2214h and impedance response for vent length thuning

[davidpou]

Hi,
I have built boxes for my 2214H that are supposed to be tuned at 29.5 Hz
I am now at vent length tuning. For this I am planning to measure impedance response (with a Dayton Audio DATS arriving soon).
If I got it correctly, I am supposed to get two peaks; the middle of the “U” being formed by them indicates FB.
And FB should be as: Fs < FB < 0.383 * Fs / Qts
if the peaks are of equal size FB = Fs, if the first peak is bigger, FB > Fs, and if it s lower FB<Fs.

From an ealier post I have seen that the 2214H impedance response looks like this, and it seems there is only one peak….. :
How then shall I tune my FB ?

Which configuration is the best : equal peaks, first one bigger, ? smaller ?

Thanks

[Mr. Widget]

From an ealier post I have seen that the 2214H impedance response looks like this, and it seems there is only one peak….. :
How then shall I tune my FB ?The impedance response you posted is likely a "free air" response. Since there is no box acting on the woofer only its own natural resonance peak is seen. In a bass reflex enclosure you will definitely see two peaks.


Widget

[Robh3606]

That graph you posted is suspect. Fs should be about 23hz with those woofers The peak is at 60Hz. If that is free air it's not a stock 2214 measurement. Could be a 2214 with an after market kit and no aguaplas although that seems high even for that.

Rob:)

[davidpou]

thanks Mr widget !

I found it here: http://www.audioheritage.org/vbulletin/showthread.php?12431-Yet-another-4425-DIY

[Earl K]

Jervs impedance plots were created within his knock-off 4425 boxes and those might not have been ported ( his graphs stop at 20 hz so it's very hard to know from the graphs if he made a sealed or ported box ).

That picture does show a very gentle rise starting at 35hz going down to 20hz ( meaning, there might actually be two peaks present ) .

LHF member "Dougie" has posted better "raw" impedance traces ( in pdf form ) within the same thread .

:)

PS; I don't think many designers actually adhere to the Altec ( so-called ) rule of holding the twin impedance peaks to the same magnitude.

[davidpou]

PS; I don't think many designers actually adhere to the Altec ( so-called ) rule of holding the twin impedance peaks to the same magnitude.[/QUOTE]

So I will have two peaks, and to tune the vent to the box the middle of the U shape shall be at 29.5 Hz, and so FB > Fs, and if so, the first peak will be bigger ?
I Have red on a US blog that the length found with measurement was between 0.64 and 0.67 of the calculated length....:blink: when the french freeware used to make the maths was saying that fine tuning was more or less a centimeter around the calculated value...

Similarly there was a concordance for the volume of the box and vent diameter / length, between this freeware and WinSD, but when it comes to dimension, I followed the french soft but got the impression that they were inspired by mystic rather than science... For a given volume are they laws to determine the dimensions ?

(see, earl I am working on these issues, so far the low end is a bit muddy, i think, but it is going damn low: ie no need for a sub... I use LE175 while waiting for brand new 2426H and 2371 horn)

[Mr. Widget]

A much better way to measure the actual tuning frequency of a ported woofer is to use a measurement mic at 1 cm from the cone. There will be a null at the tuning frequency. In the graph below (taken from the web) the green trace is a close mic on the woofer and the red trace is the output of the port. This enclosure is tuned to 43 Hz.


Widget

[davidpou]

Ho ! all right ! Measurements shall be made with or without equalization-crossover of the woofer ?
presumably without looking from the graph...but if so equalization will change the null point 's position....:confused:

[davidpou]

