Hi Owen,
I have looked closely at your cct. And it indeed looks intriguing. Very nice work. I truly hope it is a success.
I still have a bit of fear regarding transient conditions if the DAC is powered down prior to the output stage running out of steam.
I have simulated the cct and the result was a very large transient which aligns with my practical experience.
I trust your work, and I don't mean to second guess you at all. Perhaps I am missing something.
now if one can insure that the output stage always is off prior to the DAC then your are golden.
As for the input impedance of the CP-II it is well under 500 milliohms. And I think I can get it a bit lower. It has neither startup or shutdown transients.
Cheers!
Russ
I have looked closely at your cct. And it indeed looks intriguing. Very nice work. I truly hope it is a success.
I still have a bit of fear regarding transient conditions if the DAC is powered down prior to the output stage running out of steam.
I have simulated the cct and the result was a very large transient which aligns with my practical experience.
I trust your work, and I don't mean to second guess you at all. Perhaps I am missing something.
now if one can insure that the output stage always is off prior to the DAC then your are golden.
As for the input impedance of the CP-II it is well under 500 milliohms. And I think I can get it a bit lower. It has neither startup or shutdown transients.
Cheers!
Russ
hmm you would need 750ma total. So 375ma per secondary (I would run them parallel). If you use the ones we supply (15VA) you should be just fine.
Cheers!
Rus
A couple of questions for Russ while I eagerly await my Buffalo II boards.
IC7 on the new Buffalo II board, what is part # is supplied for the SPDIF comparator?
For the SPDIF connection, is there a 75 ohm termination resistor provided on the board, or does that need to be accounted for by the end user?
Many thanks!
IC7 on the new Buffalo II board, what is part # is supplied for the SPDIF comparator?
For the SPDIF connection, is there a 75 ohm termination resistor provided on the board, or does that need to be accounted for by the end user?
Many thanks!
Understanding the role of VDD (1.2V) supplies on the ES9018
It has become clear to me the some folks do not understand what the 1.2V VDD supplies on the ESS9018 and its cousins do.
The 1.2V supplies only do two things in the DAC, the first is that it drives the core of the chip. This is of course crucial. The second is it drives the gates of level shifters (a high impedance) into the quantizers.
It is important to understand what these level shifters do. They simply shift the bits from 1.2V core voltage to 3.3. They do not effect the analog reference voltage. In fact the reason it is imperative that the AVCC be an extremely low impedance is because the frequencies involved are extremely high. All the VDD supply has to do is maintain enough voltage to keep gates of the Qs saturated. The AVCC supply is crucial because it has to absorb and source current at very high frequency and with very low noise at the same time.
Now it is important to bypass the VDD pins well (as I have done) but these pins are not in any way tied to the analog reference voltage.
The key to good results is a clean very low impedance AVCC supply.
I spent many many hours researching and testing this. And I have spoken directly with the kind folks from ESS on the optimal way to supply power to the chip. All of this research and advice has been put to good use.
So the bottom line is this. All of the supplies are important, and great care has been taken in evaluating the effects of decisions made and we made decisions that were optimal for the task at hand. The very most important aspect to each of the supplies is effective routing and bypassing. VDD is no different and we have taken very great care here.
This is one reason for the 4-layer board on the BUF-II. Initially we were convinced 2-layer would be sufficient, and it is just fine, but we have had much better results in terms of effective bypassing and routing with the 4-layer layout.
Nearly three years now I have been listening and tuning this DAC. I have been very diligent to understand how it works to the point where the kind folks at ESS have probably grown tired of all my questions.
Now please understand, I don't wish to disparage those who claim to have heard extreme changes in results by doing X instead of Y here, but I will tell you that I have yet to see such a claim that stands up to either logical or practical analysis. DIY people love to tweak, let them tweak all they want I say. But please understand, all of this has been gone over with extreme diligence.
Cheers!
Russ
It has become clear to me the some folks do not understand what the 1.2V VDD supplies on the ESS9018 and its cousins do.
The 1.2V supplies only do two things in the DAC, the first is that it drives the core of the chip. This is of course crucial.
