Jump to content


  • Content Count

  • Joined

  • Last visited

About JohnSwenson

  • Rank
    Junior Member

Recent Profile Visitors

The recent visitors block is disabled and is not being shown to other users.

  1. Let me see if I can bring some clarity. There are two types of "SQ degrading" influences the EtherRegen is designed to radically decrease: leakage, both high impedance and low impedance, and clock phase noise. The clock phase noise travels on the Ethernet signal itself (every edge coming out of any digital device caries the phase noise of the clock used to "clock out" that edge). The very carefully chosen transformers on both sides play an important part in decreasing leakage. The active circuitry in the path across the moat adds a very major decrease as well. The result is that the leakage from A to B OR B to A is is decreased a huge amount. The decrease in leakage from one port to another on the A side is still quite significant but not nearly as much as when going from side to side. The circuitry across the moat is designed to essentially eliminate the signal borne phase noise from one side to the other, it doesn't matter which direction, it works identically in both directions. The circuitry between ports on the A side decreases these phase noise effects to some degree but not nearly as much as going from side to side. There is ONE small difference between directions going across the moat: The clock generator is on the B side, so the circuits on the B side get a "pristine" clock. The clock from the B side goes through a very special isolator to the A side. This isolator has extremely low additive phase noise, much lower than any other isolator I could find. (it aint cheap!) The clock on the A side has slightly worse phase noise than the clock on the B side because of this. Whether this is going to be audible, who knows. Remember all the decrease in leakage and external phase noise is still there. Going from port to port on the A side should be better than any other switch out there, but going from side to side (either way) will be a whole new world. Because the B side has a slightly lower phase noise clock it is usually better to have the B side port connected to the streamer etc. But if you need to cross the moat the other way (such as using the SFP cage to drive optical into a streamer or DAC that has an optical input) that is also fine. The same decrease in leakage and external phase noise exists either way, the only difference is the slight increase in phase noise of the clock when going from B to A. Because of this slight increase in phase noise when going from B to A, if you use a REALLY good external clock (such as a Ref10), you will only get the advantage of such a clock when connecting the B side to the streamer. I hope this makes some sense. John S.
  2. Yes the port that is on the side with the power inlet, LED etc is 100Mbs. Previously this was a "it HAD to be 100Mbs in order to work", but there are now chips JUST starting to come into being that would allow gigabit, BUT they are not in volume production, and it would take a major change in architecture. This would cause a significant increase in time to market and cost increase. We decided it wasn't worth it at this point. John S.
  3. There shouldn't be any difference going in with with optical, even an opticalModule. That is the theory. In actuality, who knows. Right now I just have a board that WORKS, but is not the final version, so very subtle things like this cannot possibly be accurately tested. John S.
  4. Yes this will work quite nicely, the EtherRegen has an SFP cage into which you can plug an optical SFP module, run fiber to an FMC (it has to match the connectors, fiber type and wavelength of the SFP module you plugged into the EtherRegen. The easiest way to do that is to use an FMC with an SFP cage and use the same model SFP module in both the FMC and EtherRegen). The single port on the EtherRegen goes to the streamer. The Roon Nucleus and NAS plug into the 4 jacks on the other side. All this is on the same subnet so it all talks to each other. If the router is WiFi you can use a wireless device to control Roon. The Roon Nuceus and the NAS do not HAVE to be connected directly to the EtherRegen. You CAN connect them to the EtherRegen if they are near to it, but they will work just as well with NO sonic degradation if connected to some other switch or the router. John S.
  5. Hi DelsFan, I think you have some misunderstanding of what the EtherRegen is and how it is designed to be used. The single port is designed to be used to connect to the high quality audio portion. So that will be connected to the streamer. The 4 ports on the other side you can consider "normal" LAN ports, think of them as extenders for the LAN ports on your wireless router. The normal way this would be used: Single port on EtherRegen connects to your streamer -- the EtherRegen should be fairly close to the streamer (like the same room). Modem connects to WAN port of the wireless router. One of the four ports on the EtherRegen connects to one of the LAN ports on the wireless router. Other devices can connect to either the LAN ports on the wireless router or the remaining 3 ports on the EtherRegen wherever they pysically fit the best. All these connections are on the same LAN, they can ALL talk to each other, the controller for the streamer can by on wifi, physically connected to a LAN port on the router, or connected to one of the four ports on the EtherRegen. You do NOT need a special power supply or clocking on ANYTHING else on the network (other than maybe the EtherRegen itself and what is down stream of it). That is what the EtherRegen is all about, preventing the effects from anything else on the network from affecting the sound. John S.
