AliExpress AD1865 R2R DAC

Status
Not open for further replies.
Member
Joined 2008
Paid Member
a more transparent system though I've found renders poor recordings more enjoyably than a lesser one.

I think I've found the opposite to be true in my experience. As my systems have improved, I have discovered some previously-cherished recordings are actually fairly lackluster in their mastering. I still love the music itself, and happily listen on the radio in the car (basic factory stereo in a Subaru) or on cheap headphones through my phone (i.e. decidedly non-hifi systems). But I don't enjoy them as much on my better equipment (which may not even qualify as hi-fi, but surely a step above typical cheap consumer stuff).

Also note that, as I've taken an interest in audio fidelity, not only have my hardware systems (electronics, speakers) improved, so have my biological systems. In other words, I believe critical listening is a skill that takes time to develop, and I'm certain mine has improved over the years. In fact, I'd almost call it a curse, because occasionally I find myself in a situation where I am unable to "un-hear" an imperfection, and it can be quite annoying!

Random anecdote on subjectivity: a few years ago, I bought an Emotiva Stealth DC-1 DAC. No one calls this an "end game" DAC, but all the reviews classify it as great value for the money (it's been a while since I read them, but I'm sure at least one gave it a punches above its weight).

When I first got it, I was like, "Yeah, great upgrade!" But over time, I found myself continually turning the volume down during extended listening sessions. And my listening levels are quite modest to begin with (that was back in my nearfield days, and free smartphone SPL apps claimed I never got above 80dBA, more typically in the 60-70dBA range). As time went on, I came to realize the DAC was giving me listening fatigue.

I brought this up with Emotiva. They actually allowed me to send it in, and run it through their AP unit. They sent it back with the AP report, which showed there was nothing obviously wrong. But I couldn't live with it. I also didn't want to sell it without a disclaimer. So I tried to start a "listening tour", to pass it around to different folks, and see if it was just me, or if there was something peculiar (albeit not measurable) about the unit itself. Well, the listening tour never happened, as a gentleman saw my post and offered to buy it after a listening audition. He received it, put it through its paces for a couple weeks, said he loved it, paid for it, and that was that.

Now, since only two people listened to this unit (the buyer and me), it's still ambiguous as to whether the problem was my ear or the unit. That DAC model seemed to be fairly popular, and I never was able to find anyone complaining about it causing listening fatigue; really, despite my best web-searching efforts, I couldn't find many complaints about its sound quality at all (save for small give-and-take nuances compared to similarly-priced DACs, or obviously if compared to much more expensive DACs). Given that, and that the person who bought mine loved it, I've come to the conclusion that the DAC wasn't a unicorn, and that my ears are the unicorn.

Even if there was something peculiar about that DAC, there is one person who can "hear" it (me) and one person who cannot (the buyer). I put hear in quotes because it wasn't something obvious I could actually hear, but rather a general sensation of discomfort that grew the longer I listened to the unit. It lessened when the volume was turned down, and went away shortly after turned off (or I switched to another DAC).

Just thought some of you might find that story interesting. When it comes to what makes the best I/V stage (opamps vs discrete vs whatever), I have neither the technical chops nor the subjective expertise to argue either way. But I can't help but wonder if some people arrive at their conclusions because they pick up on something inexplicable, such as I did with that Emotiva DAC.
 
How could you know it isn't on a recording?
Coming to the conclusion that its not on the recording is a deductive process from listening to various recordings (various labels, differing genres) and also listening to more than one DAC. Generally if I listen to the same recording on more than one DAC and some artifact is consistent then I'll assume its related to the recording and not the DAC. But that has to be cross-correlated with finding recordings which don't exhibit that same artifact on either of the two DACs.

Maybe not on the tape or other media the master was mixed down to, but how can you know the mastering ADC was perfectly transparent?
I don't think I'll ever know if the recording processing (ADC plus all the rest in the chain) is perfectly transparent. What's possible to determine though is relative transparency. From listening over time its possible to determine which DAC is more transparent because the more transparent DAC allows more visibility of the recording. Or put another way, it shows recordings to be more individual - it increases the distinguishing characteristics of recordings.
 
