Author Topic: EEVblog #738 - Yamaha DME32 Digital Mixer Teardown  (Read 13285 times)

0 Members and 1 Guest are viewing this topic.

Offline EEVblog

  • Administrator
  • *****
  • Posts: 31340
  • Country: au
    • EEVblog
 

Offline PeterFW

  • Frequent Contributor
  • **
  • Posts: 570
  • Country: de
    • Stuff that goes boom
Re: EEVblog #738 - Yamaha DME32 Digital Mixer Teardown
« Reply #1 on: April 28, 2015, 02:53:41 pm »
Dave opens a Yamaha DME32 Digital Mixing Engine with multi-channel 24bit audio ADC and DAC and custom DSP processing.

Those IO board were bautiful!

Might be wrong but i think it is designed for installations and studios, not road use.
That is one of the reasons they do not have balanced inputs and a few other things screamed installation :)

At least i never have seen something like this, modern mixers are still those huge ships with knobs, buttons ans switches as far as the eye can see.
But they do have digital(wireless) interfaces so you can control them from a tablet or notebook, during the show you need those buttons, you do not have time to fiddle around with a notebook. Every function has to be mapped to a physical button.

Have done that a few times :)
 

Offline Tothwolf

  • Regular Contributor
  • *
  • Posts: 95
  • Country: 00
Re: EEVblog #738 - Yamaha DME32 Digital Mixer Teardown
« Reply #2 on: April 28, 2015, 03:36:30 pm »
Those purple Sanyo* capacitors are early solid polymer types. I suspect they stood them up off the board because the capacitors are 8mm in diameter and have a 3.5mm lead spacing and the pads were placed for 10mm parts with a 5mm lead spacing. [* Their capacitor division is now owned by Panasonic.]
 

Offline rsjsouza

  • Super Contributor
  • ***
  • Posts: 3875
  • Country: us
  • Eternally curious
    • Vbe - vídeo blog eletrônico
Re: EEVblog #738 - Yamaha DME32 Digital Mixer Teardown
« Reply #3 on: April 28, 2015, 04:11:24 pm »
Interesting teardown; I would have expected the pads surrounding the device are for a TQFP ZIFF socket and not an ICE, but that obviously depends on the age of both the design and the device. 

The single GND connection is to avoid ground loops - a killer thing in low-noise environments.

I would be very impressed if the 60MHz of the DSP clock oscillator is distributed to the other ones without buffers. Did you see buffers on the other side of the board?

On amazing thing: the high quality power supply is "only" 40+ watts - I can't help but wonder how the durability and reliability of this compares to the modern ultra-packed 500+ watts power supplies of the PC world - oh well, it is an older design anyways... But still. ;)
« Last Edit: April 28, 2015, 06:46:34 pm by rsjsouza »
Vbe - vídeo blog eletrônico http://videos.vbeletronico.com

Oh, the "whys" of the datasheets... The information is there not to be an axiomatic truth, but instead each speck of data must be slowly inhaled while carefully performing a deep search inside oneself to find the true metaphysical sense...
 

Offline dentaku

  • Frequent Contributor
  • **
  • Posts: 849
  • Country: ca
Re: EEVblog #738 - Yamaha DME32 Digital Mixer Teardown
« Reply #4 on: April 28, 2015, 04:13:50 pm »
I like looking at Yamaha boards. They're easy to follow and well labeled.

Someday EEVblog will have a synthesizer teardown with the kind of stuff I actually know something about so I can scream at my monitor telling Dave that he's wrong :)

Does anyone in the Sydney area have an older model Dave Smith Instruments, Moog or Roland device they want Dave to show off?
An Elka synth would be fascinating. Markus Fuller hasn't opened up of one those yet.
 

Offline Fungus

  • Super Contributor
  • ***
  • Posts: 10795
  • Country: 00
Re: EEVblog #738 - Yamaha DME32 Digital Mixer Teardown
« Reply #5 on: April 28, 2015, 05:50:19 pm »
I dread to think what the development cost of something like that is...

Anybody know what the retail price was?
 

Offline mmirabent

  • Newbie
  • Posts: 2
Re: EEVblog #738 - Yamaha DME32 Digital Mixer Teardown
« Reply #6 on: April 28, 2015, 07:44:04 pm »
Very cool teardown. I wanted to share some insights coming from a live sound background. The TRS connectors on the back are very common on live sound equipment, especially when dealing with line level sources. XLR connectors can be used for line level sources, but are usually reserved for the lower level signals coming from microphones or instruments. This DME32 would be dealing with line level sources only, and as such uses 1/4" TRS connectors to reduce the size of the panels. In addition to this, you would have some pre-amplifiers to bring low level signals like those from a microphone up to line level, as well as some line level sources like a CD player or radio.

