View Full Version : Uneven Loudness

08-24-2008, 05:13 PM

I noticed the freq response of my 4430 L/R varies, when both sides mid and high lpads are set to 0, one side seems louder than the other side. Also the freq response seems uneven, in that I mean like for voice the imbalance seems ok, but for saxaphone, the sound would drift to one side.

Is this caused by the crossover or the compression driver? would changing the diaphragm works?


Fred Sanford
08-24-2008, 05:24 PM
Actually, I had similar symptoms with 4333As and it was the original equipment input terminals oxidizing. Sometimes when the music would naturally get louder the sound would balance back out, too. Cleaning them helped for a while, then setting them up for bi-amping has been OK for a while, and I'll be replacing them with new next time I have the whole mess apart.

Good luck,


08-24-2008, 05:48 PM

I noticed the freq response of my 4430 L/R varies, when both sides mid and high lpads are set to 0, one side seems louder than the other side. Also the freq response seems uneven, in that I mean like for voice the imbalance seems ok, but for saxaphone, the sound would drift to one side.

Is this caused by the crossover or the compression driver? would changing the diaphragm works?


Switch your speaker wires, if the problem moves to the other side, it's not your speakers.Then switch each components' wires until you find the culprit.

08-24-2008, 05:50 PM
Don't forget to swap the speakers (right to left) to see if the problem follows.

08-24-2008, 07:56 PM
I did swap the L/R connection and confirmed that the problem is with the speakers and not the CDP, AMP.

could it be the capacitors in the XO has reached their useful life? If that is that case , bi-amping will not solve the problem because the circuit for the HF will not be bypassed. Is it difficult to change the capacitors? Are they still avail or are there substitutes?


08-25-2008, 06:41 AM
What happens if you turn the controls up and down a bunch of times, then leave them all the way up, which theoretically is "straight-through"? What's the balance sound like then? I know the settings wouldn't be ideal, but at least you should be able to compare them.


08-25-2008, 07:06 AM
The internal speaker wires can oxidize leading to a higher resistance effecting the output level.
Strip new internal wire connections as well as clean the L-Pad. It may solve your problem.


08-25-2008, 07:13 AM
I've found it helpful to use a monophonic source (or pre-amp setting) when chasing
this sort of thing down. I don't happen to have a stereo-mono preamp switch, but can
set my FM tuner to mono (as an example).

... additionally suggest you give the biamp/passive switch a rigorous workout
(clean if practical). L-pads are probably your best target for focus, albeit a bit
of a pain to get to... (heat gun or hair drier + spackle tool or spatula to remove
name plate)

08-25-2008, 08:06 AM
If you can, pry open the Lpads and spray in first Caig Deoxit; give it a good bath. Then spray in some Deoxit Gold.

Last time I did this I destroyed the decals which covered the attenuator. But the pot worked right again.

08-25-2008, 11:47 AM
My 4430's had corroded bi-amp switch contacts. Try rotating them back and forth a few times. When I finally opened mine up, I also blasted them with some Caig.

08-25-2008, 06:50 PM
I don't think its the Lpads. when I rotate them, there is no noise which would indicate that they are in fairly good working condition. I also tried the biamp switch. I will try biamp this weekend.

Also thinking of building the XO from scratch so that I can avoid the trouble of taking out the name plate.

08-25-2008, 10:49 PM
If you're planning on going through all that (fun project that it would be) :),
why not swap horn/driver assemblies to narrow it down to crossover vs.
compression driver issue (which I believe you speculated earlier)? Only
six hex-head machine screws per speaker.

08-26-2008, 02:24 AM
Took a peek at the crossover, noticed a iron core inductor that looks like a mini transformer.

If I build my own XO, do I need to use the same type of inductor? Don't think they are avail anymore.

08-26-2008, 06:51 AM
To not alter the design, the same value and DC resistance are a minimum criteria
for a replacement inductor. Current/power-handling capability in say L102 (2.6mH),
the off-board inductor, would also be good to match (at minimum). I seem to recall
someone having posted a suggested N3134 parts list and perhaps DCR.

Something like this would work (in an overkill/feel-good sort of way):
(take off a few turns if the 4% + tolerance value bothers you... good to measure
and match them anyway)

The equivalent air-core inductor could easily cost 2x, but "different strokes..."

08-26-2008, 09:37 AM
I had a similar problem when I refoamed a set of 4425s. Since I had never heard them running right (purchased at a yard sale in sad condition), I figured that there was a bit of mid-range 'suck-out' with the speakers. What I found I had done was not wire one of the woofers correctly after I had reinstalled it after refoaming. Your symptoms sound somewhat like I was hearing.

I'd be in there confirming that the drivers are, indeed, hooked up correctly - especially if they have been opened up before.

Hope that helps.



08-27-2008, 06:46 AM
I took out the woofers to check the connections, to my surprise, one side of the cabinet is stuffed with a few extra sheet of rolled-up asbestos than the other. They looked newer than the original asbestos lining and I figured they must be inserted by previous owner and not original. He could be using them to tune the bass response, why he did not balance both speakers I do not know.

I promptly removed them and so far the sound appears to be balanced. The extra asbestos must have altered the freq response of one speaker, creating the imbalance between channels, not helped by the high crossover freq of 1 khz.

Those with 4430, would appreciate if you could post some pictures of the original internal asbestos lining. Not sure if I over-removed them.


08-27-2008, 08:56 AM
I'm sure you meant fiberglass; pics posted now.
Glad your imbalance issue was so simple to resolve (if not to determine)

08-27-2008, 02:06 PM
Thanks Grumpy, looking forward to those internal pictures.

08-27-2008, 06:18 PM
right, down, left, up views:

08-27-2008, 08:10 PM
Thanks Grumpy, really appreciate it.

Oh no, It seems I have removed the wrong fibreglass. Based on the pictures, the recess areas besides the compression driver is supposed to be filled by fibreglass. I removed those. Also, I removed the ones covering the XO. Will put them back, but I'll ensure this time its balance.

08-27-2008, 08:44 PM
I'm the second owner, but I don't believe the original owner had any work
done on them besides having the 2235's re-foamed... both appear to have
the same amount of stuffing in the same areas.

08-27-2008, 10:26 PM
I think so too. I use the WinISD software to calculate the optimum volume for 2235H and the value is 3.34 cubic feet at 33 Hz. The volume of the 4430 cabs is 5 cubic feet. I think the stuffing is suppose to get nearer to the optimum volume. I'll give it a shot.

Afternote - I do not have enough fibreglass to do everything, in the meantime, will settle for just covering the crossover. I will have to leave the cavity next to the compression driver for now. So far, things are looking positive.

I would like to express my gratitude to all, especially Grumpy, for your invaluable advice.


08-28-2008, 12:26 PM
Hmm... normally, the nominal stuffing applied would tend to compensate for
things like internal bracing, bass driver volume static displacement, and normal
sized ports, such that the nominal 5ft3 would be reduced somewhat. The
stuffing (if not grossly overdone) should make the box look a big larger in
effective volume (bringing it back to the 5ft3 number, ... more or less).

I don't have WinISD in front of me at the moment, but it wouldn't surprise
me if the 4430 tuning wasn't the same as what WinISD shows as "optimum"
... additional LF extension can be gained by loosening constraints in other
areas. Play with WinISD, you'll see what I mean.