|
|
Post by williaty on Sept 29, 2014 21:13:25 GMT -5
I'm an Amateur Radio operator and a while back I noticed that I couldn't communicate with a nearby, very powerful repeater. For a while, I just assumed it was broken. Then I realized it was local radio frequency interference (RFI) coming from somewhere near my home. Eventually, I got fed up with it and hunted it down tonight. Turns out, it's my DC-1 DAC causing the problem. The RFI is generated whenever the DC-1 has electrical power, regardless if the DC-1 is "fully" on or just in Standby. The only way to silence the RFI is to physically unplug the DC-1 or to truly power it off using the rocker switch on the back panel. The RFI is centered around 146.765MHz but is at least 20kHz wide.
Has anyone else ever had an RFI problem with their DC-1? If so, how'd you fix it?
I emailed Emotiva about this as well but if I'm hoping I might get lucky and get a quick answer from you guys.
|
|
|
|
Post by solidstate on Sept 29, 2014 22:07:23 GMT -5
BRUTAL
No way that would pass certification if this is normal operation.
|
|
|
|
Post by Gary Cook on Sept 29, 2014 22:46:53 GMT -5
Wow, 2 Metre amateur radio, that brings back memories.  I remember routers (LinkSys and maybe Cisco) being a problem around that frequency. I also recall something about Sonos also causing problems at 146. I haven't heard of a DAC with that frequency of RFI though. Cheers Gary |
|
|
|
Post by andyadler on Sept 29, 2014 22:53:04 GMT -5
Hmmm. I have a Yaesu FT-8800 that operates a few feet from my DC-1, but haven't had the rig running yet while the DC-1 is powered up. Will try it and let you know if anything seems amiss. Antenna is a J-pole set up indoors, by the way. I'm on the third floor of my house.
73...A.
N4ADL
|
|
|
|
Post by williaty on Sept 30, 2014 1:55:01 GMT -5
Wow, 2 Metre amateur radio, that brings back memories. I remember routers (LinkSys and maybe Cisco) being a problem around that frequency. I also recall something about Sonos also causing problems at 146. I haven't heard of a DAC with that frequency of RFI though. Online, I saw results of people talking about their Sirius and XM radios causing RFI at that frequency. My bet is that either there's a commonly used switching power supply or a commonly used System-on-a-Chip (SoC) that has its clock at 146.765MHz. FWIW, I actually do also have problems originating from a Sonos:Connect as well. However, it is slightly different in character (which, if it's from the SoC could be indicative of software!). The DC-1's RFI is actually audibly modulated. It hits S9+40, opens the squelch when the radio is set to FM, and buzzes like a sawtooth wave. The Sonos, on the other hand, seems to be centered about 5kHz higher with a much broader (30kHz-ish) spectrum and the noise doesn't demodulate when the radio is set to FM, you just get S9+20 static. The RFI is actually so strong that it wipes out about 2 MHz of spectrum. Once you move off-frequency from the RFI, you quit hearing the buzz and the signal meter drops to 0 but the presence of the super-strong in-band signal causes the radio to completely de-sense and you don't get any reception at all. I've got a local repeater at 146.880 that, on it's own, is about S8. When the DC-1 is on, the radio de-senses to the point that the .880 repeater can't even make the signal strength meter light the first tick. I have to get right to the top or bottom of the 2m allotment before I get "normal" reception back. Even then, I have a feeling that the filters on the RX are so wide I'm probably still suffering from the effects of the RFI. In fact, we've got a 50kW broadcast tower just down the road that doesn't even cause this big of a problem! I've been thinking about soldering together a Faraday cage out of some sort of mesh (to allow airflow for cooling) but I'm not sure what to do about the cabling. Ordinarily, you'd strip the jacket off the cabling and bond its braid to the cage as it passes through, but I can honestly say I really don't want to try that with the 8 coax, XLR, and power cables that run in and out of all this. I don't know enough about E&M to say if just clamping massive ferrites around all the cables as they come out would get the job done. |
|
|
|
Post by garbulky on Sept 30, 2014 7:20:50 GMT -5
Do you think this RFI could cause buzz from an electric guitar?
