JBL 7126 compressor info?

GroupDIY Audio Forum

Help Support GroupDIY Audio Forum:

This site may earn a commission from merchant affiliate links, including eBay, Amazon, and others.

earthsled

Well-known member
Joined
Dec 1, 2009
Messages
405
Location
Asheville, NC USA
I recently acquired a couple of these 1970s JBL compressors in non-working condition and I intend to repair and restore them. Has anyone come across this model before? I've found a few schematics posted online, but so far they have been totally unreadable. I'd love to get some info and opinions about this comp!

Here are a few images of the 7126 from an expired Reverb post:

vadmlysy80zqqijxx9ho.jpg


k9wk4o5lsrwdgc3x9xx8.jpg


nbyfmlowjjbzqi6zroie.jpg


f1twpy9t48cnamjbechn.jpg


qhcernymcf1jcwpx2mjc.jpg


Funny, it sort of looks like a DIY build inside!  ;D


 
The non-working units arrived yesterday. I've done some basic testing and the power supplies seem to be in working order. These units came with 5901 microphone input transformers which is pretty cool. I'm probably going to set them up for line level, so I picked up some 5195 matching/bridging transformers to try.

It seems one VU meter might be fried. I've contacted JBL service, but the replacement parts are long gone. Can anyone identify this meter? It was made in Japan and measures about 2 3/8" wide by 2 1/4" high (JBL part number 39961). If I can't find an exact match, I'm sure I can retrofit something.

y9p3a4hp

y9qljfzh

 
I'm hoping to get some help with this meter circuit...

The meter will adjust to zero in compression mode just fine.

The issue is with calibrating the meter for +4dB or +18dB output levels. With the proper output level set, I adjust R36 to move the needle of the VU meter, but the adjustment tops-out before the needle gets to zero.

I tried doing the calibration with a new VU meter as well, and the result was a little better, but R36 is still at the top of its range when the needle is at zero.

Screen Shot 2023-09-16 at 2.45.03 PM.png
 
Have you serviced it already? Just from the vintage of the unit in the Reverb pics I would be deeply suspicious of all of the electrolytic capacitors. If not, I would recommend a hard look at that sort of thing before trying to troubleshoot. If that adjustment worked when the unit was built, there's presumably something that's making it not work now.
 
For what it's worth, I'm having the same issue with both 7126 units, so the issue seems to be consistent. I've checked the resistance values in the meter circuit, and they all seem to be okay.

When the meter switch is in Compression mode, I can set the needle to zero. If I take the meter out of circuit and measure the feed to the meter, I get 3.23 VDC.

When I attempt to adjust the meter for +18dB, I have 6.2 VAC at the output. Turning R36 all the way up will move the needle to about -3dB on the VU scale. If I take the meter out of circuit and measure the feed to the meter, I get 0.788 VAC.
 
Last edited:
It seems like R34 and R37 make a pad to attenuate the signal going to the VU meter. Perhaps these values can be adjusted to allow for more signal at the meter? I'm not exactly sure how to adjust and keep the function of R35 constant (to reduce the signal by 14dB when switched in).

Screen Shot 2023-09-16 at 5.47.01 PM.png

Of course an active meter buffer circuit could be added to replace the original passive design. Maybe this is the best solution?
 
After more testing, I believe the retrofit of active VU buffers is the best plan for these units. I suspect the original meters may have lost some sensitivity over the years, so my hope is that the buffer circuit will boost the signal enough to set a proper reading on the meter. The buffers also have the advantage of improving the %THD by isolating the meter from the circuit.
 
The VU buffers worked out quite well. The meters can now be calibrated to the output level. I also replaced R35 with a 1k trimmer so that the +18dB meter calibration could be adjusted relative to the +4dB meter setting. (Shown below in the upper right):

IMG_3963.jpg

For the compressor calibration, I replaced the single-turn trimmers (R46 and R55) with 25-turn models. This greatly improved the accuracy of setting this part of the calibration.
 
Last edited:
To be honest, I haven't done much subjective testing / listening with this compressor yet. That being said, I'm happy to share some objective observations...

These compressors have a LOT of gain. I've been using a -20dB in-line pad at the input so that the input level control has more usable range with a line-level signal. I might consider adding an input pad switch for 10 to 20 dB, or possibly modifying the input potentiometer for a more usable range. Also, I believe a variable pad at the output would be a nice modification. This would allow for more control of the compressor considering that the 7126 has a fixed threshold.

The release time of the 7126 seems a bit slow compared to other models. At 0.5 or 1.5 seconds, one could compare the speed of the release to an LA2A or LA3A. Of course, the 7126 circuit is FET-based rather than optical compressor.

I have experimented a bit with a modification to the release time. It seems the release can be made a little faster by lowering the resistance in place of R52 and R53. A variable release control might be a nice modification as well. I'd like to explore this more...
 

Latest posts

Back
Top