Welcome Guest! To enable all features please Login or Register.

Notification

Icon
Error

Login


2 Pages12>
Options
View
Go to last post Go to first unread
viswad@gmail.com  
#1 Posted : Thursday, January 18, 2018 2:04:36 AM(UTC)
viswad@gmail.com

Rank: Member

Groups: Member
Joined: 1/18/2018(UTC)
Posts: 1
United States

Hello folks,
I have a 2 way stereo speakers and so I need 5-6 channels DAC to drive them. Is there going to be a 9038 based product that I can buy? If so when is it likely to be available.

BTW, I am an old customer, I own a Buffalo DAC that I built several years back

Edited by user Thursday, January 18, 2018 2:39:45 AM(UTC)  | Reason: Not specified

meo  
#2 Posted : Thursday, April 12, 2018 1:22:13 AM(UTC)
meo

Rank: Member

Groups: Member
Joined: 2/17/2016(UTC)
Posts: 5

Thanks: 1 times
I'm waiting for it too!!
Possum  
#3 Posted : Sunday, April 15, 2018 1:14:41 AM(UTC)
Possum

Rank: Member

Groups: Member
Joined: 8/5/2012(UTC)
Posts: 208
Australia

Thanks: 4 times
Was thanked: 16 time(s) in 14 post(s)
Couldn't wait any longer and transplanted an ES9038 into a Buffalo III:

UserPostedImage

One really cool feature of the new chip is you can map each of the 8 DACs to any input channel - which is ideal for my monoblock applications where I need to map the DACs to either all left or all right channels of multiple I2S data inputs.

You can't use the original firmware of course. I wrote an ES9028/38 Arduino Library which makes programming the firmware pretty easy - eg, the following code is the main routine to initialise a 9028/38 in 8 channel 24 bit I2S mode with Automute and lock lights working correctly:

void initDAC_ES9028(ES9028 &dac)
{
if (dac.initialise())
{
dac.mute();
dac.setVolumeMode(ES9028::Volume_UseChannel1);
dac.setVolume1(10);
dac.setInputSelect(ES9028::InputSelect_SERIAL);
dac.setSerialMode(ES9028::I2S_Mode);
dac.setSerialBits(ES9028::Bits_24);
//dac.setDpllBandwidthSerial(ES9028::DPLL_Highest);
dac.setAutoSelect(ES9028::AutoSelect_Disable);
dac.setFilterShape(ES9028::Filter_Hybrid);
dac.setAutoMute(ES9028::AutoMute_MuteAndRampToGnd);
dac.setAutomuteTime(100);
dac.setGPIO1(ES9028::GPIO_Automute);
dac.setGPIO2(ES9028::GPIO_StandardInput);
dac.setGPIO3(ES9028::GPIO_StandardInput);
dac.setGPIO4(ES9028::GPIO_Lock);

Edited by user Thursday, May 3, 2018 2:24:25 AM(UTC)  | Reason: Not specified

meo  
#4 Posted : Sunday, April 15, 2018 11:54:49 AM(UTC)
meo

Rank: Member

Groups: Member
Joined: 2/17/2016(UTC)
Posts: 5

Thanks: 1 times
Originally Posted by: Possum Go to Quoted Post
Couldn't wait any longer and transplanted an ES90386 into a Buffalo III:

UserPostedImage

One really cool feature of the new chip is you can map each of the 8 DACs to any input channel - which is ideal for my monoblock applications where I need to map the DACs to either all left or all right channels of multiple I2S data inputs.

You can't use the original firmware of course. I wrote an ES9028/38 Arduino Library which makes programming the firmware pretty easy - eg, the following code is the main routine to initialise a 9028/38 in 8 channel 24 bit I2S mode with Automute and lock lights working correctly:

void initDAC_ES9028(ES9028 &dac)
{
if (dac.initialise())
{
dac.mute();
dac.setVolumeMode(ES9028::Volume_UseChannel1);
dac.setVolume1(10);
dac.setInputSelect(ES9028::InputSelect_SERIAL);
dac.setSerialMode(ES9028::I2S_Mode);
dac.setSerialBits(ES9028::Bits_24);
//dac.setDpllBandwidthSerial(ES9028::DPLL_Highest);
dac.setAutoSelect(ES9028::AutoSelect_Disable);
dac.setFilterShape(ES9028::Filter_Hybrid);
dac.setAutoMute(ES9028::AutoMute_MuteAndRampToGnd);
dac.setAutomuteTime(100);
dac.setGPIO1(ES9028::GPIO_Automute);
dac.setGPIO2(ES9028::GPIO_StandardInput);
dac.setGPIO3(ES9028::GPIO_StandardInput);
dac.setGPIO4(ES9028::GPIO_Lock);



How did you manage the higher power requirements?

