help_outline Skip to main content
audiophile foundation logo 

News / Articles

A Beginners Guide and Other Thoughts on Audio - Chapter 5: Noise Reduction

Roger Modjeski and Anthony Chipelo | Published on 4/17/2024




This month’s Roger’s Corner focuses on a topic very near and dear to Roger. As an audio designer and repair technician Roger certainly had a few pet peeves. However, the one that annoyed him the most was noise. Roger was well known for designing quiet components. His tolerance for noise levels in his designs was well ahead of his peers especially given that Roger’s circuits used vacuum tubes. In this month’s chapter Roger details how to test for noise in your system and provides several suggestions for solving noise issues. I will note in advance that some of these methods go against what some might consider the safe operation of audio equipment in one’s household. However, I will also say that I have been following Roger’s advice on eliminating noise for years and I’m still here to talk about it. So, feel free to experiment based on your own appetite for shock potential. Finally, the chapter closes with some discussion of power conditioners and if you are willing to stash your obsessive-compulsive audiophile personality for a few minutes you might find the discussion somewhat enlightening.

 

A Beginners Guide and Other Thoughts on Audio

Roger A. Modjeski

 

Chapter 5: Noise Reduction

 

To begin, we will divide noise into two different types. There is hum, which in very low frequency is generally related to 60 cycle power. Although if the hum has some harmonics, it will sound like buzz. Hiss is very much like the air escaping from a tire. Sometimes you can get noise that crackles and might sound like a steak sizzling in a fry pan. That is also what some people like to call tube rush if it occurs with a tube component. The best way to find the source of noise in a system is to disconnect everything except the power amplifier that is connected to the speaker. The speaker itself cannot make noise. There cannot be some broken part in the speaker that makes noise. If your speaker makes noise when you are playing music that is a different thing, that is distortion. Here we are talking noise when music is not being played. We are talking about the background noise of your system when for example you push pause on your tape deck, or you cue up the arm on a record.

 

If you are assessing noise in your system, it is not correct to advance the volume control above your normal listening level. Many will set the volume control to the highest level it can go to assess noise. Again, this is not correct. The proper way to assess noise in your system is simply to adjust the volume to your typical listening level or perhaps a bit higher in case there are some situations where you might want to listen louder and then listen to the noise. If you have noise at your listening position that is something that generally can and should be cured. If on the other hand you must walk right up to your speaker and put your ear against it to hear a little bit of noise that generally should not be a problem. Although I will say this, there are systems I have had, and I can get my system to this level, where I can put my ear up to the system and I do not hear anything. It is nice, but not necessarily required. I do not necessarily believe that increases our enjoyment of music in any way. You might even say that a little bit of noise, to some extent, dithers the tweeter, and that dithering may be a good thing. I know it will be hard for some obsessive-compulsive audiophiles to accept but let us not get too worried about noise at that level.

 

We are going to talk about hiss first. If you have too much hiss coming out of your speaker, the first thing you want to do is disconnect the preamplifier and insert shorting plugs at the input of the power amplifier. We must insert shorting plugs because it is not valid to judge a power amplifier with an open input. Several things happen with an open input. In some amplifiers the gain will change. The other problem with an open input is it does not really represent what the preamplifier does to the power amplifier. The preamplifier does provide a low impedance source for the power amplifier to look at. We need to provide that same source. If you knew the output impedance of your preamplifier you could make a shorting plug with that resistance in the plug and then more accurately mimic what the preamplifier does, but generally, this is not necessary. A shorting plug is an easy thing to use. If you do not have one it is very simple to make one. Just take an old cable that you do not care about, strip the ends of it leaving a few inches to work with, then take the hot and ground and twist them together. Now you have made your own shorting plug that works just as well as one you buy.

