some more findings on ES9038Q2M boards (Netherlands), for those interested just web-translate those pages:
ES9038Q2M dac streamertje - Pagina 6 - forum.zelfbouwaudio.nl
ES9038Q2M dac streamertje - Pagina 6 - forum.zelfbouwaudio.nl
Depends what you want. There is no reason not to use modern opamps, they can give excellent sound quality. For example, the DAC in this review uses opamps and the DAC I am trying to encourage people to build if done properly can get fairly close to this: Benchmark DAC3 HGC D/A preamplifier-headphone amplifier | Stereophile.com
Or, some people like to have fun with electronics by trying to get other people to experiment around with unproven designs. If you want to do that I'm sure they would be happy to give you lots of ideas. So, it really does depend on what you want.
EDIT: Actually, if seems the latter choice above really is what a lot of people in these electronics forums do want. So, no problem there. People should be able to find what they want in DIY forums, and I don't want to bore some of those folks with my approach. My main concern would be if you want to take the first path above to a known and very good endpoint, it is a lot of work. You probably wouldn't want to redo some of it if you decided to experiment and it didn't turn out to you liking. Therefore, if you are willing to do work to end up with something very good and known to work, I would discourage experimenting with it (at least in areas where good solutions are already well known, and for this particular board everything is already pretty well known - but not as much so for boards with PRO series chips). Also, I don't think the guys who like to experiment or suggest experiments would do all that work themselves to make a good DAC like the one I did here, but with one experiment in it that would be hard to undo. I wouldn't do it either with this particular DAC project.
Last edited:
Well, my intention is to build a valve preamp to connect the DAC to, so I was thinking that a valve analogue output stage on the DAC would be the way to go, but that's just a guess and I'm always open to advice as I am very much a newbie with such things.
I always favour well proven designs, what has worked well for others as this is hardly cutting edge tech and I see little sense in trying to re-invent the wheel.
So regardless of whether I go for op amps, transistors or valves in my output stage, I will use a well proven design.
I always favour well proven designs, what has worked well for others as this is hardly cutting edge tech and I see little sense in trying to re-invent the wheel.
So regardless of whether I go for op amps, transistors or valves in my output stage, I will use a well proven design.
Looks a bit chaotic. Somebody should probably tell them about noise in tantalum caps they may be considering for LT3042 regulators. I am trying to mod one of the little LT3045 modules with 2 each 10uf X7R in parallel to get a little closer to 22uf. Haven't tested it yet, though.
Attachments
In that case my suggestion to you would be to build the I/V converters using opamps, and listen to that first to see what if anything you want to do with the sound. If at that point you still want some that valve sound then there are some places you could pick off a signal to run into a valve preamp.
I thought was to use one of these X10D clones as an output stage as it is simple solution, already being assembled and it looks to be a faithful copy of the original, so with a couple of good NOS tubes should sound good.
X10Db Tube preamp board base on X-10D Musical Fidelity ( no tubes) | eBay
However, I'm always open to advice and suggestions for alternative solutions. I haven't done much looking into op amp solutions so I don't know much about them.
X10Db Tube preamp board base on X-10D Musical Fidelity ( no tubes) | eBay
However, I'm always open to advice and suggestions for alternative solutions. I haven't done much looking into op amp solutions so I don't know much about them.
X10D have a schematic?
It is probably a voltage amplifier, and for low distortion with these DACs you need a virtual ground, or virtual offset of about +1.65v, which is called the reference voltage. You need four of those circuits. Then the current from the DAC runs into that offset virtual short circuit and is converted into a voltage signal. Done properly one can get down to -120dB distortion by such means.
On the other hand, the DAC output can be taken as a voltage, but the distortion is then about -75dB, which to me is unlistenably bad, and awful.
Making the offset current to voltage converters out of valves probably require development of a new and somewhat complex valve circuit. I don't think anyone I know of has been very successful with that so far. And you would need to make four of such circuits to do it the proven way we do it with opamps.
So, I would advise that the 'proven' approach requires opamps for that initial analog DAC interface circuitry. After that tubes would be fine.
