JBL 4345 Wisdom wanted

[NavnFugl]

Hi,

I've acquired a set of replica JBL 4345 made with no Internal crossover, ment to be used with a 4 way active crossover. 1 amp for each driver.

I'm using the Deqx Pre-8 as my active crossover.

The sound is good, but I feel like something is off.
I've tried to read the forums to see, if there's something I should watch out for in my configuration. Some posts have mentioned something about the polarity of the JBL 2245h driver, is it different from the rest of the drivers? If so, how do I flip it to be the same?

Furthermore, on the Deqx Pre-8 I've set the crossover point for the 2245h at 290hz. Once all calibration and such is done, the Deqx shows me a massive dip on the frequency curve at 290hz.

Any wisdom or tips to getting this unusual set of 4345 running at their best is much appreciated.

/Navnfugl

 

[hjames]

Exactly which drivers are in your replicas?
Here is a view of the pair of 4345 clones Heritage member Giskard built ...

4344 Mk II network

System information. Pages 3, 4 and 5 are blank.

www.audioheritage.org

Looking at that you can see the original woofers are out of phase and which drivers are in phase to get the sharper cutoff between them.
You may find some other interesting info in that 11 page thread ...

2245H
2122H
2445J/2311/L94 (aquaplased)
2405
Charge-coupled, bi-amp only 3144/3145 equivalent networks

 

[BMWCCA]

I run 4345 clones with cabinets built by Kenrick's US cabinet maker and partner "Rick". I'm also running a bi-amp system usig CC networks from the Giskard/Timbers schematic Heather linked to, but only the HF three-way section. I'm using an Ashly XR1001 set at roughly 290Hz fed directly to the woofer. Crown PS-200 and PS-400 are the amps used.

 

[BMWCCA]

Not sure if this was among Heather's links provided in this thread, but this is from a compilation she made some time ago you will likely find helpful, too.
https://www.audioheritage.org/vbulletin/showthread.php?16967-4345-References-Thread

 

[NavnFugl]

Exactly which drivers are in your replicas?
Here is a view of the pair of 4345 clones Heritage member Giskard built ...

4344 Mk II network

System information. Pages 3, 4 and 5 are blank.

www.audioheritage.org

Looking at that you can see the original woofers are out of phase and which drivers are in phase to get the sharper cutoff between them.
You may find some other interesting info in that 11 page thread ...

2245H
2122H
2445J/2311/L94 (aquaplased)
2405
Charge-coupled, bi-amp only 3144/3145 equivalent networks

Mine has the following drivers:
077
LE 85
2122h
2245h

I'm not too clever to understand the information in that link you've sent.
Since my replicas have no Internal crossover, the driver is connected directly to its amp, how do I know if one driver is out of phase with the others? And what's the best/easiest way to solve it?

Thank you.

/Navnfugl

 

[malden]

these have to be the nicest looking JBL studio monitors I have seen, the black wood-grain baffle is a major improvement over the standard JBL blue, in my not so humble opinion ...

 

[hjames]

Mine has the following drivers:
077
LE 85
2122h
2245h

I'm not too clever to understand the information in that link you've sent.
Since my replicas have no Internal crossover, the driver is connected directly to its amp, how do I know if one driver is out of phase with the others? And what's the best/easiest way to solve it?

Thank you.

/Navnfugl

Basically, in the JBL design the Woofer is in reverse polarity to the other 3 drivers, so you get a tighter crossing between the 2245 woofer and the 2122 Mid-bass driver (usually around 290Hz).
Since you have 4 drivers in your box, can we assume there are 4 pairs of Banana jacks on the back of your cabinets? Are the pairs marked or color coded in any way?
We gather you have 4 sets of speaker wires going to each of the 4 amps you are using?

It's anyone's guess what the builder of your systems did when they wired the drivers in the cabinets. The best way to be sure how they are wired is to open the cabinets
and track the wires from each driver to whatever connection is on the back of the cabinets. (That usually means carefully pulling the woofers to see inside).