Okay, good enough I guess. However, the perception you described as an emphasis on certain frequencies, so far as I can tell it is caused by a type of nonlinear distortion present on all frequencies, but that is perceptually most noticeable to humans in the frequency range you mentioned. It certainly doesn't seem to be a linear distortion of frequency response. And, IME, as I make multiple small improvements to dacs expected to reduce low levels of distortion to even lower levels, I find that, for example, as mid-high frequencies sound less forward and less clinical, bass frequencies also sound less bright. It's just that on bass, the original effect didn't sound bad or implausible, it didn't attract attention.
 
Last edited:
Fair enough - it could well be that the distortion happens equally across the band and its perceived as being at the frequencies where our ear is most sensitive (2k-4k, roughly).

What I have found is I can reduce it by lowering the impedance of the power supply. I demonstrated this most recently with my gf listening - I'd played a track with female vocal and she'd remarked 'her voice does sound a bit harsh' so I said 'wait a minute, let me twiddle with something'. I slapped on extra 47uF 0805 caps between the supply pins of the AD8065s (4 in total) and powered up again. Her face showed complete surprise - 'how did you do that?' (and no, she wasn't in the kitchen at the time so the cliche can't quite apply here :D).
 
In this case, they already were biassed into class A. I have two-transistor current sources between outputs and GND. Incidentally these opamps are running from a single (not dual) supply so GND = the -ve supply pin.


Biassing the opamp with a CCS to GND doesn't make the supply current signal independent. It helps but doesn't eliminate (even to a first order) the load-induced supply ripple. In order to achieve that, one has to run balanced - that's my next experiment. I'm part way through building a balanced I/V+filter.
 
Last edited:
But I don't find that detail is really on the recording, rather its a slightly unnatural emphasis on certain upper mid-range frequencies.

For a while I too thought this was the case, but I am now fairly sure that it is on the recording.

I notice this on analogue recordings, and sometimes on some early digital recordings, but generally not on modern digital recordings.

With regard to classical recordings, I think it's "hard" sounding microphone preamps in conjunction with older microphones which cause the effect we hear. I've heard the same hardness when playing vinyl. (Not often, but I have heard it)
I just think that digital playback more easily reveals these recording problems.
 
I've read that article by Lynne Olson. After years of reading people say that op-amps are no good for IV, and trying valve IV stages and discreet transistor IV stages, and op-amp IV stages Ive come to the conclusion that op-amps are perfectly fine, in fact best for IV.

I don't like valves because each valve sounds different and consistent performance cannot be obtained. Discrete sounds fine, but no better than op-amps. I recently found a Yamaha CD-500 CD player with PCM54 DAC and slow old NJM op-amps and it sounds a little dark, but very smooth. Hard to say if the sound is due to the DAC or output stage, but I like it none the less.

I'm not convinced that op-amp bandwidth matters as much as some people say. My best sounding DAC uses AD711. I haven't tried the 844 or the Sen/zen IV stage. Perhaps they are markedly better?


I like the AD811 for an I/V stage. Installing a CLR filter in front of it helps lower the out of band energy. The small R (1-2 ohms) is in series with the L as the 811 will osc if it sees a C on its input.
 
That sounds good, given the filter is very close to the DAC IC so there is a short return path for the HF. What values did you use for your filter? Have you tried a very small series R, a shunt C, then another small series R?

Basically, I think a circuit tailored to the op amp/DAC IC with excellent layout and power is the best I/V solution but I've never seen anyone build a range of circuits and test each of them. It's all rather subjective without this.
 
Last edited:
I'm part way through building a balanced I/V+filter.

The AD8067s came in and they've been built into a balanced I/V stage (totally 4 I/V opamps for stereo). The board piggy-backed onto a lingDAC DAC board has these latest opamps plus two current sources each - one for DAC offset adjust and the other for classA biassing.