Also, this unit isn't likely to see use in touring, it's intended for permanent installations. Think restaurants or conference rooms where the layout is fixed. The scenes would be preset mixes used for different types of events. In a restaurant for example, maybe you have a private room and want to separate the music going to the rest of the restaurant from the slideshow playing in the private room. There would be a scene for that, so the maitre d' can make that change without worrying about messing up the sound for the rest of the restaurant, or having to know about signal routing and mixing.

The inputs for something like would likely be a building wide intercom system, probably some type of muzac, and maybe some panels located around the room for direct connection for a visiting sound engineer to plug his or her mixer into the installed sound system.
 

Offline Jr460

  • Regular Contributor
  • *
  • Posts: 117
Re: EEVblog #738 - Yamaha DME32 Digital Mixer Teardown
« Reply #7 on: April 28, 2015, 08:16:32 pm »
Echoing what others have said that a DME is not used as a live mixer.

I do a bit of live sound, the most I need for the size of things I do is a Driverack.  I have some friends that have done hig end tours, and others that work at large venues.  My venue/theater friends are the ones using DMEs.  They have a board and it provides one feed to the DME, the others feeds would be a lobby paging mic, Muszack feeds, etc.  the outputs would be the flown speakers, the subs installed under the stage, under balcony feeds, e tra subs when doing heavy acts, the lobby speakers.

Also some small control panel that allows simple changes/scenes.  One the speaker feeds inside seating area you would configure parametric EQ to correct for speaker issues, delays so the sound from the under balcony lines up time wise with the mains.  Sum the L+R for the lobby feed., etc.

Software on a PC allows you to layout all the blocks of funtiona you want, and then it generate a file that downloads into the DME.

Anymore, you put a Dante card in the DME and then all the audio stays digital from the mixer, no extra conversions and the added delay.

In case anyone wonders about delay and latency, it can be a real killer to players.  It gets worse when using in ear monitors.  Having bounded latency is a must.  A few milliseconds changes the groove of a song.
 

Offline smjcuk

  • Frequent Contributor
  • **
  • Posts: 464
  • Country: gb
Re: EEVblog #738 - Yamaha DME32 Digital Mixer Teardown
« Reply #8 on: April 28, 2015, 10:17:43 pm »
I love seeing Japanese equipment in teardowns. The construction is often a work of art.  :-+

Apart from Korg. Whichever cretin designed the Triton series of synthesizers needs to poke themselves in the eyes. Getting inside one is easy. Getting to the front panel controls (the ones that wear quickly) is a two hour long slog. Also the PSUs are incredibly weak and tend to smoke transistors. Plus the user interface sucks. Sold now but nothing but £2400 of regret.
 

Offline R_Gtx

  • Regular Contributor
  • *
  • Posts: 52
  • Country: gb
Re: EEVblog #738 - Yamaha DME32 Digital Mixer Teardown
« Reply #9 on: April 28, 2015, 10:57:43 pm »
Motorised faders are not strictly the domain of digital audio mixers. In 1977 Neve electronics, a UK professional audio manufacturer introduced NECAM (Neve Computer Assisted Mixdown). If memory serves me correct, the installation I first encountered in the early 1980s comprised a DEC PDP-11 minicomputer controlling a 48-channel Neve mixing console with motorised faders and twin Studer A80 16-track analogue tape recorders, plus timecode reader.
 

Offline loneoceans

  • Contributor
  • Posts: 36
  • Country: us
    • Loneoceans Laboratories
Re: EEVblog #738 - Yamaha DME32 Digital Mixer Teardown
« Reply #10 on: April 28, 2015, 11:05:48 pm »
Hi Dave, thanks for another fascinating video!

Since you mentioned the interesting grounding system used, I thought people might be interested if you did a 'fundamentals Friday' video talking about grounding such as why designers sometimes separate grounds from digital and analog parts of the circuit, what exactly is star grounding, and as a designer, what should one look out for when designing and laying out boards specifically on grounding issues.

Also on a random note, I thought it might be fun to discuss RF soldering irons sometime which heat up super fast!
 

Offline dentaku

  • Frequent Contributor
  • **
  • Posts: 849
  • Country: ca
Re: EEVblog #738 - Yamaha DME32 Digital Mixer Teardown
« Reply #11 on: April 29, 2015, 12:48:36 am »
Hi Dave, thanks for another fascinating video!

Since you mentioned the interesting grounding system used, I thought people might be interested if you did a 'fundamentals Friday' video talking about grounding such as why designers sometimes separate grounds from digital and analog parts of the circuit, what exactly is star grounding, and as a designer, what should one look out for when designing and laying out boards specifically on grounding issues.

