• 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!

JTR Noesis 210HT speaker (review by Erin)

Head_Unit

Major Contributor
Forum Donor
Joined
Aug 27, 2018
Messages
1,341
Likes
688
The speaker is designed for very high sensitivity
But it doesn't seem to be achieving that, that's what puzzles me. Not even their own spec (which as you pointed out manufacturers often exaggerate). And while manufacturers each measure differently, I've measured one big-magnet tweeter (no horn) literally a bit above 95 dB, and a matching 5x7* I built also at 95, then bigger PA type woofers somewhat above that. That was with old school 2.0V sine sweep in an actual anechoic chamber, no conversions in software to lead astray. (2V as the speakers were 4 ohm).

So since the JTR are not designed to go super low...I still don't get it. Big woofers, big box, big horn, only 92-93 dB? The only other big thing I see at Erin's is a Danley, whose sensitivity also looks lower in sensitivity than I would expect. Puzzling. :confused:

*(which OK had a ginormous magnet, and was 95 only at higher frequencies since the Q was probably like 0.05 or something :p. But it was still cool/weird enough to get a bunch of business out of Mazda-for a much cheaper version of course.)
 

ctrl

Major Contributor
Forum Donor
Joined
Jan 24, 2020
Messages
1,616
Likes
6,088
Location
.de, DE, DEU
But it doesn't seem to be achieving that, that's what puzzles me. Not even their own spec (which as you pointed out manufacturers often exaggerate). And while manufacturers each measure differently, I've measured one big-magnet tweeter (no horn) literally a bit above 95 dB, and a matching 5x7* I built also at 95, then bigger PA type woofers somewhat above that. ...
So since the JTR are not designed to go super low...I still don't get it. Big woofers, big box, big horn, only 92-93 dB?

As I said, because of completely exaggerated manufacturer specifications (Klipsch can be mentioned as a negative example), people have lost the "feel" for what is realistic in terms of sensitivity values.

But nowadays that's no longer a problem, since one can check what sound pressure levels are possible under ideal conditions with just a few clicks.
Let's take any 10'' woofer that gives us comparable f3 and f6 values as the 210HT in a sealed cabinet (closed box, CB). The frequency response should be linear/flat down to the CB typical second order filter high pass slope.

On an infinite baffle (half-space, 2pi) we would be able to achieve 92dB@1m@2,83V sensitivity with an 8 ohm driver.
1647293763632.png
With two drivers connected in parallel, we get +3dB because of twice the cone area and +3dB because of twice the power (8Ohms --> 4Ohms, same voltage), so 6dB more sound pressure level.

But our loudspeaker does not play in half-space, but in the free field. Therefore, the "baffle step" (transition from the 2pi to the 4pi room) reduces the sound pressure level in the low frequencies by -6dB. Since we aim for a linear frequency response, this limits the possible achievable sensitivity.

This results in a sensitivity of 92dB@1m@2,83V (92dB + 6dB - 6dB) for our CB loudspeaker, under ideal conditions and in free field, with two 8 Ohm 10'' drivers connected in parallel.

The fortune cookie saying, said a million times, always annoying: Nobody can trick physics.
 
Last edited:

Head_Unit

Major Contributor
Forum Donor
Joined
Aug 27, 2018
Messages
1,341
Likes
688
But our loudspeaker does not play in half-space, but in the free field. Therefore, the "baffle step" (transition from the 2pi to the 4pi room) reduces the sound pressure level in the low frequencies by -6dB.
Mmmm maybe this is the smoking gun. Stuff I was measuring in an anechoic chamber was on a half-space baffle. Which phenomena can be seen on various big tours like U2s Songs of Experience, Arcade Fire's last go-round and others, the bass is just not up to par with the rest, it doesn't have the "feel" even out of quad hanging arrays. What I think you're saying is that at the lowest frequencies in free-field it's less, so the horn for example has to be pulled down in sensitivity to make the whole thing flat?
 

ernestcarl

Major Contributor
Joined
Sep 4, 2019
Messages
3,106
Likes
2,313
Location
Canada
@ctrl

How does one convert the sensitivity in dB when different volts is specified? Any easy formula to remember?

I'm looking specifically at Fulcrum Acoustic's RM28 and RM22 where voltage is spec'd differently.

