I’m adding a sub to my 2ch setup.
Im wondering if I can omit U1, U2, R1 and R2 without degrading the signal to the preamp.
Is the buffer on the input necessary? I want to avoid any interaction of the two channels going to the preamp? Thanks
Im wondering if I can omit U1, U2, R1 and R2 without degrading the signal to the preamp.
Is the buffer on the input necessary? I want to avoid any interaction of the two channels going to the preamp? Thanks
Last edited:
U1 and 2 are useless. The inverting input of the last is a "virtual ground" and no merge of one signal in to the other.
I suggest to try to substract both channels in place of add them. Better sounds appears in live recordings and with the 3rd speaker placed back in the listening site. Add a volume control to this 3rd channel too.
I suggest to try to substract both channels in place of add them. Better sounds appears in live recordings and with the 3rd speaker placed back in the listening site. Add a volume control to this 3rd channel too.
Interesting. Using a differential amp to merge right and left for the sub wasn’t something on my radar. Thanks for the info.
So would this work as a unity-gain subtraction circuit for me?
Attachments
I think the virtual ground mixer (original post) is the way to go tbh and yes, you can omit the first opamps as long as you are OK with the 10k's setting the input impedance for each input.
I'm not sure if Osvaldo is thinking of the old Hafler set up where you could connect a speaker between between the two 'plus' or live speaker terminals of a left/right stereo setup and get a 'difference' channel to a third speaker placed at the rear.
I'm not sure if Osvaldo is thinking of the old Hafler set up where you could connect a speaker between between the two 'plus' or live speaker terminals of a left/right stereo setup and get a 'difference' channel to a third speaker placed at the rear.
Exacly. I had implemented this idea using a "rester" in place of an adder. I saw this one in an old magazine about 1972 my dad collected, and I use the same idea but add a third volume control independent of the main and an own amplifier. Too low the setting of it, nulls the effect, and too high, strange sounds are created in general ugly. But there is an optimum setting that let you feel you are in the middle of the recording set.
This effect is much more pronounced in good sources recorded live and with, say, an MP3 of 320KB/S. Lower quality signals are very unpleasant and this pot must be at 0. The rightmost pot is my 3rd channel volume ctrl, and the middle TL082 in the second pic is my "rester".
Attachments
Still more. I rescued my schematic 20 years old. In such magazine, the author suggested to add some kind of bass boost in the 3rd channel, this is why I added a second stage to behave as low pass but in practice and in my experience, it isn't good and the JP remains shorted (plain gain).
Attachments
Yep. Just go directly into the 10 kΩ resistors. As others have pointed out, the virtual ground at the inverting input to the summing opamp will prevent the channels from interacting. Do note, though, that the virtual ground is only good as long as the circuit has substantial negative feedback, so you'll want a modern opamp with high open-loop gain for the summing amp to ensure a good virtual ground through the audio band.
Also note that the circuit is inverting. If you want the sub to have the same phase as the rest of the system you'll need to add an inverter (or to swap the polarity on the sub driver).
Tom
U1 and U2 are not useless. They raise the input impedance way beyond to
10K Ω of the actual summing amp. Not useless but also not needed if 10K Ω
is OK. I used them in the drawing to get back up to 10K Ω. You do not want
a differential amp as your summer as the output is only the difference which
is the opposite of what you want for the sub.
I included the 5.1 preamp for my TV project which has been used for almost
4 years. The upper left corner is what you want but my 80Hz filter is actually
100 Hz. Change the 33k2 resistors to 39k2 to correct that. The gain on my
summing amp is 1/2 of your initial drawing to be sure it doesn't clip.
https://www.analog.com/media/en/technical-documentation/data-sheets/ADG1633_1634.pdf
G²
Attachments
Thanks. I also think having the buffers is worth having an additional IC. I just see so many posts where people suggest using just resistors, I started wondering if I’m overcomplicating things for no benefit.
The differential circuit is an interesting concept for a third channel. I wonder what that sounds like.
It's OK. I played with it way back in the '70s but it is no match for a true 5.1 system.
G²
Have successfully used single series 1 to 2 k (whatever is lying around) resistors for summing with good result. No power supples to fool with or parts to fail.
Thanks! That circuit looks like exactly what I’m looking for.
What type of filter is that low pass? Butterworth?
Also, is that a volume control IC that the inverting inputs from the first stage go to?
The filter is a Butterworth which you can tell by the 2:1 ratio of C15 and C8.
The LP filter is for when it is in Stereo (3.1) mode where the mono mix is for
the center and sub channels. They are switched by U3 controlled by an Arduino.
The original writeup from 8-29-2019 if you're interested. Search
5.1 channel preamp.
G²That's incorrect. A virtual earth configuration is widely used in audio mixers.
Yes, but you misunderstood the main question and my answer. It is clear that two or more signals can be mixed and this scheme is very old, from tube era. But the guy in the main question asked if one signal can go into the other creating only 3 mono signals. Or almost I undertand so.
Attachments
You have misunderstood, although why is beyond me.....
He just wants a mono signal from L + R so his schematic, with out the input buffers will work, but bearing mind the output is inverted so may need the sub woofer cable polarity swapped.
This is the main question sir:
"Is the buffer on the input necessary? I want to avoid any interaction of the two channels going to the preamp?"
Clearly an inverting mixer without the input buffers do the task thanks to virtual ground established at NI input of the amp. This also can be done with passive mix but there will be interaction unless both buffers are added.
"Is the buffer on the input necessary? I want to avoid any interaction of the two channels going to the preamp?"
Clearly an inverting mixer without the input buffers do the task thanks to virtual ground established at NI input of the amp. This also can be done with passive mix but there will be interaction unless both buffers are added.
Much depends on the output impedance of the source. It its low - i.e 600r of a preamp output, then there should be no interaction, and also, the 10k resistors of the summer could be increased to 22k or more.
If the input is from a high impedance source, then yes, the input buffers would be beneficial.
If two dual opamps were used, the 4th one could be used to invert the summer output to restore the polarity.
Also, since the summing point is a virtual earth, there can be no interaction.
Last edited: