The JBL 4345 Club

[hjames]

Hey - thanks bunches!
Great stuff to study, even for those of us with a lesser budget!
Talk care!

[Ian Mackenzie]

If the truth be known it probably makes the ambiguous Japanese amplfier sound listenable, at least 95 % of them that are chip amps in some way or another.

The thing is the 4345 does not need CC network to sound good and there are no electro capacitors in the stock networks.

Its always been my understanding and that of many loudspeaker engineers as published in many well regarded text books that electros are not the preferred variety.

http://www.audioheritage.org/vbullet...0&postcount=37

[yggdrasil]

Hi Ian

Regarding the active crossover you are looking into - if you are looking at discrete opamps, this could be an idea.

I have been playing with the thought of a small power opamp at the output of the crossover. This could be a variation of the MOX.

This one is biased at 0,3A, which should at least be enough for the slot, maybe even the CD.

[Ian Mackenzie]

Yes,

That is a possibility, the Fet follower adds current gain.

The Borbely headpone amplifier is similar.

Give it a try.

regards

Ian

[yggdrasil]

I will, but a little different. Was not able to get the 2SK389 at a reasonable price. Had made a deal with some chinese, but they turned the price way up after receiving my money.

I have several other dual JFET's to try, a couple of MOSFET's and a couple of BJT's. Will try with higher rail voltages as per NP's advice.

Hope to get i playing before christmas...

[Ian Mackenzie]

Pm me your details and I will mail some 2SK389s for you to try.

best

Ian

[Ian Mackenzie]

Please note the slot energiser black box is still on my list but there has been a delay due to some renovations here at home.

I propose to make the black box using a diy X Aleph and include a Mox hi pass filter. The reason for going this route is I prefer to use identical amplifiers for mid and HF and this design is easily scaleable.

[Ian Mackenzie]

http://www.audioheritage.org/vbullet...&postcount=109

[Ian Mackenzie]

Here is a link to an interesting modification

http://www.audioheritage.org/vbullet...&postcount=153

Over time I have also tried different placement postions and have found elevating the system on a stand about 8 inches off the floor can help with balance pending you room dimensions.

Ian

[Ian Mackenzie]

news

http://www.audioheritage.org/vbullet...&postcount=194

[Chas]

Ian, I hope you don't mind me posting this here, but I think it is relevant. Please feel free to have the Moderators move it.

Now that my 4345 restoration is complete, I have upgraded all the internal capacitors necessary for biamping, bypassed the switch and along the way, bi-wired the MF and HF/VHF. Lately, I have been thinking about something Greg Timbers mentioned in his 4345 "You Guys are Amazing" post.

By the way, the results to date, are nothing short of amazing.

What I am concerned about - and the reason for this post, is Greg's comment about the 2245 and 2122 drivers not having flat response at the crossover frquency (290 Hz) when bi-amping the 4345.

I am using L-R 24dB/octave at 290 Hz and it sounds fine but I have yet to run my RTA and actually see what is going on. I have changed the polarity of the high pass signal so that all the drivers are the same polarity as the 2245 and there was a huge improvement. I guess that makes sense due to the original passive network being 12 db/octave.

Has anyone addressed this issue? My understanding is that the JBL active crossovers were standard third order (18 dB/octave) and there was no other signal manipulation. What gives?

Charles.

[boputnam]

I have changed the polarity of the high pass signal so that all the drivers are the same polarity as the 2245 and there was a huge improvement.

Describe the result?

My understanding is that the JBL active crossovers were standard third order (18 dB/octave) and there was no other signal manipulation.

Yes, the 5234a has 4345 specific cards with a -18dB slope. I have wondered about the difference between the two slopes myself. I have no clear explanation of why they were so different except I understand the 5234a was done on a "best efforts" basis.

[Chas]

Hi Bo. Just so I am not misunderstood, I have positive going signal on the low pass side attached to the black terminal on the 2245's. Now, if you look at the 3145 schematic, the positive signal is reversed to all the rest of the crossovers feeding everything above. I think this was bcause the slopes are 12 dB/octave with the attendant 180 degree phase shift. Strictly a guess here on my part...

In my case, I simply reversed the high pass network input (actually I did it at the amp), since a L-R fourth order network does not have the same issue.

The effect was to bring everything together in a more homogeneous way. It definitely improved imaging as well. I haven't tried returning to the previous polarity, I'll give it a try, when I have time just to be sure I wasn't hearing things.

[boputnam]

Hey, Chas...

Yea, I understood that. Interesting. I too would be keen to know if the previous polarity was so different - with all the changes you've made it gets complicated attributing improvement!

[Mr. Widget]

In my case, I simply reversed the high pass network input (actually I did it at the amp), since a L-R fourth order network does not have the same issue.

Do you know if both amps you are using invert phase or are non phase inverting? This is something that always needs to be kept track of in a multi-amp situation. Obviously it is a non issue wen you are using multiple identical amps.

Bo, we did some EQ tweaking near the crossover frequency on your system when we were setting up your Bryston 10B... I don't remember what we did or the slope of the Brytson, but if you could add those data points to Chas's discussion it would be interesting.


Widget