Behringer BSY600 Bass Synthesizer

A forum devoted to mod, tips and suggestions for upgrading and rehousing your VERY CHEAP commercial stompbox to near boutique excellence.
Post Reply
User avatar
JiM
Diode Debunker
Information
Posts: 967
Joined: 11 Mar 2008, 22:56
Completed builds: Completed builds :
Proco Rat
MXR MicroAmp in a volume pedal
TubeDriver (w/ NoS russian tube and big muff tone contol) + Phase 45 (w/ univibe cap ratio)
Dallas Rangemaster (w/ noisy OC75, negative ground)
SubCaster tube booster (w/ NoS russian tube, PtP)
Hot Harmonics
Music From Outer Space SubCommander in progress
Crackle Not OK
Simple bass blender in a 1590A
Bazz Fuss with a photo-darlington
Location: France
Has thanked: 70 times
Been thanked: 66 times
Contact:

Post by JiM »

Hi folks !
I've recently bought this little synth toy, and i couldn't resist : i had to have a look inside. And do some reverse-engineering.

Here is the result : gutshots and preliminary schematic.
There are still missing parts, and probably errors, but most of the working principles are visible :
  • The ATmega16 takes care of the "user interface", aka pots and leds, and also holds the program for the DSP (because he has the only Flash memory onboard)
  • The Freescale DSP56364 gets the audio from the AKM4552 codec, and processes it in its internal memory
  • Both processors are linked by a SPI link, which is mostly accessible at the programming header
  • There are some oddities, such as Reset, NMI and MODB tied together, or the ATmega driving the output transistors ... the feedback loop of the first opamp looks strange too.
  • The clock generator IC8 is unknown, the makings on the SOT23-6 package looks like CJ4R, or CU4R
Does anyone have improvement ideas, advice on dumping the SPI link, or error corrections ?
Attachments
bsy600_top.jpg
It's not really red, i've gimped it for see-through tracing
It's not really red, i've gimped it for see-through tracing
unverified
unverified
I only give negative feedback.

User avatar
snyder80
Information
Posts: 45
Joined: 19 Aug 2007, 21:05
Has thanked: 5 times
Been thanked: 10 times

Post by snyder80 »

Hi Jim!

Awesome tracing job! Saves me a lot of time.
I try to compare my RV600 Pedal to the Tonecore Schematic (inside TCDDK Hardware Guide.pdf), to get rid of the slight constant hissing noise. Briggs simple and easy LPF-Filter mustnt be the only way to handle the noise in these Pedal.

I changed the TL062 to TL072 but that didnt do realy a thing.

Either i suspect the hissing (coupled with some varying rumbling growl, depending on the Pedals settings) is there due of cheap/lack of supply filtering (1591 or ak4552), PCB-trace design.

Sadly some Parts in my pedal i couldnt identify ( 2x CCB Transistors and the IC8 - mine is coded "VU4"). Did you make any progress so far?

Greetings,

snyder80

EDIT: C21,22, 23 should be 100n according to L6 and Freescale documents.

User avatar
JiM
Diode Debunker
Information
Posts: 967
Joined: 11 Mar 2008, 22:56
Completed builds: Completed builds :
Proco Rat
MXR MicroAmp in a volume pedal
TubeDriver (w/ NoS russian tube and big muff tone contol) + Phase 45 (w/ univibe cap ratio)
Dallas Rangemaster (w/ noisy OC75, negative ground)
SubCaster tube booster (w/ NoS russian tube, PtP)
Hot Harmonics
Music From Outer Space SubCommander in progress
Crackle Not OK
Simple bass blender in a 1590A
Bazz Fuss with a photo-darlington
Location: France
Has thanked: 70 times
Been thanked: 66 times
Contact:

Post by JiM »

snyder80 wrote:I try to compare my RV600 Pedal to the Tonecore Schematic (inside TCDDK Hardware Guide.pdf), to get rid of the slight constant hissing noise.
Thanks for pointing to the TCDDK documents, it helps a lot to understand the difference between ToneCore and RSM.
snyder80 wrote:Either i suspect the hissing (coupled with some varying rumbling growl, depending on the Pedals settings) is there due of cheap/lack of supply filtering (1591 or ak4552), PCB-trace design.
You may have a look at the switch-mode DC-DC converter (MP1591) maybe it's producing some HF-induced noise.
The pre-emphasis thing is also something to look at, it's intended to reduce noise, unless the noise comes in before boosting treble.
I only give negative feedback.

User avatar
snyder80
Information
Posts: 45
Joined: 19 Aug 2007, 21:05
Has thanked: 5 times
Been thanked: 10 times

Post by snyder80 »

Hi!
Good to know i could help you with that doc, to me its "slightly" too much because of my personal EE/DSP knowledges...