It is probably a voltage amplifier, and for low distortion with these DACs you need a virtual ground, or virtual offset of about +1.65v, which is called the reference voltage. You need four of those circuits. Then the current from the DAC runs into that offset virtual short circuit and is converted into a voltage signal. Done properly one can get down to -120dB distortion by such means.
On the other hand, the DAC output can be taken as a voltage, but the distortion is then about -75dB, which to me is unlistenably bad, and awful.
Making the offset current to voltage converters out of valves probably require development of a new and somewhat complex valve circuit. I don't think anyone I know of has been very successful with that so far. And you would need to make four of such circuits to do it the proven way we do it with opamps.
So, I would advise that the 'proven' approach requires opamps for that initial analog DAC interface circuitry. After that tubes would be fine.
I think this is the correct schematic:
Taken from this thread:
Musical Fidelity X10D schematic
The X10D is a low gain buffer rather than an amplifier.
I don't understand why you would want or need to use op amps, the whole point of using valves is to avoid op amps as they are sonically less good, which is the entire basis of the work of Lampizator, for instance.
TECHNOLOGY
Taken from this thread:
Musical Fidelity X10D schematic
The X10D is a low gain buffer rather than an amplifier.
I don't understand why you would want or need to use op amps, the whole point of using valves is to avoid op amps as they are sonically less good, which is the entire basis of the work of Lampizator, for instance.
TECHNOLOGY
Lampizator looks like all marketing and not much substance to me. In any case, I have not done any work with valve IV stages and particularly for use with this type of Sabre DAC. (I do use valves very happily for guitar amps though.) So, sorry, I think you may be talking to the wrong person for what you want to do. Maybe one of the other guys can help.
EDIT: Some discussion of Lampizator here: Some final thoughts on the Lampizator Level 4 DAC – Part-Time Audiophile
Seems like it works for some music but not some others, and that it doesn't sound exactly like what's on the CD even though it can sound exceptionally good on some source material. Those things I can believe. For $10,000 or so, whatever, most of it for power supplies and implementation details, it is possible to get a single tube stage down to reasonably low distortion, and the distortion it does make can sound nice on some material. All that is in keeping with known facts about the physical world.
EDIT: Some discussion of Lampizator here: Some final thoughts on the Lampizator Level 4 DAC – Part-Time Audiophile
Seems like it works for some music but not some others, and that it doesn't sound exactly like what's on the CD even though it can sound exceptionally good on some source material. Those things I can believe. For $10,000 or so, whatever, most of it for power supplies and implementation details, it is possible to get a single tube stage down to reasonably low distortion, and the distortion it does make can sound nice on some material. All that is in keeping with known facts about the physical world.
Last edited:
Lampizator has a lot of substance behind it, he's examined virtually every DAC chip available and converted probably a hundred different models of CD player, I find his thinking makes a lot of sense to me:
Lampizator page index
The simple concept behind Lampizator is to replace all the active crap in an output stage with a single active component - the triode of a tube, and that makes sense to me, simplify things in order to get rid of all points where distortion and noise could be added to the signal by actives.
Lampizator page index
The simple concept behind Lampizator is to replace all the active crap in an output stage with a single active component - the triode of a tube, and that makes sense to me, simplify things in order to get rid of all points where distortion and noise could be added to the signal by actives.
Last edited:
A valve is active crap, unfortunately. If you get rid of all of it you wont have any dac at all.
Not sure where some information comes from but opamps used for DACs have much, much lower distortion than any valve. Valves are good for adding a little distortion, exactly what gives Lampizator its special sound. It is a little distorted, but a kind many people seem to like.
Not sure where some information comes from but opamps used for DACs have much, much lower distortion than any valve. Valves are good for adding a little distortion, exactly what gives Lampizator its special sound. It is a little distorted, but a kind many people seem to like.
Last edited:
We don't do that kind of stuff here with these DACs. It would ruin the sound. They are as minimalist as they can be, nothing not essential for the best possible sound by any means, accurate sound that is, whatever is on the CD as it was meant to be by the artist. If valves could do it better we would use valves. We don't think they can do it better than the way we do it.
Last edited:
I knew it would come to this as soon as I saw the word "Lampizator". Totally different paradigm for sound reproduction.
Tubes can be nice and sound great at times but at the same time it imparts a different sound. It is controversial to say the least and not for everyone. At the same time SS is not for the tube crowd.
