• WANTED: Happy members who like to discuss audio and other topics related to our interest. Desire to learn and share knowledge of science required. There are many reviews of audio hardware and expert members to help answer your questions. Click here to have your audio equipment measured for free!

Subwoofer integration at both high and low volume

matching a sub with the satellites
Sub drivers with main speaker drivers, either in the same cabinet or external. It that a little clearer. I'm not a HT guy, BUT
I like 60-65" TVs that's for sure. We still concert with LPs and stereo for the most part. The occasional DVD or MC concert.

I don't use satellite speakers. 2.4/2. Two mains, 4 subs and 2 bass columns. Pure analog stereo signal, no processor at all.

I use a DCX2494 but just the OXO no correction, GEQ, or PEQ. Pretty simple really.

Mainly valves for the monitor section and SS for sub/bass duty. I do use tone control for crappy source material, like
my old 78s and a few home band recording on RtR through the years.

Lets put it this way, the wife uses it all the time actually more than me, and she is VERY picky about gear and music. If it's not simple
to use, she in in my ear about it. RtR, Cassette, Records, Streeming, CDs/SACD, FM, MP3s. She uses everything.

Me, I hum a lot and listen to a transistor radio in the shop with the dog, A LOT! LOL

Besides it's time to feed the chickens.

Regards
 
 
Look it's pretty simple

OHM,

I'm afraid there's much for you to learn, and your confident assertions aren't doing anyone any favours.

First, cone loudspeakers are lucky if they exceed a few percent efficiency over a useful bandwidth. NB - sensitivity (expressed as dB@w) is the same as efficiency (%), but written as a different number. It's most convenient to talk in dB.

Next, mis-matched efficiencies can easily be fixed via attenuation or EQ. While I can see that you perceive it as a problem, reality is more complex. The following is a pretty simple/typical example: I'm using a LF driver that's rated [email protected]@8ohm (1w), and when I look at the datasheet, I see that it has some peaks in the kHz range which are pulling the average sensitivity up. In the range I'm interested in, it's more like 90dB@1w. My baffle-step calculator, however, shows that it'll lose around 6dB below 400Hz as it transitions to full-space radiation.
If I'm to choose a tweeter, what should the sensitivity be?


The answer is this: anything over [email protected]. That could even be 81dB@1w, if it's a 4ohm tweeter.


Now, tweeters are often considerably more efficient than that, typically coming in at or above 90dB@1w. Because of their high sensitivity and bandwidth being restricted to the HF region, tweeters generally receive relatively little power. This means that using resistors to attenuate them will result in minimal real power loss: the low-pass section of the crossover is the part that's seeing of the order of 90% of the power from the amplifier. The HF network gets the remaining 10%, and even if we throw away half of that, that's 1w out of 20 that we're actively "throwing away".



Next up, we're going to talk about Hoffman's Iron Law: small size, LF extension, high sensitivity - pick any two.

If you've ever spent time with PA subwoofers, you will know that getting a useful amount of LF extension as well as reasonably-high sensitivity (low-90dB-range, minimum) requires a big box. In a domestic environment, then, matching the sensitivity of the main speakers simply isn't an option. The market generally wants LF extension and small size, resulting in low sensitivty, so large power inputs become mandatory. This is why you'll see lots of 500w+ amps feeding low-sensitivity drivers in small sealed boxes.

If you'd like to read more, here's an excellent resource: https://sound-au.com/articles/index.htm


Chris
 
In short, my system sounds great, but I only get the right subwoofer level at the volume I calibrate at. If I play at a lower volume, the sub disappears, and at higher volumes it overpowers the monitors.

Interesting problem.

You mentioned that you have REW. I'd recommend putting the mic somewhere near your listening position, and then running sweeps at a few different levels, just as a helpful poster did earlier in the thread.

I have a suspicion that the measurements will come back fine, and that there's something weird going on in software.


Chris
 
Remember the loudness contour button on ye olden equipment? Is there a modern DSP equivelant?

