Cheap and (Pretty) Easy JBL 4430 Upgrade

Zonker92

All shiny and chrome
Last week I made an essentially free, reversible, and fairly easy modification to my 4430s that dramatically improved their sound.

As a preface, I would say that if you've read many of my postings here, you may know that I'm pretty skeptical about the audible benefits of some changes involving things like cables, caps, coils, etc. I may believe that a change makes an improvement, but I'm also quick to doubt my own ability reliably to perceive the improvement, as compared to wanting to believe I hear a change.

With this modification, though, there is no question in my mind that I have achieved a dramatic sonic improvement, as I will explain below.

So what is it? It's getting rid of the rotary switch in the N3134 factory passive crossover and direct-wiring the inputs. It's very easy to do.

The back of the 4430 has high level and low-level inputs, with a rotary switch to select between using an external crossover and the internal crossover. Here is the factory schematic. As you can see, turning switch does not bypass the internal crossover all the way. Instead, it breaks the LF and HF inputs apart and changes the internal crossover circuit to work with an OEM JBL external network.

attachment.php


And here's how it looks from the factory, with the switch in the middle:

index.php


So what does that mean? It means that you cannot directly, passively biamp (or biwire for that matter) the 4430 using the factory switch and configuration, unless you're using the JBL external network. That's not a big deal, because the internal network is excellent. However, it does prevent you from using separate amps for the LF and HF, and it does require you to use the rotary switch.

I wanted to bypass that switch AND enable passive biamping, and I found that by removing the woofer, I could remove the crossover simply by unscrewing six screws that held it to the back panel of the cabinet. The wires all unplug, and on my speakers, here is the wiring color code that I wrote up:

index.php


I figured out how to hard-wire the switching points thusly, and reassign (break out separately) the HF input (E3), so as to "lock in" the internal network while separating the LF and HF inputs from each other:

index.php


To do this, I just unscrewed the printed circuit board, unscrewed the nut holding the switch to the back panel, gently unplugged all the wires (none is soldered to the board), then unsoldered the switch. Then I installed three jumpers (using gold-plated wires left over from some film cap leads) and moved the E3 (HF) input to one of the switching points:

index.php


I put it all back together and plugged the hole using a black plastic (3/8"?) plug from the hardware store.

The result? I found a massive increase in lower bass response. Previously, the lower octave was present but not particularly strong. Now it's up several decibels (to my ears) and it will actually cause objects in the room to vibrate audibly, and the walls to creak, under high volume. This never happened before, at any output levels. My subwoofer has gone from being my major source of lower bass, to being merely a supplement from 20-35 Hz or so.

Why the difference? The only thing that occurs to me, is that there was a lot of oxidation in the crappy OEM rotary switch. I examined it and found that while it looked pretty good overall, the blades were oxidized and must have been choking off a lot of current.

Sure, I could have just cleaned it (although some of the blades are extremely hard to reach in a tiny gap above the board), but now I don't have to worry about it oxidizing again, and I can passively biamp the speakers if I so desire. (Right now I have the speaker cables connected to both sets of inputs.) If you don't want to break out the inputs, then you can just put a jumper in the final position on my diagram, and omit reassigning E3.

Now, if anything, the bass is a bit much, whereas before it was too little. But I understand that from the factory, the cabinet is tuned to around 32 Hz (or is it 34?), and along the lines of what Zilch once suggested, I plan to re-tune it to around 29 Hz for a smoother bass response. I also plan to raise the cabinets on stands maybe 4-6 inches to get the woofers away from the floor and have them behave more like the OEM soffit-mounting that the engineers intended for the speakers. (That is a mod suggested on the LH forum by JBL experts.)

But, I can assert with confidence, this inexpensive mod made a huge difference with the speakers.

(I have not looked at the 4435 schematic, but I'm sure there's a very similar mod to be done there, too.)
 