Also on a random note, I thought it might be fun to discuss RF soldering irons sometime which heat up super fast!

You might like the Amphour interview Dave and Chris did a while ago with Henry Ott.
http://www.theamphour.com/165-an-interview-with-henry-ott-forced-fcc-filtering/
 

Offline NiHaoMike

  • Super Contributor
  • ***
  • Posts: 6212
  • Country: us
  • "Don't turn it on - Take it apart!"
    • Facebook Page
Re: EEVblog #738 - Yamaha DME32 Digital Mixer Teardown
« Reply #12 on: April 29, 2015, 03:40:40 am »
Great example of how to design high end audio stuff without the audiofools getting in the way. Switching PSUs are perfectly fine to use, as are seemingly ordinary JRC opamps.
Cryptocurrency has taught me to love math and at the same time be baffled by it.

Cryptocurrency lesson 0: Altcoins and Bitcoin are not the same thing.
 

Online coppice

  • Super Contributor
  • ***
  • Posts: 5464
  • Country: gb
Re: EEVblog #738 - Yamaha DME32 Digital Mixer Teardown
« Reply #13 on: April 29, 2015, 04:57:10 am »
Its interesting how many through hole parts they have on those boards.
 

Offline EEVblog

  • Administrator
  • *****
  • Posts: 31340
  • Country: au
    • EEVblog
Re: EEVblog #738 - Yamaha DME32 Digital Mixer Teardown
« Reply #14 on: April 29, 2015, 05:51:23 am »
Anybody know what the retail price was?

Someone mentioned 4000 euro.
 

Offline PinheadBE

  • Regular Contributor
  • *
  • Posts: 65
  • Country: be
  • Pinball Freak
    • The Belgian Pinball Forum
Re: EEVblog #738 - Yamaha DME32 Digital Mixer Teardown
« Reply #15 on: April 29, 2015, 06:43:16 am »
In such a teardown, you can see how much Yamaha cares about their customers: this specific one seems to be specially designed for the Southern hemisphere.
Therefore, most of their IC's and silkscreens are labeled upside down, because we all know that Aussies read with their heads upside down.

Now, my neck is hurting.  |O

More seriously (or is it?) : Dave.... seriously.... a screwdriver as pointer :wtf:   I always thought you were a professional ....  :palm:
The Belgian Pinball Forum: http://ericpinballforum.be/
Please keep our planet clean
 

Offline Fungus

  • Super Contributor
  • ***
  • Posts: 10795
  • Country: 00
Re: EEVblog #738 - Yamaha DME32 Digital Mixer Teardown
« Reply #16 on: April 29, 2015, 07:19:18 am »
Anybody know what the retail price was?
Someone mentioned 4000 euro.
Less then I imagined...

OTOH the development costs are probably spread across various models in the range (and maybe over a couple of generations of the product).
 

Offline Tothwolf

  • Regular Contributor
  • *
  • Posts: 95
  • Country: 00
Re: EEVblog #738 - Yamaha DME32 Digital Mixer Teardown
« Reply #17 on: April 29, 2015, 10:59:22 am »
More seriously (or is it?) : Dave.... seriously.... a screwdriver as pointer :wtf:   I always thought you were a professional ....  :palm:

That's not a screwdriver... THAT's a screwdriver...  :-/O



Dave, any chance of revisiting the 2225 at some point? I've always wondered if -8.6V was out of spec causing the rest of the cal to go out. The power supply section on the 2200 series is notorious for having all sorts of weird issues.
 

Offline PeterFW

  • Frequent Contributor
  • **
  • Posts: 570
  • Country: de
    • Stuff that goes boom
Re: EEVblog #738 - Yamaha DME32 Digital Mixer Teardown
« Reply #18 on: April 29, 2015, 12:47:20 pm »
Maybe for cheap consumer crap, but not for pro gear.

What is pro gear?
All professional amplifiers i have seen are Class-D with a switching power supply and the ADC right at the input jacks to take that digital as soon as possible.
Every professional will be very happy if he sees them, why?

Because i can carry the 12kW RMS powersoft digicam with one hand, that is a 5K Euro power amplifier with all the bells ans wistles in the world.
I wanted to buy it dinner and take it to bed the first time i held one in my arms.
That is one seriously awesome piece of gear, dsp, power monitoring, impedance tracing, filters, crossover, ethernet, rs422, power limiting and so much more.

Edit: And the damn thing is only 1U high, seriously, if you want to see high integration, open one of these. There is no air left inside except for the cooling duct.
Every square millimeter is stuffed with electronics.

