Something stirs in the Undergrowth

Dedicated to those large boxes at one end of the room
User avatar
IslandPink
Amstrad Tower of Power
Posts: 10041
Joined: Tue May 29, 2007 7:01 pm
Location: Denbigh, N.Wales

#2581 Re: Something stirs in the Undergrowth

Post by IslandPink »

steve s wrote: Thu May 21, 2020 11:32 pm With regard to the horn, with a folded horn the sound would have to still travel the 12 ' to get out of the horn giving the same time delay ?
The problem is the relative time delay for different frequency sounds. When the horn is doing its job, you don't have that. It's like 100Hz waves and 1000Hz waves all come from 12' away.
"Once you find out ... the Circumstances ; then you can go out"
steve s
Shed dweller
Posts: 2829
Joined: Wed May 30, 2007 6:19 pm
Location: east yorks

#2582 Re: Something stirs in the Undergrowth

Post by steve s »

IslandPink wrote: Fri May 22, 2020 1:12 am
steve s wrote: Thu May 21, 2020 11:32 pm With regard to the horn, with a folded horn the sound would have to still travel the 12 ' to get out of the horn giving the same time delay ?
The problem is the relative time delay for different frequency sounds. When the horn is doing its job, you don't have that. It's like 100Hz waves and 1000Hz waves all come from 12' away.
Sorry Mark I was thinking of a back horm where the bass (horn) output has a time delay.. My apologies
The tube manual is quite like a telephone book. The number of it perfect. It is useful to make it possible to speak with a girl. But we can't see her beautiful face from the telephone number
cv
User
Posts: 40
Joined: Mon Jun 08, 2009 7:39 pm

#2583 Re: Something stirs in the Undergrowth

Post by cv »

Here's my very rough interpretation of this business. Not because I'm convinced it's correct but if it is, I've found it helpful for my own intuition.

First aspect:
We know from electronics that if you are driving a purely reactive load with a voltage, the current is 90 degrees out of phase and as a result, no power is transferred, work (in the physics sense) is not done

Second aspect:
If you wave your hand furiously in air, it won't produce much of a breeze.
But if you gently wave an enormous fan, things are obviously very different.

The point here is that direct radiating drivers are usually much smaller than the wavelength they are trying to reproduce, so it's like the first case.
There's a mismatch to the load so little power is actually transferred. The load presented by the fluid is reactive, not resistive.

The horn (or something like a large area ESL given the above comments about wavelength) gradually couples the small driver to the air in a way that the driver effectively looks much larger.
I think of this as enabling the mouth of the horn to act as a new driver of much larger dimensions. This can only work if the mouth is big enough for the lowest frequencies of interest - which we all know, right?

And also only work properly if it is long enough to enable to provide that gradual coupling/transition - which again, we instinctively understand. (The lower the frequency, the longer the wavelength and the less the wavefront "notices" small features like our pathetic little horns that do actually barely fit in the living room.)

When the horn is working properly, the driver sees a more resistive load (with respect to power pumped out as acoustic radiation) as opposed to a reactive one.

So the acoustic radiation follows the input signal more closely with regard to phase, by analogy with the electrical case above.

Figure 6 here may be helpful:
https://www.grc.com/acoustics/an-introd ... theory.pdf

Did that help in the slightest?!
JamesD
Old Hand
Posts: 997
Joined: Tue Oct 09, 2007 4:26 pm
Location: North Yorkshire

#2584 Re: Something stirs in the Undergrowth

Post by JamesD »

Hi Chris,

Nice simple explanation of the situation - thank you for that it helps :-)

Thinking it through it suggests to me that as the wavelength gets to the horn mouth dimensions that a horn gets less and less the right answer as the response drops away and the phase starts to rotate so we do need to look at an alternative loading method for low end reproduction. If that is correct it suggests the quest is for a horn compatible alternative loading method and then how to blend the two together as seamlessly as possible...

This suggests that looking at some form of tapped horn or TWQT style bass loading might have the best chance of working with mid and hf horn loading but from my listening tests I find them to be not ideal for bass but then lf horns aren't my favourite either so maybe they are a good match for one another. I guess these are all back loading systems rather than front loading and that seems to make a difference in sound too..