First I was happy to find the L6 Schems, but sadly they differ alot to the SYB/RV600. Especially in the IN/Out Buffer section and the "PSU".

I first thought "CCB" could be some kind of dual Diodes or Mosfets, but google tells it should be 2SC3326B BJT´s.

The hissing is a lot reduced if the "output"-IC´s r30/32 in your schem are bridged (for less gain). But it´s not the wet signal that´s hissy, its the dry.
Also the Input jacks are slighty microphonic ("knocking test").

Maybe you could be right with the 1591 oscillating. Any hints to prevent that? I added a 220uF II with C2 (VCC out after L3), but it didnt do anything, too.
Another idea could be a 10R between AGND and DGND, but i think thats too hard to mod in the Behringer Design, since both GND´s seem to be identical.

Should i try to mod/adept the 1591 to the specs in the datasheet (except for +8,04 V DC)?

Greetings,
snyder80

User avatar
jonphilpott
Information
Posts: 3
Joined: 28 Feb 2012, 23:29

Post by jonphilpott »

I'm interesting in getting one of these pedals for some hackery (considering its similarity to the line6 tcddk), I'm thinking that it's likely the ISP disable fuses in the AVR16 have not been set given that there's an ISP header on the board.. or maybe that's a foolish assumption! Has anyone attempted to hook up to this header and see whats up?

User avatar
JiM
Diode Debunker
Information
Posts: 967
Joined: 11 Mar 2008, 22:56
Completed builds: Completed builds :
Proco Rat
MXR MicroAmp in a volume pedal
TubeDriver (w/ NoS russian tube and big muff tone contol) + Phase 45 (w/ univibe cap ratio)
Dallas Rangemaster (w/ noisy OC75, negative ground)
SubCaster tube booster (w/ NoS russian tube, PtP)
Hot Harmonics
Music From Outer Space SubCommander in progress
Crackle Not OK
Simple bass blender in a 1590A
Bazz Fuss with a photo-darlington
Location: France
Has thanked: 70 times
Been thanked: 66 times
Contact:

Post by JiM »

jonphilpott wrote:I'm interesting in getting one of these pedals for some hackery (considering its similarity to the line6 tcddk),
I would suggest getting one of the delays, reverb or pitch shifter, for the presence of external memory. Then it depends on the number of knobs, switches and either stereo input or mono + expression.
jonphilpott wrote:I'm thinking that it's likely the ISP disable fuses in the AVR16 have not been set given that there's an ISP header on the board.. or maybe that's a foolish assumption! Has anyone attempted to hook up to this header and see whats up?
I'm doing the opposite assumption (but i'd love being wrong on that question :blackeye). The headers are used in factory once, and Behringer is likely to set the lock bits, to avoid further chinese copies. In case of problem, they can erase the chip to clear the lock bits and program it again.
I did not try an ISP there, i'm still planning to dump the DSP code from the SPI line. I tried with an Arduino loaded with BusNinja with little success, i guess a real BusPirate would be much better for the job.
I only give negative feedback.

User avatar
jonphilpott
Information
Posts: 3
Joined: 28 Feb 2012, 23:29

Post by jonphilpott »

JiM wrote:
jonphilpott wrote:I'm thinking that it's likely the ISP disable fuses in the AVR16 have not been set given that there's an ISP header on the board.. or maybe that's a foolish assumption! Has anyone attempted to hook up to this header and see whats up?
I'm doing the opposite assumption (but i'd love being wrong on that question :blackeye). The headers are used in factory once, and Behringer is likely to set the lock bits, to avoid further chinese copies. In case of problem, they can erase the chip to clear the lock bits and program it again.
I did not try an ISP there, i'm still planning to dump the DSP code from the SPI line. I tried with an Arduino loaded with BusNinja with little success, i guess a real BusPirate would be much better for the job.
Right - my understanding would be that the device could still be reprogrammed by using the high-voltage mode to rewrite the firmware, or if worse comes to worst, unsolder the device and replace it with a new one :twisted:

User avatar
JiM
Diode Debunker
Information
Posts: 967
Joined: 11 Mar 2008, 22:56
Completed builds: Completed builds :
Proco Rat
MXR MicroAmp in a volume pedal
TubeDriver (w/ NoS russian tube and big muff tone contol) + Phase 45 (w/ univibe cap ratio)
Dallas Rangemaster (w/ noisy OC75, negative ground)
SubCaster tube booster (w/ NoS russian tube, PtP)
Hot Harmonics
Music From Outer Space SubCommander in progress
Crackle Not OK
Simple bass blender in a 1590A
Bazz Fuss with a photo-darlington
Location: France
Has thanked: 70 times
Been thanked: 66 times
Contact:

Post by JiM »

IIRC the ATMega can be reprogrammed with the ISP anytime.
BUT if the lock bits are set, you can't readback the program. You have to erase the whole Flash to reset the lock bits.
This is intended to forbid reverse-engineering of the program :slap: but one could still load his own firmware.
I only give negative feedback.

