Need to know about audio processing

[iso9001]

Hey,

Have a project that is kinda frustrating since I know little about analog circuits.

Have a line level output coming from a dac (0-4V) but its a little odd b/c it seems to have a flat carrier signal of about 2.75V. And I would like to output a crossed over output at about the same 0-4V but with some basic EQ added. So 2 stereo pairs come in, 4 stereo pairs go out (Left-Hi, Left-Low, etc etc)


Its some stupid custom dj equipment that was made by god knows who and I have a friend that wants to get this hooked up to a car amp.

Now, the problem is for one I dont know what this carrier is and if its ok to feed it along to the amp or not.

If I can get that sorted out I would like to add a simple EQ to the output (bass, treble, volume control, fade, balance, etc etc), nothing real complicated.

But I'm not sure if there are ICs that handle this or its all going to be separate circuits for each function. I know there are some ICs to do this when its a digital format, but don't really know what to look for if the signal is already in analog.

Anyone have any ideas ?

 

[audioguru]

The "flat carrier" might be some of the digital clock leaking out. It might destroy tweeters or even kill an amplifier.

You could use an LM1036 tone controls IC:

 

[iso9001]

I'm pretty sure the 2.75V carrier is not a problem. Rather a design intent. The whole system came with an amp but its gone now. So I don't really know what was in it, but it was a modified car amp. I think it may have had a dsp or some sort of other unusual processor to find in an amp. But it was really low wattage anyway, so.... a new car amp would be a perfect replacment, but I have no idea why they would have wanted a carrier


I understand why a flat signal like that might be harmfull to a tweeter. I'm thinking maybe it might be removed prior to amping or for all I know a good amp might remove that already.

I think this was all designed so some sort of standard. But maybe just some made up goofiness. Either way, I'm pretty certain its supposed to be there.


Thanks audioguru, I like that chip. It has balance but it seems like its missing fade though (?). I guess if they only planned two speakers then they wouldn't need fade.


I have a left +/- and right +/- so I think I would need 2 of these chips. Then take those two outputs and use some sort of crossover circuit to cut them up so I have a front highs, front mids, rear mids, and maybe even a sub output. Should just be where and how many crossovers I have I would think.

I'll try searching for other audio control chips, maybe someone has an expensive all-in-one ?

 

[audioguru]

I have never seen an audio processor IC that has crossover circuits and is analog. Car audio equipment all use DSP now.

 

[iso9001]

I guess I could do this with 4 of these ICs (if I want the fading, I think I need to use 4), or DSP.

I know more about DSP then I do audio so thats good, but still not enough to feel I can do crossovers right now. That'll take some more reading.

With DSP the idea would be to sample in the audio then apply my 'effects' (fade balance volume crossover) to each sample then output that (ac97?) to a codec ? Or am I way off ?

I think I could also fix the carrier problem with dsp.

Too bad, I like these little IC's so much more


I think I need to find out why there is a 2.75 signal there at all. I'm starting to think the original system used a DSP.

 

[audioguru]

You only need one stereo tone controls circuit. Then the faders are just a stereo potentiometer that feeds the 4 crossover circuits.

I wouldn't know how to connect nor how to program a DSP.

 

[iso9001]

Now thats the thing that confuses me.

I have left + and - and right +/-. These are very similar to preamp signals (rca back of a stereo or something) so I thought that they were a differential pair

I don't have the unit yet so I'm going off what my firend is telling when he hooked his scope to it. I really need to get my scope on it i think (!)

Are preamp signals that travel over rca cables usually a signal and a dedicated ground or differential pair or AC ?

When measured against each wire its 4V DC but thats all I was told.


Ifs is signal and return then I see I only need 1 chip. If its differential pair or AC I need a circuit to combine that signal (opAmp ?)


If its DC I think I'll look into DSP and tone chip. DSP allows much more flexibility, but I gotta be snappy on the sampling. Its actually pretty easy I think. I beleive there are a lot of DSPs that have native support for reading a continuous stream of samples in as AC97 or I2S format, that gives you a packet with your 'sound' in it. Apply an effect (bass +1) and ship the packet in whatever format to a coder/decoder that intergrates a dac. Feed that to an amp.

