- Posts
A little off-topic, but perhaps one of you could help. My transmitter is fed with a single audio cable (mono). My mixer and everything else is stereo because we stream. Thus, in order to preserve both left and right channels, I need a y-adapter going to the transmitter.
Problem: every time I use the y-adapter, from the “tape out” or even the “output 2” from my equalizer, it makes everything coming out of the mixer MONO. When I unplug it, it’s back to STEREO.
So, even though that’s the only wire going to the transmitter, when it’s plugged in, my stream and studio monitors automatically become MONO.
It automatically sums the left and right channel on everything… I just want it to sum the output to the transmitter.
What’s the easiest way around this?
Total posts : 45366The best way would probably be to get an audio DA (distribution amp) with isolated outputs. That way you could (in theory) ru one set of outputs into a dead short and not effect the other outputs. I haven’t found a passive device that I like yet, but maybe someone will come up with something.
Total posts : 45366An isolation transformer like one used in car stereo applications for ground loop issues would work. When you combine the lines after the transformer, They won’t be shorted togather at the source..
Total posts : 45366I tried that with some Altec audio transformers, no luck at all. It did not provide the amount of isolation that I needed. I forget the model number of those transformers, but they are the expensive model 600:600 ohm and 20-20K reponse.
Total posts : 45366The transformers won’t work because they pass the audio signals in both directions.
You might try this which worked for me. It depends on the output impedance of your source being low impedance (hundreds of ohms) and the input impedance of your load being high impedance (47 k or so). Connect a 47 k resistor in series with the center conductor of the L channel, and another 47 k resistor in series with the center conductor of the R channel. Connect the other ends of the resistors together and connect this junction to the center conductor of the load (your transmitter audio input). Connect all the grounds (outer shells of the plugs) together. If it works then you can get some RCA panel mount jacks, wire the resistors to them, mount them in an Altoids tin and you will be good to go.
There will be a little cross mixing but I doubt you will hear it.
Another way involves an operational amplifier summing circuit but this requires dual power supplies and is much harder to build. If available, the distribution amp. suggestion of Greg’s probably would work.
Neil
Total posts : 45366The resistors will work if you increase the value to more than 50K, probably around 100K. I tried this for another mixer I had to make, and the audio level was far too low for me to use. Even at 10K the signal was too low. Now maybe your transmitter has enough gain in the audio section to make up for the loss. If it does, then you will be all set. At any rate it’s ridiculously cheap to test. There are also other active devices that you can buy, but I’m not sure of the cost. Check Whirlwind or Benchmark to see what they might have for active (or passive) summing devices.
If you decide to build and active device, there are many good op-amps that can be powered with a single ended power supply. And if you are dealing with unbalanced audio, then the cost should be pretty low.
Total posts : 45366Greg,
How do you know the resistors will work if you increase them to 100 k which implies that the 47k won’t? Since you don’t know, as neither do I, the input and output impedances in his system it appears that we both are making educated guesses and I suppose we can honestly differ in our opinions.
My advice is to try it. Adjust from there if it doesn’t work. The 47 k resistors, if they work, will result in less attenuation than the higher resistance components.
True, the loss in signal will have to be recovered with the transmitter gain. This may or may not work but it is worth a try. By the way, this resistor mixer is very similar to that used in the SSTRAN AM transmitter where it works just fine.
Regarding the single supply op-amps, I assume, perhaps incorrectly, that your are thinking of the Norton amps which I have found can cause problems unless the audio signal is properly DC biased which really complicates the design. Perhaps that is a topic for another time.
Neil
Total posts : 45366With the smaller value resistors, you still get that little bit of audio going back the other way. I remember trying somewhere around the 47K, so 50 is indeed just a guess. And really so is 100K. In theory 10K should do the job, but hasn’t worked for me. I would start higher and work down to lower than start the other way around.
As far as the op-amps, many are just as happy running single ended supplies as dual, the key is to have the differential voltage with the operating range. If the range is 12 volts, it doesn’t care if you have +6 and -6 or a full +12 to ground. Some of the older designs did want to have the audio biased to a DC level, but many of the newer designs don’t really care. In other words you can use gorilla engineering to get the job done. It may not be pretty, but it works.