The best by far, excellent loudness implementation, with complete adjustability/customisation and progressive in and out effect according to volume I know of, is the one in RME ADI DACs line( both 2 and Pro). Actually, togheter with the great phone amp, it is The reason it is the the centerpiece of my setup.
I don't know of any software that has this fully customisable implementation, everything else I know is just dumb heavyhanded on/off (bar some older amps that had manual level loudness, ask @restorerjohn) . I keep hammering this but just can't understand why playing software like Jriver or Roon hasn't implemented this long ago, it looks easy enough and its a game changer (plus some tilt or bass/treble to adjust on the fly for really bad recordings that it also has and).
Get around is using two or three presets for different level playing with corresponding loudness levels profiles dialed in, but thats a faff..
 
Here is the loudness compensation part of my EqApo setup:

1730730808846.png


I just switch them on and off.
Here is -20 for example:

1730730882383.png
 
The best by far, excellent loudness implementation, with complete adjustability/customisation and progressive in and out effect according to volume I know of, is the one in RME ADI DACs line( both 2 and Pro). Actually, togheter with the great phone amp, it is The reason it is the the centerpiece of my setup.
I don't know of any software that has this fully customisable implementation, everything else I know is just dumb heavyhanded on/off (bar some older amps that had manual level loudness, ask @restorerjohn) . I keep hammering this but just can't understand why playing software like Jriver or Roon hasn't implemented this long ago, it looks easy enough and its a game changer (plus some tilt or bass/treble to adjust on the fly for really bad recordings that it also has and).
Get around is using two or three presets for different level playing with corresponding loudness levels profiles dialed in, but thats a faff..

CamillaDSP has the same customizable loudness implementation as RME -> https://github.com/HEnquist/camilladsp?tab=readme-ov-file#loudness, along with a tilt filter and customizable bass/treble controls.

I agree that it is odd that this feature set is not more common.

Michael
 
here you go
 

Attachments

  • -35.txt
    1.2 MB · Views: 16
  • -30.txt
    1.2 MB · Views: 16
  • -25.txt
    1.3 MB · Views: 19
  • -20.txt
    14.2 KB · Views: 13
  • -15.txt
    19.5 KB · Views: 14
  • -10.txt
    13.7 KB · Views: 15
  • -5.txt
    19 KB · Views: 17
Now that OHM has described his system I have a better idea of how he views the OP problem from a passive analog point of view. I don't agree with him on any number of points he made, especially on using padding resistors and matching nominal impedance drivers.

No disrespect intended to OHN but matching drivers across an analog DIY build driven with tubes isn't really applicable to the OP and his problem using powered near field speakers and a separate powered subwoofer with their built in electronics.
 
I'm afraid there's much for you to learn, and your confident assertions aren't doing anyone any favours.
Well Chris there is no doubt there is a lot for me to learn, but what I have learned is how to builds HiFi speakers for MY home and many others through the years that don't complain about the subject at hand, the OP posed about.

I just happened to focused on the question and then responded to "use a tone control fix" for a design flaw.

I'd doubt you would find a single speaker manufacturer that didn't have his/her/their own little quirks built into their brand.

Speaker building is NOT written in stone any more than analog vs digital is anything more than a preference.

Your now talking about PA speakers, what does that have to do with a HiFi speaker in a home and most equipment without DSP or processors to
semi fix a PROBLEM that wouldn't be there if the speakers drivers were matched.

I match drivers for a reason why one earth wouldn't you? The highs are a little hot (I use a L-Pad and resistor if needed and the mids are robust and still
to be attenuated

If we measure efficiency, that is the reason WHY. To match speaker drivers, it's a piece of data that is used by most speaker manufactures. You are saying
it doesn't matter evidently once you play with a processor and adjust it out of the system.

I'd say you need a lesson is how the cow ate the cabbage. AT LEAST.

I guess you like to fix problems that shouldn't be there by design.

Let make it simple, you guys build speaker the way you like and I'll design speakers the way I have for 45 years. BTW I don't need to use room correction
it's mechanically addressed by design. Helmholtz adjustable resonators. I use 8 tube modules in 2560 square feet 16X20X8 rooms. It's a simple math thing.

No different than matching drivers and STAYING out of trouble vs adjusting your way out of trouble with a processor. I don't have to.

It seems many people have replaced good setup and planning with an afterthought called a DSP processor. LOL DSP the last thing I want in the signal
path if can at all help it. I just have to turn the volume up or down and adjust for the source. Simple tone controls.

