/lil-erebus/
cool, thanks for posting! :tu:

hm, that's odd: the manual says about the time cv input:

"DELAY TIME CONTROL (0V - 5V)"

but on my unit, the effective cv range for time cv is +/-1v. there is nothing happening beyond 1v...

anybody else getting any delay time changes between 1v...5v?

User avatar
bartleby
Ultra Wiggler
Posts: 812
Joined: Wed Oct 20, 2010 2:20 pm
Location: berlin

Post by bartleby » Sat May 19, 2018 10:45 am

well, at least the manual confirms my observation about warm-up times and vco stability:

"TUNING AND STABILITY
Lil’ Erebus needs at least 20 min of warm up until the oscillators are stable.
If your module is out of tune, leave it to warm up for 35 min and
follow the calibration procedure described on the constraction manual."

vertflyer
Common Wiggler
Posts: 103
Joined: Mon Nov 21, 2011 3:46 pm

Post by vertflyer » Sat May 19, 2018 11:25 am

bartleby wrote:i've finished my lil erebus, and everything seems to be working fine. i love that creamy filter and the very playlable delay section.

but the envelope seems to be behaving a little strange in some ways. i'm not sure if these are just some quirks of this envelope design, or if there is something wrong with my particualr build:

1. even with decay set to minimum, there is still some noticeable decay time.

2. with high attack and only slightly lower decay settings, a very short gate signal will often result in the envelope not decaying at all, but staying high until re-triggered with a new gate signal.

3. maximum decay time is much longer (ca 6s) than maximum attack time (ca 2s).

4. the envelope won't retrigger while the decay phase is still lasting.

i suspect that the last item is probably normal. but i'm not sure about the other ones. could anybody with a lil erebus check if their envelopes are showing similar behavior?

(these are best tested with the cutoff set to minimum, env lvl to maximum, and the env output patched into the vca input)
Maybe some else should check too. But I’m not seeing what you are seeing. My max attack and max decay are the same. Both about 2 seconds. And I definitely can retrigger at any time. I think you should double check that you don’t have that 470p cap on. Check to see if you left out the correct one?

vertflyer
Common Wiggler
Posts: 103
Joined: Mon Nov 21, 2011 3:46 pm

Post by vertflyer » Sat May 19, 2018 12:36 pm

Throwing out some info that cost me about an hour. Hopefully it helps someone down the line.

If you need to replace a TL074CN chip, as much as possible, keep an original from dreadbox in IC1 and IC2.

I ordered some Chips online that were called TL074CN, but when used in the osc section, you cant get the osc to scale more than one octave. And if you scale it across 4 octaves, the middle 2 aren’t scaled.

When I use the replacements on the other sections of the synth and put an original onto IC1 and 2, the synth scales perfectly.

User avatar
bartleby
Ultra Wiggler
Posts: 812
Joined: Wed Oct 20, 2010 2:20 pm
Location: berlin

Post by bartleby » Sat May 19, 2018 12:57 pm

vertflyer wrote:I’m not seeing what you are seeing. My max attack and max decay are the same. Both about 2 seconds. And I definitely can retrigger at any time.
ok, thanks for cheking! :hail:
i'll take a good look tomorrow to see if i can find anything amiss with my build.
(it might actually have to do with a slight mod i did to the env cv output, although i don't see how that could affect env performance)

User avatar
bartleby
Ultra Wiggler
Posts: 812
Joined: Wed Oct 20, 2010 2:20 pm
Location: berlin

Post by bartleby » Sat May 19, 2018 2:23 pm

vertflyer wrote:I think you should double check that you don’t have that 470p cap on. Check to see if you left out the correct one?
oh. em. gee.
this is beyond embarrassing.
i was positive that i had left that 470n unpopulated.
turns out i hadn't. :doh:
i removed it, and now the envelope is behaving just like it should.

so you were absolutely right, many thanks for keeping pointing this out! :hail:

now that the env is working like it should, i can recommend that mod i did to the env output: as it is, that output (like the two lfo outputs) is the signal attenuated by the 'lvl' fader. but this seemed impractical to me, as i wanted to use the envelope to completely open the vca and still be able to control the env->vcf amount independently from the vca.

so here is what i did to get the unattenuated env scv ignal to the env output: just cut one trace (red) and connect the pre-fader signal to the output instead (blue).

Image
Last edited by bartleby on Sat May 19, 2018 2:38 pm, edited 2 times in total.

User avatar
bartleby
Ultra Wiggler
Posts: 812
Joined: Wed Oct 20, 2010 2:20 pm
Location: berlin

Post by bartleby » Sat May 19, 2018 2:30 pm

now that i have been able to fix the envelope, there seems only one issue open with my build:
bartleby wrote:the manual says about the time cv input:

"DELAY TIME CONTROL (0V - 5V)"

but on my unit, the effective cv range for time cv is +/-1v. there is nothing happening beyond 1v...

anybody else getting any delay time changes between 1v...5v?

do you guys really notice any delay time changes with 'time' cv voltages beyond a little over +1v? because i don't on my unit. maybe i messed up something else?