The second is it drives the gates of level shifters (a high impedance) into the quantizers. It is important to understand what these level shifters do. They simply shift the bits from 1.2V core voltage to 3.3. They do not effect the analog reference voltage.
In fact the reason it is imperative that the AVCC be an extremely low impedance is because the frequencies involved are extremely high. All the VDD supply has to do is maintain enough voltage to keep gates of the Qs saturated. The AVCC supply is crucial because it has to absorb and source current at very high frequency and with very low noise at the same time.Now it is important to bypass the VDD pins well (as I have done) but these pins are not in any way tied to the analog reference voltage.
The key to good results is a clean very low impedance AVCC supply.
I spent many many hours researching and testing this. And I have spoken directly with the kind folks from ESS on the optimal way to supply power to the chip. All of this research and advice has been put to good use.
So the bottom line is this. All of the supplies are important, and great care has been taken in evaluating the effects of decisions made and we made decisions that were optimal for the task at hand. The very most important aspect to each of the supplies is effective routing and bypassing. VDD is no different and we have taken very great care here.
This is one reason for the 4-layer board on the BUF-II. Initially we were convinced 2-layer would be sufficient, and it is just fine, but we have had much better results in terms of effective bypassing and routing with the 4-layer layout.
Nearly three years now I have been listening and tuning this DAC. I have been very diligent to understand how it works to the point where the kind folks at ESS have probably grown tired of all my questions.
Now please understand, I don't wish to disparage those who claim to have heard extreme changes in results by doing X instead of Y here, but I will tell you that I have yet to see such a claim that stands up to either logical or practical analysis. DIY people love to tweak, let them tweak all they want I say. But please understand, all of this has been gone over with extreme diligence.
Cheers!
Russ
Last edited:
It does make me chuckle when I see people asking Russ and Brian whether or not ripping out this capacitor or adding that capacitor or randomly adding distorting tubes hither and thither will improve the sound. Clearly if they thought it would, it would already be there.
I also consider it to be rather ill-mannered to tell Russ and Brian that you think their DAC does not sound good on their own forum and then ask why, or whether the next one will not sound so bad!
Anyway, I'm looking forward to receiving my IVY III and Buffalo II. Any idea when the documentation will be on the site by the way? I like to be prepared.
I also consider it to be rather ill-mannered to tell Russ and Brian that you think their DAC does not sound good on their own forum and then ask why, or whether the next one will not sound so bad!
Anyway, I'm looking forward to receiving my IVY III and Buffalo II. Any idea when the documentation will be on the site by the way? I like to be prepared.
Yes not all the components are shown in the pictures.
glt, if you are referring to the filter caps that were at the differential input of the old IVY, those are not required (or in any way desirable) in the new design.
The IVY-III has a lot of differences from the first generation IVY.
The power supply rails are both well decoupled and well bypassed.
Cheers!
Russ
glt, if you are referring to the filter caps that were at the differential input of the old IVY, those are not required (or in any way desirable) in the new design.
The IVY-III has a lot of differences from the first generation IVY.
The power supply rails are both well decoupled and well bypassed.
Cheers!
Russ
Yes not all the components are shown in the pictures.
glt, if you are referring to the filter caps that were at the differential input of the old IVY, those are not required (or in any way desirable) in the new design.
The IVY-III has a lot of differences from the first generation IVY.
The power supply rails are both well decoupled and well bypassed.
Cheers!
Russ
Yeah, not including the components on the back of the board.
I was referring to bypass caps (100uf) at the Vin for the analog regulator daughter board. [You can see the cost comparison in the blog]
I have not looked at the new IVY in any detail (not much to look at yet
)Q: Are there plans of making a single version of the analog regulator daughter board so one can just plug it into the 3 pins of the clock supply
?
Ahh yes that AVCC is well bypassed too. No worries there. Remember too that it is pre-regulated and presents a constant current load..
Yes there are plans for a single version of that reg. I actually almost have it done.
But, if you ever take apart one of those clocks(it is a pcb with a circuit of its own), you will see why the regulator we use is actually quite well suited.