OK - I've demonstrated that I don't properly understand horns - I'll stick to sealed boxes and OBs, and go for minimal variation in group delay...

James
User avatar
ed
retired
Posts: 5384
Joined: Thu Jun 21, 2007 4:01 pm
Location: yorkshire
Contact:

#2585 Re: Something stirs in the Undergrowth

Post by ed »

cv wrote: Fri May 22, 2020 11:16 am
The horn (or something like a large area ESL given the above comments about wavelength) gradually couples the small driver to the air in a way that the driver effectively looks much larger.
I think of this as enabling the mouth of the horn to act as a new driver of much larger dimensions. This can only work if the mouth is big enough for the lowest frequencies of interest - which we all know, right?

Did that help in the slightest?!
yes, thanks.

Sorry to cherry pick the post but I was warmed by your mention of large area ESL, of which I am still a great fan after nearly 20 years. Not withstanding the capacitance argument I was always swayed by the large air coupling which I thought caused the 'presence' effect.

as you were, back to horns.....
There's nowhere you can be that isn't where you're meant to be
vinylnvalves
Old Hand
Posts: 1074
Joined: Sat Jan 26, 2013 12:08 pm

#2586 Re: Something stirs in the Undergrowth

Post by vinylnvalves »

This got me thinking.... looking back at when I did a direct comparison of OB loaded to Horn loaded for these midrange AMTs I am using. I am not sure whether my measurements agree or contradict what you are saying http://www.audio-talk.co.uk/phpBB3/down ... &mode=view. There seems to be less phase change as the frequency increases in the OB horn compared to being attached just to the baffle.
User avatar
Scottmoose
Needs to get out more
Posts: 1802
Joined: Sat Sep 08, 2007 11:03 am
Contact:

#2587 Re: Something stirs in the Undergrowth

Post by Scottmoose »

Well, a horn is just an expanding pipe which may (or may not) be impedance matched to its QW cutoff frequency. A simplistic way of looking at them is that length determines the nominal lower tuning frequency, bulk or expansion the amount of gain provided, and usually its linearity. It will be impedance matched down to whatever frequency where the wavelength is equal to the circumference of the mouth (more or less). Below that point, if the horn is acoustically longer, it will operate as an expanding resonant pipe.

A point worth keeping in mind about Tom's post is that he was referring to a driver's rising response (aka 'acceleration') and mass controlled (aka 'flat') bandwidths. The transition frequency is taken as 2Fs/Qes (or Qts, with Qe typically being more appropriate in the context of horns). It's usually called Fhm, or the '-3dB mass corner frequency'. This is the upper limit of a moving coil driver's usable horn loading bandwidth, above which in the mass-controlled ('flat') region, per Tom's summary, you start to encounter the nominal phase lag referred to. So short version: if you want the flat phase TD is talking about, you need to use your driver in its nominal rising response region with a horn to acoustically equalise its amplitude response (i.e. boost it so its flat, rather than falling as frequency decreases), and then cross at, or before, that upper limit to the next driver, whatever that might happen to be.

Minor point: don't forget to include speaker wire loop resistance, any passive component series R, connector resistance & the amplifier output impedance as relevant in this, because these all raise the effective Qe (and thereby Qt) of the driver.
Last edited by Scottmoose on Fri May 22, 2020 3:28 pm, edited 1 time in total.
'"That'll do," comes the cry of the perfectionist down the ages.' (James May The Reassembler)
Website www.wodendesign.com
Community sites www.frugal-horn.com & www.frugal-phile.com
User avatar
IslandPink
Amstrad Tower of Power
Posts: 10041
Joined: Tue May 29, 2007 7:01 pm
Location: Denbigh, N.Wales

#2588 Re: Something stirs in the Undergrowth

Post by IslandPink »

Scottmoose wrote: Fri May 22, 2020 3:11 pm Minor point: don't forget to include speaker wire loop resistance, any passive component series R, connector resistance & the amplifier output impedance as relevant in this, because these all raise the effective Qe (and thereby Qt) of the driver.
Yes, good point, and you can ( and should) include that in any Hornresp simulation.

