Resistive mixing

[RobbieD]

I often use an mp3 player or CD walkman with a pair of powered speakers to provide back ground music in reception rooms. I use a a 3.5mm stereo jack to phone lead, connected to a home made phono to XLR leads (phono pin & ring to XLR hot & cold).

 

But next week I have limited space for speakers and intend to use a single speaker. The music will be 50's & 60's and includes songs with instruments, or even vocals in only one of the stereo channels, so obviously I need to mono the two channels before feeding in to the single powered speaker. (The speaker only has one XLR input.)

 

Can I make up a cable to resistively mix the two channels? My guess is that as the headphone output is low impedance, using the same resistor value on each channel will effectively see a potential divider down to ground, giving a 6dB (50%) signal drop from each channel. (But will the addition of the two channels give a 6dB increase in signal to make up to a similar level?)

 

Will such an arrangement work OK?

 

I was thinking resistors of around 100ohms each.

 

Many thanks in advance for the advice.

[paulears]

I had a headphone splitter that had 20 Ohm isolation with 2W ratings, but with speakers 2W won't be enough for even moderate volumes, and size wise they get a bit big as you go 'meatier' - 100 Ohms is great for isolation, but rather wasteful of available power.

[Rob_Beech]

But surely this split will be done at "line" level so the output of the speaker is irrelevant.

 

the same value resistor on each leg will be ok. its not ideal, a small passive mixer would be more suitable.

 

 

Rob

[RobbieD]

I had a headphone splitter that had 20 Ohm isolation with 2W ratings, but with speakers 2W won't be enough for even moderate volumes, and size wise they get a bit big as you go 'meatier' - 100 Ohms is great for isolation, but rather wasteful of available power.

As Rob says, this it at line level, combining the headphone output left and right to feed in to the powered speaker input.

 

the same value resistor on each leg will be ok. its not ideal, a small passive mixer would be more suitable.

Thanks, I'll make a cable up later.

 

I did consider using a line mixer, but wanted to keep this simple. (The mp3 player will be velcroed to the rear of the speaker.)

[dbuckley]

I have a very similar cable, stereo 3.5mm to mono standard jack, that I use to plug into a fostex speaker box, and I think the resistors are 1K each. The input impedence of most line inputs is several K, so there is not too much level lost from potential divider effect, a speaker out socket has enough voltage drive capability for it all to work Ok.

[david.elsbury]

I made one for my DVD player to combine into a mono TV recently, and used a pair of 1,1 k resistors as that's what I could easily get. Works just fine!

[ojc123]

I wonder if I can ask for a little education here? I'm intrigued.

 

I occasionally have to do something similar when I put a stereo source (often the line out from a minidisc) into a mono amp (usually small PA Mixer amp). I have a 3.5mm mono jack to 1/4" mono Jack lead which I plug in to the slave input of the PA mixer amp and the sound is fine. Both channels seem to be present in the mono 'mix'.

 

Have I completely missed the point of the question? Should I be doing something else? Thanks in anticipation.

 

A few hours later...

 

I’ve googled a bit and thought a bit. I think I see the issue. If I connect two inputs in parallel there will be a voltage difference between the two inputs. This will lead to a current between the two inputs. This would mean that either of the two inputs may be asked to deliver a higher output than it is designed to or may be accepting a current it isn’t designed to accept at all. This will cause distortion. I assume there are also all sorts of more complicated things going on involving reactance and capacitance too.

 

I suppose that putting a reasonable level of resistance into the path of each signal reduces the voltage between the two inputs thereby reducing the effect to negligible.

 

Am I on the right lines? I also guess I’ve got away with it because the two stereo tracks haven’t been too different. Also because the PA isn’t that brilliant in the first place so the distortion hasn’t been so noticeable.