|
|
hemster
Global Moderator 
Particle Manufacturer
...still listening... still watching
Posts: 51,951
|
Post by hemster on Sept 30, 2014 7:38:39 GMT -5
Do you think this RFI could cause buzz from an electric guitar? I'm no expert but I'd say that is unlikely. |
|
|
|
Post by audiobill on Sept 30, 2014 10:01:05 GMT -5
Do you think this RFI could cause buzz from an electric guitar? I don't think so. |
|
KeithL
Administrator
Posts: 10,269
|
Post by KeithL on Sept 30, 2014 11:10:21 GMT -5
I can tell you why this happens, and give you a few tips.... but no promises. Normal (analog) radio basically operates at a single frequency (the one you select). However, digital circuitry (which includes everything from the digital audio signal itself to switching supplies), and which works by "switching on and off", operates on square waves. As you probably know, an "ideal" square wave is made up of an infinite number of odd-order harmonics. This means, for example, that a 1 mHz square wave consists of 1 mHz, 3 mHz, 5 mHz... 99 mHz... theoretically continuing on forever, with the higher frequency harmonics gradually decreasing in amplitude. Of course, in reality, there is some limit to the highest frequency a given circuit will produce, but with modern digital circuitry that limit could be up in the high gHz range. Most switching supplies operate at a principal frequency of somewhere between a few tens of kHz and a few mHz, but, since they operate on square waves, the harmonics of those switching frequencies can easily reach into the hundreds of mHz (or even higher). Personal computers and such operate with clock frequencies in the high hundreds of mHz (or higher) to begin with, and their harmonics can easily reach into the gHz (which is why computers can interfere with cordless phones). And, in case you were wondering, some modern PC processors consume up to 100 watts - which is all being switched at those frequencies. Also, since many digital devices use assorted multiples of a few "master clocks", there is a good chance that different circuit segments, even those that operate at different speeds, may "hit" the same harmonics. Since most complex digital equipment has a lot of switching going on, at many different frequencies, the result is that noise is radiated at a wide variety of frequencies and multiples. And, since radio reception often involves very low signal levels, which are easily interfered with, it's not unlikely that some multiple of one of them will interfere with a given signal. While there are some limits to how much RF junk a given piece of equipment is allowed to radiate, in reality most consumer equipment is required to "protect itself" from being interfered with. A lot of modern wireless equipment - like phones and WiFi routers - is "spread spectrum" - which simply means that it has the option of using a whole batch of frequencies and "skip around" or decide what ones to use to best avoid the interference that is present... so it can "dodge" most interference, even interference that changes from minute to minute. The DC-1 actually uses a linear power supply, which minimizes power supply emissions, and it has a metal case (some other devices have plastic cases AND switching supplies), and the processing going on is of much lower power than a typical PC, but it will still radiate some noise. Noise will also leak from your source computer or device, and even the USB cable itself. Putting ferrite beads on all the wires involved may indeed help considerably, and shouldn't hurt anything. Since the DC-1 already has a metal case, I'm inclined to think that adding another Faraday cage around it wouldn't make much difference (although solidly grounding the DC-1's case might help); I suspect ferrites on the wires would make more of a difference... Wow, 2 Metre amateur radio, that brings back memories. I remember routers (LinkSys and maybe Cisco) being a problem around that frequency. I also recall something about Sonos also causing problems at 146. I haven't heard of a DAC with that frequency of RFI though. Online, I saw results of people talking about their Sirius and XM radios causing RFI at that frequency. My bet is that either there's a commonly used switching power supply or a commonly used System-on-a-Chip (SoC) that has its clock at 146.765MHz. FWIW, I actually do also have problems originating from a Sonos:Connect as well. However, it is slightly different in character (which, if it's from the SoC could be indicative of software!). The DC-1's RFI is actually audibly modulated. It hits S9+40, opens the squelch when the radio is set to FM, and buzzes like a sawtooth wave. The Sonos, on the other hand, seems to be centered about 5kHz higher with a much broader (30kHz-ish) spectrum and the noise doesn't demodulate when the radio is set to FM, you just get S9+20 static. The RFI is actually so strong that it wipes out about 2 MHz of spectrum. Once you move off-frequency from the RFI, you quit hearing the buzz and the signal meter drops to 0 but the presence of the super-strong in-band signal causes the radio to completely de-sense and you don't get any reception at all. I've got a local repeater at 146.880 that, on it's own, is about S8. When the DC-1 is on, the radio de-senses to the point that the .880 repeater can't even make the signal strength meter light the first tick. I have to get right to the top or bottom of the 2m allotment before I get "normal" reception back. Even then, I have a feeling that the filters on the RX are so wide I'm probably still suffering from the effects of the RFI. In fact, we've got a 50kW broadcast tower just down the road that doesn't even cause this big of a problem! I've been thinking about soldering together a Faraday cage out of some sort of mesh (to allow airflow for cooling) but I'm not sure what to do about the cabling. Ordinarily, you'd strip the jacket off the cabling and bond its braid to the cage as it passes through, but I can honestly say I really don't want to try that with the 8 coax, XLR, and power cables that run in and out of all this. I don't know enough about E&M to say if just clamping massive ferrites around all the cables as they come out would get the job done. |
|
|
|
Post by williaty on Sept 30, 2014 22:10:08 GMT -5
But wait, there's more!