[edit:] well, depending on the 'echo time' knob setting, i'm actually getting some delay time changes with control voltages between -1v...+2v. but nothing beyond +2v. [/edit]

Mattbartlett303
Learning to Wiggle
Posts: 11
Joined: Tue Jul 30, 2013 5:14 pm
Location: London

Post by Mattbartlett303 » Sun May 20, 2018 6:27 am

Hi I'm after a bit of advice. My midi still isn't working properly, my keyboard triggers the same note on every key.
I don't have much experience with atmega or any digital circuits, but it seems that the crystal isn't oscillating (no reading with frequency counter at pins 9 or 10) I've replaced the crystal, but this had not helped.
I have 5v to the atmega and 0v at pin 8, so no obvious shorts. Does this sound like a problem with the atmega itself? :despair:
Any help greatly appreciated.

vertflyer
Common Wiggler
Posts: 103
Joined: Mon Nov 21, 2011 3:46 pm

Post by vertflyer » Sun May 20, 2018 6:35 am

bartleby wrote:now that i have been able to fix the envelope, there seems only one issue open with my build:
bartleby wrote:the manual says about the time cv input:

"DELAY TIME CONTROL (0V - 5V)"

but on my unit, the effective cv range for time cv is +/-1v. there is nothing happening beyond 1v...

anybody else getting any delay time changes between 1v...5v?

do you guys really notice any delay time changes with 'time' cv voltages beyond a little over +1v? because i don't on my unit. maybe i messed up something else?

[edit:] well, depending on the 'echo time' knob setting, i'm actually getting some delay time changes with control voltages between -1v...+2v. but nothing beyond +2v. [/edit]
Not sure if this helps but when I put an external envelope signal in to the vca it peaks out at a really low voltage meaning full volume is reached. I had to really drop the envelope voltage down real low before I even got a decent shaped envelope response.

vertflyer
Common Wiggler
Posts: 103
Joined: Mon Nov 21, 2011 3:46 pm

Post by vertflyer » Sun May 20, 2018 1:07 pm


User avatar
autodafe
Wiggling with Experience
Posts: 372
Joined: Mon Oct 02, 2017 3:14 am
Location: Italy
Contact:

Post by autodafe » Sun May 20, 2018 5:46 pm

vertflyer wrote:A quick track using 4 lil erebus

https://soundcloud.com/vertflyer/1-audi ... -20-135120
excellent!!!

you bought 4 ??? :woah: :woah: :woah:

thement
Learning to Wiggle
Posts: 6
Joined: Thu May 10, 2018 6:47 am

Post by thement » Tue May 22, 2018 1:13 pm

Mattbartlett303 wrote:Hi I'm after a bit of advice. My midi still isn't working properly, my keyboard triggers the same note on every key.
I don't have much experience with atmega or any digital circuits, but it seems that the crystal isn't oscillating (no reading with frequency counter at pins 9 or 10) I've replaced the crystal, but this had not helped.
I have 5v to the atmega and 0v at pin 8, so no obvious shorts. Does this sound like a problem with the atmega itself? :despair:
Any help greatly appreciated.
I'm not sure you can read the frequency on the crystal like that. If you can trigger note via MIDI (no matter what pitch), then the atmega is definitely working (and also the crystal - MIDI baudrate is 31250 which gets clocked from the crystal).

The "MCP4922" chip is responsible for generating the voltage for oscillators. Measure the pins 10 and 14 of the MCP4922, they should read voltages 0...5V depending on the note you press. They should also both have the same value unless you play two notes at the same time.

vertflyer
Common Wiggler
Posts: 103
Joined: Mon Nov 21, 2011 3:46 pm

Post by vertflyer » Thu May 24, 2018 6:14 am

Is it just me or is the vca control extremely non linear? It’s hard to coax a smooth decay or release from it, even using external envelopes. At long decay or release settings, the vca stays at full volume then just falls to zero really quick instead having a slow drop in volume.

All my units behave this way. Even with my intellijel envelope.

Eta: playing with it more it seems like they programmed the vca to just drop at a constant rate after a certain voltage threshold is reached. It doesn’t matter how slow your release or decay is, it’s full volume until that voltage drops then it’s a constant time decay of about 1.5 seconds


Eta2: played with it more and it does have teatime control, it’s just that the voltage range from full volume to zero is on a very very narrow voltage range. It’s very hard to get an envelope to control it well.

User avatar
Northpaw27
Learning to Wiggle
Posts: 30
Joined: Sun Oct 02, 2016 9:44 pm
Location: Glasgow
Contact:

Timelapse build

Post by Northpaw27 » Fri May 25, 2018 8:29 am

Hey wigglers!