Possum  
#5 Posted : Saturday, April 21, 2018 2:08:18 AM(UTC)
Possum

Rank: Member

Groups: Member
Joined: 8/5/2012(UTC)
Posts: 208
Australia

Thanks: 4 times
Was thanked: 16 time(s) in 14 post(s)
simply replace the old tridents with the new series regulators
Lucaslucas  
#6 Posted : Monday, April 23, 2018 6:04:51 AM(UTC)
Lucaslucas

Rank: Member

Groups: Member
Joined: 3/7/2017(UTC)
Posts: 2

Eagerly awaiting... Any update, vague ETA or just info and about the IV would be very interesting. thanks
TioFrancotirador  
#7 Posted : Wednesday, September 5, 2018 6:39:43 AM(UTC)
TioFrancotirador

Rank: Member

Groups: Member
Joined: 11/8/2016(UTC)
Posts: 10
Poland
Location: Krakow

Russ, any update on Buffalo es9038pro 8 channel?
Possum  
#8 Posted : Tuesday, September 11, 2018 11:43:51 PM(UTC)
Possum

Rank: Member

Groups: Member
Joined: 8/5/2012(UTC)
Posts: 208
Australia

Thanks: 4 times
Was thanked: 16 time(s) in 14 post(s)
Well, simply exposing the individual DAC input and output channels on the circuit board is pretty straightforward, and the firmware would only require a minor change.

So I'm guessing Russ is either fundamentally changing the design (there are a lot more options like channel mapping that would be very useful for multi-channel applications but would not be configurable with the limited dip switches on the current board) - or he's busy with other things (multichannel IVY perhaps?)

In one way it would appear to make more sense to bring out multichannel first as its easy to turn it into stereo. The BIII had precisely such a capability via optional jumpers. However, it appears that the bulk of the customers want stereo by default - and with spdif and hirose connectors to boot (God only knows why as both provide inferior digital comms for music)
akras  
#9 Posted : Wednesday, September 12, 2018 1:00:13 AM(UTC)
akras

Rank: Member

Groups: Member
Joined: 12/13/2016(UTC)
Posts: 30
United States
Location: PA

Thanks: 2 times
Was thanked: 1 time(s) in 1 post(s)
If I were to guess, I would say a multichannel Mercury that doesn't have to deal with the extreme current output of the 9038 chip in the stereo configuration and thus can be cheaper...

Still, it would be nice if they were responding to the most general of the questions (like is it coming?-- which is especially relevant given that TPA no longer sells the 8-channel 9018-based board).
Possum  
#10 Posted : Wednesday, September 12, 2018 1:07:54 AM(UTC)
Possum

Rank: Member

Groups: Member
Joined: 8/5/2012(UTC)
Posts: 208
Australia

Thanks: 4 times
Was thanked: 16 time(s) in 14 post(s)
or do what I do and transplant a ES9038 into the old 8 channel board. Pin compatible - done a few of them. They work fine with an Arduino replacing onboard firmware (need I2C and reset pin isolation too tho).

You can pick up second-hand BIIIs relatively cheaply as many people want to upgrade
akras  
#11 Posted : Wednesday, September 12, 2018 1:42:17 AM(UTC)
akras

Rank: Member

Groups: Member
Joined: 12/13/2016(UTC)
Posts: 30
United States
Location: PA

Thanks: 2 times
Was thanked: 1 time(s) in 1 post(s)
I'm considering this option, that's why I was trying to understand if TPA is still planning to make the new 8-channel version. If not, then that's what I'll have to do. Of course, having a plug-and-play 9038-based solution with a reportedly superior Mercury-like output would be better, the question is if it's going to materialize...
Possum  
#12 Posted : Wednesday, September 12, 2018 2:04:10 AM(UTC)
Possum

Rank: Member

Groups: Member
Joined: 8/5/2012(UTC)
Posts: 208
Australia

Thanks: 4 times
Was thanked: 16 time(s) in 14 post(s)
TBH I didn't notice a staggering improvement in the Mercury over Legato - more midrange presence perhaps. I actually have one of my monoblocks still using legatos for top end and bottom end with ES9028s and I can't really tell any obvious diff. Am using DSPs tho, but the original calibration sweeps looked pretty similar and top end detail as well as bottom end punch/timbre are pretty similar

I do use ferrite beads on dac outputs to filter some of the high frequency gunk as the Legato doesn't have any filtering (but I use them on the Mercury inputs too)

I prefer to use Arduino to control the DAC as it gives you more flexibility with all the new registers (particularly input mapping), and I can also use it to control other things (like startup sequence, speaker protection, wifi control etc) - I actually had no choice in any case as the ES9038PRO board no longer has the ability to operate in mono mode by itself.
akras  
#13 Posted : Wednesday, September 12, 2018 7:24:17 PM(UTC)
akras

Rank: Member

Groups: Member
Joined: 12/13/2016(UTC)
Posts: 30
United States
Location: PA

Thanks: 2 times
Was thanked: 1 time(s) in 1 post(s)
Looks like in some time I may start asking you questions about your 9038 transplantation...

