Simple Stereo Audio Mixer (Two RCA in, One RCA out)

Discussion in 'The Projects Forum' started by erasei, Aug 12, 2009.

  1. erasei

    Thread Starter New Member

    Aug 12, 2009
    9
    0
    Hello all, First off, just wanted to say what a great forum this is. I just found it a few hours ago while searching the web looking for help solving the problem I'm posting about. I've spent the last little while here going through all of the older posts that are at all related to my problem. I found a few close ones, but nothing that helps me directly. With that said, on to my problem.

    For the sake of example, let's say I have two ipods I want to be mixed together and sent out one pair of speakers. I want to preserve as much of the music quality as possible, but also to be as simple as possible. I know my way around a soldering iron but I'm still pretty new to circuits. I do a lot more 'repair' than 'creating'.

    There are a ton of schematics online (and some in the forum here) that deal with audio mixers, but most of them are geared towards muscians and deal with mics and mono outputs. That doesn't help me.. but I was thinking that a mono signal is (for my practical purposes) just half of the stereo signal. So if a 4 channel mono mixer worked, why wouldn't that same circuit work as a 2 channel stereo mixer?

    Well.. it doesn't. I built one last night and one input always sounds great, while the other one sounds terrible. I could change ipod1 to input1, ipod2 to input2, and no matter what combination, input1 would always sound great and input2 would always sound crappy.

    I actually built the circuit twice, with the same result on both.. so I'm hoping you guys can help me out. Is my circuit design wrong? I wouldn't be surprised as I'm just sort of hacking at it. Can you give me a good fundamentals lesson on why its wrong? I'm eager to learn.

    I've attached a rough drawing of my circuit design for reference.
     
  2. bertus

    Administrator

    Apr 5, 2008
    15,648
    2,347
    Hello,

    The problem is perhaps DC on the signal.
    You can get rid of the DC using capacitors in series with the resistors.
    A capacitor of 10 uF will work. ( + to the input ).
    Another problem can be the output and input impedance.
    Did you try other values for the resistors?
    What will happen when you make all resistors 1K?

    Greetings,
    Bertus
     
  3. erasei

    Thread Starter New Member

    Aug 12, 2009
    9
    0
    Hi Bertus, thanks for the reply.

    >The problem is perhaps DC on the signal.
    >You can get rid of the DC using capacitors in series with the resistors.
    >A capacitor of 10 uF will work. ( + to the input ).

    DC being Direct Current? Would this be the DC coming down the Signal line on the RCA jacks? Otherwise, this is a passive circuit.. no added DC.

    If so, I'll try this tonight. I'm sure i have a few baby caps laying around I can test with.

    So instead of what I have currently of:

    Input Signal >> resistor >> Output Signal

    it would be

    Input Signal >> resistor >> +Cap- >> Output Signal, correct?


    > Another problem can be the output and input impedance.

    If the input impedance was the problem wouldn't using a different source cause a change? I tried last night with my iphone playing and with my old favorite standby mp3 player.. a 1998 Diamond Rio PMP300 with 32mb of flash ram. (yes, it's awesome). It didn't matter which one was playing, they both sounded either good or bad, depending on which input they were connected to. However, i always used the same output, my old sony receiver connected to a few speakers.


    >Did you try other values for the resistors?
    >What will happen when you make all resistors 1K?

    Yeah, my first circuit I made with 4k7 resistors as per one of the schematics i found online. The sound came out very muffled and not very good at all. The second time i made the circuit i used the 10k resisters and it sounded better.
     
  4. bertus

    Administrator

    Apr 5, 2008
    15,648
    2,347
    Hello,

    I mean the capacitor all at front.

    input signal >> +cap- >> resistor >> output signal .

    It can be that there is a little DC at the output of the MP3 player.
    This can influence the input of the amplifier.

    Greetings,
    Bertus
     
  5. erasei

    Thread Starter New Member

    Aug 12, 2009
    9
    0
    Sweet! I'll try that tonight.
     
  6. erasei

    Thread Starter New Member

    Aug 12, 2009
    9
    0
    No Luck!

    I just tried it with every cap and resister combination I could think of. All with more or less the same result. I did get a little more information. Previously I said that one input was stronger than the other the other, when actually whichever one is plugged in is strong when the other input is unplugged, regardless of which input it is.

    For example: If input1 is active and playing it sounds great. As soon as I plug in input2, even if its powered off, input1 gets muffled. My understanding of the circuit was that both inputs would get mixed together without signal degradation. I know the resistors will cause /some/ impact, but it should not be audible.

    I first started off with the caps like you said. I don't have 10uf caps, so I started off with 4.7uf caps, which i have a bunch of. When that didn't work, I tried to find some other caps, and only really have some 2.2uf caps and some 220uf caps. None of the three work. Really no difference between any of the caps or just using bare wire connections (ie, no caps).

    Next i tried different resistors, going down to 10 ohm, then up through the series of 220 ohm, 1k, 10k, 22k, 100k and 470k. For 10o - 10k there was no difference. 22k and above were apparently too large because no sound got through at all, regardless of the other input being plugged in or not.

    I really have no idea what's wrong. Any ideas?
     
  7. jj_alukkas

    Well-Known Member

    Jan 8, 2009
    751
    5
    I guess you will need an active mixer. Its the input impedance s that is causing the problem. OR you can use 2 I/O audio transfomers to mix the signals.
     
