AVS Forum banner
  • Get an exclusive sneak peek into our new project. >>> Click Here
  • Our native mobile app has a new name: Fora Communities. Learn more.

Never Boost a Null = Voodoo Science

2041 Views 22 Replies 12 Participants Last post by  Eddie Horton
I read over and over to never boost a null.


The first reason given is that you lose amp headroom. The other reason is that nulls do not respond to boosts.


After more extensive reading and my own limited experience I find that the first reason often falls apart after even preliminary analysis. In my case I cut a 12 Db peak and then boosted a 7 Db null. How did that result in a loss of headroom? Especially since the peak was at a lower frequency than the null and therefore takes more amp power to produce. If you have no peaks and then boost a null you will lose headroom. This reasons truthfulness will vary from room to room and will be true maybe half the time and therefore is not even reliable to go by as a "rule of thumb."


The second reason is only true in certain rooms and is the result of the sound waves just cancelling each other more violently (somebody elses description) as power is added and therefore not resulting in any more SPL even though a boost has been added. Once again this is only true in some rooms. Also, this may be true of one of a rooms nulls and not true of another null in the same room. Only attempting the boost and then measuring will tell you whether the null will respond to a boost. For instance, the null in my room responded perfectly well to a boost.


I guess the point of this topic is to say that I believe that "never boost a null" is basically just an old wives tale. It is based on only half the story and stifles people even attempting to make what are reasonable and perfectly acceptable improvements.


Another point. I wonder if any of those who are in the "never boost a null" also have a "house curve" set up in their system. Guess what, you have applied a boost to your entire bass range, including any nulls that you have. :eek: (If you follow the guides I have read on the internet you will have set up a very wide filter at a higher frequency that will result in a falloff of SPL from about 30 Hz down to the point of the center of the cut. You then turn up the gain on the sub to match the SPL levels at the point of the crossover. In effect you boosted the entire range of the subwoofer.) Also, if you cut all your peaks and then turn up the gain nob on the sub to rematch your subs level with your main speakers you just boosted a null. Know something else? If you don't own an EQ, have a null in your system response, decide you want some more bass so you turn up the sub-level in the receiver: You just boosted a null. ;)


You may now post your graphs refuting my points that prove beyond a shadow of a doubt that you know how to use Excel. :D
See less See more
Status
Not open for further replies.
1 - 20 of 23 Posts
Quote:
Originally Posted by dae3dae3
I read over and over to never boost a null.

...snip...


The second reason is only true in certain rooms and is the result of the sound waves just cancelling each other more violently (somebody elses description) as power is added and therefore not resulting in any more SPL even though a boost has been added. Once again this is only true in some rooms.
AFAIK, these are the nulls that rule of thumb applies to. You can have frequency response variations that respond to boosts, but then you also have the nulls that you can't boost. If you can boost it, it's not a true null that this rule applies to.


Hehe. So, I guess the rule of thumb is don't boost if it doesn't respond to boost. Make sense?


Regarding running out of amplifier head room - what sub(s) do you have and at what spl do you typically run them? Maybe yours has/have enough amp room to respond to your boosting. If so, great.


I'm no expert, but that's my guess on all of this.
See less See more
There is a difference between a relative low spot in the FR curve, and a true null. If the relative low responds to EQ boosting, then it is not a null. There is nothing wrong with boosting lows and cutting peaks.


In my experience, nulls are pretty easy to spot - they are usually sharply sloped, very deep, and narrow in bandwidth. For those, bass traps are best, or experimenting with subwoofer placement and/or the listening position.


So maybe the best advice is "try to boost a low and if it responds - great; if it doesn't respond, then try other tactics".
I'm not sure how serious you are about this so i will respond with only limited strenth.

If you are refering to a typical null then this is what i know.Lets say for conversation the null is at 40hz and spans about 15hz in either direction.Now you hook up your "BFD"(for example reasons) and you build a filter to compinsate for this null.Theoretically the speaker is pretty flat throughout it's operating range.Now you begin to apply gain to this filter.The response at the seating position changes about +1db for every +15db applyed.You are in a true null and you can jack up the gain until the cows come home with no result to the positve at the seating position.This will have two effects.You will cause most of the amps resourses to produce your new filter gain of (of +15db/+30db) who knows how much.Second, you cause greater excursion at 40hz due to the new filter you created.


Now, if you choose to create a house curve you will also create demands on both your driver and amp.The difference is that you may only be boosting the bottom end +1/+7db.You are right about the effect of this new curve, but it in no way resEmbles the problems created by boosting a nulled frequency.


I think boosting gets a bad rap.There is nothing wrong with boosting with the knowlegde of what is occuring with your driver and amp.Should you have the proper amp in place driving a top grade driver, I say "house curve" away.


I know your thread will get some attention so I will now give the floor to someone who has the numbers to go along with what I tryed to say.