If it were me, I'd have the Pos (Red) lead of each driver go to the appropriate RED (pos) jack on the back of the cabinet, so I could play with polarity of the wires to each amp ...

 

[toddalin]

There are two most common reasons for the "massive" dip.

If the drivers are out of phase in the area of the crossover frequency, their signals can cancel each other out creating a dip.

The other reason is that the sound from the 18" is bouncing off the floor and when it arrives at the listening position it is out of phase with itself, again creating a dip.

 

[Mr Boochie]

The other reason is that the sound from the 18" is bouncing off the floor and when it arrives at the listening position it is out of phase with itself, again creating a dip.

I guess testing and modifying the room's acoustics would help.

 

[toddalin]

To determine if it is a phase issue between the two drivers, it is simple enough to disconnect the mid and see if the dip goes away. It probably won't entirely go away because if the mid is in phase, it should be helping to raise this dip up to near flat.

I have a big dip in the same area with my JBL 18"s too (about 10 dB IIRC), but I attribute a good portion to floor bounce cancellation. In my case, I think there could also be some cabinet resonance/internal standing waves at play.

 

[BMWCCA]

these have to be the nicest looking JBL studio monitors I have seen, the black wood-grain baffle is a major improvement over the standard JBL blue, in my not so humble opinion ...


View attachment 3378373

I'm fond of that look, too, even if I can't take any of the credit for them.
Rick apparently made four pairs of these with the black ash baffle. The first was last known to be complete but awaiting drivers to be returned from re-coning as of 2008! The second one was my pair, a third one of which went to France. The fourth he prepared and had not yet assembled for himself, but that was quite some time ago. He did comment that the black baffle was a real pain to accomplish and he had no wish to produce any more of them!

 

[Mr Boochie]

I'm fond of that look, too, even if I can't take any of the credit for them.
Rick apparently made two pairs of these with the black ash baffle, one of which went to Europe, then mine. I believe he prepared and kept another that he'd not yet assembled for himself, but that was quite some time ago. He did comment that the black baffle was a real pain to accomplish and he had no wish to produce any more of them!

Those wooden slats!

 

[hjames]

Yep - the wooden JBL Lens over the horns ... There were a couple sources for those once, but now I think just KenRick offers them.

 

[BMWCCA]

Yep - the wooden JBL Lens over the horns ... There were a couple sources for those once, but now I think just KenRick offers them.

Yep, mine came from Kenji but were not produced by Kenji, according to Rick who said all he knew was they came from Japan. There was also some discussion at the time regarding the reason for "trimming" the corners from the upper four plates. Sharp-eyes may also notice the lens from Rick had nine thin wooden plates where the ones in Heather's photo shows eight plates that appear to be a bit thicker.

 

[Chad Michael]

Oh my goodness, give them the juice!!! Pretty sure they will haul the mail!!!

 

[Pre-8]

Hi,

I've acquired a set of replica JBL 4345 made with no Internal crossover, ment to be used with a 4 way active crossover. 1 amp for each driver.

I'm using the Deqx Pre-8 as my active crossover.

The sound is good, but I feel like something is off.
I've tried to read the forums to see, if there's something I should watch out for in my configuration. Some posts have mentioned something about the polarity of the JBL 2245h driver, is it different from the rest of the drivers? If so, how do I flip it to be the same?

Furthermore, on the Deqx Pre-8 I've set the crossover point for the 2245h at 290hz. Once all calibration and such is done, the Deqx shows me a massive dip on the frequency curve at 290hz.

Any wisdom or tips to getting this unusual set of 4345 running at their best is much appreciated.

/Navnfugl

I have also JBL 4345 and Pre-8 fully activated.

As Pre-8 is in beta test phase the software is doing sometimes some weird things. It sounds like your phases are twisted and it is caused by the software.

1. Load the latest softwsre version! It seems to be working better than the previous.
2. When profile is created listen first and if the sound is odd test changing the polarity of each driver(one by one) in "Levels" tab. You hear when the phase is twisted or not.
3. It would be best way to verify the correct phases with another measurmenet tool like REW.

Anyways, there is coming the officiel software soon so this problem will go away.

