Splitting, attenuating cable?

I’m looking to buy or make a cable that I thought I first learned about on one of the NPR Radio training web pages. Unfortunately, I can’t find it now to save my life.

The cable connected to a line-out jack from an audio system. It split into two plugs. One plug delivered the full line-out from the upstream system. The other attenuated the signal, I thought by -12dB. These could be used in a recorder like the Tascam DR-40, with it’s two external line-in jacks. By recording both tracks, if the full line-out clipped in one section, it could be spliced with the -12dB track to recover the un-distorted audio.

I have three questions:

  1. Where would I buy such a cable? Better yet, does anyone have any plans or schematics for such a cable that I could build myself?

  2. -12dB sounds like quite a lot of attenuation. If I remember my math correctly [-3db = half the signal], -12db is 1/16 of the signal. Is this correct? Would anyone suggest a different value for the amount of attenuation?

  3. How would I actually use the two tracks? I envision sync-locking the two tracks, cutting out the clipped section from the full-line-out track, cutting out everything else from the attenuated track, and then mixing them together and Normalizing. Is this the procedure you’d recommend?

Thanks so much for your advice and guidance.


I’m not sure that I need this equipment now. Both of the recorders I use, the TASCAM DR-40 and the Zoom H4n Pro, have settings to record from the Input 1 and 2 jacks, with one set at some level below the other. I think that this does in software what I envisioned the hardware cable doing.

I think that this just means that I need a splitting cable. Is this just as simple as two male plugs wired in parallel?

Question 3 still stands: how would I mix the two tracks to cut out a clipped section?

Thank, again, for your help.


I doubt it exists ready made. I’ve never seen one.

The specs say (http://tascam.com/product/dr-40/specifications/) that the line input is 10 kOhm, so just putting a 33 kOhm resistor in series will give approximately the right gain drop.

-6 dB is (approx) half amplitude. -12 is 1/4 amplitude.

What are you wanting to record? For music, it would be easier and better to just record at a low enough level that clipping does not occur.

Steve, thanks for your reply and suggestions.

My full story is that I’m trying to record the homilies or sermons from my church, and turn them into podcasts. The audio panel is behind the altar, and I can’t get into and out of that area without distracting the whole congregation. So, I have to set my levels in advance of the service starting. I can’t do an audio check. Then, some celebrants are loud, others soft. Sometimes the mic is placed correctly, sometimes it’s off.

I’m recording WAV at 44.1k, 24 bit. I thought that if I reduced the levels too much, I was wasting most of the 24 bits of data, and the sound would be degraded.

I appreciate your advice and welcome any suggestions you have.


24-bit has a huge amount of dynamic range. You can afford to “waste” quite a bit as a safety margin. (a peak level of -20 dB will still have more dynamic range than a CD.)

Indeed that sounds like something that might have done back in the day of analog tape recording to deal with live sources that have a lot of dynamic range. But hardly seems useful with modern equipment.