Well, I performed some measurements without filtering, whether active or passive.
First, a comparison of the same woofer in both boxes (green my box, red stock JBL 4425).
The extra SPL is not due to the box but to amp tuning. My box seems to go a little bit higher and lower, but also seems to have slightly more bumps….
76640
Second, the stock 4425 box close mesurements (red woofer, right port):
Tuning freq is 64 Hz…
The woofer measurment shows a 58Hz to 880 Hz (after eq ) response, before going down at both ends.
The port measurment is matching in SPL the woofer’s, it is going stiffly down, being a bit chaotic afterwards, and indeed puts the 0dB low freq at 43Hz (annonced 45Hz +/- 3dB, so all good)
76641
Third, my box close measurements (blue woofer, red 50 cm long port, green 35 cm long port):
Tuning freq is einther 48Hz with the long port and 40.5 Hz with the short port, so seems better and I was going to shorten them even more (they are two 7.5 cm diameter) when I realised the port SPL was diminishing with its length…If I want to make it match the SPL level of the woofer’s response, I should then increase the port length…. So what is your take ?
The woofer mesurement shows a 52 Hz to 1 kHz (after eq) response, before going down at both ends, so I guess an improvement over the 4425 box...of 6Hz at the bottom ? slope there, is swetter aswell.
The port measurement, on top of not matching the woofer’s SPL, doesn’t really go stiffly down and seems very chaotic after 220 Hz or so…but it is maintaining a nice response down to 20 Hz….
76642
And I am not even started with damping ! I red a post here with a consensus over rockwool, which I put inside my boxes, but I saw this post showing the impedance response going left with extra damping (10 mm felt, and 2 sheets of Monacor MDM-3) or instead of felt polyester foam same thickness…. I d like to check for myself….

[davidpou]

A much better way to measure the actual tuning frequency of a ported woofer is to use a measurement mic at 1 cm from the cone. There will be a null at the tuning frequency.
Widget

Well, the close measurements did not validate my calculated 29.5 Hz tuning frequency. So I started to doubt the french soft, and tried four different box simulators and none of them gave the same results: Vb, Fb, port length, etc.... none agreed and none were confirmed by measurments as stated above.... I am still waiting for the Dayton Audio DATS V2 ....to see if it confirms the tuning freq as observed by the close FRC...

In the mean time since I had no concordance on my box tuning freq, I decided to tune the vent at 49.5 Hz, just when the woofer starts to go down: by trial and error I finally have a 30 cm vent length, straight, because the radius version (90 °) , allthough having the same length, didn't give a good result.....:
76663

76664

[davidpou]

I should take reading lessons before getting into this.... I understood that the null was at the crossing of the two FRC :banghead:

well, that made me cut a lot of plastic.... but here we are 29.5 Hz.....
76704

How does one know that the calclated Fb is correct ?

[RMC]

Bonjour Davidpou, and other members who replied here.

You seem to be giving yourself a lot of trouble to confirm the calculated box tuning frequency (Fb). The method I use is probably the simpler one and requires no instruments nor a microphone as suggested by Widget. I use Audio test CD by the Editors of Sound and Vision Magazine made in co-operation with Nippon Columbia, copyright 1993. This Test CD is filled with numerous sound system equipment tests, one of which I use to verify my boxes' Fb, is a frequency response sweep from 20hz to 20 khz. The sweep can be started at any point you want by your CD player.

The test tones are announced by a voice one after the other and each one lasts a few seconds (e.g. 40hz, 35hz, 30hz, etc.). While this CD is playing through your audio sytem/speakers and sweeping the frequencies, you barely touch the driver's cone near the surround with a finger (no pressure on cone) just enough to feel the cone's vibrations as the frequencies going lower are being called. The frequency at which there is the least cone movement or none at all IS the boxe's tuning frequency. Voilà ! As simple as that.

Naturally, there isn't a 29.5hz or a 36.3hz frequency called since it would be an endless process... However, all of them go by in the 20hz-20khz range but frequencies are named by the voice in increments of 5hz in the low end, and in your specific case of 29.5hz the 30hz tone would do the trick. I have used this test CD method for many years with the numerous boxes I've made, including with my pair of JBL 2214H, and it works quite well in confirming my calculated ducted port dimensions/Fb.

Now for the million dollar question: if you insist on confirming say a 32.5hz Fb, how do you proceed with such a CD ? Since the frequencies are ALL swept (20hz-20khz) but only called in increments of 5hz in the bass range, by keeping your finger lightly on the cone ALL the time during the sweep from say 35 to 30hz you WILL feel the absence of vibration at one point in between those two frequencies... This is pretty damn close (just a few hz error margin), though there is no readout to indicate the exact number. If you missed the feeling of non-vibration on your finger or if you're not sure, you can repeat the process from your CD player from anywhere you want on the disc and as many times as you want... You'll see after a few passes how to recognize it easily.