 

[NavnFugl]

I have also JBL 4345 and Pre-8 fully activated.

As Pre-8 is in beta test phase the software is doing sometimes some weird things. It sounds like your phases are twisted and it is caused by the software.

1. Load the latest softwsre version! It seems to be working better than the previous.
2. When profile is created listen first and if the sound is odd test changing the polarity of each driver(one by one) in "Levels" tab. You hear when the phase is twisted or not.
3. It would be best way to verify the correct phases with another measurmenet tool like REW.

Anyways, there is coming the officiel software soon so this problem will go away.

I'm relieved to hear that you're in the same boat as me.

After a bunch of googling, and testing with a 1.5v battery, I've learned that on the 077, 2122h and 2245h the black terminal on the driver is positive. For the LE 85 it seems that the red terminal is positive.
Getting this right I've managed to get rid of all the major dips at each crossover point, except for the 077 at 10khz which is acting extremely weird as you can see from the pictures below.

I believe the purple line is the initial frequency response, the black is how the Pre-8 is trying to correct it.


This is the best result I've managed yet, still not a straight line and I've no clue what is going on from 6khz up to 16khz.


I don't think I've got anymore phase issues. I've tried swapping the red and black on the tweeter but the response remains as the one above.


How does your final frequency response curve look if I may ask/see? Do you have any tips in the software that you've learned perhaps?

Thank you all for your contribution of knowledge.

/NavnFugl

 

[Pre-8]

I'm relieved to hear that you're in the same boat as me.

After a bunch of googling, and testing with a 1.5v battery, I've learned that on the 077, 2122h and 2245h the black terminal on the driver is positive. For the LE 85 it seems that the red terminal is positive.
Getting this right I've managed to get rid of all the major dips at each crossover point, except for the 077 at 10khz which is acting extremely weird as you can see from the pictures below.

I believe the purple line is the initial frequency response, the black is how the Pre-8 is trying to correct it.
View attachment 3379894

This is the best result I've managed yet, still not a straight line and I've no clue what is going on from 6khz up to 16khz.
View attachment 3379893

I don't think I've got anymore phase issues. I've tried swapping the red and black on the tweeter but the response remains as the one above.


How does your final frequency response curve look if I may ask/see? Do you have any tips in the software that you've learned perhaps?

Thank you all for your contribution of knowledge.

/NavnFugl

You don't need to change the polarity from the driver because the DEQX software does it anyways. Looks like you need to change the polarity of horn.

Here is my measurement, done with REW.
There is 55hz dip caused by room. Otherwise this is quite perfect.

Hey, contact Alan Langford from DEQX. He can help you remotely and will fix if there is something.

 

[NavnFugl]

You don't need to change the polarity from the driver because the DEQX software does it anyways. Looks like you need to change the polarity of horn.

Here is my measurement, done with REW.
There is 55hz dip caused by room. Otherwise this is quite perfect.

Hey, contact Alan Langford from DEQX. He can help you remotely and will fix if there is something.

View attachment 3379923

I know you can switch the polarity in the Deqx software, but I really don't trust the software 100% as it's still in beta. I don't think the changes in the levels tab are carried over when you enter setup to make a new calibration.

If the horn had the wrong polarity, shouldn't it also be evident where the horn takes over from the mid on the frequency curve?

I will try and reach out to Alan as well, doubt I'll hear anything before next year though:=)

Thank you for your input.

/NavnFugl

 

[Pre-8]

I know you can switch the polarity in the Deqx software, but I really don't trust the software 100% as it's still in beta. I don't think the changes in the levels tab are carried over when you enter setup to make a new calibration.

If the horn had the wrong polarity, shouldn't it also be evident where the horn takes over from the mid on the frequency curve?

I will try and reach out to Alan as well, doubt I'll hear anything before next year though:=)

Thank you for your input.

/NavnFugl

Actually it is opposite way. You need to do the configuration first. After that go to the Levels and test changing the polarity.

Easiest way would be to measure with REW after creating the configuration. And based on the REW measurement change the polarities correct.

But anyways. Alan will do all this for you.