Attachments

  • Direct Wiring (2).jpg
    Direct Wiring (2).jpg
    74.1 KB · Views: 2,617
  • Direct Wiring (3).png
    Direct Wiring (3).png
    92.8 KB · Views: 2,560
  • Direct Wiring (5).jpg
    Direct Wiring (5).jpg
    54.6 KB · Views: 2,615
  • Direct Wiring (4).jpg
    Direct Wiring (4).jpg
    55.6 KB · Views: 2,550
Last edited:
I don't have a lack of lower octave on my 4430's. My crossovers are stock, untouched.
 
Great info Zonk, when I get to building my new cabinets and installing the drivers I will look into that mod. The pristeen pair I had that sold, had new 1% caps from Dayton that I installed, I did an A / B with one that had the new caps, and the one that didn't. There was improvement, not huge, but the new caps sounded nicer.
 
Thanks for the comments; and I'd forgetten that quote from Greg Timbers. :thmbsp:

I don't have a lack of lower octave on my 4430's. My crossovers are stock, untouched.

That's funny; a week ago I would have said the exact same thing. But seriously, all that tells me is (1) your switch has lower resistance than mine (i.e., less oxidation); (2) I'm imagining the improvement; and/or (3) you're getting less bass than your speakers can provide. I'm OK with any of those possibilities. :D
 
The pristeen pair I had that sold, had new 1% caps from Dayton that I installed, I did an A / B with one that had the new caps, and the one that didn't. There was improvement, not huge, but the new caps sounded nicer.

Cool. I replaced the caps with Audiocap Thetas (2, 6, 12 and 14 (12 + 2) uF), the resistors with 12 watt Mills (7.5 and 20 ohms) and the L2 coils with lovely air-core foil coils of identical DCR. :D Everything was from Sonicraft. The new coils went where C7 was, since I don't need C7 in this configuration. But it's all fully reversible.

I haven't had a chance to really sit and listen, but A-B'ing one old one against one new one (white noise and music on mono), the new caps seemed to sound smoother and less edgy than stock.

index.php
 

Attachments

  • DSC_4847.jpg
    DSC_4847.jpg
    85.4 KB · Views: 2,328
Last edited:
On the old switches, the blade contacts in the middle were virtually covered in black oxidation, with tiny contact points though which all your signals must attempt to pass. (Desoldering the switches is really the only way to gain full access to the underside contacts.)
 
Last edited:
Really interesting and compelling stuff. and with quotes from Zilch and a JBL engineer. Thanks for the thread.

I don't have interest in or see a need to bi-amp these, and I'm very happy with the LF in my system. so much that I need to install the Rega wall mount for my TT. ...but will investigate the mod to take out the switch or at least attack any oxidation especially if I need to get at the woofers.

:beerchug:
 
Nice work, Zonk.
One less set of contacts to mess up the signal.
My 3110A CO are quickly approaching 30 years. May be time to clean them and see if there is a difference.
 
Thanks! I'd say of all the many speaker mods I've done, this was the most noticeable and beneficial. :D
 
Great mod Zonk, one that I'll put into practice when I get moving on restoring the 4430/4435. And I do have a couple of 5235, along with 5234/5234A, and the 4430/4435 cards to play with.

Then considering the charge coupled crossovers. Decisions!
 
couple of other comments that worked on the 4345 that probably apply to the 4430/4435:
"The network components should be updated and this isn't easy. All the capacitors should be polypropylene and you would want to use air core inductors where ever possible. I believe the 4345 used some tapped iron cores. Those are hard to replace with aircores because the surrounding network topology would have to change and all of the values have to be re-engineered. I would suggest leaving them alone unless you are really good at this stuff."

"If you can swing it, go to a biased network. The difference is unbelievable."


"The L-pads aren't so hot either, particularly after all of these years. Once you have your preferred balance, it is fairly easy to measure each leg of the L-pad and replace it with fixed resistors."

"I notice from many of the pictures that the system is elevated on blocks. It is very good to get the 2245 up off of the floor to minimize midbass fatness."

Your 4430 stands will do the trick for those speakers, lots of stuff to do to update the 4345, and 4430 will work similarly, although the controls operate a bit differently with the horn and CD compensation.
 
Great mod. Thanks for all the details.

Are you still happy with it? Have you tried passively bi-amping?

Also, I wonder if similar results could be heard by bypassing the CDs l-pads.

