JBL 2445_46 frq.response

[ivica]

good to have that 2421 measurement for comparison (not likely to be an inherent
measurement problem). I'm happy to do another independent measurement and
it would be good to dust off the test gear, so no problem I will not be using a
complete network though (3133 or otherwise), just a large capacitor.

For sure just large capacitor would be OK, to protect driver from LF. I did my first measurement without network, just in order to be sure that there was no inherent measurement problem.
Your work would be very informative, and any suggestion is well come.

[ivica]

In my experience a notch that narrow and deep is usually due to a "bounce" somewhere. Since you get it at different distances, it is probably not a floor bounce. The obivous question (and you probably already know this) is whether there is a hard surface about 2-4 inches from the mic.

I have measured 2446's mounted on tractrix horns (Klipsch) and CD horns (Electro-voice HP series) and I have not seen that behavior.

Just an oddball possibility, is the seal between the driver and horn nice and snug (is there a gasket)?

BTW, I am a fan of the 2446. It is a great driver, but you need to stay from after-market diaphragms.

-Tom

The mic was mounted about 1.5 m from ceiling and about 1.5m from speaker, no other reflective object near. That was my first "idea" that there was kind of 'reflective cancellation' due which would make about 5cm sound delay ( 3kHz ~ 10cm wave length). yesterday I have done measurements with 2421 (&2307) fixed on the same place as 2445 and there was no notch bear 3kHz, so "ambient" influences is out of.....

I do not know is there any special part which is missed....

[ivica]

Tom,

Thanks for answering that queston . Your diaphragms certainly seem legitimate .

I'm not aware of any known ( official JBL ) restriction against putting a D16R2450 diaphragm into the 2445 body .

For now, I guess I'm out of ideas .

<> cheers

ps ; One last idea, have someone else measure a single unit of yours ( with a 2311 horn ) and see if they can confirm your findings .

As you can seen on this thread, i have done measurements with 2421 with 2307 on the same place as 2445 with 2311, i either case with 2308 lens and with 2421 everything is OK, with expect able response.

[ivica]

Also, inexplicable notches can be caused by a poorly fitted diaphragm .

Heres' one such example .



I've never had any success in getting a proper fit with this one Radian diaphragm .
The pic shows separate attempts on two different Altec driver bodies ( an 802 & a 902 ) .

<> cheers

Some differences are expect-able between the drivers and diaphragms but I have almost the same behavior around 3kHz on either driver.

[ivica]

Sorry I meant to say I also have a pair of 2446's (with the original diaphragms), I could measure one of them also, they are in service so I would have to pull one out to measure it. I believe it will measure the same as the 2445 however.

I am very restless to see your data, even I think that 2446 is much better then 2445.

[grumpy]

graphs as promised, noting that below 10KHz, in a 2445 body, the difference between
diaphragm responses is minimal. Above 10KHz, there -appear- to be some resonances added
(2446 'ribs') to effectively extend the response. I see no deep/wide notch, and see no need
to attach a 2308 lens and remeasure (there is nothing to mitigate). At this time I did not
try a 2446 body to see if the "Coherent Wave" phase plug improves possible summation/
cancellation issues in the 11-15KHz range.

[grumpy]

As the plots of your drivers appear consistent from unit to unit and diaphragm to diaphragm,
I can only think that either both drivers or both diaphragms are self-consistent, and perhaps
precision-machined (just not correctly), non-JBL "copies" ... or JBL units that are somehow
damaged in a similar fashion.

Can you compare some of the specs with other measurements?:

http://www.jblpro.com/pub/components/2446.pdf

http://www.lansingheritage.org/html/...comp/2445h.htm

[ivica]

graphs as promised, noting that below 10KHz, in a 2445 body, the difference between
diaphragm responses is minimal. Above 10KHz, there -appear- to be some resonances added
(2446 'ribs') to effectively extend the response. I see no deep/wide notch, and see no need
to attach a 2308 lens and remeasure (there is nothing to mitigate). At this time I did not
try a 2446 body to see if the "Coherent Wave" phase plug improves possible summation/
cancellation issues in the 11-15KHz range.