User avatar
jonphilpott
Information
Posts: 3
Joined: 28 Feb 2012, 23:29

Post by jonphilpott »

JiM wrote:IIRC the ATMega can be reprogrammed with the ISP anytime.
BUT if the lock bits are set, you can't readback the program. You have to erase the whole Flash to reset the lock bits.
This is intended to forbid reverse-engineering of the program :slap: but one could still load his own firmware.
There's a set of lock bits that also disables the SPI programming interface, see SPIEN bits in this sheet:

http://www.avrfreaks.net/modules/Freaks ... DN_020.pdf

User avatar
Siro78
Information
Posts: 19
Joined: 02 Sep 2009, 07:28
my favorite amplifier: Crate VC5212
Location: Parma Italy

Post by Siro78 »

Hi guys.
I have disassembled my DD600 for investigate the constant noise that it has so. The schematic is very similar to the showed on this post, except for the presence of an external SDRAM needed for long delay time.
A correction on schematic posted is that the 1591 switching regulator output 3.3V and not 5V!
I try to replace the 1591 with a 9V->3.3V linear regulator but the result is that the regulator overheats a lot, the current absorption from 80mA become 200mA and the noise not disappear.
I think that the problem is to search on inductive filter or the pcb layout.

Also my DD600 eat a lot of high frequency in all mode. It's possible because it not have the classical FET input signal buffer?

Bye

User avatar
Seven64
Information
Posts: 42
Joined: 22 Mar 2011, 06:14
Has thanked: 1 time
Been thanked: 7 times

Post by Seven64 »

How can you tell the pot values?

User avatar
modman
a d m i n
Information
Posts: 4890
Joined: 19 Jun 2007, 16:57
Has thanked: 4394 times
Been thanked: 2131 times

Post by modman »

[smilie=a_cookiemonster.gif]
Attachments
P0524 Service Manual_2011-04-27_Rev.0.pdf
Behringer Bass Synth schematic
(103.85 KiB) Downloaded 231 times
Please, support freestompboxes.org on Patreon for just 1 pcb per year! Or donate directly through PayPal

User avatar
benagain
Information
Posts: 1
Joined: 19 Dec 2014, 19:23

Post by benagain »

pot values for this synth?

User avatar
JiM
Diode Debunker
Information
Posts: 967
Joined: 11 Mar 2008, 22:56
Completed builds: Completed builds :
Proco Rat
MXR MicroAmp in a volume pedal
TubeDriver (w/ NoS russian tube and big muff tone contol) + Phase 45 (w/ univibe cap ratio)
Dallas Rangemaster (w/ noisy OC75, negative ground)
SubCaster tube booster (w/ NoS russian tube, PtP)
Hot Harmonics
Music From Outer Space SubCommander in progress
Crackle Not OK
Simple bass blender in a 1590A
Bazz Fuss with a photo-darlington
Location: France
Has thanked: 70 times
Been thanked: 66 times
Contact:

Post by JiM »

Besides being listed in the schematic above, the pot values do not matter in such a design. They are used as voltage dividers on the supply rail for an ADC to read. The only thing to look at is the power consumption of too low a value...

(And does it hurt your keyboard to type "please" when you ask a question ?)
I only give negative feedback.

User avatar
Andy2No
Breadboard Brother
Information
Posts: 82
Joined: 23 Sep 2015, 16:33
Has thanked: 154 times
Been thanked: 7 times

Post by Andy2No »

I'm impressed that you traced all that out, Jim.

Does the design have a shared analog and digital ground, or an inadequate ground plane?

I can't claim to understand grounding fully, for microcontroller designs, but I understand there is provision for separating the analog circuitry ground from the digital one, at least on the Atmega chips. The analog side needs a big enough, well designed ground plane, to stop it picking up noise from the CPU.

There's potential for the CPU to generate a constant stream of digital noise, if it isn't done properly.

E.g. http://www.eetimes.com/document.asp?doc_id=1272289

[EDIT: Correction; AVRs don't have a separate analog ground, as such, only a separate analog Vcc. This article explains how it should be done: http://avrprogrammers.com/howto/ground ]

I was wondering if these could be encouraged to work work better with a guitar - I don't own one, so I haven't tried it, but they seem interesting. Presumably there was no easy way, or they would probably have done it - and provided separate guitar and bass inputs like on some of their other pedals.

The capacitors around the input could be the culprits for cutting off higher frequencies. The series one is part of a DC blocking high pass filter of sorts, so the value of that probably doesn't matter. There is one connected to the non inverting input of the first op amp though. The trace doesn't show where to, but possibly that's acting as part of a low pass filter.

