DIYstompboxes.com

DIY Stompboxes => Digital & DSP => Topic started by: aon on April 27, 2021, 12:21:08 PM

Title: Cuvave DIG Delay
Post by: aon on April 27, 2021, 12:21:08 PM
Hello,

Occasional lurker, first time poster here. I was intrigued by the low price of this pedal, and was thinking that maybe it could serve as a DSP learning platform, and if not, at least it'd be interesting to see what sort of tech is inside. Also my 2-year-old son needs an echo pedal to mess around with using a microphone as I sold the EHX Oceans 11 he was previously using for that.

So here's what's inside one of these, in case anyone is interested:

(https://i.postimg.cc/yk7Tz6SX/IMG-20210427-170504.jpg) (https://postimg.cc/yk7Tz6SX)

(https://i.postimg.cc/BXNBZWTW/IMG-20210427-170527.jpg) (https://postimg.cc/BXNBZWTW)

The bottom board hosts the bulkier components and has two TL072C dual op-amps that probably serve as buffers.

The top board has the following IC's:

There is a datasheet available for TG401 that can be found by searching for TG401_Datasheet_v09.pdf on Baidu. Haven't figured out how to download it yet. Also looks like there is a development board available that does have an I2S audio codec on it.

This leads me to believe that one of the mystery chips U8, U9, U19 must be an audio codec that is interfacing with the main MCU with I2S.
One more interesting thing is that there's an USB port and when you plug it in it shows up as:

SG-Audio
USB\VID_0418&PID_1095&REV_0005&MI_00

SG-Midi
USB\VID_0418&PID_1095&REV_0005&MI_03

Nothing can be done over the audio device as far as I can tell. It won't record or play back through it.

Once I'm bored enough I might trace the board a bit to figure out which chip might be the audio codec, and the audio path in general interests me. But it probably isn't worth the effort to figure out how to program the chinese MCU.

Still, the overall construction and implementation seems alright and the delay algorithms sound pretty good. Too bad there's no mix control so the effect is always at a constant level which renders this a bit useless as a guitar effect for me... But like I said, for my son's "singing echo" it's good enough. And unless I'm mistaken about the chips, the overall topology of having a general-purpose microcontroller to do the DSP stuff and an I2S DAC is interesting.
Title: Re: Cuvave DIG Delay
Post by: aon on April 27, 2021, 12:21:54 PM
Oops, accidentally posted twice. Oh well, might as well make use out of it.

Looking around a bit more, I think U8 might be a Cirrus Logic CS5340 (ADC) and U9 might be a Cirrus Logic CS4344 (DAC), or clones of them. Meaning U19 might be flash?

EDIT: U19 could also be a Rainbow RB55P045A ADC. Maybe it's used to read the potentiometers. There's a single ADC on the main MCU so if that' used for the program select, a separate ADC would be needed for the time and feedback pots.
Title: Re: Cuvave DIG Delay
Post by: Digital Larry on April 30, 2021, 10:04:47 AM
Looks like a cool little platform, if only you could figure out how to push firmware into it and how to write said firmware!

DL
Title: Re: Cuvave DIG Delay
Post by: ElectricDruid on May 05, 2021, 07:08:47 AM
Looks like a 4-pin programming header up by the pots. Two power lines, plus PDAT and PCLK seems pretty convincing. God knows what it's expecting to see there though.

What does the 10-pin box header socket do? Is that where the footswitch plugs in?
Title: Re: Cuvave DIG Delay
Post by: aon on May 06, 2021, 11:53:32 AM
Quote from: Digital Larry on April 30, 2021, 10:04:47 AM
Looks like a cool little platform, if only you could figure out how to push firmware into it and how to write said firmware!

Yeah, to be fair I kind of gave up on using this pedal for anything. The tools are only available on the Chinese Internet (I suspect there might be some intellectual property naughtiness involved and that might be why there's almost no mention of the chips anywhere in English... wouldn't be the first time for that), but the idea of using eg. STM32F4 and an audio codec for something intrigues me. That way the code wouldn't be so tied to that particular platform as with the dedicated DSP's.

Quote from: ElectricDruid on May 05, 2021, 07:08:47 AM
Looks like a 4-pin programming header up by the pots. Two power lines, plus PDAT and PCLK seems pretty convincing. God knows what it's expecting to see there though.

What does the 10-pin box header socket do? Is that where the footswitch plugs in?

Yep, the 10 pin header connects the boards together. I think the footswitch is true bypass but audio, power and maybe something else travel through that header.