Cheers,

Warren
 
Thanks! Still quite happy with it, and I am now passively biamping:

Horns: VTA Dynaco/Dynakit MK3 monoblocks (60 WPC) with a 500 Hz high-pass filter on their inputs

Woofers: Adcom GFA-5800 stereo amp (250 WPC)

I would consider bypassing the adjusters (one L Pad and one rheostat each); it should be extremely easy with some jumpers and maybe some resistors, but I can't recall whether that would screw up the CD compensation or change how the HF crossovers operate, and it's been months since I looked at that issue. I suspect it would be fine, though.
 
OK, Zonk...

I'm afraid to even ask...

How does one "passively biamp" my extraordinary sounding 4435s???

I tried using a DBX (active) crossover (since the JBL M522 Crossover is still missing and likely gone forever...) with a Crown on the bottom crossed over at 1K, and let me tell you, the stock internal crossover and one amp blew it away...
 
Last edited:
:) Your 4435s have slightly different crossovers, but I bet the basic idea is the same: You remove the rotary switch and slightly remap the inputs so instead of having both HF and LF circuits being controlled by that switch (and being joined together or separated by it) you wind up with two separate circuits, one for the LF and one for the HF.

My first post above goes through the process for the 4430. It also removes one cap from the LF circuit that would have been used in the "external crossover" mode on the switch.

The bottom line is, the crossover operates as it normally does in the "internal crossover" mode, except the switch is gone and the lower external input terminals are connected only to the woofer circuit; same deal for the upper ones, which now feed only the horn circuit. It's now set up as many modern speakers are that have a set of "biwire" terminals joined by removable straps.

Clear as mud? :D If you want to do this for your 4435s, I could stare at the schematic and try to draw it out for you, or discuss further options by PM; maybe take a stab at modifying your crossovers sometime.

Mainly, I would do it only if (1) you want to passively biamp them without the OEM JBL external crossover; and/or (2) you want to get rid of the rotary switch, which tends to get oxidized and in my experience, reduce the woofer output. You could also do it (3) for biwiring.
 
Last edited:
:) Your 4435s have slightly different crossovers, but I bet the basic idea is the same: You remove the rotary switch and slightly remap the inputs so instead of having both HF and LF circuits being controlled by that switch (and being joined together or separated by it) you wind up with two separate circuits, one for the LF and one for the HF.

My first post above goes through the process for the 4430. It also removes one cap from the LF circuit that would have been used in the "external crossover" mode on the switch.

The bottom line is, the crossover operates as it normally does in the "internal crossover" mode, except the switch is gone and the lower external input terminals are connected only to the woofer circuit; same deal for the upper ones, which now feed only the horn circuit. It's now set up as many modern speakers are that have a set of "biwire" terminals joined by removable straps.

Clear as mud? :D If you want to do this for your 4435s, I could stare at the schematic and try to draw it out for you, or discuss further options by PM; maybe take a stab at modifying your crossovers sometime.

Mainly, I would do it only if (1) you want to passively biamp them without the OEM JBL external crossover; and/or (2) you want to get rid of the rotary switch, which tends to get oxidized and in my experience, reduce the woofer output. You could also do it (3) for biwiring.




If it were not for about 3,000 miles, I'd invite you over for a beer or two and we could discuss it.

Even as a kid I knew when my buddy Joe used to solder caps and resistors in circuits and his old man played with his vacuum tube CB stuff I should have paid more attention...MUCH more attention in fact.
 
Hi Zonker,
I have recently bought a pair of JBL 4430. I have not bought amplifier yet, and at the moment, I am connecting the 4430 to my Pioneer LX73 receiver by bi-amping mode. The sound not so OK, both LF and HF.
I am now finding a suitable amp, not bi-amping (it's difficult to me to do technical things to follow this :). Could you please advice me if the following amps (that are available now in my local) are suitable?
- Marantz PM-15, 150w/ch, 8ohm
- Luxman L509s, 160w/ch, 8ohm
- Mcintosh MA-6800, 150w/ch (most expensive)
Thank you very much!
 
Back
Top Bottom