  6. This is just for the high impedance leakage generated by SMPS. Linear supplies still have leakage but it behaves very differently. Some LPS can have significant amount of leakage others extremely small amounts. BTW SMPS have both types, the same as LPS AND the high impedance type which is just SMPS produce. John S.
  7. I don't really know what is in a Gigafoil so it is hard to speculate, but my guess is that the EtherRegen would make it superfluous. Remember that the etherRegen has two primary things: provide isolation from upstream leakage current (in all forms), AND prevent upstream clock noise (phase noise, jitter, whatever you want to call it) from propagating through. To my knowledge nobody else is doing both those. John S.
  8. I don't know what this statement is about, DDS has nothing to do with clock generation. DDS is a way to generate (note, not play but generate) an analog wave form such as a sine wave. You have a counter which counts up and down a certain number of bits (8 bits, 16 bits, 24 bits etc). If you feed the output of the counter into a DAC chip you get a triangle wave. Converting into a sine wave is usually done with a look up table. But if the number of bits is high that lookup table can be HUGE. FPGAs are getting good enough that a trigonometric processor can actually do the calculations on the fly, even at very high bit depths. There are a couple of different ways of getting that initial triangle wave. A common one is say have a 32 bit accumulator (register and adder) and a fixed sample rate (44.1, 192 whatever). Every click of the sample clock you add a number into the accumulator. This number changes with the frequency of the sine wave you want to generate. For a higher frequency you use a higher number. This type of circuit has nothing to do with jitter attenuation, the jitter in the clock is directly transmitted to the output. I don't get what DDS has to do with clocking, it's all about generating a specific frequency sine wave, why would you use that for clocking? DDS is great for building a synthesizer (thing with a keyboard and music comes out) but for playing back music I don't get it. I supposed they could be meaning that they use the part about using different numbers into an accumulator to generate word clocks from a high frequency oscillator rather than using an analog PLL. This is usually called a DPLL. This works by having circuitry in the FPGA or processor code change the number sent to the accumulator such that the time that the accumulator overflows is somewhat close to the reference. The problem with this is that if you need fairly fine frequency change you need a very high frequency local clock. The frequency resolution is usually fairly coarse so the output frequency is jumping around rather than smoothly following the input reference. But personally I do not call this "DDS" there is no analog waveform coming out of this. If they are using a Digital Phase Locked Loop, just call it that. They COULD be meaning that they are using a DPLL to generate the internal word clock, that can very well have lower jitter than an analog PLL, but the output frequency will probably be bouncing around. But the jitter is still not immeasurable. The info in the quote doesn't give anywhere near enough info to figure out what is actually happening in the circuit. John S.
  9. There is no way to make a general "one is always better than the other". There is far more to a power transformer than just if it is EI or toroid. How the design handles high current pulses (does the magnetic circuit saturate) is very important for most linear supplies. The standard way these supplies work (large cap after diode bridge) produces high current pulses at the peaks of the waveforms, the transformer needs to be able to handle these. NOBODY specs this for power transformers. So you either have to measure this yourself or significantly derate the transformers. This is why it is usually not good to spec the VA of a transformer near what steady state ohms law calculations say. Exactly how much? Who knows, it can vary all over the place depending on the circuit and the transformer. Some people say you should rate the transformer at 1.5 what the simple calculations say. Most of the time that works pretty well, but not always. Several years ago I was working on a fairly high power PS design and decided to go with the empirical approach, I ordered a whole bunch of different transformers and tried them out in the real circuit. I found that most of the EI transformers were very close to their rated VA (almost no margin). Many of the large high power toroids had fairly large margins. Thus for similarly speced transformers I got much better performance from the toroids. Thus for the specific PS design I was able to get good performance (properly handling high current pulses) from a toroid with a lower VA than a EI. This meant the toroid transformer was much cheaper. On the issue of high frequency passing from one winding to another, a basic toroid has much more capacitance than