Ed, good point, I would expect ESL's to have low group delay and good dynamic performance in the region where they are operating efficiently.
"Once you find out ... the Circumstances ; then you can go out"
User avatar
IslandPink
Amstrad Tower of Power
Posts: 10041
Joined: Tue May 29, 2007 7:01 pm
Location: Denbigh, N.Wales

#2589 Re: Something stirs in the Undergrowth

Post by IslandPink »

I should just finish off TD's thoughts on the subject, it wasn't quite the end of that topic :

"A proper LF horn can have flat acoustic response AND roughly zero degree
acoustic phase.

For a person more sensitive to "time errors", they will likely find an
over damped system more realistic.
For a person more sensitive to "amplitude errors" the traditional "flat
response" system will be more satisfying.
For the person lucky enough to have heard a proper lf horn system, you
have heard that one can have "lightning fast" sounding bass and still
make your pant legs flap.

A normal "flat" response point source speaker HAS this kind of delay
built right in and it is unavoidable (currently).
All  conventionally driven point source speakers MUST have the phase
shift / delay if they are to have flat frequency response (dictated by
the falling velocity, acceleration controlled response needed to offset
the changing radiation resistance with frequency).

Additional reactances can alter this phase relationship.
For example above midband at the point in the impedance called Rmin, the
electrical series "L" is equal but opposite the reflected moving mass
(capacitive reactance) of the driver thus cancelling each out and being
resistive (no phase shift).
Above that frequency, the series Inductance dominates and produces a
roll off with an inductive reactance.
At the bottom end of the response, for a simple sealed box, there is a
point where the parallel spring or "compliance" of the box and driver
are equal but opposite the moving mass and this point is also resistive
(at box resonance). Below that frequency, the spring constant dominates
(an inductive reactance) and the acoustic phase leads"
"Once you find out ... the Circumstances ; then you can go out"
JamesD
Old Hand
Posts: 997
Joined: Tue Oct 09, 2007 4:26 pm
Location: North Yorkshire

#2590 Re: Something stirs in the Undergrowth

Post by JamesD »

Regarding OB versus Horn Phase response - for constant group delay I would expect a linear slope of increasing phase as a function of frequency - to me looking for constant phase response is "wrong" as it implies decreasing group delay with frequency...

Maybe I'm looking for the wrong thing but I find varying group delay more of an issue and particularly rapid changes in group delay as can happen in the transition region between drivers...

To elaborate a little - through a transition region region referring to constant phase can be a useful way of discussing the crossover region as the change in frequency response is relatively small but as applied across the whole frequency range it becomes less useful at least to me...

J.
chris661
Shed dweller
Posts: 2556
Joined: Mon Feb 22, 2010 7:29 am
Location: Sheffield

#2591 Re: Something stirs in the Undergrowth

Post by chris661 »

Interesting stuff. Thanks for digging that out, Mark, and thanks to Scott for the additional clarity.

The more I read, the more I'm convinced that FIR is the only way to go for sensible-sized speakers with minimal phase shifts. The huge multi-way horn systems have their place, but that place is not in my living room.

FWIW, I've done some tinkering with FIR stuff again. To reiterate, all the software is free and easily runs on a 5-year-old i5 laptop. The laptop is connected to the amplifier's built-in DAC via USB.

Here's what I did:
- Measure in REW
- Export to RePhase
- Import impulse correction into EQ APO
- Re-measure with REW

Aside from the struggle of getting REW to route signal through EQ APO (ie, show the differences on the graphs), it went pretty smoothly. The before/after graphs are attached.

A few notes:
- The mic isn't a measurement mic. It's a Beyer MC930, which is a cardioid instrument mic. The bump around 12kHz is due to the mic, and the LF rolloff might not be accurate (with directional mics, the LF response depends on distance). The advantage is that the directional characteristic means I'm excluding some of the room reflections.
- The more FIR taps you use, the lower the frequency you can effectively manipulate. There is an overall delay penalty, though: the delay is proportional to the number of taps you use, and inversely proportional to the sample rate. So, if you want to fix low-frequency stuff, you need a lot of taps = long delay. You can offset that, though, by using high sample rates. I've gone for 192kHz on mine (as high as the DAC supports), which has meant the 1000 taps only has a delay of 2.6ms - negligible, and means I don't need to mess around with lip-sync settings when it's time for a movie.
- The speakers are 2-way boxes, with an 8" midbass and a HF horn. Passive rossover around 900Hz. I haven't added anything extra to the system - it's still a laptop feeding into an amplifier feeding a pair of passive speakers - but now I have FIR processing.