What Markw4 was eluding to is something I have on my build schedule. To add a Tube Based Balanced to Single ended converter to my SS DAC. Most times, I will be using the single ended op amp based output. However, when I want to feel even more relaxed and want some euphonic sweetness, I'll turn on the tube based converter and switch my input on my preamp to run it in. (I could also build a tube buffer and run it through that Tape Monitor Loop. Remember those? What about 3 head tape decks. Remember those?) That way I can get some tube "sweetness" when I want and not worry about too many tube changes. Tubes have some weaknesses, and my approach is to use a hybrid approach that Borbeley suggests in this article which is hosted on Pearl Hifi website which was shown in Glass Audio in 1998. I've never had the desire to own tube stuff because of the constant rebiasing shifts and expense of changing tubes but it will be an adventure.
Last edited:
I choose to use a tube buffer/preamp to add some character. If the original sound is clean enough then the character will be an additional color by just selecting tubes. So for me it is of a better value to have tubes as an option, not as a final solution.
An other serious option is to use software tube DSP emulations, VST format in players like foobar2000... You can also fine tune with an eq, I choose to use PTEq-X by Ignite Amps
KVR: PTEq-X by Ignite Amps - Tube EQ VST Plugin, Audio Units Plugin and VST 3 Plugin
An other serious option is to use software tube DSP emulations, VST format in players like foobar2000... You can also fine tune with an eq, I choose to use PTEq-X by Ignite Amps
KVR: PTEq-X by Ignite Amps - Tube EQ VST Plugin, Audio Units Plugin and VST 3 Plugin
i'm sorry, but I don't remember the exact state of your DAC. Did you post some pics already? If so, I will try to find them.
About "tubes", this is a personal note and is not specific to any sound system.
I have ordered a su5 but not received it yet, at the moment I use other equipment than those referenced here. But considering the bad comments about su5 I will probably abandon it already, and choose an other path, probably a completed one with an other ess chipset this time. I am not specialized at electronics, just do some homework here to avoid bad choices like the one mentioned... with the potential to implement some minor tweaks if needed...
I have ordered a su5 but not received it yet, at the moment I use other equipment than those referenced here. But considering the bad comments about su5 I will probably abandon it already, and choose an other path, probably a completed one with an other ess chipset this time. I am not specialized at electronics, just do some homework here to avoid bad choices like the one mentioned... with the potential to implement some minor tweaks if needed...
Last edited:
It's not marketed for audio use first off, and second off, why would you, in any way, assume it gets worse below the measured frequency, which includes the tidy region where loop gain falls off and output stages become less linear? What red flag lights up? I can't think of an opamp that doesn't operate better at lower frequencies (or at least equal!). If anything, I'd be far more worried the opposite direction (not showing the opamp when its performance is "falling apart"). It's already stressing the floor of distortion test equipment at 100k! (roughly -125 dB HD3 into 1k with 2VPP)
I mean, I'd be pretty surprised if there was any audible difference with a 49710, and I'd be more worried about the surrounding circuit being sub-optimally designed for one or both of said opamps. They both seem quite up to the task here. Current vs voltage noise, power supply decoupling, and RF sensitivity are the things I'd be concerned with.
I guess you can't win them all.
Last edited:
If they thought it would sell at lower frequencies then why not show it in the data sheet? That doesn't seem odd for an opamp? If it were a transistor then sure, that could be.
One thing that is not unusual for devices optimized for HF is they may have a high 1/f noise corner. Don't know about this particular device, but my gut feeling is it doesn't smell right.
But, hey, I can hardly wait for one of you guys to build a ES9038Q2M DAC with it and do A/B comparisons with the opamps ESS recommends. However, I don't think we should be giving newbies the idea that they should be the ones to try it first if they want a good DAC and are putting a lot of work and some money into it. That's all. If one or two of the knowledgeable and experienced people want to do it to show the beginners how one should do such testing that would be great. Good learning opportunity for them to watch your progress as you build and test it. Please post plenty of pics if you would be so kind, I am sure the beginners would appreciate all they can learn from you.
Last edited:
The noise performance at audio freqs is shown in the DS, nothing to worry about from that pov.