Also, If you look at a cone excursion graph for a vented box in a speaker design software, you will SEE what your finger on the cone is expected to FEEL in terms of vibrations: more above Fb, none or little at Fb and more below Fb...

This test CD also has MANY other goodies, like 20 different spot frequency test signals from 4hz up to 20khz... I doubt this CD would still be available for sale, but you may beg, borrow or steal(?) as they say from someone who has it. Moreover, I suppose that in very large markets such as the European Union where you are or the USA, there MUST be something equivalent available for sale for the audio people.

If not, I wish I could share that CD with others here, but there is a copyright on it and I don't want to be the guy nailed to the cross , have my head on the log ready to be axed or being sued for copyright material theft (violation). Maybe Senior members here, whom I respect and enjoy reading, in addition to being much smarter than me (e.g. Widget, Barry, Earl K, Robh, Ian, Giskard, Ruediger, Horn Fanatic, etc.) have an idea on how to go about sharing that good stuff with others here if there is an interest... I have an idea but keeping it to myself for the time being for the reasons mentioned above, plus others may have an even better one ! BTW I do not know if there is a "lock" on this CD to prevent copying it, never tried . Let's just hope Webmaster/Policeman Don McRitchie wouldn't hit my fingers with his large ruler about this offer... If you would rather discuss in private, my e-mail is on my profile page here.

Salutations,

Richard

[edgewound]

Bonjour Davidpou, and other members who replied here.

You seem to be giving yourself a lot of trouble to confirm the calculated box tuning frequency (Fb). The method I use is probably the simpler one and requires no instruments nor a microphone as suggested by Widget. I use Audio test CD by the Editors of Sound and Vision Magazine made in co-operation with Nippon Columbia, copyright 1993. This Test CD is filled with numerous sound system equipment tests, one of which I use to verify my boxes' Fb, is a frequency response sweep from 20hz to 20 khz. The sweep can be started at any point you want by your CD player.

The test tones are announced by a voice one after the other and each one lasts a few seconds (e.g. 40hz, 35hz, 30hz, etc.). While this CD is playing through your audio sytem/speakers and sweeping the frequencies, you barely touch the driver's cone near the surround with a finger (no pressure on cone) just enough to feel the cone's vibrations as the frequencies going lower are being called. The frequency at which there is the least cone movement or none at all IS the boxe's tuning frequency. Voilà ! As simple as that.

Naturally, there isn't a 29.5hz or a 36.3hz frequency called since it would be an endless process... However, all of them go by in the 20hz-20khz range but frequencies are named by the voice in increments of 5hz in the low end, and in your specific case of 29.5hz the 30hz tone would do the trick. I have used this test CD method for many years with the numerous boxes I've made, including with my pair of JBL 2214H, and it works quite well in confirming my calculated ducted port dimensions/Fb.

Now for the million dollar question: if you insist on confirming say a 32.5hz Fb, how do you proceed with such a CD ? Since the frequencies are ALL swept (20hz-20khz) but only called in increments of 5hz in the bass range, by keeping your finger lightly on the cone ALL the time during the sweep from say 35 to 30hz you WILL feel the absence of vibration at one point in between those two frequencies... This is pretty damn close (just a few hz error margin), though there is no readout to indicate the exact number. If you missed the feeling of non-vibration on your finger or if you're not sure, you can repeat the process from your CD player from anywhere you want on the disc and as many times as you want... You'll see after a few passes how to recognize it easily.

Also, If you look at a cone excursion graph for a vented box in a speaker design software, you will SEE what your finger on the cone is expected to FEEL in terms of vibrations: more above Fb, none or little at Fb and more below Fb...

This test CD also has MANY other goodies, like 20 different spot frequency test signals from 4hz up to 20khz... I doubt this CD would still be available for sale, but you may beg, borrow or steal(?) as they say from someone who has it. Moreover, I suppose that in very large markets such as the European Union where you are or the USA, there MUST be something equivalent available for sale for the audio people.