Sound-wise, the improvement is subtle, but I don't think I'm particularly sensitive to phase shifts. I can say that there's a certain clarity that I like, which I'm not sure was there before. Backing vocals in Fleetwood Mac's Dreams, for instance, are obviously distinct voices, rather than the layer of background mush that lesser systems render them as.

Chris

PS - If anyone's seriously interested in this stuff, I'll make a how-to video.
Attachments
Before FIR.png
After FIR.png
User avatar
Scottmoose
Needs to get out more
Posts: 1802
Joined: Sat Sep 08, 2007 11:03 am
Contact:

#2592 Re: Something stirs in the Undergrowth

Post by Scottmoose »

JamesD wrote: Fri May 22, 2020 5:30 pm Regarding OB versus Horn Phase response - for constant group delay I would expect a linear slope of increasing phase as a function of frequency - to me looking for constant phase response is "wrong" as it implies decreasing group delay with frequency...

Maybe I'm looking for the wrong thing but I find varying group delay more of an issue and particularly rapid changes in group delay as can happen in the transition region between drivers...

To elaborate a little - through a transition region region referring to constant phase can be a useful way of discussing the crossover region as the change in frequency response is relatively small but as applied across the whole frequency range it becomes less useful at least to me...
And muggins here. Presumably I'm one of the deaf (I accept we're all different) but linear-phase systems I can personally take or leave -to be honest, I've never heard much / any benefit from it, and a gradual / progressive shift is not something I really even think about on that basis. No problem with them, so long as they're done well. Agreed about rapid rotations around the XO & the GD too -that can / may be more of an issue, although if possible, I tend to target XO frequencies where our hearing acuity runs into some issues.
'"That'll do," comes the cry of the perfectionist down the ages.' (James May The Reassembler)
Website www.wodendesign.com
Community sites www.frugal-horn.com & www.frugal-phile.com
User avatar
IslandPink
Amstrad Tower of Power
Posts: 10041
Joined: Tue May 29, 2007 7:01 pm
Location: Denbigh, N.Wales

#2593 Re: Something stirs in the Undergrowth

Post by IslandPink »

JamesD wrote: Fri May 22, 2020 5:30 pm Regarding OB versus Horn Phase response - for constant group delay I would expect a linear slope of increasing phase as a function of frequency - to me looking for constant phase response is "wrong" as it implies decreasing group delay with frequency...

Maybe I'm looking for the wrong thing but I find varying group delay more of an issue and particularly rapid changes in group delay as can happen in the transition region between drivers...
Well as a starting point, here's Hornresp's analysis for a 285GMF on an 85L sealed box - which is just a calculation via MH audio, for a Qtc of 0.56, based on the new 2019 T/S parameters that Supravox have on their website - slightly different to the ones we know and love .
285GMF_84L_i.JPG
285GMF_84L_ii.JPG
"Once you find out ... the Circumstances ; then you can go out"
User avatar
IslandPink
Amstrad Tower of Power
Posts: 10041
Joined: Tue May 29, 2007 7:01 pm
Location: Denbigh, N.Wales

#2594 Re: Something stirs in the Undergrowth

Post by IslandPink »

steve s wrote: Thu May 21, 2020 11:32 pm 12 '
At 25Hz , yes, 1/4 wave ie. 90° phase lag would be 11.3 feet to be pedantic - was that the calc ?
"Once you find out ... the Circumstances ; then you can go out"
steve s
Shed dweller
Posts: 2829
Joined: Wed May 30, 2007 6:19 pm
Location: east yorks

#2595 Re: Something stirs in the Undergrowth

Post by steve s »

IslandPink wrote: Fri May 22, 2020 7:36 pm
steve s wrote: Thu May 21, 2020 11:32 pm 12 '
At 25Hz , yes, 1/4 wave ie. 90° phase lag would be 11.3 feet to be pedantic - was that the calc ?
Yea.... Got there...
The tube manual is quite like a telephone book. The number of it perfect. It is useful to make it possible to speak with a girl. But we can't see her beautiful face from the telephone number
Post Reply