 

After you insert the shorting plugs and turn on the amplifier, you should then judge if the hiss is significantly lower. It may even be eliminated. The noise level in a good power amplifier should be quite low and be very quiet at the speaker, even to the point of saying you could put your ear up against the speaker and hear nothing. It turns out that in most systems the greatest source of noise is the preamplifier. The power amplifier will just amplify the noise of the preamplifier and that is what you are hearing. However, from a testing perspective we must eliminate the power amplifier because it is the closest component to the speaker. Then we work our way back through the system towards the sources. If it turns out the power amplifier does exhibit unacceptable noise, then have it serviced as power amplifier noise can be measured. For a point of reference, a quiet power amplifier is down to about 100 microvolts (uV) of noise. You can even have up to 1/2 millivolt (mV) of noise before the noise becomes a problem. On the other hand, if you have very sensitive speakers, say a horn speaker with104 dB sensitivity, you may hear 1/2 mV of noise quite clearly.

 

If you have determined that your power amplifier is quiet, insert the preamplifier back into the system and reconnect your cables from the preamplifier to the power amplifier. Do not have sources connected to the preamplifier as we just want the preamplifier by itself plugged into the wall. To start with you should have the volume control turned down and once the preamplifier is powered on you may very likely hear some noise with the volume control in this position. If you want to experience the noise of the preamplifier as you increase the volume control, then you should put shorting plugs into the input that you will be listening to because the source going into the preamplifier also stimulates a very low impedance which will be stimulated by the shorting plug. Although reviewers like John Atkinson often measure a preamplifier at full volume thinking that is the highest noise level, that is not the case. The highest noise level of a preamplifier will be around the 3 o'clock setting of the volume control because that is when the volume control presents its largest impedance. As such, the volume control impedance itself can be a source of noise. However, technically It is not the volume control contributing to the noise, it is the resistance of the volume control contributing to the noise. For now, let us just evaluate the noise with the volume control all the way down.

 

If you have significant noise with the volume control all the way down, you either have a noisy preamplifier - if it is tubed it may just be the tube - or you have too much gain in the system. When I say too much gain that could also mean the gain of the preamplifier and gain of the power amplifier combined is too high for the speakers. In a sense, you could think of the sensitivity of the speaker contributing to the gain in the system. A speaker that is 20 dB more sensitive than some other speaker means that speaker contributes 20 dB more gain to the system. If your system is now very quiet with the volume all the way down it would be worth putting a shorting plug in one of the inputs and advancing the volume control to about the 3 o'clock position and seeing if the noise rises significantly. If it does, I would say that is a bad preamplifier design. Although the noise will increase somewhat as you change the volume setting, it should not really change more than about 6 dB in a well designed preamplifier.


Assuming we have gotten past the preamplifier and our system is still quiet it is time to add a signal source. Of course, if you add a CD player and it is functioning properly the noise output of it should be so low that it does not cause a problem. If you have a noisy digital source, then I would say it is broken. When adding a phono stage as the signal source it is much more important that you know where the volume control is positioned for normal listening. To find the position put on a record and play it until you reach normal listening volume, then just cue up the arm. At this point now you will be listening to the total noise of the system with a phono stage as the signal source. Given the nature of phono stages this is also where you will most likely find the noise source. If this is the case, there is no other solution but to get a higher output cartridge or lower noise phono stage. Nothing else is going to help you because you have already determined that the power amplifier, preamplifier, and digital source are quiet. I should also say at this point that hiss will not be affected by any cable changes or any power cord changes. Do not even bother with that.

 

Now on to the subject of hum. Hum is a more prevalent problem and can come from many different sources. You also have the advantage of being able to cure most hum problems. Whereas a lot of hiss problems will require the offending component to be serviced or replaced. You may also be able to solve the problem by attenuating the signal somewhere in the chain. The best place to attenuate the signal would be between the preamplifier and the power amplifier if the preamplifier is the source of hum. It is very difficult to attenuate between the power amplifier and the speaker because of the loss of power.

 

The largest source of hum in a system is ground loops. If you have a totally balanced system that is one way to avoid ground loops. However, it is not necessary to have a totally balanced system to have a system that is hum free. With careful wiring and proper cabling, you can have an extremely quiet, hum-free system with unbalanced cables. In an unbalanced system the problem you have is that the shield of the interconnect also shares ground with the chassis and the power cords. This is how you get ground loops. To find and correct these ground loops we still go through the same procedure. We disconnect everything except the power amplifier connected to the speaker, we put shorting plugs into the input of the power amplifier, and we listen to see if we have hum. If we have hum at this point you have a defective power amplifier or you just have a power amplifier that is not quiet enough. I would like you to know that there have been manufacturers of power amplifiers, both tube and solid state, who believe that 2 mV of hum is an acceptable level. I will assure you that 2 mV is not an acceptable level of hum for most audiophiles with reasonable speakers. You will hear it. A good power amplifier will have hum down at around a level of about a1/2 mV or even less.