If the RM28's equalized sensitivity is 91dB @ 4.0v, does that mean at 2.0v (twice the voltage/power?) = +3dB ... so 94dB at 2.0v ?

Am I also correct in assuming "free air/space" and "whole space" are the same?
 

azzy_mazzy

Member
Joined
May 31, 2020
Messages
69
Likes
73
@ctrl

How does one convert the sensitivity in dB when different volts is specified? Any easy formula to remember?

I'm looking specifically at Fulcrum Acoustic's RM28 and RM22 where voltage is spec'd differently.

If the RM28's equalized sensitivity is 91dB @ 4.0v, does that mean at 2.0v (twice the voltage/power?) = +3dB ... so 94dB at 2.0v ?

Am I also correct in assuming "free air/space" and "whole space" are the same?
remember Voltage squared divided by resistance equals power

so that RM28 at 2.0V its at quarter the power
 

ernestcarl

Major Contributor
Joined
Sep 4, 2019
Messages
3,106
Likes
2,313
Location
Canada
remember Voltage squared divided by resistance equals power

so that RM28 at 2.0V its at quarter the power

I can sort eyeball voltage together with the watts to get a sense of how the sensitivity in dB rating would change in relation — or not… maybe it’s all just wishful thinking on my end — I’m pretty shit with maths. Sorry, but is there some way to calculate and show what the equivalent difference in relative efficiency between these speakers are in more understandable numbers if they were specifying the same voltage and impedance, for example? Based on the provided data and reasonable assumptions, say, between the RM28 and 210HT. I know, this may sound stupid, but I’m really maths challenged!
 

Chromatischism

Major Contributor
Forum Donor
Joined
Jun 5, 2020
Messages
4,767
Likes
3,706
Also something interesting I found in the JBL SR-X Series User Guide:

1647931601646.png


Note the ** at the bottom implicating different measurement voltage depending on ohm ratings. Maybe this is useful to someone.
 

ernestcarl

Major Contributor
Joined
Sep 4, 2019
Messages
3,106
Likes
2,313
Location
Canada

ernestcarl

Major Contributor
Joined
Sep 4, 2019
Messages
3,106
Likes
2,313
Location
Canada
how much exactly @ 2.83v?


Thinking about this more, perhaps the rated sensitivity is going to be close to ~91dB all the same as already written since all measurements were taken at 1m and at 1 watt with respective impedance into account.
 

gnarly

Addicted to Fun and Learning
Joined
Jun 15, 2021
Messages
992
Likes
1,390
Here's my take on how to compare the sensitivity of the 210HT and RM22....

Looking at Erin's chart I see he shows a mean SPL (in a very relevant range) of 92.5dB @ 2.83V.

Since 2.83V is used for specifying 1W/1m sensitivity at 8 ohms, and the 210HT is listed as 4 ohms, a -3dB adjustment need to be made if I want a 1W/1m spec.
So I'd call the 210HT's 1W sensitivity about 89.5dB, based on it being a 4 ohm nominal box.

Here's the test voltages necessary for common nominal impedances and 1W specs.
2 ohm, 1.4V; 4 ohm, 2.0V; 8 ohm, 2.83V; 16 ohm, 4V.
If manufacturers aren't matching their test voltage to those nominal impedances, dB adjustments need to be made with each step being 3 dB.
(Note that JTR used a correct 2.0V for the specified 4 ohm impedance)




JTR 212HT sens.JPG



Looking at Fulcrum's RM22 charts, on top we see the raw response for both the LF and coax sections, both specified as 8 ohm nominal.
Fulcrum chose 91dB sens for the lower section, and 104 dB sens for the upper section. Both appear realistically well chosen (not inflated).

2nd down is processed response.
Note the lower section's green trace is about the same as the raw, other than raised a bit from 45-100Hz, smoothed, and hpf and lpf added.
It's sens probably is unchanged at 91dB.
But note the coax section's black trace has been brought down throughout it's entire range. Basically, the it's curve and smoothing necessary to sum well below 1000Hz with the lower section, define the overall sensitivity of the entire speaker. This is because the coax has more than enough to match above 1000Hz.
Fulcrum specifies that combined sum sensitivity as 95dB, which again looks very realistic, if not conservative, vs the Axial Processed Response.
Fulcrum RM22.JPG


So I'd say the 210HT is 89-90dB 1W, and the RM22 96dB 1 W. That's a pretty huge difference, with at least 4x the wattage needed for the 210HT.
It's the dang passive xovers in the 210HT....passives always have to bring sensitivity down to the lowest producer........
The RM22's active bi-amping of course avoids that.