User avatar
Andy2No
Breadboard Brother
Information
Posts: 82
Joined: 23 Sep 2015, 16:33
Has thanked: 154 times
Been thanked: 7 times

Post by Andy2No »

modman wrote:[smilie=a_cookiemonster.gif]
I missed seeing that service manual attachment before. Thanks. I don't see a "Thank" button today, or I'd press it - and for JiM's initial post too. Maybe they're rationed?

So, they've tied the Vcc and AVcc together, on both the Atmega16L and the DSP chip. The DSP chip Vcc/AVcc is separated from the main 3.3V supply by an inductor, but that doesn't prevent its own digital side interfering with its analog side - adding digital noise.

I guess the Atmega16 might not need separate Vcc and AVcc particularly, because it's only reading the pots, though it could contribute to the general noise if the inductor separating the DSP from 3.3V isn't an adequate protection.

I would guess the biggest noise improvement would be got from giving the DSP chip a separate AVcc, protected from noise from its Vcc pins... There's actually three sets of Vcc, all tied together - DVcc (digital, presumably), AVcc and SVcc. I could go and find the datasheet, to see what SVcc does (sampling, maybe?), but I don't expect my soldering skills would be up to doing much about it.

I watched a review of the Behringer DW400 auto wah, which seems very impressive, apart from quite a lot of self interference going on - it sings to itself as the pots are adjusted, but keeps on singing to itself in at least some positions:


User avatar
JiM
Diode Debunker
Information
Posts: 967
Joined: 11 Mar 2008, 22:56
Completed builds: Completed builds :
Proco Rat
MXR MicroAmp in a volume pedal
TubeDriver (w/ NoS russian tube and big muff tone contol) + Phase 45 (w/ univibe cap ratio)
Dallas Rangemaster (w/ noisy OC75, negative ground)
SubCaster tube booster (w/ NoS russian tube, PtP)
Hot Harmonics
Music From Outer Space SubCommander in progress
Crackle Not OK
Simple bass blender in a 1590A
Bazz Fuss with a photo-darlington
Location: France
Has thanked: 70 times
Been thanked: 66 times
Contact:

Post by JiM »

Andy2No wrote:I'm impressed that you traced all that out, Jim.
Thanks. It's not that hard when you understand a bit what the designer had in mind.
Andy2No wrote:I was wondering if these could be encouraged to work work better with a guitar - I don't own one, so I haven't tried it, but they seem interesting. Presumably there was no easy way, or they would probably have done it - and provided separate guitar and bass inputs like on some of their other pedals.
It works reasonably well with a guitar as it is (as far as the note tracking goes). Of course, you need to play very neatly, and avoid chords at all cost when using all but the last two modes.
Andy2No wrote:So, they've tied the Vcc and AVcc together, on both the Atmega16L and the DSP chip. The DSP chip Vcc/AVcc is separated from the main 3.3V supply by an inductor, but that doesn't prevent its own digital side interfering with its analog side - adding digital noise.

I guess the Atmega16 might not need separate Vcc and AVcc particularly, because it's only reading the pots, though it could contribute to the general noise if the inductor separating the DSP from 3.3V isn't an adequate protection.

I would guess the biggest noise improvement would be got from giving the DSP chip a separate AVcc, protected from noise from its Vcc pins... There's actually three sets of Vcc, all tied together - DVcc (digital, presumably), AVcc and SVcc. I could go and find the datasheet, to see what SVcc does (sampling, maybe?), but I don't expect my soldering skills would be up to doing much about it.
There is nothing analog in the DSP, all conversions are done by the AKM chip (which has both Vcc and Vdd supplies).
The various power supplies for the DSP are all explained in the datasheet, they are for different parts of the digital chip (core, address and data busses, I/O).
I only give negative feedback.

User avatar
Andy2No
Breadboard Brother
Information
Posts: 82
Joined: 23 Sep 2015, 16:33
Has thanked: 154 times
Been thanked: 7 times

Post by Andy2No »

Thanks, JiM. I'd wondered about where the audio signals went to - looking at a large diagram on a small screen is a bit disorienting at times.

I have a few Behringer pedals and they all have some noise. Some sounds digital (SO400 octaver) but some just sounds like analog hiss (SO300 octaver and RV600 reverb).

From a brief scan of the AKM chip datasheet, it seems low sample rates can be a problem for noise. Maybe sampling at uneven intervals too?

There's nothing we could realistically do about noise produced by software problems, but some analog noise could possibly be reduced. Personally, I wouldn't attempt anything involving replacing SMD chips, but that might not always be needed.

User avatar
Andy2No
Breadboard Brother
Information
Posts: 82
Joined: 23 Sep 2015, 16:33
Has thanked: 154 times
Been thanked: 7 times

Post by Andy2No »

... That should have said UO300 octaver not SO300. No edit button, I see.

Post Reply