For some time, I have been dissatisfied with the sound from my CD setup. I used an old S.A.T. CD-player as transport and a heavily modified Musical Fidelity M1 for DAC. For various reasons, I suspected that the transport might be the problem.
To test my idea, I bought an old DVD-player with a SPDIF digital sound out socket (Pioneer DVDs apparently have such). Price equivalent to 25 USD.
Surprise surprise, the improvement in sound quality was amazing. Clearly, the old transport was the problem and the DVD is staying in the system.
Does anyone else have experience with DVDs as CD transports?
To test my idea, I bought an old DVD-player with a SPDIF digital sound out socket (Pioneer DVDs apparently have such). Price equivalent to 25 USD.
Surprise surprise, the improvement in sound quality was amazing. Clearly, the old transport was the problem and the DVD is staying in the system.
Does anyone else have experience with DVDs as CD transports?
I used DVD players quite a lot as sources going back a few years - mainly because in those days they were too cheap to pass off. What I found was they could work fine provided I mitigated the effect of common-mode noise from their el-cheapo switching supply. Prior to discovering the noise on the digital output I'd been of the view that the only difference between transports was jitter. DVDs as transports do tend to have higher jitter because the clocks used for video (typically 27MHz) don't divide down exactly to audio clocks so a PLL needs to be used.
Thanks for your reply. I have considered the jitter problem, but that should be taken care of by having a DAC, which reclocks the incoming signal? I will have to check on the psu noise problem.
Merely from listening, I've not had any subjective problems with jitter using DVD players. But that could be because I'm not using S-D type DACs which output a lot of HF energy. They're considered to be more jitter-sensitive than multibit DACs. The PSU CM noise problem though was clearly audible until I fixed it (partly through re-layout of DAC and partly adding an isolation transformer).
So far your questions seem intelligent ones to me, so nothing to worry about!
As CM noise was really a new concept for me at the time, no I didn't add an extra CM choke. I could have done but wasn't up to speed on them. I've since discovered that many CM chokes don't work all that well i.e. don't provide a high enough CM impedance at the frequencies of interest (typically in the region 40-80kHz for the fundamental switching freq).
I've since then experimented with CM chokes on the mains side and developed some of my own segmented versions which cut down on the parasitic capacitance (hence improve noise rejection). However I prefer to put chokes on the low-voltage side as then the wires can be straightforward enamelled copper (this is given that I like bi-filar wound chokes).
As CM noise was really a new concept for me at the time, no I didn't add an extra CM choke. I could have done but wasn't up to speed on them. I've since discovered that many CM chokes don't work all that well i.e. don't provide a high enough CM impedance at the frequencies of interest (typically in the region 40-80kHz for the fundamental switching freq).
I've since then experimented with CM chokes on the mains side and developed some of my own segmented versions which cut down on the parasitic capacitance (hence improve noise rejection). However I prefer to put chokes on the low-voltage side as then the wires can be straightforward enamelled copper (this is given that I like bi-filar wound chokes).
Thanks a lot. I will experiment with the psu noise, as I do experience a bit of "listening tiredness" after extended listening, and that may well be psu noise - based on experience. Jitter gives me that same feeling and I initially thought it was caused by that, but because of reclocking it is unlikely.
Yes in my experience too these kinds of noise issues show up in part by generating 'listening fatigue'. They can also be more explicit and generate sibilance, particularly on vocals. From reading audio forums over many years I've come to the conclusion that a high proportion of what people think is jitter actually is CM noise issues. There does seem to be a fairly high awareness of jitter out there but awareness of CM noise is close to zero.
Until this last week, I had barely heard of "common mode noise", but you are now the second person to bring the subject up in a week (the other was "MEPER" ,also on diyaudio). I will have to investigate common mode noise, it seems to be important, not least with swiched mode psu's.
Happy to have piqued your interest, CM noise has been one of my hobby horses for the best part of a decade. You're right - while CM noise is an issue for the usual kinds of power supplies its a lot more important when dealing with SMPSUs - which pretty much all have CM noise generators built in.
One last question. If I put a common mode choke in, how big should it be. I believe I have heard 3A mentioned?
The DVD players I bought had maximum power draws under 15W if I recall correctly. So then you'd want the lowest rated CM you could get your hands on. Typically that might be a 0.3A one. The lower the current rating the better the rejection characteristics in general.
Incidentally, I'm talking here about mains side CM chokes as pretty much all the off-the-shelf ones tend to be as it saves on copper.
This one should be fine - SS21V-R031380 KEMET | Mouser
Incidentally, I'm talking here about mains side CM chokes as pretty much all the off-the-shelf ones tend to be as it saves on copper.
This one should be fine - SS21V-R031380 KEMET | Mouser
Last edited:
You sort of also need to know the frequencies of interest to select the choke with the appropriate core. It's probably in the kHz range for an off-line SMPS as Abraxalito mentioned, though. Don't pick one with something like a NiZn core that has an impedance that peaks at like 50 MHz.
A final word. I have now added an iron core choke (price 5 USD), and as the chassis/cabinet was very flimsy (tin-can like) I also added some bitumen damping to the lid and baseplate. Result is, that I now have an excellent cd-transport for just 30 USD. Brilliant and highly recommended, if you have an external DAC.
Last edited:
Richard, I agree. Blocking CM noise from transferring across the digital signal ingress is a frustrating choice, typically entailing an trade-off of CM reduction versus jitter increase.
A transformer coupled solution is commonly utilized, but introduces leakage inductance must be accounted for, else the interface impedance at the high frequencies of S/PDIF will not be stable. The the result, of course, is increased interface jitter. Correction typically requires use of an TDR setup to ensure an accurate impedance match. Transformers purposely designed for low leakage inductance obtain that at the cost of increased parasitic interwinding capacitance, which is a path for CM noise transfer right through the transformer.
Utilizing a CM choke at the digital signal ingress has the issue you've already mentioned.
TOSLINK perfectly blocks common mode noise transfer, yet typically introduces the most link jitter.
Arggg!
Last edited:
- Status
- Not open for further replies.