Initial listening results are promising, no immediate signs of harshness on female vocals or violins even with just a single 47uF cap per opamp as decoupling. Seems like theory and practice might be in agreement :)
 

Attachments

  • AD8067IV.jpg
    AD8067IV.jpg
    255.8 KB · Views: 352
That sounds good, given the filter is very close to the DAC IC so there is a short return path for the HF. What values did you use for your filter? Have you tried a very small series R, a shunt C, then another small series R?

Basically, I think a circuit tailored to the op amp/DAC IC with excellent layout and power is the best I/V solution but I've never seen anyone build a range of circuits and test each of them. It's all rather subjective without this.

I use a cap of around 330pf then a single layer 10uh choke with a DCR of 1 ohm. I have not used the RCR you describe as I like the 12db/oct of the CL filter. The most important thing is to roll off the high frequency out of band energy of the DAC chip before it goes into the I/V op-amp.
 
Thanks. I've seen a few people post about using a similar sized cap across the differential outputs for current out DACS with great (subjective) success. Apparently it doesn't seem to have much effect with voltage out DACs like AKM. I have an ES9038 I plan to try it with but so far haven't got round to it - if it ain't broke, why hack it thing.
 
I agree @Rick its very important to roll-off the HF before it gets to the I/V. Here's the filter I'm using with my AD8067 I/V stage - its a 3rd order LPF with enough peaking to provide NOS droop correction.

The 13ohm resistor (R3) is the termination for the filter and its value is fairly critical to getting the correct FR. With opamp I/V the input impedance at the -ve pin of the opamp is normally very low (created by feedback) so the filter can be used as-is with an opamp following. It can also be used with a discrete stage with the proviso that R3 will need to be reduced so that the sum of it with the input impedance of the I/V stage equals 13ohm.

The caps should ideally be NP0/C0G - use 4*100n in parallel for 400n and 2*100n + 2*33nF for 266n. The inductor isn't too critical in terms of ESR, I'm using SLF7045T-151 (TDK).
 

Attachments

  • NOS-droop-correcter.png
    NOS-droop-correcter.png
    4.3 KB · Views: 451
Abraxalito, I like your choice of I/V op-amp, FET input and high speed. I also like your LPF for the input of the I/V stage. I too have noticed the slight roll off of treble and have tried other things to bring up the highs but would like more and this filter might do the trick. I am using a TDA 1541 DAC and I know they don't like to see much voltage on their outputs. Maybe the 13 ohm would be ok, but I like to keep the voltage low at the output of the DAC.
Thanks for posting the filter.
 
Hi Rick - there's one caveat with this droop-correcting filter. That is the impedance seen by the DAC's output rises sharply at the filter's corner frequency. I'm seeing about 8dB boost meaning that close to 20kHz there's an equivalent load of about 33ohm. With the 4mA total swing that results in a worst case 131mV peak-peak at the DAC output.

If that turns out to be too high the solution is to scale the C and L values to give a lower characteristic impedance. For example if L was reduced to 100uH the 400n -> 600n and 266n -> 400n. The 13ohm becomes 8.7ohm and the worst case swing is 87mV.
 
What's missing is a connection to GND for R1, right-hand end. The far end of R1 is designed to go to the virtual earth (pin 2, -ve input) of an opamp. Then you'll get current flowing in R1 - right now looks like you've an open circuit.


In the second plot you're seeing an extremely high Q due to it being an undamped circuit. R1 provides necessary damping, but only when current can flow in it.
 
Last edited:
My AD1865 board arrived today. It looks very well made, same maker as the "little giant" 1387 board. Seems to have some quality parts if they are real. I can plunk it into the test box I have the TDA 1387 board in to give it a listen, but I need to get some connectors for the board first, I ran out in my parts bin. I'm out on vacation next week so I'll post back with my impressions when I get back.
 
Status
Not open for further replies.