I'm a fan of JTR, having owed a pair of 3TX, and still owning a pair of Orbitshifter subs.
That said, Fulcrum's stuff is exemplary imo,.....great boxes, accurate extended specs, great support with processing packages etc...
(Fulcrum put's out DSP presets for almost all major prosound processor/amps. https://www.fulcrum-acoustic.com/support/processor-configurations/ )
I likes what they do !!!!
 

ctrl

Major Contributor
Forum Donor
Joined
Jan 24, 2020
Messages
1,616
Likes
6,088
Location
.de, DE, DEU
What I think you're saying is that at the lowest frequencies in free-field it's less, so the horn for example has to be pulled down in sensitivity to make the whole thing flat?
Exactly, for a full-range loudspeaker with a flat FR the overall sensitivity is determined by the driver with the lowest sensitivity (in free field).

Many loudspeakers with high sensitivity are basically only mid-high units and not full-range speakers. That's okay, if you're aware of it and what it means for the crossover frequency to the subwoofer - for example, no low crossover frequency of 50Hz or 60Hz.

The JTR Noesis 210HT has an f3 of 79Hz. If the speaker is used as a full-range speaker, the specification of 92.5dB sensitivity is not of much use, since the low bass range has to be boosted extremely by EQ.
A 10W tube amplifier is not capable of this, although the sensitivity specification suggests it.

If a loudspeaker like the 210HT is operated as a full-range loudspeaker and equalized accordingly, for example to an f3 of 35Hz (which would be good for a floorstanding loudspeaker), then of course the woofer limits considerably.

1648023907966.png
 
Last edited:

AnalogSteph

Major Contributor
Joined
Nov 6, 2018
Messages
3,338
Likes
3,278
Location
.de
So I'd say the 210HT is 89-90dB 1W, and the RM22 96dB 1 W. That's a pretty huge difference, with at least 4x the wattage needed for the 210HT.
It's the dang passive xovers in the 210HT....passives always have to bring sensitivity down to the lowest producer........
I very much doubt it's the passive XO by itself. I would expect to be losing a dB or two in the LF section (inductor DCR, mostly), maybe 3 dB on a high-order job, but definitely not as much as 6 dB.

The RM22 is also recruiting its coax driver for helping out in the low end (and it has 2x 12" in a generally larger box). Now while 92 dB SPL and 88 dB SPL in parallel may be adding up to about 96 dB SPL, you have to refer to the raw axial sensitivity graph to get an idea of what this actually means in terms of amplifier power needed.
At 300 Hz, we're at 2.83 V for LF and about 9 dB below that for coax, so with pretty much exactly 8 ohms on both drivers that would be about 1.1 W, for 95.5ish dB/W.
At 45 Hz, we'd need +2.5 dB above 2.83 V on LF and +5 dB above 2.83 V on coax. That's about +7 dB above 2.83 V for 96 dB, so about 89 dB/W given that we are close to 8 ohms once again.
At 80 Hz, there's some massive help from bass reflex (another thing the JTR does not have), so power per watt gets up to 10 dB of boost there, peaking at 104ish or so.
 

gnarly

Addicted to Fun and Learning
Joined
Jun 15, 2021
Messages
992
Likes
1,390
I very much doubt it's the passive XO by itself. I would expect to be losing a dB or two in the LF section (inductor DCR, mostly), maybe 3 dB on a high-order job, but definitely not as much as 6 dB.

I didn't mean losses from the passive xover itself. Although that said, on my 3TX's there was a sizeable bank of resistors in the xover section feeding the pair on 10"s. Had to make some padding of some sorts on the 10"s too, i think. (i'm not a passive guru at all ! ...hate em lol)

I meant passives inevitably need to pad all driver sections down to the lowest sensitivity section.


The RM22 is also recruiting its coax driver for helping out in the low end (and it has 2x 12" in a generally larger box). Now while 92 dB SPL and 88 dB SPL in parallel may be adding up to about 96 dB SPL, you have to refer to the raw axial sensitivity graph to get an idea of what this actually means in terms of amplifier power needed.
Yes. I was trying to point out the summation of the coax and 12"s. It can be easily seen in the Axial Processed Response.