While playing around with this, I noticed something kind of funny. While the ERC-3, XSP-1, and and XPA-1G2 monoblocks don't seem to emit any RFI, the monoblocks, at least, misbehave in the presence of RFI. If you have the ham radio in the 145MHz-ish range and key the mic, the XPAs start clicking. Hold the mic key down and the XPAs will go through their normal startup sequence and come full-on. Let go of the mic key and the XPAs will immediately go back to Standby. Seems the trigger input on them is not well shielded and is sufficiently high impedance that the presence of RF in the vicinity will "fake out" the trigger circuit and turn the amps on.
|
|
|
|
Post by ansat on Sept 30, 2014 22:23:25 GMT -5
I wonder if grounding out the chassis on both units would be enough to prevent the interference in an out. also I highly doubt it Is the trigger itself that is receiving the RFI instead I think it might be the antenna plugged into it.
Tony
|
|
|
|
Post by ansat on Sept 30, 2014 22:28:07 GMT -5
@145 mhz the wavelength is 2m long. I take it you have a 6 ft trigger cable?
|
|
|
|
Post by solidstate on Oct 1, 2014 18:32:02 GMT -5
I'm an Amateur Radio operator and a while back I noticed that I couldn't communicate with a nearby, very powerful repeater. For a while, I just assumed it was broken. Then I realized it was local radio frequency interference (RFI) coming from somewhere near my home. Eventually, I got fed up with it and hunted it down tonight. Turns out, it's my DC-1 DAC causing the problem. The RFI is generated whenever the DC-1 has electrical power, regardless if the DC-1 is "fully" on or just in Standby. The only way to silence the RFI is to physically unplug the DC-1 or to truly power it off using the rocker switch on the back panel. The RFI is centered around 146.765MHz but is at least 20kHz wide. Has anyone else ever had an RFI problem with their DC-1? If so, how'd you fix it? I emailed Emotiva about this as well but if I'm hoping I might get lucky and get a quick answer from you guys. When the unit is in operation and creating this RFI issue I'd try to disconnect the USB or SP/DIF cable from your PC and see if the issue stops. I would then try a TOS/link connection as this will isolate the PC from the DAC. |
|
|
|
Post by solidstate on Oct 1, 2014 18:34:56 GMT -5
@145 mhz the wavelength is 2m long. I take it you have a 6 ft trigger cable? Good point! Using shorter trigger cables might help attenuation. |
|
|
|
Post by williaty on Oct 1, 2014 23:05:56 GMT -5
Yeah, they're the stock trigger cables that come with the devices. I think they are around 6'. Anything that's an integer multiple or integer fraction of 2m will resonate at least some though. Frankly, the getting switched on by the RF from the radio doesn't really worry me too much as I know I can do things like try different length leads or discard the leads and tap the power buttons with my toe when I want to wake the amps up. The DC-1, on the other hand, can't be easily turned on and off like that and the standby state still causes massive RFI. So there's really no convenient way to make it shut up.