Where is Duke, he has built and helped a lot of people, ask him what he thinks and how he builds speaker and select drivers?

Ask Parts Express personnel if they wouldn't match drivers, if they had a choice?

Should mains and subs be matched (within reason) if we have a choice?

Regards
 
I'd suggest swapping components one at a time to see if you can isolate the issue or variable. Can you plug your speakers and subwoofer into another setup in your house like an AVR to see if the same behavior occurs?
Or can you swap in a different set of mains+amp with your sub and dac to see if the same behavior occurs?
 
Well Chris there is no doubt there is a lot for me to learn, but what I have learned is how to builds HiFi speakers for MY home and many others through the years that don't complain about the subject at hand, the OP posed about.

I just happened to focused on the question and then responded to "use a tone control fix" for a design flaw.

I'd doubt you would find a single speaker manufacturer that didn't have his/her/their own little quirks built into their brand.

Speaker building is NOT written in stone any more than analog vs digital is anything more than a preference.

Your now talking about PA speakers, what does that have to do with a HiFi speaker in a home and most equipment without DSP or processors to
semi fix a PROBLEM that wouldn't be there if the speakers drivers were matched.

I match drivers for a reason why one earth wouldn't you? The highs are a little hot (I use a L-Pad and resistor if needed and the mids are robust and still
to be attenuated

If we measure efficiency, that is the reason WHY. To match speaker drivers, it's a piece of data that is used by most speaker manufactures. You are saying
it doesn't matter evidently once you play with a processor and adjust it out of the system.

I'd say you need a lesson is how the cow ate the cabbage. AT LEAST.

I guess you like to fix problems that shouldn't be there by design.

Let make it simple, you guys build speaker the way you like and I'll design speakers the way I have for 45 years. BTW I don't need to use room correction
it's mechanically addressed by design. Helmholtz adjustable resonators. I use 8 tube modules in 2560 square feet 16X20X8 rooms. It's a simple math thing.

No different than matching drivers and STAYING out of trouble vs adjusting your way out of trouble with a processor. I don't have to.

It seems many people have replaced good setup and planning with an afterthought called a DSP processor. LOL DSP the last thing I want in the signal
path if can at all help it. I just have to turn the volume up or down and adjust for the source. Simple tone controls.

Where is Duke, he has built and helped a lot of people, ask him what he thinks and how he builds speaker and select drivers?

Ask Parts Express personnel if they wouldn't match drivers, if they had a choice?

Should mains and subs be matched (within reason) if we have a choice?

Regards
Sorry for repeating it again: you talk about how to designs passive speaker. Here efficiency does matter insofar as the woofer must not have higher sensitivity than mids and tweeters - those can be matched using an L-Pad, as you wrote.

The problem at hand is matching an (presumably, that is on 99% of cases) active sub to active or passive satellites. In this case the efficiency of the sub does not matter at all as long as you can (a) set its volume pot such that the resulting frequency response is flat, and (b) that the sub does not start clipping earlier than the satellites.

If this difference is not clear to you further discussions are fruitless.
 
Last edited:
The problem at hand is matching an (presumably, that is on 99% of cases) active sub to active or passive satellites. In this case the efficiency of the sub does not matter at all as long as you can (a) set its volume pot such that the resulting frequency response is flat, and (b) that the sub does not start clipping earlier than the satellites.
AGAIN we are not talking about satellite speakers or surround systems. I'm talking 101 speaker building from a friggin' book you can buy at most bookstores or at
Parts Express.

I'm not saying you can't make mismatched drivers work, at all. I'm saying it's EASIER to start from a place that requires NO DSP intervention or manipulation
from PEQ which usually has the ability to boost the volume at a given frequency vs GEQ which normally (not always) has no gain. I don't have to use EITHER.

The OP clearly stated he can get it right at one position but not above or below it, without changing things.

That is the question he posed, I didn't. I just gave him a way that most people, I know in the business address the situation from a speaker-building perspective
not how people fix a problem after the fact.

You and I are NOT talking about the same thing. You have NOT given the OP a way other than stored setting for different volumes which is fine if that is how
you want to fix the problem the OP is talking about. I'm not suggesting you run out a buy new equipment or that your fix hasn't been used 1000s of times.
I'm saying there is a BETTER way FOR ME, that I found out about many many years ago. It's the way every speaker builder that I know, designs speakers for
the most part.

