View Full Version : DIY Air Core Autofomer Replacements for JBL Crossovers?

11-23-2013, 11:05 AM
Hi Everyone,

I am working on modifying/rebuilding some old JBL Crssovers. My goal is to better match the crossover and my specific driver choice.

(This is for a DIY Hartsfield stereo pair).

The issue is: the components are very old and while I can readily buy new air core inductors, and caps to suit my crossover design, I want to rebuild the autoformers to air core units as well as understand how they are built in the event I want to modify them or play with the design down the line.

The crossovers I am starting with can be found here (the JBL 3115):

http://www.jblproservice.com/pdf/Net...%20Network.pdf (http://www.jblproservice.com/pdf/Network%20Schematics/3115%20Network.pdf)

The autoformer for the HF driver is specified as a 3.85mH unit with -6 , -8 & -10dB attenuation taps.

For example, if I wanted a -3db attenuation tap, which is a voltage gain of 1/2, would I simply put a tap exactly splitting the 3.85mH inductor?

Or its it more complicated that that since the effective impedance of that new center tap to the autoformer common tap is 1/2 the inductance of the total coil, in parallel with the load of my loudspeaker?

This then leads me to my next point: The voltage on the new center tap will only be -3db or 1/2 at a single specific frequency since the impedance of the inductor moves higher with frequency. At very high frequencies, the parallel combination of my speaker load and 1/2 the inductance of the total autoformer coil will asymptote to the impedance of just the driver.

Thus, the voltage across the HF driver at high frequencies would be low since it is divided down by the top 1/2 of the autoformer coil inductance, (now very high since frequency is high) and essentially the impedance of the driver (low around 8-16ohms)!

Thanks for any help,


11-23-2013, 11:34 AM
If you are DIY what's wrong with the original especially if you are using E-145's which is the recommended replacement driver?

http://manuals.harman.com/JBL/HOM/Technical (http://manuals.harman.com/JBL/HOM/Technical Sheet/D30085 Hartsfield ts.pdf) Sheet/D30085 Hartsfield ts.pdf

Thanks to Techbot and David Smith



11-24-2013, 06:25 PM
I while back I ran some tests on the possibly of winding an air core tapped inductor (Auto-transformer) and while my test results were not super conclusive. I was most certainly under the impression that an Auto-transformer really should be made with an iron core. Particularly, if you donít have the test equipment for measuring the voltages drive levels. Here is a link to my old test results: http://www.audioheritage.org/vbulletin/showthread.php?33057-Alternative-to-tapped-inductor-in-3110a-or-3115a
My recommendations for anyone wanting to make an autotransformer would be to use an iron core from an Erse I-Bar coil and then follow the percentage of Inductance formula, and to wind the secondary windings (taps) closest to the core and then wind the primary winding on top.
This should give you the correct voltage drive levels, or at least be fairly close.