KG
See less See more
We've actuallly discussed this quite a bit here and elsewhere. You can often achieve a very similar if not identical frequency response curve through boosting frequencies as you can from cutting. The resulting EQ curve is of more importance than what filters you used to get there.


There are times when cutting vs. boosting will have different effects, especially if you are not paying attention to a wide enough bandwidth. In the context of this old advice in the title of this thread, what is important to remember is that you should be concious of using lots of EQ to fix response variations, and make sure that you are in fact improving the response. There are some times where you will boost a region 3dB and only see 1dB of change. Before you keep pushing harder, move the mic around and measure some other locations to see what is going on elsewhere in the room.


When I start measuring a system, after I get a first measurement to see what I'm working with, I then move the microphone in each axis (W, D & H). It is important to get some understanding of how fast some of the observed problems are changing with seating position, and what dimensions they may be tied to. From there you can experiment and make sure that the adjustments you make are making subjective improvements, or at least not making anything worse.


As with most things, the first pass and basic adjustments are very simple, resulting in noticable improvements. The dirty nuances lie in trying to take things to the next level as we look at finer details.
See less See more
Actually, so far you all agree with me on my core point. It all depends on the room and the equipment.


My problem is that "never boost a null" is just flat misleading. I have read many posts where people didn't even attempt to fix a problem because they were led to believe that the Earth would fly out of orbit or in lesser cases the house would burn down if they tried to give a little boost. ;) (I submit to the possibility that this may be an exageration.) :eek:
An example. Post # 4 in the following thread.

http://www.avsforum.com/avs-vb/showt...06#post6748606


He doesn't know anything more that that there is a "nasty valley" between 32Hz and 40Hz and tells the OP that the SMS-1 will not be able to fix it. How does he know? He didn't ask how deep it is or how wide or anything else. It may very well be a null that will not respond to EQ but why should one not even try just because some nulls will not respond. Well, everyone knows you can't boost a null. ALL nulls are invulnerable to EQ.


I apologize to the poster for using him as an example but this is exactly what I am talking about. That new user of the SMS-1 is not even going to attempt to fix the dip because in the very infancy of his learning curve he is given advice based on half-truths.
I think that boosts are ok as long as it just a few db (if it is helping). If more than 5-6 db then other solutions should be looked into.
Hi Darren,


I think I understand your overall point, especially when put in context by the other knowledgeable posters.


However, in fairness to Tim I think his advice quoted below is right on target, and not necessarily at variance to your premise.

Quote:
Originally Posted by Tim916
I'm sorry to say that the SMS-1 will not be able to cure your "nasty valley". You will have to move the subwoofer around the room and try to find a location that has the smoothest overall response BEFORE eq. Once you have found a better spot, then you can let the SMS-1 further smooth out the bass. Also, have you tried adjusting the phase on your sub, and the polarity on the SMS-1?
IF you have the luxury of adjusting the placement of subwoofer and listening locations, then it absolutely makes sense to do so BEFORE attempting equalization. That equalization attempt could include attempting to boost "nulls" that are cooperative. ;)


Larry
See less See more
Quote:
Originally Posted by Mark Seaton
You can often achieve a very similar if not identical frequency response curve through boosting frequencies as you can from cutting.
I want to reply to this because I think it could be interpreted in a way I don't think Mark intended. This topic was actually touched upon in another thread, where it was suggested that you could simply "move" the baseline. I don't think that's the case, at least not with the equalizers I'm familiar with. Let's say you have a 12db dip at 25hz, and a 12db peak at 40hz. It has been suggested that you could use only cuts to bring everything down to the level of the 12db dip (so the peak would actually require a 24db cut). The problem with this though, is that it's not as easy as it sounds. Since each cut or peak is actually a single point, with varying degrees of magnitude and slope, to be TRULY flat, you'd have to have an infinite number of EQ points to bring all the other frequencies down to the level of the dip. Of course, to be close enough, you don't need anywhere near "infinite" if you use broad enough slopes, but it still takes a lot more filters, and is much more difficult to do. Since each dip or peak tends to be similar to a bell curve, it's much easier to correct them with a single filter, which also happens to be a bell curve, than it is to try to approximate the same adjustment with numerous inverse (relative to the dip/peak) bell curves. Furthermore, I think it's best to use as few filters as possible. It's easier to understand what's happening with phase.


So while you can get similar results by "moving the base line", they aren't typically the same, and it's much more difficult to do it. And the other question is: why bother? Even if you are successful in moving the baseline, you still end up with the same boosts/cuts relative to each other, even if you didn't technically apply any boosts. If you adjust to a flat line using all cuts, you will simply need more overall gain to play at the same level as you would have with a mix of boosts and cuts.


At least that's my take on it. :)
See less See more
I believe post #4 is correct (at best i'll give him a 98%). The machine has the ability to protect him from the null.


My advice to anyone who thinks they need an automated device is to STOP. Read and build knowledge.Ask questions and learn what it is that the device is meant to do...and do it yourself with a $99 BFD.He could have saved 500.00 and truly understood standing waves,nulls and ect.