Even Danny at GR matches drivers in his OB/servo subs to his main speakers. Wilson, Infinity, VMPS (did), JBL, Polk, Klipsch, Jensen Imperial, Altec, WE, and
dozens of speakers I've never owned. I have owned and do own the speaker I mentioned and built over 100 SETS of speakers in the last 45 years that adhere
to those basic building blocks of good practice.

I'll give you two one thing for sure, you have a fix. If loading a preset bunch of settings is possible for people and they can access those settings from the
seated position bravo, it's just NOT the way I would do it IF I had a choice.

I'm going through a 4-way build right now that all the subs, bass columns, mids, and highs are closely matched and can gain match with the simple use of
L-Pads to turn DOWN the HF and mids for the customer. He has pretty serious hearing loss above 12KHz. I'm making sure he can preset his speakers and
set down and enjoy his speakers without jumping up and down adjusting stuff after were set up for the room and amps he will be using.

You have to set that parameter also with the weight (mass) on the passive radiators for his room and amps. It's not just a given, You have to TUNE every set
of speakers, I build for the room, amps/cables, and individual needs for "A" one remote fix, a volume control. I use putty as the final adjustment for the
lowest octave, then remove putty a thumbnail at a time to remove BOOM. It's the lowest possible mechanical solution for a simple valve gear setup
with ZERO intervention from DSP or any other processor. It's a "set and forget" until there is a little wear on the drivers.
Then you just remove a small amount of putty to accommodate for that spider wear every 2 years or so.

I'll give you this, if the way you fix the problem makes you happy you would be an EASY customer to deal with. I guess I'm not so lucky. LOL I have a few
friends/customers that I've dealt with for 30+ years. I just happen to know what they expect. Horn nuts are even worse, one of the main reasons I stick
with small planar/ribbon builds. It's just my knitch. I did build my own planars for a while. It's just easier to use premade driver off the shelf just in case
I kick the bucket. I prefer to leave the old customers with a fix, not a problem like many custom speaker builders do, Wilson is one off the top of my head.
You pretty much have to use their very expensive drivers if you lose one. I've fixed a few Wilsons, they are expensive to fix. The XOs are usually set in
an epoxy, even worse for the repairman. PITA is a better analogy.

In any case, I wish you the best, I have 36 planars/ribbons (matched) drivers to mount and wire, while I listen to the election results. BTW they were tested
and matched out of over 120 total. You would be surprised how much drivers vary. 6.2 to 8.4 ohms on Neo 10s alone.
Like I said I match drivers top to bottom and left to right for gain. .5% l/r spl match when testing. It's all part of tuning the finished speaker sets correctly.

Regards.
 
OHM, your position is puzzling. Nobody disagrees with you that if you are building a speaker with a passive crossover, it is important to match the sensitivity of the drivers. As you say, it is Speaker Building 101. But that is not the situation here, the OP has clearly stated that he is using Neumann KH-80 monitors (these are powered) and a Kali WS 8.2 sub (also powered). The efficiency of any of the individual drivers in these speakers is of no consequence. All that matters is whether the speakers respond in a linear way to volume input or not.

I am afraid that while you are correct, your observations for the OP's situation is also irrelevant and likely to confuse him. I believe he has received the correct advice in this thread. If he thinks that the gain structure of the sub and speakers is mismatched, the first step is to test it by taking measurements of subwoofer with speaker at different volumes.
 
Not sure exactly why, but it seems most of the problem is eliminated now that I removed the Topping DAC and ran it through a MiniDSP DDRC-24 instead.

dump.png

This is a single point measurement with the mic placed on the desk just to illustrate the general increase in sound level as I increase the volume. It sounds ok both at low and higher volumes now. The average curve around my ears is a bit flatter than this.

Could also be that moving the sub to a better position fixed the worst room modes that were overpowering everything at higher volumes as Sigberg wrote (if I understood his opinion on this correctly).

Next step is to improve the correction a bit, but I am getting some new furniture in the next couple of days, so it would be a waste of time at the momement.
 
Glad you got a handle on your problem. The graphs look great in terms of a nice linear response between the sub and speakers.
 
Back
Top Bottom