Total posts : 45366To convert Stereo to mono, a 10K resistor in the high side (+) of both left and right feeding a 10 K resistor high side (+) of an unbalanced mono audio feed will only drop the audio level by 6db.
If your mixer has 0dbu audio output or better, you should have enough audio to modulate the transmitter.
If your transmitter uses balanced audio, you will need to build the same network for the low side of the line output.
Also, many mixers have balanced audio ouputs. That means you will also have to convert the balanced output will need to be converted to unbalanced inorder to properly interface the mixer with the unbalanced audio input of the transmitter.
Knowing the output impedance of the mixer is key to being transmitter friendly audio. If the output is 600 ohms, then a 10K ohm input will cause the audio output to act like the audio level has been elevated. It involves the loading of the output amp of the mixer. Impedance matching of both devices will allow for better quality audio.
My AM transmitter is 600 ohm balanced audio input. The Mackie mixer has a 1K balanced audio out impedance. So, the resistive network above has served my purposes.
Marshall Johnson, Sr.
Senior Pastor, President
Rhema Christian Fellowship, Inc.Rhema Radio – The Word In Worship
AM 1660 – FM 93.5
http://www.rhemaradio.org
Total posts : 45366But are you using that same output to also feed a stereo device and keep proper stereo separation? That’s where I’ve had the issues. When I spilt the signal to another stereo device, and to the mono summing network, then the stereo path suffers.
If keeping a stereo signal is not important then you can use around 100 to 120 ohm resistors to make the network, this will keep the signal level approximatly at the correct level.
For active devices, I just looked for a couple things and found this:
http://www.opamplabs.com/va14sum6.htm
http://www.opamplabs.com/4s4x10.htmI’m not sure if the second one will do the job, but if it does the price is right.
Total posts : 45366To the original poster, you have received many suggestions here. Don’t get too wrapped up in the technology. Just try what looks reasonable and see if it works. I would like to know if our suggestions were helpful.
Neil
Total posts : 45366
Great point Neil.I’ve used this circuit (adapted from THIS ARTICLE) to combine stereo output into a mono input with a passive circuit.
I have also used resistors of lower Ohm values for inputs and outputs lower than line level.
1/8th watt resistors seem to work OK at low audio power levels
The circuit seems to work equally well as a splitter or a combiner.
Experimental broadcasting for a better tomorrow!
Total posts : 45366I tried something similar in my Kluge Box that combines 3 1/8 inch stereo sources to a single stereo source for use with BSI’s Simian automation. It might have been the sound card I used, but the level got so low that I had to change to a lower value. I think I ended up with something around 250 ohms as a good compromise between level and what should be proper engineering. I tried something similar with a stereo to mono where the stereo side still needed to feed a stereo device. This didn’t work so well because the stereo side lost a lot of separation.
For stereo to mono where you are not trying to protect the stereo side for other devices, these types of passive devices work really well, especially considering the price.
Total posts : 45366The reason you want to use a high resistance in the network is that the stereo separation depends on low source impedance and high mixing impedance to keep the source channels separated. There will always be some mixing with these resistive networks and it becomes a matter of experimentation or calculation to tradeoff the signal attenuation in the mono channel against the loss of stereo separation in the stereo channels. Higher mixing resistance equals more separation but less mono signal. A quick off the cuff calculation shows that with 100 ohm source Z and 47 K mixing and mono load resistances the stereo cross channel mixing will be below -57 dB. This also shows that the attenuation of each stereo channel signal into a mono load of 47 k will yield a loss of 9.5 dB for each channel but since the two channels add in the mono circuit the overall loss is 3.5 dB. This can easily be compensated if the mono circuit has reasonable gain.
This does not apply to a 600 ohm balanced system in which case a new set of calculations would be needed and it is my guess, without having done the calculations, that if the input and output Zs are 600 ohms, the mixing vs. mono signal loss will be unacceptable.
My experience is that using a sound card and 47 k mixing resistors in a single ended low output Z, high input Z system works adequately. Your results may vary.
Neil
Total posts : 45366Maybe that’s my problem. Everything I do is balanced audio, though very little is actually terminated into 600 ohms. The trend is to leave it unterminated which gives better headroom, better high frequency response, and lower power consumption. Most devices are now some sort of active buffered high impedance inputs and active balanced outputs.
- You must be logged in to reply to this topic.