 

If you put in the shorting plugs and there is no hum in the system, then your power amplifier is okay. Now you want to reconnect the preamplifier. We will take out the shorting plugs, connect the interconnects from the preamplifier to the power amplifier, turn the preamplifier on, then plug in the power amplifier and turn it on. Prior to turning the power amplifier on make sure to reduce the volume to zero on the preamplifier. Now listen for any hum. Let us make it clear that your speaker cables cannot cause hum unless for some reason you have a large transformer parked next to them. Your interconnect cables certainly can cause hum. Interconnect cables really should be shielded. An unbalanced interconnect is meant to be shielded. The shield should be quite heavy. The heavier the shield, the better it will connect the interconnect chassis of the preamplifier and power amplifier.

 

At this point if you hear hum the next thing to do, which is very easy, is to simply lift the ground on the preamplifier. Some preamplifiers have a ground lift switch, and you lift the ground that way. Another option is to get a ground lift at the hardware store, which merely has the three prongs coming in and only two prongs going out. These are better known as “cheater plugs”. An even better solution is to take a standard power cord from a computer that you have lying around and take a pair of pliers and just break the ground pin off the end. Now you have an ungrounded power cord. I would suggest you have several of these. When you connect the preamplifier with the ungrounded power cord and the hum goes away, now you know you have a ground loop. The simplest solution is to just leave that power cord on there and leave the preamplifier ungrounded. There is no other way really to get around it. Even if you plug the preamplifier and the power amplifier into the same power strip or power conditioner it will not break the ground between the two chassis.

 

Now if you have lifted the ground on the power cord of the preamplifier and you still have hum, then you have a defective preamplifier or you have a preamplifier that hums and that is just the way it is. Some preamplifiers are just poorly designed that way. However, if the system is still quiet at this point you again will need to reconnect sources one at a time. I would suggest you connect your CD player first. Your CD player may or may not have a ground pin on its power cord. Frankly, it should not. When you connect the CD player, turn the volume up to your normal listening level. If you have hum at this point, then the ground on the CD player should be lifted to see if that reduces or eliminates the hum. At this point the hum will most likely change with the volume setting. The volume setting is now an important part of the test. Again, you need to turn the volume not higher nor lower than your normal listening level. If the system is quiet with the CD player connected, leave it connected and add your next source, which may be a tape recorder, tuner, phono stage, or another digital source. One by one you bring in these extra sources and find the one that gives you a hum problem and you disconnect the ground on the offending component.

 

As a side note, your television is most notorious for causing hum problems. Especially a television that is connected to a cable, because the cable coming into your home and then coming into the television can carry a very strong current. This is very simple to understand. The cables running to your house from the cable company are running right underneath the power lines for many miles. They are inducing an AC current into the television cable, which then must find a way to ground. The simplest way to correct this is to get what is called a cable ground isolator. At this point you really want that isolated. You would rather have that than float the whole system. You can float the whole system, but at this point you can get a shock because many television cables have enough energy on them from the power lines to cause a problem. Some cable installations ground the shield of the antenna cable at your home to a water pipe, but that might not be a sufficient ground. I do advise you get what is called a cable isolator.

 

Regarding cables, as noted previously, in general, you want heavily shielded cables, especially in an unbalanced system, because the heavy shield on a cable will conduct the hum currents better. In addition, it will connect the chassis’ together better than an unshielded or lightly shielded cable. A good, shielded cable has a braided shield that is quite visible, there is a lot of copper there and it does a nice job connecting the chassis together. When you get a hum loop which are the cause of the ground loops we have been talking about, they are generally reduced by having heavily shielded RCA cables. If you wanted to keep the three pin power grounds for your own peace of mind, you could experiment with putting a very heavy gauge wire from chassis to chassis of whichever component causes the hum. Something like a 14 or 15 gauge wire screwed down to the chassis really well in addition to the RCA cable. That might allow you to reduce the hum enough to keep your power ground if that makes you feel better.