I received a reply back from Emotiva's tech support today. It was... frankly very, very disappointing. There's a HUGE difference between the response I got from Emotiva and the response I got from Sonos when I contacted each of them about RFI problems. I feel like Sonos's response shows that they are concerned about the problem and want to make sure that they've done everything they can from their end to help resolve it.
|
|
|
|
Post by garbulky on Oct 2, 2014 0:33:33 GMT -5
Yeah, they're the stock trigger cables that come with the devices. I think they are around 6'. Anything that's an integer multiple or integer fraction of 2m will resonate at least some though. Frankly, the getting switched on by the RF from the radio doesn't really worry me too much as I know I can do things like try different length leads or discard the leads and tap the power buttons with my toe when I want to wake the amps up. The DC-1, on the other hand, can't be easily turned on and off like that and the standby state still causes massive RFI. So there's really no convenient way to make it shut up. I received a reply back from Emotiva's tech support today. It was... frankly very, very disappointing. There's a HUGE difference between the response I got from Emotiva and the response I got from Sonos when I contacted each of them about RFI problems. I feel like Sonos's response shows that they are concerned about the problem and want to make sure that they've done everything they can from their end to help resolve it. Just to make sure it is the DC-1 and not something else that is interacting with it somehow....do try unplugging the trigger wires. Just unplug em see what happens. Also try plugging it into the wall with NO inputs connected or anything else on that circuit. See if that makes any difference. Though this does not fix the problem, the DC-1 can be turned completely off from the switch at the back of it. |
|
KeithL
Administrator
Posts: 10,269
|
Post by KeithL on Oct 2, 2014 9:43:04 GMT -5
Just FYI - anything with a "normal power" processor in it is going to leak at least some RF (as well as lots of other things - like microwave ovens). About the only exception (sort of) would be the processors in some LCD devices (simply because they use such a tiny amount of power that the leakage from it is equally tiny). The frequencies that are leaked are specifically related to the clocks used in the device, and multiples of those clocks. General purpose computers like PCs are the worst in that regard because they have lots of clocks, running at lots of frequencies, many of them switching relatively high power. (In contrast, things like microwave ovens leak a proportionally massive amount of power - but only at one frequency, or a narrow range of frequencies.) Whether this bothers you will therefore depend on whether any of the frequencies used by a particular device happen to be at the same frequency you're interested in.
Since field strength drops off as the cube of the distance (from a pure point source), and somewhere between the square and the cube of the distance for more complex geometries, even a very powerful but distant source will have trouble competing with a much less powerful source that is very close. This means that simply moving things around can often eliminate a problem. Also, a radio receiver itself (the unit) should not be especially sensitive to interference if IT is properly shielded... in which case, moving the antenna further away from the source of interference should yield a major improvement.
In some cases, simply placing a sheet of metal (which could simply be a 12" square of aluminum foil) between a source of interference and the receiving antenna or equipment that is sensitive to it might entirely eliminate the problem. Also, since a significant proportion of the leakage usually occurs via cables that enter and leave the unit, using well shielded cables is a good idea, and putting ferrites on them will probably help a great deal.
Incidentally, many devices like analog audio amplifiers are sensitive to RF because the silicon junctions in many transistors and other components can sometimes "pick up" RF, rectify it, and turn it into audio - which may then interfere with the normal analog signal being intentionally processed by the unit. All manufacturers of equipment do their best to resist this type of interference, but, as they say "nobody's perfect".
|
|
KeithL
Administrator
Posts: 10,269
|
Post by KeithL on Oct 2, 2014 10:58:07 GMT -5
On virtually all of our other equipment, including the XDA-2, most of the operating circuitry is shut down when you're in Standby mode. However, on the DC-1 (which is part of our "professional line"), most of the circuitry remains powered up (and "warmed up"), even in standby. You can, of course, switch the DC-1 fully off using the rear-panel power switch. (When you turn it back on that way, it will come on in Standby mode, but you can then turn it full on using the remote control.) Yeah, they're the stock trigger cables that come with the devices. I think they are around 6'. Anything that's an integer multiple or integer fraction of 2m will resonate at least some though. Frankly, the getting switched on by the RF from the radio doesn't really worry me too much as I know I can do things like try different length leads or discard the leads and tap the power buttons with my toe when I want to wake the amps up. The DC-1, on the other hand, can't be easily turned on and off like that and the standby state still causes massive RFI. So there's really no convenient way to make it shut up. I received a reply back from Emotiva's tech support today. It was... frankly very, very disappointing. There's a HUGE difference between the response I got from Emotiva and the response I got from Sonos when I contacted each of them about RFI problems. I feel like Sonos's response shows that they are concerned about the problem and want to make sure that they've done everything they can from their end to help resolve it. |
|
|
|
Post by garbulky on Oct 2, 2014 14:12:40 GMT -5
Are you by any chance using ethernet over the AC lines?
|
|