a split bobbin EI. But you have to be sure the EI is split bobbin. This can mostly be alleviated by a toroid with a shield between windings. I found in my tests that a toroid with inter winding shield fared just as good as the split winding EI, and given the much lower price, it was the way to go. For lower power PS I found the opposite. For some reason the small toroids were not nearly as good as the large ones. So for me at least I would go with a split bobbin EI for lower power (say less than 30 watts) but derate it to 1.5 simple calulations (say 48 VA instead of 30VA) and an inter winding shielded toroid for larger supplies. Of course that is not always true for all designs. I would not swap out a transformer in an existing design unless you really know what you are doing and can measure what is happening. There are so many ways you can make it worse, and only a few that will make it better, its usually not good odds of accidentally finding the better. John S.
  10. Hi Johnseye, I personally do not like those 3 to 1 adapters for two reasons: 1) you never know what is inside one unless you buy two and destroy one so you can see what is inside 2) when you plug and unplug cords in you are producing highly unbalanced forces on the outlet this thing is plugged into. This can be very bad for the outlet. I have seen these destroy an outlet. For number 2 you can theoretically plug all three cords in before plugging the whole thing into the wall, but I still think you wind up putting a lot of stress on that one outlet. I personally think that your best bet is to have the electrician put in the quad outlet. Other options are as has been mentioned using a good power strip. There is a whole range here. I personally do not like the ones with all the surge suppressors and filters etc, I find they harm the sound more than they help. I really like the Tripp Lite ones with the aluminum housing, such as the PS120410. This is just a straight simple power strip, they are built well and work well. Another option is the VH Audio Hot Box, it is a an electrical box with 4 good outlets and a power cord, no switch etc. It works very well but costs $250 so not cheap, but if it is cheaper than the electrician than it might be a good way to go. It is probably a little better than the Tripp Lite above for 5 X the price. John S.
  11. Most of the time this happened the source had a connection to safety ground and something on the other side of the USB ground cut also had a connection to safety ground, such as DAC, preamp, power amp etc. One system I heard about had several of these special USB cables with various injection schemes which wound up making a ground connection through one of the power supplies to something on the other side of the USB ground cut. That one took a long time to figure out what was going on since there were 4 different "special" USB cables involved. It took tests of all the cables to figure out what was actually connected to what to figure it out. John S.
  12. The term "VBUS handshake" is not an official term. There are two things that may fall under this name. According to the USB spec a device (such as a DAC) cannot connect to the D+,D- pins unless the VBUS pin is near 5V. Some DACs power some or parts of their circuitry from the VBUS pin, but even if they are self powered they are still supposed to look at the VBUS and not connect if VBUS is less than 4.75V or so. SOME DACs completely ignore VBUS and connect whether VBUS is powered or not, this is not spec compliant. There is another use of VBUS, for OTG ports (which can be either host or device) under some circumstances VBUS is actually used as a communications line, the voltage is pulsed, this is used to resolve which type of functionality the port should take on. This is rarely if ever used with DACs, thus "VBUS handshake" usually refers to the above, the device should not connect if VBUS is not near 5V. This all assumes that the ground line is always connected. Things get really wonky when people try and disconnect the ground line. Some aspects of the USB signaling will not work without the ground line, and the VBUS detect will also not work without the ground line. There are a few very specific instances where some people have gotten away with cutting the ground wire, but in these cases there has ALWAYS been some other path where the computer ground gets connected to the USB receiver chip's ground. John S.
  13. I agree with mansr on this. Unless something is really done wrong the reflections on USB are quite small. This all started for S/PDIF interfaces using RCA plugs and jacks which DO have significant relections (RCA plugs are nowhere near 75 ohms). The 1.5 meters was based on the round trip time of a cable relative to the cycle time of the 44.1 S/PDIF stream. Even if there ARE reflections on USB this is only going to be appropriate when using full speed NOT high speed. Using this logic the minimum would be an inch or so for high speed USB. I don't see this as a criteria at all. John S.
  14. The issue there is the SBT, IT doesn't have a standby mode either. When it is "off" everything is still powered up, there is just no audio data flowing through it. So the power draw when "off" is almost the same as running. Thus whatever is powering it is still supplying almost the same amount of power. John S.
  15. I was listening to headphones, how could that happen? John S.
  • Create New...