What a nice and valuable work.
Everything is under expect-able tolerance. This second pole "resonance" is even expected and explained in the JBL Tehn.Notes Vol.1 no.8 ,
JBL_Durbin Patanet no.4324312 and in Murray_Durbin JASE paper "Three-Dimensional Diaphragm Suspensions for Compression Drivers"-1979-1980.
Unfortunately I can not reach such good results.
Yesterday night I open both 2445 divers, to see if anything can be a reason for such unexpected behaviors, but everything seems to be on the right place. The diaphragm is fixed tightly by 8 screws. When the screws are removed you need some amount of force to take the diaphragm assembly out of the body. The diaphragms are in perfect condition without any visible damages.

Mechanical grave for accepting diaphragm assembly does not allow any "radial" misalignment of the diaphragm. Rotation of the assembly seems that does not give any audible differences.
Phasing plug is fixed firmly. On the plyg there are about (as remember) 6-six (concentric openings). Everything is clean......, so now I have no ANY ideas., but any help is well come . Say: photos of the open 2445 body, your 2446 diaphragm (specially from "bottom side"). What for me is unexpected is relatively short voice coil (former) about 4mm......
Ivica

[grumpy]

I'll take and post some driver and diaphragm pictures this evening, as well as
measuring the voice coil former length... 4mm seems much too short (from memory).

[ivica]

I'll take and post some driver and diaphragm pictures this evening, as well as
measuring the voice coil former length... 4mm seems much too short (from memory).

well, may be measured from the diaphragm itself is 5mm maximum 6mm

[grumpy]

Meanwhile, here are some images that may help:
http://www.speakerrepair.com/mm5/mer...ct_Code=10-257

[4313B]

Yesterday night I open both 2445 divers, to see if anything can be a reason for such unexpected behaviors, but everything seems to be on the right place. The diaphragm is fixed tightly by 8 screws. When the screws are removed you need some amount of force to take the diaphragm assembly out of the body. The diaphragms are in perfect condition without any visible damages.

Mechanical grave for accepting diaphragm assembly does not allow any "radial" misalignment of the diaphragm. Rotation of the assembly seems that does not give any audible differences.

As nice and tight as it all seems it's only a ruse. The eight screws should be tightened while feeding a 350 Hz test tone (4-inch diaphragms, 500 Hz for 1.75-inch diaphragms) into the compression driver. Once the screws are tight enough to mitigate the various resonances a slow sweep from 350 Hz to 20 kHz is made to ensure that there are no resonances at any other frequencies. It can take some time to get a diaphragm properly torqued down.

Post a picture of the diaphragm installed in the core, preferably with the backcap wires still connected to the diaphragm. The wires should stay carefully routed around the outsides of the diaphragm ring when the backcap is closed (and there shouldn't be anything in contact with the diaphragm, for instance, rotted foam).

How does the impedance curve look?

[ivica]

Meanwhile, here are some images that may help:
http://www.speakerrepair.com/mm5/mer...ct_Code=10-257

Yes, the box and diaphragm looks similar to mines, and the prices too

[ivica]

As nice and tight as it all seems it's only a ruse. The eight screws should be tightened while feeding a 350 Hz test tone (4-inch diaphragms, 500 Hz for 1.75-inch diaphragms) into the compression driver. Once the screws are tight enough to mitigate the various resonances a slow sweep from 350 Hz to 20 kHz is made to ensure that there are no resonances at any other frequencies. It can take some time to get a diaphragm properly torqued down.

Post a picture of the diaphragm installed in the core, preferably with the backcap wires still connected to the diaphragm. The wires should stay carefully routed around the outsides of the diaphragm ring when the backcap is closed (and there shouldn't be anything in contact with the diaphragm, for instance, rotted foam).

How does the impedance curve look?

while installing the diaphragm I tried do my best
first cleaning the voice-coil gap in 2445 body. I have done that in the way that is suggested to be done while changing diaphragm on 2405 ( presented on:

http://www.youtube.com/watch?v=POHs1n_PXX4

as:"JBL 2404, 2405, 077 and 075, 2402 Installation Video:" from SoundSpeakerRepair.com
). sweeping tons....and could not recognize any imperfection in tones. Of course the back-cap wires were carefully arranged out of the diaphragm assembly. I will try to make a photo with "half removed back-cap". I wonder how almost the same problem is present to both drivers. If some imperfection in work was present on one driver, there is low level of probability that the same is present to the other, but may be I have make some systematic ERROR , but what ????

[grumpy]

voice coil depth (from underside of diaphragm 'diamond' surround to edge of former)
measured at 4.5-4.6mm, best I could determine.

pics. D16R2450