I'd look more to processed sensitivity myself, and how it compares to raw, than going back to raw.


At 300 Hz, we're at 2.83 V for LF and about 9 dB below that for coax, so with pretty much exactly 8 ohms on both drivers that would be about 1.1 W, for 95.5ish dB/W.
That brings up an interesting point (to me anyway) about how to assess overall sensitivity of multiway actives.
Fulcrum's footnote for the 95dB processed sensitivity says:

5. Equalized Sensitivity The 1-meter-referenced SPL produced when an EIA-426-B signal is applied to an equalized loudspeaker system, at a level which produces a total power of 1 watt, in sum, to the loudspeaker subsections. (emphasis mine)

I've never noticed the total power of 1 watt before.

I've made many sensitivity measurements of processed multiway sections at some constant voltage, done the math to see the resultant wattage of each section, and summed them up to get a total wattage. Then i do the math to pull that wattage sum down to 1W, to see what efficiency that implies.

Although when i say constant voltage, i really mean time-averaged voltage of full-range pink, run with all section processing in place.
Equated against time averaged SPL (Leq) over the same time period, to get an integrated sensitivity measurement of the entire passband.

Most honest way to assess sensitivity and efficiency i know of. Sorry if OT, just one of my interests in high dynamic DIY speaker building.
 
Last edited:

Jon AA

Senior Member
Forum Donor
Joined
Feb 5, 2020
Messages
465
Likes
905
Location
Seattle Area
I don't think there's any obvious mistake here. Very high efficiency speakers like these often require additional EQ processing
Do you have an explanation for or are you able to point to any research documenting one thing has anything to do with the other?

From a 30,000 ft perspective, efficiency of the speaker is usually limited by the bass section and the potential bounds are limited by Hofmann's Iron Law (make the box bigger, tune it to the same frequency, more efficiency is available if you select the right driver....).

How does that have anything to do with roughness of frequency response--particularly at high frequencies?

I don't mean to nit-pick, but unless there's a fundamental causation/correlation (as opposed to anecdotal examples) that can be shown, it shouldn't be stated that there is or people get the wrong idea as the below example shows.

This just means high effiency speakers are not for me. Im not going to want to use eq.....
 

Chromatischism

Major Contributor
Forum Donor
Joined
Jun 5, 2020
Messages
4,767
Likes
3,706
How does that have anything to do with roughness of frequency response--particularly at high frequencies?
It's probably based on the observation that most high efficiency speakers available today don't have the best frequency responses. Think PA speakers, or speakers with > 100 dB sensitivity. Part of the way they achieve the high efficiency is by not filtering anything out in the crossover network. They also do not attempt low bass extension. Using the best drivers available and utilizing smart speaker design can mitigate (not eliminate) the need for equalization however this increases the price substantially.
 

ernestcarl

Major Contributor
Joined
Sep 4, 2019
Messages
3,106
Likes
2,313
Location
Canada
Do you have an explanation for or are you able to point to any research documenting one thing has anything to do with the other?

I primarily mean very high-efficiency compression drivers where if you look at the unprocessed response, they're often very uneven. A combination of waveguide plus cabinet optimization and especially response equalization is required. Depending on the design restraints/goals, it may be very difficult to get an optimally "flat" response via purely passive means alone. See any of Fulcrum Acoustics' spec sheets, for example.

You can see this in the documented response graphs of such drivers.
 
Last edited:

ernestcarl

Major Contributor
Joined
Sep 4, 2019
Messages
3,106
Likes
2,313
Location
Canada
Just as an example, I think the ff. high-efficiency speaker system may be further improved upon (even if just frequency response-wise) if FIR filters were to be used: 215-DCX

1650501938222.png


1650501945702.png
 

Sancus

Major Contributor
Forum Donor
Joined
Nov 30, 2018
Messages
2,923
Likes
7,616
Location
Canada
I admit I don't fully understand the use of compression drivers in speakers like this that still have relatively weak woofers. My understanding is that a good 1" dome tweeter is capable of 120dB SPL, and midranges at least that much. And that's ignoring any increase from a waveguide.

Considering the downsloping FR of content, it seems to me that most compression drivers are unnecessary for output.

Presumably there is some other advantage but I don't know what it is.
 
Top Bottom