  8. erasei

    Thread Starter New Member

    Aug 12, 2009
    9
    0
    Yeah, I think you might be right.

    Do you guys have any schematics for an active mixer? The only ones I can find online are ones that are more complicated, ie more features, than I need. As a newbie I'd like it to be just as complicated as it needs to be to meet my needs.
     
  9. bertus

    Administrator

    Apr 5, 2008
    15,648
    2,347
  10. flat5

    Active Member

    Nov 13, 2008
    403
    17
    Have someone else check your wiring, or you wait awhile and check it. Check any plugs and sockets and cables for shorts.
    Do you have an ohm meter?

    Your simple resistive mixer should work.

    You could also use a multimeter to see if there is any dc voltage coming from the Ipods
     
  11. jj_alukkas

    Well-Known Member

    Jan 8, 2009
    751
    5
    If you need it to be passive, try using 2 Audio transformers and mix them. Then input impedance will be no problem.

    However, if you need an active one, here is one. Sorry, its the simplest one I could design for 2 inputs. You will need a pair of these for stereo.

    [​IMG]
     
  12. jj_alukkas

    Well-Known Member

    Jan 8, 2009
    751
    5
    Hey here's also a simple passive mixer. Adding a cap in front of each input would also be good...


    [​IMG]
     
  13. Audioguru

    New Member

    Dec 20, 2007
    9,411
    896
    Aaron Cake's FET mixer circuit (copied at Electronics-Lab) is a lossy piece of junk.
     
  14. erasei

    Thread Starter New Member

    Aug 12, 2009
    9
    0
    Thanks for all the replies.

    >flat5: Have someone else check your wiring,

    This is a great idea. I'll try to breadboard it up tonight in a way that makes for easy and clear photographing and post it up here tomorrow, assuming I have time tonight.

    >jj_alukkas: <<schamatic>>

    That looks pretty much like what I have already, except I omitted R1-4 as I understand those pots/attenuators to just be for gain/volume control which I don't need.

    One difference is that I keep the Shield of the Left channels of both Inputs connected only to the Shield of the Left output channel, and the same for the Right inputs to the Right output, and in the schematic you attached all shields are connected together. Think that matters? I thought the shields were just basically grounds (if anything).

    As far as active/passive it doesn't need to be either really, passive is just easier.. don't have to worry about another wall wart.
     
  15. jj_alukkas

    Well-Known Member

    Jan 8, 2009
    751
    5
    All shields for line level inputs connect to common ground. So it would remove any hum. The pots R1-4 are not simply for volume control. They adjust the levels such that both the audio sources give similar output. Otherwise one will be louder than the other.
     
  16. erasei

    Thread Starter New Member

    Aug 12, 2009
    9
    0
    I didn't have a chance to work on it last night (too busy drinkin beer and watchin football *go Pack!*), but I'll try to wire up all of the shields together and see if that changes anything.

    I think I might have a few pots lying around, I'll play with adding them to the mix as well, but I don't think my primary problem is pot related. Either input sounds great until the other input is plugged in, even if the source is turned off. So it sounds more a problem of impedance than volume of the source. The common grounding might help though.. we'll see.
     
  17. hgmjr

    Moderator

    Jan 28, 2005
    9,030
    214
    A source of signal attenuation that is frequently overlooked in these passive mixers is that the resistor in series with input1 and the resistor in series with input2 cuts each of the inputs in half when the resistors are both equal. This assumes that the source impedance of the signal sources are very low.

    You can think of it as the output at the common point is the superposition of the two input signals each of which has been attenuated by 50%. Again, this assumes that the two series resistors are equal. Also this assumes that the input impedance to the stage into which the mixed signals are be fed is very high.

    hgmjr
     
  18. erasei

    Thread Starter New Member

    Aug 12, 2009
    9
    0
    hgmjr: You might be on to something. All 4 resisters (each of the two series) are all 10k. Input impedence is likely very small, both coming from portable music players. Output impedence is probably fairly high, going into a Sony 100watt receiver driving 2 12" woofers (both with Low crossovers), 2 5" Fosgate Punche midranges, and two generic 2" tweeters. (This is all in my workroom, not the ideal setup, but ya work with what ya have)

    I just got a few minutes to re-breadboard for clarity. The circuit photo is attached. All advice, including 'its all wrong stupid' is very welcome!
     
  19. jj_alukkas

    Well-Known Member

    Jan 8, 2009
    751
    5
    I have another suggession. Take the L of 1st input and L of the 2nd and give each through a 100n disc cap and combine the outputs to the amp. You might get through thatway. If it doesnt work, replace it with a 0.1uf disc or 1uF electolytic. GND needs not have the capacitor.

    Another possibilty is that the outputs of your mp3 players might be reversed. Test that also. That really causes noise
     
  20. erasei

    Thread Starter New Member

    Aug 12, 2009
    9
    0
    Sorry for the delay in following up.. just haven't had a chance to get back to this project.

    I had to track down some 100n caps and I just tried it out. Apparently the 100n is too large because it cuts off the sound completely. I tried it with the cap and a resister (2.2k) and without a resister, and neither worked (both blocked the sound entirely).

    I don't have any 0.1uf caps so I can't try that.

    I also tried all of the combinations of inputs and outputs to see if any were reversed.. no difference there either.

    I'm about to just give up on this project.. *sigh*
     
Loading...