Find a smart friend and pick his brain, why do you think I have Ed Mullen as a friend.I'm still wiring my subs out of phase :D .
Quote:
Originally Posted by kgveteran
I believe post #4 is correct (at best i'll give him a 98%).
LOL, I looked up at post #4, and thought you were agreeing with yourself. Then I realized you were talking about another thread. :D
Quote:
Originally Posted by Darin
LOL, I looked up at post #4, and thought you were agreeing with yourself. Then I realized you were talking about another thread. :D


I know, I thought the guy was talking about me at first but I actually looked up the thread.I'm thinking to myself.....My example was at 40hz with 15hz in either direction. :D :D
Quote:
Originally Posted by kgveteran
The machine has the ability to protect him from the null.
Is that why you're not supposed to boost.


:eek: The Null will come out of the closet and get you. :eek:


I just had a flashback of the "Attack of the 50 foot Woman" type movies my dad always used to watch. :)
A null is definately a dip in frequency response, but a dip in frequency response is not necessarily a null.


jonnyozero3 AFAIK is about right...


Think of what a strict null is, the place where the primary wave and reflected wave collide to form an area of destructive interference and will be in a narrow band. Now the distance and loss is not going to be the same between the primary and reflected (and there are more than one reflection) so you won't get perfect cancellation and a 0 dB output, but whatever EQ is applied to one will also be applied to the other.


This link http://www.glenbrook.k12.il.us/gbssc...es/u10l3c.html


shows what happens when waves overlap, frequency will determine the length of the waves and the room dimensions will determine where waves of different freq will interfere. You can move yourself forward or back (or side to side) a bit and change the freq at which the null takes place. Remember the room has more than one boundary and more than one surface for reflections. (uh oh!!)


Check out the excellent Harmon white papers and calculator on the subject:

http://www.harman.com/wp/index.jsp?articleId=131.0
See less See more
Quote:
Originally Posted by dae3dae3
An example. Post # 4 in the following thread.

http://www.avsforum.com/avs-vb/showt...06#post6748606


He doesn't know anything more that that there is a "nasty valley" between 32Hz and 40Hz and tells the OP that the SMS-1 will not be able to fix it. How does he know? He didn't ask how deep it is or how wide or anything else. It may very well be a null that will not respond to EQ but why should one not even try just because some nulls will not respond. Well, everyone knows you can't boost a null. ALL nulls are invulnerable to EQ.


I apologize to the poster for using him as an example but this is exactly what I am talking about. That new user of the SMS-1 is not even going to attempt to fix the dip because in the very infancy of his learning curve he is given advice based on half-truths.
I'll grant you that my reply in the thread you reference may have been a bit hasty, but a "nasty valley" between 32Hz and 40Hz sounds like a null to me. Also, I got the impression that the OP in that thread assumed that the SMS-1 would be able to magically create flat bass in his room no matter where he placed the sub which most of us know is not true.


The SMS-1 is a great tool (I do not own one, but I have a DD-15) that is wonderful for identifying bass peaks and nulls, helping one figure out where to place one's sub(s), and adjusting phase and polarity. Frankly, I consider the EQ to be a bonus feature.


I was simply trying to give the best advice I could based on the information that the OP provided, and I think that advice was sound.
See less See more
Quote:
Originally Posted by Tim916
I'll grant you that my reply in the thread you reference may have been a bit hasty, but a "nasty valley" between 32Hz and 40Hz sounds like a null to me. Also, I got the impression that the OP in that thread assumed that the SMS-1 would be able to magically create flat bass in his room no matter where he placed the sub which most of us know is not true.


The SMS-1 is a great tool (I do not own one, but I have a DD-15) that is wonderful for identifying bass peaks and nulls, helping one figure out where to place one's sub(s), and adjusting phase and polarity. Frankly, I consider the EQ to be a bonus feature.


I was simply trying to give the best advice I could based on the information that the OP provided, and I think that advice was sound.
Hi Tim,


As I posted earlier, I believe your advice was right on target. Even taken out of context from the original poster's question, I still think it makes sense to experiment with subwoofer and listening locations to get as flat a response as possible BEFORE equalizing.


Equalizers have the ability to boost frequencies and following the thrust of Darren's premise it can be helpful to boost a dip. However, if as others have already pointed out, we are really really with a true null, the attempts to boost it will have very little useful effect. There of course is nothing wrong with experimenting to find out, so long as we know when to call it quits. ;)


Larry
See less See more
Quote:
Originally Posted by dae3dae3
Is that why you're not supposed to boost.


:eek: The Null will come out of the closet and get you. :eek:


I just had a flashback of the "Attack of the 50 foot Woman" type movies my dad always used to watch. :)


:confused:
Quote:
Originally Posted by kgveteran
:confused:
That was supposed to be a joke.
1 - 20 of 23 Posts
Status
Not open for further replies.
Top