I put together a little timelapse of my build with some noodles over the top


Its a big build this one! I'm thinking about modding it with a switch that normalises the gate input to the 5V rail so can use it for drones.

Mattbartlett303
Learning to Wiggle
Posts: 11
Joined: Tue Jul 30, 2013 5:14 pm
Location: London

Post by Mattbartlett303 » Sat May 26, 2018 8:35 am

thement wrote:
Mattbartlett303 wrote:Hi I'm after a bit of advice. My midi still isn't working properly, my keyboard triggers the same note on every key.
I don't have much experience with atmega or any digital circuits, but it seems that the crystal isn't oscillating (no reading with frequency counter at pins 9 or 10) I've replaced the crystal, but this had not helped.
I have 5v to the atmega and 0v at pin 8, so no obvious shorts. Does this sound like a problem with the atmega itself? :despair:
Any help greatly appreciated.
I'm not sure you can read the frequency on the crystal like that. If you can trigger note via MIDI (no matter what pitch), then the atmega is definitely working (and also the crystal - MIDI baudrate is 31250 which gets clocked from the crystal).

The "MCP4922" chip is responsible for generating the voltage for oscillators. Measure the pins 10 and 14 of the MCP4922, they should read voltages 0...5V depending on the note you press. They should also both have the same value unless you play two notes at the same time.
Thanks Thement. I'm getting 0v on pin 10 and 14 of the 4922. I've already tried replacing it, but this hasn't helped. I'm completely stumped by this - Are you able to check the voltages on the other pins of the 4922 for me so that I can compare. Cheers

csaban
Learning to Wiggle
Posts: 38
Joined: Sun Aug 07, 2016 4:55 pm
Location: Glasgow

Post by csaban » Sat May 26, 2018 1:02 pm

vertflyer wrote:Is it just me or is the vca control extremely non linear? It’s hard to coax a smooth decay or release from it, even using external envelopes. At long decay or release settings, the vca stays at full volume then just falls to zero really quick instead having a slow drop in volume.

All my units behave this way. Even with my intellijel envelope.

Eta: playing with it more it seems like they programmed the vca to just drop at a constant rate after a certain voltage threshold is reached. It doesn’t matter how slow your release or decay is, it’s full volume until that voltage drops then it’s a constant time decay of about 1.5 seconds


Eta2: played with it more and it does have teatime control, it’s just that the voltage range from full volume to zero is on a very very narrow voltage range. It’s very hard to get an envelope to control it well.
I think this is relevant form the documentation:
When I patch something to the VCA , i get no sound
-The VCA needs at least 3,5V so as the synth starts producing sound

mrcase
Learning to Wiggle
Posts: 6
Joined: Fri Jul 24, 2015 5:27 am

Post by mrcase » Sat May 26, 2018 4:17 pm

I finished my build, all working except VCO2 isn´t doing anything, same issues two other members have. Did anybody solve this?

vertflyer
Common Wiggler
Posts: 103
Joined: Mon Nov 21, 2011 3:46 pm

Post by vertflyer » Sat May 26, 2018 4:28 pm

mrcase wrote:I finished my build, all working except VCO2 isn´t doing anything, same issues two other members have. Did anybody solve this?
Swap IC 1 and 2 and see if you still have vco2 missing or vco1 missing.

mrcase
Learning to Wiggle
Posts: 6
Joined: Fri Jul 24, 2015 5:27 am

Post by mrcase » Sat May 26, 2018 5:33 pm

Thanks, will do. Time to put the sockets in as well

BradshawsWindmill
1-Post Wiggler
Posts: 1
Joined: Mon Nov 13, 2017 11:37 pm
Location: NZ

Post by BradshawsWindmill » Sun May 27, 2018 3:40 am

I've just noticed something that might be helpful to people having midi problems;

On my build at least, switches 1 & 3 on the dip switch are transposed,

so: midi channel 1 = switch 3 on,
channel 2 = switch 2 on
channel 3 = switch 2+3 on
channel 4 = switch 1 on
channel 5 = switch 1+3 on
channel 6 = 1 + 2 on
channel 7 = all on

vertflyer
Common Wiggler
Posts: 103
Joined: Mon Nov 21, 2011 3:46 pm

Post by vertflyer » Sun May 27, 2018 3:52 am

BradshawsWindmill wrote:I've just noticed something that might be helpful to people having midi problems;

On my build at least, switches 1 & 3 on the dip switch are transposed,

so: midi channel 1 = switch 3 on,
channel 2 = switch 2 on
channel 3 = switch 2+3 on
channel 4 = switch 1 on
channel 5 = switch 1+3 on
channel 6 = 1 + 2 on
channel 7 = all on
I’ll have to try this! I wonder why this would happen.

Post Reply

Return to “Music Tech DIY”