Getting a bit off topic here, but when you say that you don't notice much difference between Mercury and Legato, what kind of setup do you have (amplifier, speakers)?

Thanks!
Possum  
#14 Posted : Wednesday, September 12, 2018 11:15:46 PM(UTC)
Possum

Rank: Member

Groups: Member
Joined: 8/5/2012(UTC)
Posts: 208
Australia

Thanks: 4 times
Was thanked: 16 time(s) in 14 post(s)
Off topic doesn't matter on this thread given Russ is apparently not going to respond :-)

I actually did first listening tests with headphones (AKG K701 reference). I built a headphone rig so I could more easily detect the differences made by some build variants I had in mind (like adding an extra tier of power regulation and using chokes and ferrite beads)

I initially compared an ES9028 and Legato to ES9038 and Mercury in stereo mode (Legato is unsuitable in that config with a 9038 but Russ confirmed it can be used with a 9038 in 8 channel mode)

Not exactly an apples-to-apples test, but the midrange/low midrange sounded more pronounced/forward/fuller with the Mercury/ES9038 combo. Overall Resolution/detail sounded about the same. BTW: the 701s have a very flat response (they do not over-emphasise bottom end like most other headphones). I am not saying that a more pronounced midrange/low midrange is necessarily a good thing BTW - I prefer the mids to 'disappear'.

For speaker transponders I am using ScanSpeak Revelator tweeters and 5" midrange Revelator drivers in their own unported cabinet and 12" Peerless XXL extra-long-throws in a separate open dipole configuration (you can see pics I recently posted in the "projects" topic). Tweeter and midrange are powered by separate TPA Sympaticos. The 12" woofers are powered by a Chinese class-D 400W module I found on EBay.

I modded the Sympaticos to add filter caps and inductors leading to an array of 4 Jensen 4-pole 10,000mf caps. Bottom end PS uses 2 Mundorf Supreme 33,000mf caps (so thats a combined total of 212,000mfs in left and right front monoblocks). The midrange sounds particularly snappy - good recordings of drum toms and snare sound quite life-like.

3 way crossovers are all done by DSP so no loss caused by passive xover circuits. Xover points are currently set at 140Hz and 1450Hz. Each transducer is individually calibrated via DSP to give a flat response, and I am also using Dirac Live PC DSP to calibrate the room acoustics. The result is a very flat response from 20KHz down to 40Hz (where the 12" drivers start to fade dramatically) when you are listening in the sweet spot. It sounds incredibly good if I do say so myself. Very transparent except for the very low bottom end.

The right monoblock uses mono-mode ES9028/Legatos for both top and mids, and a mono-mode ES9018/Legato for the lows. The left monoblock uses mono-mode ES9028/Legato for mids and a single ES9038/Mercury for lows and highs. There is no obvious difference in the highs (>1450 Hz) between ES9028/Legato and ES9038/Mercury. There is an audible difference (just) in bottom end between half a ES9038/Mercury and mono-mode ES9018/Legato, but it is very close.

Unfortunately I am using the same mono-mode ES9028/Legato config for both midranges so can't do any comparison in the 140-1450Hz range and this is where you might hear the difference, but if youre using DSP speaker room correction any difference may well vanish anyway.

The Legato is nearly half the price of a Mercury, and I am very satisfied with its results. IMHO a bigger difference can be achieved by powering the DACs analog left/right by independent untainted PS rather than via the onboard VD supply (default config on Buffalo)

Edited by user Wednesday, September 12, 2018 11:40:27 PM(UTC)  | Reason: Not specified

akras  
#15 Posted : Thursday, September 13, 2018 12:38:54 AM(UTC)
akras

Rank: Member

Groups: Member
Joined: 12/13/2016(UTC)
Posts: 30
United States
Location: PA

Thanks: 2 times
Was thanked: 1 time(s) in 1 post(s)
Thank you for the details: you have a nice setup there...