I do not have to push around a 150kg rack that will crush me if i hit the ramp at the wrong angle.
« Last Edit: April 29, 2015, 12:49:59 pm by PeterFW »
 

Offline smjcuk

  • Frequent Contributor
  • **
  • Posts: 464
  • Country: gb
Re: EEVblog #738 - Yamaha DME32 Digital Mixer Teardown
« Reply #19 on: April 29, 2015, 01:27:59 pm »
Great example of how to design high end audio stuff without the audiofools getting in the way. Switching PSUs are perfectly fine to use, as are seemingly ordinary JRC opamps.

Maybe for cheap consumer crap, but not for pro gear. Even if you are one of those lucky people who can't hear any difference between an Amstrad hifi and a Technics one, there are good reasons for using a high end linear PSU and high end op-amps. That PSU is going to get a lot of hours on it, and you need it to remain consistent over that time. Same with the op-amps, you want them to be well matched and tested for many years of operation. The cheap stuff tends to be hand matched or they just don't bother, and good lucky getting data on ageing.

Sorry got to jump in on this one...

My Korg M3 and Nord Lead 2 weigh enough without adding linear supplies thank you :)

As for the whole reproduction argument, the majority of stage/studio instruments use switchers, shit 7/0.2 wire, shit connectors and shit op amps. Once the signal leaves that kit, it must be already irrecoverably tainted and sound like farts down a wire according to the pro-linear and audiofool crowds.
 

Offline Tandy

  • Frequent Contributor
  • **
  • Posts: 372
  • Country: gb
  • Darren Grant from Tandy, UK.
    • Tandy
Re: EEVblog #738 - Yamaha DME32 Digital Mixer Teardown
« Reply #20 on: April 29, 2015, 04:29:11 pm »
The interface boards fit a number of Yamaha devices, you can get ADAT optical digital, firewire audio, Analogue and a fancy network audio card. They fit in things like the 02R mixing desk that has been around for decades.
For more info on Tandy try these links Tandy History EEVBlog Thread & Official Tandy Website
 

Offline dtweed

  • Contributor
  • Posts: 9
Re: EEVblog #738 - Yamaha DME32 Digital Mixer Teardown
« Reply #21 on: April 29, 2015, 06:35:27 pm »
24:10 Regarding passing rework wires through connector pads -- I learned long ago to include extra holes in my PCB layouts for exactly that purpose, especially on projects whose requirements were not completely known or likely to change. Saved my butt more than once!
 

Offline Tek_TDS220

  • Regular Contributor
  • *
  • Posts: 66
  • Country: 00
Re: EEVblog #738 - Yamaha DME32 Digital Mixer Teardown
« Reply #22 on: April 30, 2015, 02:48:24 am »
As Dave said, the YSS910-S is a 44 bit DSP, although I couldn't find the datasheet.   Anybody know why?  Since they are using 24 bit A/D's, wouldn't a standard 32 bit DSP be more than enough?  Or may use a floating point DSP?  They've were available for a couple of decades, I think.
 

Online coppice

  • Super Contributor
  • ***
  • Posts: 5464
  • Country: gb
Re: EEVblog #738 - Yamaha DME32 Digital Mixer Teardown
« Reply #23 on: April 30, 2015, 02:57:19 am »
As Dave said, the YSS910-S is a 44 bit DSP, although I couldn't find the datasheet.   Anybody know why?  Since they are using 24 bit A/D's, wouldn't a standard 32 bit DSP be more than enough?  Or may use a floating point DSP?  They've were available for a couple of decades, I think.
Consider the dynamic range of the numbers in an IIR filter that is filtering bass frequencies and you'll wonder how they can get away with just 44 bits. They are probably still doing double precision calculations for some of the processing.
 

Offline Tek_TDS220

  • Regular Contributor
  • *
  • Posts: 66
  • Country: 00
Re: EEVblog #738 - Yamaha DME32 Digital Mixer Teardown
« Reply #24 on: April 30, 2015, 03:10:41 am »
As Dave said, the YSS910-S is a 44 bit DSP, although I couldn't find the datasheet.   Anybody know why?  Since they are using 24 bit A/D's, wouldn't a standard 32 bit DSP be more than enough?  Or may use a floating point DSP?  They've were available for a couple of decades, I think.
Consider the dynamic range of the numbers in an IIR filter that is filtering bass frequencies and you'll wonder how they can get away with just 44 bits. They are probably still doing double precision calculations for some of the processing.
Thanks, of course, I wasn't thinking of the accumulator...  And maybe floating point processors were too new at that time.   I'm not sure what you mean by 'double precision' for an integer processor, though.  I'm not an expert on DSP.
 


Share me

Digg  Facebook  SlashDot  Delicious  Technorati  Twitter  Google  Yahoo
Smf