Agreed Fran! Now the question is - does this mean that you have an impedance mismatch that the attenuator is somehow alleviating or is it some other issue being addressed by the attenuators. It seems the suggestion here is that you take out the attenuator & find your impedance mismatch & fix it. So get your scope out (or buy one) & learn how to use it & get yourself some 75ohm BNC terminations & start to test your cable, DAC output stage,SPDIF receiver input stage, etc. I know this will cost you a thousand or so & take forever & you may not be sure how to correct it but this is the correct way to do it & it's a learning experience, right! Please don't be harsh with me in your reply!
It would be good if SY talked about his measurements & we might be able to tease out why they are significantly different to the previous scope shots?
Indeed!
Last edited:
Well, no, I'm not going to do that. If Nelson Pass, Peter Daniels, Brian Donegan/Russ White and Doede Douma can't design their product properly, then I know I certainly can't. Lets see some of the more vocal contributors here take on some of those names
At the end of the day, some here are going to absolutely disbelieve that either the attenuators or the battery power supply to the hiface make any positive difference at all (and lets remember that they are 2 different things). Likewise for the other side. Personally I would like to think I sit in the middle and trust my own ears, in what I have experienced for myself, not making comments on something that I say can't work without ever having heard it!!
Sy, I look forward to your measurements (I said this already) - but don't think they will show anything (said this already).
For those who say I am biased and have expectation bias. Well, at least I have heard this stuff. The likes of wakibaki have exactly the opposite bias, but haven't heard any of the things being discussed here. Now thats an interesting experiment!!!
At least Sy will have heard this equipment. For that he deserves respect. He may be wrong, he may be right, but at least he has heard it.
Fran
They are exactly what one would expect with properly functioning attenuators using the right termination (NOT 175 ohms!) and the right characteristic impedance cable.
The right termination isn't 72ohm as you used! So why do you think the other scope shots are so different? Can you mis-terminate the impedance (50ohm) & show the scope shots?
As I said, if reflections are generated they will return back to the transmitter (passing through an attenuator) & assuming mis-termination at that end, some of the reflection will be bounced back to the DAC (through the same attenuator again). This to me means that the attenuation will be doubled - certainly not what is shown on your scope shots so I assume you aren't seeing any reflections?
Can you do so?
As I said, if reflections are generated they will return back to the transmitter (passing through an attenuator) & assuming mis-termination at that end, some of the reflection will be bounced back to the DAC (through the same attenuator again). This to me means that the attenuation will be doubled - certainly not what is shown on your scope shots so I assume you aren't seeing any reflections?
Can you do so?
Last edited:
Best to ask the people who got unusual results. Mine was set up properly and got exactly the results expected.
Because it's correct, within the tolerances of real-world cables and source impedances, and my measurements confirm that. The 175 ohms you suggested seems to be an odd choice, a deliberate mis-termination. I have no interest in wasting time looking at incorrect terminations- it's trivial to do it right, as I've demonstrated.
So you did carefully avoid eflections (by selecting a non standard 72R), great which of course invalidates anything I have to say about reflections.
Thorsten's words are certainly ringing true - you demonstrate no interest in testing this as per the listeners reports or previous tests. You have a different agenda, it seems. So what have you proven - that you can terminate properly given a scope, etc - bravo!
If you are not interested in testing the premise about reflections why are you wasting all our time? That was the purpose of the test!
You know my 175ohm was a typo, I wasn't suggesting this as a mis-termination.
Try an RCA termination or 50ohm - you already said you tried 110R - no shots?
Last edited:
Yes, with one qualification. "Carefully" implies "painstaking," which this wasn't- it's not rocket science to terminate a line with a resistance close to the characteristic impedance. I grabbed a resistor out of the box, checked it with an ohmmeter (I have old eyes and digits are easier to see than tiny markings), and soldered it in. That's all it takes to make things work textbook.
Yes you measured the characteristic impedance (how did you do this, btw?) & terminate properly, well done! Now why not do the test that was intended - to prove if these attenuators reduce reflections which probably occur in a lot of implementations (certainly ones that haven't been carefully measured & setup to avoid reflections)
Last edited:
John, why are you trying to make this sound much harder and more complicated than it actually is? I grabbed some off the shelf RG59 (bought it at Fry's, a large US chain store). I grabbed some off the shelf BNC connectors (also from Fry's). I grabbed the first resistor out of the bin that was close enough (gamma is something like 0.02). OK? Nothing fancy, nothing esoteric, exactly what any average audiophile will do.
Result? Attenuators attenuate exactly as textbook calculations predict.
See, none of this stuff is mysterious. And doing basic, simple, trivial things in the setup costs nearly nothing (I'm out $10 for the cable and some connectors). Of course, as someone who has nothing to sell, I would say that.
Result? Attenuators attenuate exactly as textbook calculations predict.
See, none of this stuff is mysterious. And doing basic, simple, trivial things in the setup costs nearly nothing (I'm out $10 for the cable and some connectors). Of course, as someone who has nothing to sell, I would say that.
Maybe you are missing the basis of what's gong on here. It is claimed that these devices reduce jitter. They may, or they may not. But what is important to most of us is the resultant analog signal. That is what we listen to - not the spdif signal.
FFT is a very powerful tool. Jitter can be seen in an FFT of the analog signal. A singe tone makes it very easy to see. Have a look at the signal below. It should be a single peak at 11Khz. The extra spikes are because of the jitter. On a decent system, you can hear it. If jitter affects music, it will affect a single tone. If jitter is reduced, then you will hear the difference and see it on the FFT. It's that simple. The effect would be the same on a complex musical signal, but it would be much harder to see. And probably much harder to hear.
The scope shots are also interesting. The Hiface interface seems to have a lot of overshoot and ringing on the spdif signal. Does that effect the decoding of the signal? Probably not.
For me the only question here is: Are the test instruments sensitive enough to detect small amounts of jitter?
Attachments
Is it? You take jitter as a single number! Maybe looking more into the type of jitter & the spectrum of it would prevent it being treated as THD was treated in the past - a single number
Maybe
Well that certainly is one question. If the measurements show nothing then of course any listening differences will be put down to boogaboo as already stated - neatly wrapped up, isn't it?
If you lay out what the mis-termination commonly is, then that could possibly be tested. I certainly understand what you're getting at, that attenuation seems to improve mis-termination.
Be specific.
BTW - I thought it was the Hiface that was being tested.
Let's just use RCA termination as I already suggested (30 - 50ohm)
Well this thread seems to be the catch all for both now - remember you closed down the other thread about the Hiface & measurements
Last edited:
OK. Please give us a typical signal flow and cable layout where this could be used. You've probably done it before - but if you could again, please.
I won't be testing it any time soon, but SY will. A cable and connection layout that this can be effective on is most welcome. Just give us something you've found to work before. Spell it out in good detail. Don't put anything else in the post. That would be a big help. A diagram would be nice if you have one.
- Status
- Not open for further replies.