If not, I wish I could share that CD with others here, but there is a copyright on it and I don't want to be the guy nailed to the cross , have my head on the log ready to be axed or being sued for copyright material theft (violation). Maybe Senior members here, whom I respect and enjoy reading, in addition to being much smarter than me (e.g. Widget, Barry, Earl K, Robh, Ian, Giskard, Ruediger, Horn Fanatic, etc.) have an idea on how to go about sharing that good stuff with others here if there is an interest... I have an idea but keeping it to myself for the time being for the reasons mentioned above, plus others may have an even better one ! BTW I do not know if there is a "lock" on this CD to prevent copying it, never tried . Let's just hope Webmaster/Policeman Don McRitchie wouldn't hit my fingers with his large ruler about this offer... If you would rather discuss in private, my e-mail is on my profile page here.

Salutations,

Richard

An easier way is to download a free app to your smartphone called "Signal Gen". It does sweep tones from 20-20KHz in 0.1 increments while displaying the exact frequency.

It's a great gizmo for such applications, as well as finding the resonant frequency of hanging pictures and other breakables.

[RMC]

Hi Ken,

Thanks for the tip. Too bad I don't have a smartphone, just a good old Samsung Galaxy cell phone. I don't even know how to send text messages on it... I guess I'm pretty old fashion but that's OK for my needs. I'm not really into the world of gizmos...

However, in my text where I talk about using the Test CD, Davidpou may replace the CD by the more modern Signal gen. app and the rest of the procedure remains pretty much the same in terms of feeling the cone's vibrations...

Richard

[Lee in Montreal]

Room EQ Wizard is free and allows to sweep all day long with 0,1Hz increments. Plus dozens of other very useful functions. ;-)

[davidpou]

Bonjour Davidpou, and other members who replied here.

You seem to be giving yourself a lot of trouble to confirm the calculated box tuning frequency (Fb). The method I use is probably the simpler one and requires no instruments nor a microphone as suggested by Widget.

Thank you so much ! Indeed I felt like lost in the middle of the sea with just a tiny buoy thrown by Mr Widget...

An other great advantage of this method is that one doesn't have to take the woofer off the box many times in order to tune the vent length: in my case provided it was radius I coudn't use paperboard or something to push forward or backward for thesting: I had to take off the 2214H and this is risky even done carrefully.... So yes excellent !

[davidpou]

Room EQ Wizard is free and allows to sweep all day long with 0,1Hz increments. Plus dozens of other very useful functions. ;-)

Thanks Lee I do use Rew for the measurements you see.
Just did not want to make the small circuit they suggest to measure TS parameters: not shure to do i right , afraid to blow my drivers, coudn't be bothered....
so I bought the Dayton thingy...

[RMC]

Hi Davidpou,

You're right about handling (install and remove) of large or heavy drivers is a risky business even when done carefully. Many years ago in removing a large and heavy driver (not my 2214H) the Philips screwdriver accidentally slipped off my hand... and went right through the surround :banghead: :banghead: :banghead:

It cost me a genuine recone job $$ so now I proceed differently and ALWAYS put the screwdriver away before even thinking about touching the driver...

Richard

[Mannermusic]

An easier way is to download a free app to your smartphone called "Signal Gen". It does sweep tones from 20-20KHz in 0.1 increments while displaying the exact frequency.

It's a great gizmo for such applications, as well as finding the resonant frequency of hanging pictures and other breakables.

Yes. And, even better, do the freq vs voltage (impedance) plot - need a good voltmeter. However, truth be said, I doubt anyone could hear +/- 10% difference. It is a very broad null point. The technique for this is on site in the JBL Enclosure Information Manual.

[jerv]

Jervs impedance plots were created within his knock-off 4425 boxes and those might not have been ported ( his graphs stop at 20 hz so it's very hard to know from the graphs if he made a sealed or ported box ).

That picture does show a very gentle rise starting at 35hz going down to 20hz ( meaning, there might actually be two peaks present ) .



Hi.

This is a long time ago - but this is definitely my measurement. And as Earl says: in my knock-off boxes, which is ported and at about 54 litres internal volume.
And indeed there are two peaks present - you cannot see the other one - as I remember: it is quite damped, and is down at 16-18 Hz.

Reflex tuning is at the impedance minimum between the two peaks: in this case about 35 Hz.