Cheers!
Russ
Yes there are plans for a single version of that reg. I actually almost have it done.
But, if you ever take apart one of those clocks(it is a pcb with a circuit of its own), you will see why the regulator we use is actually quite well suited.
Cheers!
Russ
Last edited:
Nice comparison.
I want to point out that the XO on the production Buffalos are actually custom built 20ppm units (when you order a lot of them...)
We are also now sold out of this batch, despite having a lot of them made.
We will begin mustering to get another batch going soon. It's amazing how low part inventories are right now.
I want to point out that the XO on the production Buffalos are actually custom built 20ppm units (when you order a lot of them...
)We are also now sold out of this batch, despite having a lot of them made.
We will begin mustering to get another batch going soon. It's amazing how low part inventories are right now.
Which connector is best for I2S/DSD into the Buffalo II
I posted on the digital board and didn't get a response, so I thought I'd post here.
I've ordered a Buffalo II and am planning the digital input scheme. I am thinking of connecting the SPDIF in and ground to a BNC, and then the 3 I2S/DSD inputs and the same ground to either a 4 pin XLR, RJ45, Lemo, or 2 more BNC's.
Would any particular connector work better? I have read that DSD should use CAT5, and some manufacturers are using CAT5 for I2S, so are RJ45's the best connectors?
Would it be a problem to connect the digital input ground to a BNC for the SPDIF and also to another connector for I2S/DSD? Should this ground also go to chassis ground at the connector?
Also, are there any problems with running I2S or DSD on a short CAT5 cable? I was thinking of modding one of my SACD players to output DSD, and I'm still thinking about the transport for an I2S connection.
Thank you for taking the time to read my post. I searched, but couldn't find a good answer, so if anyone happens to know what is ideal for these digital connections, I would appreciate the help.
Best,
Aaron.
I posted on the digital board and didn't get a response, so I thought I'd post here.
I've ordered a Buffalo II and am planning the digital input scheme. I am thinking of connecting the SPDIF in and ground to a BNC, and then the 3 I2S/DSD inputs and the same ground to either a 4 pin XLR, RJ45, Lemo, or 2 more BNC's.
Would any particular connector work better? I have read that DSD should use CAT5, and some manufacturers are using CAT5 for I2S, so are RJ45's the best connectors?
Would it be a problem to connect the digital input ground to a BNC for the SPDIF and also to another connector for I2S/DSD? Should this ground also go to chassis ground at the connector?
Also, are there any problems with running I2S or DSD on a short CAT5 cable? I was thinking of modding one of my SACD players to output DSD, and I'm still thinking about the transport for an I2S connection.
Thank you for taking the time to read my post. I searched, but couldn't find a good answer, so if anyone happens to know what is ideal for these digital connections, I would appreciate the help.
Best,
Aaron.
Chiaki and I have connected I2S signals of up to 352.8 kHz / 24 bit or 192 kHz / 32 bit from our DIY transport to ESS ES9018 2 CH Evaluation Board.
In our case, we use a flat cable that is usually used in digital circuits and its length is less than 5 inches. We have not experienced co-existence of S/PDIF and I2S. When we try I2S, we remove a S/PDIF line and vice versa.
As you may know, usual I2S is not a differential but a single-ended digital signal. Therefore, I do not believe a combination of CAT5 cable and RJ45 plug/jack is suitable. They are for a differential transmission.
Anyway, I think we should select appropriate one by observing waveform with an oscilloscope and listening actual sounds. Sometimes, we may need a trimmed termination resister for a line.
My friend uses a 6P DIN connector for I2S signals(including MCLK).
Bunpei
I'll say that for short runs, any cable will work. Cat 5 is actually very good copper and they come in twisted pairs that provides shielding (if you want more shielding, you can get cat-6). You can connect all the "white" wires to gnd.
If you look at the RJ45 connectors, it is basically two wires touching each other, except that if you use the Ethernet-ready sockets, they have a little pcb with some funny looking trace patterns which I think is for high speed issues (reflection and so forth) which I'm not sure matters or not in audio.
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.