I use UcD180HG with active 4-way speakers (you can find there description in the OPLUG forum under "DIY/experimental"); my goal is to see how far I can get with a better DAC solution (I'm using miniDSP now). With my system, the realism in the reproduction of sibilant sounds can certainly be improved, and this is what I want to try to achieve with the new DAC. The DAC solution is to include a Singxer F-1 USB receiver, a few TPA boards for optical and SPDIF inputs and switching between sources, a no longer available TPA ASRC board, feeding a modified miniDSP miniSHARC board (I didn't want to use the built-in miniSHARC ASRC); the miniSHARC board is connected a Cronus reclocker (which provides the clock signal for the ASRC and the miniSHARC). So all that has been assembled and tested (with a scope) more than a year ago, now just sitting and waiting for the actual DAC...

If I'm to go with a modified B-III board, do you have any suggestions on how to arrange the 4 output boards in a reasonably neat fashion? I saw someone was selling 8-channel IVY-III boards, but I don't think those are available now...
Possum  
#16 Posted : Thursday, September 13, 2018 1:18:27 AM(UTC)
Possum

Rank: Member

Groups: Member
Joined: 8/5/2012(UTC)
Posts: 208
Australia

Thanks: 4 times
Was thanked: 16 time(s) in 14 post(s)
I did a minisharc/Buffalo build last year - loaded with the Dirac Live firmware update - Correct me if I'm wrong, but it can only do at best a 2.1 digital xover - so you'd only need 5 channels (2.5 Legatos)

If you don't require SE output you could cut the Legato boards clean down the middle and only populate the balanced part of the circuit so you could stack each half on either side of the DAC to make the connections as short as possible - and it would only be a bit wider footprint than a DAC on top of a Legato/Mercury

I haven't tried this with a Legato but there's no reason I can see why it shouldn't work. The balanced L/R circuits are completely independent (you won't be able to cut a Mercury in half and have it still work :-) )
akras  
#17 Posted : Thursday, September 13, 2018 2:01:12 AM(UTC)
akras

Rank: Member

Groups: Member
Joined: 12/13/2016(UTC)
Posts: 30
United States
Location: PA

Thanks: 2 times
Was thanked: 1 time(s) in 1 post(s)
There is a plugin that allows running miniSHARC with 8 channels out, with a considerable number of IIR filters and FIR banks, independent for each of the 8 channels.

I like the idea of using truncated Legato boards (I don't need SE output). I'll have to think about it. Do you remember how large is the Legato output common mode DC bias (the real question: do I need DC-blocking capacitors when using UcD180HG amplifiers)?



Possum  
#18 Posted : Thursday, September 13, 2018 2:08:09 AM(UTC)
Possum

Rank: Member

Groups: Member
Joined: 8/5/2012(UTC)
Posts: 208
Australia

Thanks: 4 times
Was thanked: 16 time(s) in 14 post(s)
The later Legato version has a trimpot to manually adjust DC offset - no caps required.
akras  
#19 Posted : Thursday, September 13, 2018 2:18:05 AM(UTC)
akras

Rank: Member

Groups: Member
Joined: 12/13/2016(UTC)
Posts: 30
United States
Location: PA

Thanks: 2 times
Was thanked: 1 time(s) in 1 post(s)
Do you mean Legato v3.1? The manual ("draft") on the site says there is no differential DC bias, but there is a DC bias relative to the GND (this may potentially overload the amplifier's differential input). Am I getting something wrong?
Possum  
#20 Posted : Thursday, September 13, 2018 2:46:54 AM(UTC)
Possum

Rank: Member

Groups: Member
Joined: 8/5/2012(UTC)
Posts: 208
Australia

Thanks: 4 times
Was thanked: 16 time(s) in 14 post(s)
From the manual:

"It is important to note that while the balanced output differential offset can be completely nulled there will still be common mode DC bias on those outputs relative to GND. The user must be careful about this and if the next stage requires that there be no common mode DC bias then external AC coupling caps should be added to the balanced output. The single ended outputs DC offset can be completely nulled, so that if you using these you can have no AC coupling caps in the signal path at all."

I've used the Legatos with a number of amps with balanced inputs without issue. I typically only connect hot and cold and leave ground floating. I do not know if your particular amps have an issue with ground differential tho.
Rss Feed  Atom Feed
Users browsing this topic
GuestUser (2)
2 Pages12>
Forum Jump  
You cannot post new topics in this forum.
You cannot reply to topics in this forum.
You cannot delete your posts in this forum.
You cannot edit your posts in this forum.
You cannot create polls in this forum.
You cannot vote in polls in this forum.