3 ways buffered signal splitter based on TL072

[HeaD]

Hi, I recently built a 3 ways buffered signal splitter based on two tl072 opamp's.

Here's the simple schematics I sorted out:



As you can see I used only the first stage of one of the opamps.

I actually built two of these, for two different channels. The goal I would like to achieve is to split the +4dB output signal of a mixing console.

At low volume levels the circuit does its job, but by increasing the output volume, the sound begins to get distorted.

I wonder if it could be a problem due to a physical limit of the opamps, or maybe in my poor design?

Thanks to all!

[Kipper4]

At low volume levels the circuit does its job, but by increasing the output volume, the sound begins to get distorted

How are you increasing the output volume?

If you have a spare opamp in the tl072 why not use it to make a buffered (1/2V) V bias?
then you can lower the value of the 2 x 10M used as your V bias in the above schematic and maybe it will have less thermal noise too.

[HeaD]

How are you increasing the output volume?

If you have a spare opamp in the tl072 why not use it to make a buffered (1/2V) V bias?
then you can lower the value of the 2 x 10M used as your V bias in the above schematic and maybe it will have less thermal noise too.

Hi Kipper4, thanks for your reply.

I have no spare opamps as I built two of this circuits using 3 TL072s. In fact I was not very clear: I meant the output level of my mixing console, which is the input volume of the splitter. I increase the volume directly from the master output of my mixing console.

Any help is appreciated, thank you.

[Kipper4]

Maybe a 0db output rather than +4db would work better if you have one. just guessing and clutching at straws here with regard to the distortion when you bump up the volume.

I guess it depends on what your feeding this to and from also !!!!!!!!!!

I see what you mean now and understand why you have no spare op amps, my bad.
It was just an idea.

[HeaD]

Maybe a 0db output rather than +4db would work better if you have one. just guessing and clutching at straws here with regard to the distortion when you bump up the volume.

Yeah, you're right, it works nice with a standard -10db input... already tried this out. Unfortunately this is an old analog mixing console that works only at +4db.

So you think that might be a physical limitation of the TL072? Maybe it doesn't tolerate a so high input level...

Thanks!

[Kipper4]

"So you think that might be a physical limitation of the TL072? Maybe it doesn't tolerate a so high input level..."
Not really sure
Have you tried a resistor to ground on the input to attenuate the the +4db signal?
or a series input resistor which would help isolate the op amp input some too.
Easy enough to do temporarily with some patch cables and some alligator clips and said resistor
This might affect the circuits input impedance but then its quite high anyway.
Have you tried any other resistors than the 10M to create the V bias?
or was that a trial and error, what worked best on the breadboarding stage of the design?

[slacker]

Yeah +4dB level signals could be high enough to clip a TL072 running on 9 volts. If you can try running it from a higher voltage supply.

[HeaD]

Yeah +4dB level signals could be high enough to clip a TL072 running on 9 volts. If you can try running it from a higher voltage supply.

Hi, slacker. Thank you for your reply.

I tried with a 12V supply too. Same result.

Maybe I should try to reduce the 10M resistors value as suggested by Kipper4.

Anyway, could you suggest me some TL072 alternatives for this purpose?

[samhay]

If the problem is that your signal is strong enough to hit the rails (op-amp clipping), then you can try a rail-to-rail op-amp (google will find lots of options). This will only buy you a few extra V headroom. The better solution is to use a higher supply voltage and/or turn the signal down.

[karbomusic]

If you are going to the trouble to make a splitter, you should go through the additional trouble of make some or all of them phase/polarity invertible.  Eventually, your going to wish you had that feature unless this is a one off specific and static solution. See my post in this thread.

http://www.diystompboxes.com/smfforum/index.php?topic=112268.0

I'm not suggesting my design is some broad stroke of genius - surely far from it, but it's the only example I have built that I know works really well. However, that design is going to get more complex if you needed to add gain. As a side note, I think I ended up using an OPA2134 in mine for no particular reason.

[Kipper4]

Would a ne5532 work here
Lm358 or the lm324 ?

[PRR]

"+4dBu" is nominal 1.228Vrms or 3.474V peak to peak.

*However* the "+4" assumes an averaging meter. Speech/music peaks do not fully register.

We normally allow 10dB to 20dB over nominal level to account for peaks. In live broadcasting 10dB is accepted (we only hear a distorted peak once, so it is not annoying). In multi-track studio work we need 16dB or 18dB headroom above nominal level (because we hear the same peaks distorted over and over).

18dB above +4dBu is 9.76Vrms or 27.6V peak to peak.

With a "perfect" amplifier you need a 27.6V (not 9V) battery.

Real amplifiers lose some both ways. Allow 2V each side, 4V total. On top of 27.6V this is about 32V total.

Indeed opamp based consoles use +/-15V or +/-18V (30V to 36V total) power supplies.

And they often work two opamps in opposition, so a balanced load can get double voltage.

> TL072 alternatives

Different opamps make "NO" difference. Yes, some may only lose 2V instead of 4V, but that's little help if you really need 28V total swing.

BTW, TL072 is happy with a 30V supply, 36V if it "never" goes higher.

> an old analog mixing console that works only at +4db.

Sure it works at -10dBV..... it works with signals from -70dB (quiet) to over +4dB (loud). Pot-down so the meters show around -10VU to -6VU max.

1) any decent console CAN drive multiple loads, especially modern >10K loads. Just use Y-adapters. If an output may be shorted and you can't risk losing signal to another output, add a 1K resistor in series with each. (This is what most Distributions Amps do.)

2) If you can't stand to see meters never touching red, then add a pad between the console output and your 9V distribution amp. 4:1 may work. 33K plus 10K. But you still can only get about 2.5V rms output on peaks with just a 9V supply.

[HeaD]

1) any decent console CAN drive multiple loads, especially modern >10K loads. Just use Y-adapters. If an output may be shorted and you can't risk losing signal to another output, add a 1K resistor in series with each. (This is what most Distributions Amps do.)

Hi PRR, thank you for your reply.

So, if I understand you right, I can simply use a passive "Y" or three ways adapter for my purpose. I had not considered that at all...

[ashcat_lt]

Hi PRR, thank you for your reply.

So, if I understand you right, I can simply use a passive "Y" or three ways adapter for my purpose. I had not considered that at all...

Or use the jacks and box that you had for this project and just connect them with straight wire.