Quote:
Originally Posted by
mogorfÂ
Quote:
Originally Posted by
kbarnes701Â
Feri, it's possible that mains hum may find its way into the graphs via the laptop is it not? Â In which case it won't be audible in any circumstances. For this reason I always make my measurements with the laptop on battery power (not that I know of any mains-induced 50Hz hum here, but it pays to be cautious).
Hi Keith,
The hum actually finds its way into the graph via the test mic that picks up SPL of the sub in case affected by mains hum. Of course troubleshooting can start by swithching laptop to battery, but usually that doesn't resolve the problem. Each and every severly audible hum problem is different, in case user can not solve it the power companies are willing to come to your house and the technician can make a thorough investigation and can get rid of the source.
Although my major is in telecommunications, not power electricity, but some general considerations I'll try to describe. First off, hum is always caused by a "ground loop". Since obviously one ground can not make up a loop it always needs at least two separate grounds in a household. In the beginning I also had serious (read: crazy loud) hum in my sub with my old Toshiba laptop (non-HDMI) when I connected it to the Denon AVR via stereo headphone jack. Also started out with laptop on battery, but to no avail.
Then I found the second ground in my home which was an independent ground from the one the electricity company provides. Guess where this is? In the cable TV box. Yeap, the ground of the cable TV incoming coax cable was on a different potencial than the ground of the electricity company. From there the solution in my case was very easy. I have a Belkin power strip with surge protector and it also came with a plastic imitation plug with plastic prongs but metal ground terminals and coax-in/ coax-out sockets. Installing it into one of the power sockets galvanically connects the power company ground to the cable TV ground. Result? Crazy hum went away.
The installed imitation plug looks like this:
Feri has brought up some good points and a few great questions.
Here are some tips to figuring out "if" you have a ground loop/actual electrical hum, discovering where it resides, and what to do about it/how to fix it.
First, if you see evidence of 60 cycle hum (60hz in US and 50hz in UK) in the waterfall, you first need to verify it actually does exist and isn't an anomaly from your test equipment.
This is just one of many good reasons to always take several measurements "WITHOUT moving the mic." REW will let you average all those measurements together for the FR graph which is good to do as it will get rid of a lot of comb filter "noise" as I like to call it and show the true response allowing you to see what is really comb filtering vs noise or anomalies.
REW doesn't let you "average" Cumulative Spectral Decay plots (Waterfall) though, so you simply need to compare 4-6 different measurements, (without moving the mic) and see if you have the hum issue apparent in all, or only some of the measurements.
If it's only in some, then you can most likely ignore it and it most likely is coming from the mic or the computer, especially if you don't have any regularly audible noise/hum in your system, you can simply discard this as measurement artifacts and only use/view the cleanest looking measurements while deleting the others.
If, however, the artifacts show up consistently in each measurement, you need to determine if this is a mode or actual noise.
The best way to do this is to look at level and decay. It's noise if it doesn't seem to decay. If you alter the window length out to 1 second or more (1000ms) and the noise at 60hz (50hz UK) doesn't seem to dissipate over time, then this is definitely evidence of noise on the system, but not "necessarily" an actual ground loop. (If you don't hear any hum in your system when the volume is turned up all the way with no signal going to the speakers, then simply ignore this part of your waterfall.)
Another way to tell if this is in fact just noise and not a mode is to look at the level, starting at around 10db above where the anomaly begins and verify the lines that make up the slices in the waterfall are straight/even and flow smoothly just as it does from around 55hz - 65hz (adjust for UK) and if they are smooth and without peaks, this is noise.
If, however, the line slices are peaky around 60hz, plus the "mountains" actually eventually decay into the noise floor over time, and this is present in 100% of the measurements, you have an actual room mode.
It is also possible you could have a room mode at 60hz AND noise, with or without an audible hum.
So what do you do to fix this?
Well, again, if you take several measurements and a few of them don't show this noise at 60hz and you don't have any audible problems, simply ignore it and use the prettiest measurements.
If all the measurements have the noise yet you do not have any audible issues and have also verified you don't have a mode, you "can" ignore it or you can try to clean up your power a bit.
If you do have noise/hum, then follow the below suggestions, IN ORDER to isolate and get rid of the problem.
If you hear hum coming from your speakers and/or subwoofer(s) you aren't going to have a very good dynamic range and will constantly be annoyed. Even if you don't hear it directly once you listen to actual program material, it can absolutely destroy the dynamics of what you're listening to and make taking accurate measurements very difficult.
The first step is to find out what is causing the hum.
9 times out of 10 it will be the CATV or Satellite TV box. It is NOT good enough to simply unplug this device. You must actually UNSCREW all of the RG-6 Coax cables from the box and unplug it.
If this fixes the problem, it's because the Cable/Satellite company used their own ground instead of the house ground. It is LAW in the USA that they must come and fix this without charge if this is the case. It's a good idea for everyone to check this and even if you aren't having problems now, make the call while it's fresh in your mind and so you don't start having problems down the road.
if this doesn't fix the problem, leave the cable box and cables disconnected then start unplugging each peripheral device, one at a time (so when the hum goes away you know what caused it and obviously save the AVR/external amps for last
) until the hum goes away.
It's possible you'll find that the hum is generated not by the devices in your rack but your powered speakers, subwoofer, or other household appliances.
For the cleanest power with no chance of a hum, there is one simple way to take care of it.
Plug EVERY device, subwoofer, speaker, etc into a dedicated circuit and also into the same outlet.
Due to the way devices are built and capacitors storing energy, even though you'll see amps and other power hungry devices individually rated at up to a few thousand watts (or more) and when you add up all those watts you may think I'm crazy for recommending you plug, for example, 4500 "watts" worth of equipment into a standard 1800 watt outlet, please understand that you will never be pulling "IE: 4500 watts" at one time and that those are peak levels.
It is important to have a dedicated circuit for your HT, though. If you don't have one specifically for the HT, you can make it "dedicated" by simply not plugging anything else, not related to your HT, into that circuit and keeping everything plugged in via power conditioners and power strips to one outlet.
That is sure to fix your problem.
It is okay to temporarily use "cheater plugs" (3 prong grounded to 2 prong ungrounded adapters) until the CATV/Satellite company comes to fix your problem or until you can get more heavy duty extension cords to plug everything into one outlet but this is NOT safe at all and I wholeheartedly recommend this for diagnostics purposes or temporary use only. BTW, a chain of dollar stores called Dollar Tree based here in the US sells 3 of these adapters for $1.00!
Once you have your power cleaned up, everything on the same circuit, no other household items on your "dedicated" circuit, preferably in the same outlet (unless you can afford an electrician and multiple truly dedicated circuits which is best of course, for the most headroom and possible future use/expansion) it is time to take another set of measurements.
The next measurement you should take is with the mic plugged in and in the MLP but with the AMPS/Speakers OFF which will give you a more specific noise floor reading, letting you know which frequencies are worse (which will help in determining how hard it will be and what materials to use to help quiet the ambient noise in the room for a lower noise floor and better dynamic range to your system).
Finally, take a measurement without the mic to see what the electrical noise floor is. This should be as quiet as possible and under 15db. Once you've taken the above steps you shouldn't see any issues at 60hz (50hz UK) and you will only be looking at the Waterfall for this. This tells you the ultimate electrical noise floor for your system.
TIP: IF you decide to have an electrician add dedicated circuits, understand that you can STILL have noise/hum and even add more issues than you began with if he/she doesn't install your circuits on the same Phase. Be SURE to instruct the electrician to install all new circuits for your HT on the same phase. An easy way to make sure you get what you ask for is to ask that each new circuit is installed in your panel separated by one blank spot. So if you have a circuit, then a blank, then another circuit in your breaker box, the two new circuits will be on the same phase. Just make sure there is a blank (no circuit installed) between each new circuit for however many you are adding and they should all be on the same phase and you will enjoy a very quiet noise floor with no issues.
Hope this helps some folks...
--J
Should the reason for hum be misunderstood by some doesn't mean it is neither there, nor is it negligble.
On a second note, this questions how accurate the frecuency scale of REW is. Any thoughs welcome.
