Mute with THAT1512

GroupDIY Audio Forum

Help Support GroupDIY Audio Forum:

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

n0npr0phet

New member
Joined
Jan 2, 2008
Messages
4
Location
Columbus, Ohio
Hi All,
I am building a 6 channel preamp using THAT1512 and THAT1646. I didn't invent anything and I was mostly datasheet diving. It sounds and measures good but when nothing is connected it is obviously noisy (johnson?) when the gain is cranked up. I know.. I know.. turn down the gain when nothing is connected. :roll: but... I used the Nuetrik Combo XLRs that have switches that open when something is plugged in so I rigged up the following to mute the output of the THAT1512 when nothing is inserted into either jack.

gn.jpg


The mute I think I found a link for in this forum to this edn soft mute article. I modified the suggested circuit as I don't need indicators etc.

mute.jpg


So finally here are my questions. Is this the best place to do this (pin 6 of the 1512)? Should I provide a mute bypass for critical recording in case the JFET might be introducing distortion into the signal path? Using the mute I have used a series resistor that makes a voltage divider when the fet gate is not open. In this configuration with a 1K resistor the mute is -40db but it mutes more (-50db) with a 5K. Dare I use a higher value -60db would be nice? I was worried about using a 5K resistor here but I didn't see or hear any problems. What series resistor would you use?

Has anyone else solved this problem another way? I thought about somehow opening the circuit at pins 4 and 8 which would take the gain to -6db but I was concerned about introducing noise or other problems by messing with the gain loop. Obviously relays could be used in this loop or on the output but they are expensive and perhaps inelegant. My PCBs are made and I will have to lift traces to make the mute but my stepped gain selectors are not all made yet so suggestions on the latter approach if possible would also be great.

Thanks, This is my first post.
np.
 
Hey I'm in no condition to give you any help--and am not familiar with the parts/design in question. I DO commend you on the well presented (first) post. Surely our THAT prof.s could provide some guidance. I' not familiar with your switching input connect, but if there are closed contacts(s) with nothing plugged-in; and they open when something is plugged in, can't you short the input nodes when nothing is plugged in? that usually gives best case noise output.
 
Thanks shabtek,
By default all the switches are closed unless something is plugged in and in this case the hot and cold are grounded and the mute is active.
np
 
I'm not sure if your Neutrik connector works the way you believe it does. I think it just switches between jack and XLR--which means that you can detect if there's a jack plugged in or not, but not if there's a XLR connected or not.

I didn't study the 1646 datasheet very closely but it looks to me as if it would like a low source impedance drive, which makes your approach unsuitable. You'd either need a buffer (which is too much work just to get rid of "unplugged noise") or a series FET.

Samuel
 
I could be wrong but I did test the Combo jack and my circuit appeared to work. I will test it again. Here is the diagram from the datasheet of the jack.

normal.jpg


The 1600 datasheet focuses, obviously, on the output and says very little about the input except the swing. The input impedance is described as 5K and this is why I was concerned. I have tested and it mutes and sounds good (I haven't done critical listening tests yet).

So if the input impedance is 5K then by rule of thumb (and I am no audio guru by any means) we should have 500 ohms or Zsource < Zin x 10 for bridging and so ideally a 5K resistor isn't best but what is the downside in this case? This is the crux of my question. It didn't mute just pulling it to ground btw.

I happen to have an NE5532 buffer right there because I have a built in VU meter. Your suggestion actually means I would not have to cut traces. Would the introduction of this opamp in the signal path or a series FET affect the signal more than using impedance matching over bridging? Are there FETs that "sound" better over others? If using this buffer would you use a series FET or the series resistor?
 
I am adding a schematic of one channel of my project in hopes that it will help to explain what I am asking. This is before the two diagrams in the first post which are a proposed mod. Any other suggestions or comments would be great too! :)

channel2.jpg

thanks,
np
 

Latest posts

Back
Top