 

If safety is a concern, connect your chassis together when you are floating the power cords for additional safety. Having a heavy wire connecting the chassis together to the power amplifier removes the dependency on the RCA cables to conduct a fault current. Remember though that in the early days of Hi Fi all equipment had two wire power cords and there was not any great problem with that. We have been a little bit skewed in our thinking that now everything must have a grounded power cord on it these days. While this does promote consumer protection it also affords manufacturers liability protection. You may find if you open them up, that some preamplifiers and signal sources have the ground pin of the power cord unconnected at the IEC inlet. You may think you have a grounded component when you do not. The reason that manufacturers do that is because they want to stop this ground loop problem before it happens.

 

Other noise issues are caused by wireless transmitters and portable phones. Even a portable phone that is not the main station but the remote station, puts out RF signals. You do not want these close to your audio cables. You should remove or turn them off if you still have some noise problems, especially when you are adding in sources. They will not affect the power amplifier test very much, but once you start adding interconnects, they will possibly introduce noise into the interconnects, and you should turn them off or move them as far away as possible. Some of these devices can cause noise even at a 10 ft. distance.

 

The other noise we want to address is noise coming from the component itself. Previously we have been discussing noise coming from the speaker. If you have noise coming from power amplifier, it is going to make that noise whether the speakers are connected or not. This mechanical noise is typically caused by the power transformer. There is only one solution to this. There is a component called a hum buster and it merely connects a couple of diodes in series with the power cord. You can make one of these yourself that will block DC on the power line. This is because if you have a little bit of DC on the power line, and we often do, this causes saturation in the transformer. The saturation causes the transformer to buzz. An isolation transformer would also correct this problem. Keep in mind some power amplifiers just have buzzy transformers. The larger the transformer the more chance there is of it buzzing. If you have a buzzy transformer, you should talk to the manufacture, and they may be able to replace it with one that is quieter. In many cases the best solution is just to get the power amplifier as far as you can from where you listen. Then you will not hear the buzz.

 

In general, power conditioners will not help your noise problems because the noise is not coming over the power line. Power conditioners usually add some resistance to the circuits because it has small chokes in it to do its job. Chokes are little coils of wire and the little coils of wire have resistance. Your power amplifier does draw a lot of current and then these little coils cause the power amplifier not to be able to draw as much current. In a way you might say they choke it off a bit. Many people have also gone to the point of plugging the power amplifiers directly into the wall, and sometimes using power conditioners on their preamplifiers. Again, unless you have a bad noise problem coming in on your power line, power conditioners are not going to make any difference. I know sometimes audiophiles have this silly notion that if they put a power conditioner in the system, they feel like it creates a darker background. I would have to say this about noise: if you do not hear noise, there is not any noise. You are not going to get any darker background by removing noise that you do not hear.

 

There are also problems in some homes where there is a great deal of voltage variation. I've lived in a home where the voltage varies from as much as 110 V to 125 V. Power amplifiers especially can sound differently with these different voltages because the bias idling current changes. Some power amplifiers are more sensitive to this than others. If you do have a large voltage variation one thing you can do is call the power company to see if they can do something about it. Anything that varies from 110 V to 125 V is really a bit too much. If you cannot get the power company to do anything about it there are a few different things you can buy. There is something called a power regenerator where the AC power comes in, is converted to DC, then there is an oscillator and an amplifier to recreate the AC sine wave at very low distortion and very constant voltage. Getting one of these though that is large enough to run a power amplifier can be expensive, but you may want to use one for your source components. Although the power amplifier may be the thing that would benefit the most from it.

 

A less expensive solution is an automatic voltage regulator which do not regenerate, but merely are called tap switchers. A company called Triplet makes these. Some have a motorized variac. The variac is constantly adjusting itself to change the line voltage and make it correct. You may find one of these helps. In general, well-designed equipment, especially preamplifiers and signal sources, have built in voltage regulation that is quite adequate to the job, very much more accurate than regulation you could add. Adding regulation to regulation does not necessarily make things any better. If something is well regulated, it is well regulated. Also, a good power amplifier will be immune to voltage variations if it is well designed.

 


AF_Logo_white