On the other hand tone chip might add more noise, but its damn simple.

 

[audioguru]

Two pairs of outputs marked "+ and -"?
4VDC on the outputs?
"Its some stupid custom dj equipment that was made by god knows who"?

You need to find out what the signals are, I sure don't know.

 

[Nigel Goodwin]

The description is next to useless, but it sounds like it might just be sat on a DC level - no coupling capacitors, they may have been in the second half of the unit. Adding coupling capacitors may be all that's needed?.

 

[iso9001]

Yea,

I know I need to get my scope on it. I can't go off what I'm being told cause its incomplete.

But, I did figure out that is DC signal, left and right channels. 0-4V with a 2.75 carrier. I don't really understand the carrier, what its for or how to get rid of it.

I also looked into DSPs, I don't really have the time for that. Its not really complicted, and I may look into it later but for now I think tone control IC is the way to go for now.

Just need to loose the carrier. Not sure how to do that. Maybe ground loop isolator, ? but I'm thinking thats not it.

I looked into crossovers too, but thats a lot of opAmps and r/c parts, think maybe I'll leave that out. Maybe just a low pass for a sub.

Any idea on that carrier ? All I really need is something that will let the AC pass and stop the DC. I think I should know the answer to this.

 

[Nigel Goodwin]

But, I did figure out that is DC signal, left and right channels. 0-4V with a 2.75 carrier. I don't really understand the carrier, what its for or how to get rid of it.

Try reading my post immediately above your last one!.

I suspect you're confusing everybody (especially yourself!) by keeping referring to 'carrier'.

 

[audioguru]

Is "the carrier" actually 2.75VDC?

You mentioned mids and highs. A tweeter and a midrange speaker need a crossover network that can be either active or passive.

 

[iso9001]

Nigle: I misread your post sorry. I see what you're saying. If I add a coupling cap to it I can block the dc but its going to take some work to figure out the size, otherwise I may end up blocking audio or on the other end allowing some of the dc to get to the amp. When I was explained about the project the term carrier was used. But you're right I'm confusing everyone.

Why would they do this ? Add in a 2.75V signal only to cut it out once its inside the old weak amp ? Noise immunity ? The whole thing ran over an rca cable, center pin signal, shield ground, I can't imagine they would do this for no reason, only thing I can think of is noise. Its always about noise


Guru: I need to look into whether I want to add the crossovers in or use store bought stuff. Most amps have either a high or low pass filter anyway. I saw some good crossover circuits but they have a lot of parts and I wasn't thrilled about that.

I want to get this thing running first

 

[Nigel Goodwin]

Nigle: I misread your post sorry. I see what you're saying. If I add a coupling cap to it I can block the dc but its going to take some work to figure out the size, otherwise I may end up blocking audio or on the other end allowing some of the dc to get to the amp. When I was explained about the project the term carrier was used. But you're right I'm confusing everyone.

Why would they do this ? Add in a 2.75V signal only to cut it out once its inside the old weak amp ? Noise immunity ? The whole thing ran over an rca cable, center pin signal, shield ground, I can't imagine they would do this for no reason, only thing I can think of is noise. Its always about noise

You're confusing yourself again - it hasn't been 'added in', it's how amplifiers work - you have coupling capacitors to pass the AC and block the DC levels.

Usually, for seperate units, you would have a blocking capacitor on the output of the preamp, and another on the input of the power amp - however, only ONE is actually required (as long as they are only ever used with each other).

In your case it sounds like the blocking capacitor was in the following section, and not in the preamp.

There's no calculations really required, ANY capacitor will block the DC, it's just a question of a big enough one to pass the lowest frequencies you need. To calculate it you would need to know the input impedance of the following stage - but I suggest you stick a 10uF in there, with an electrolytic connect the +ve to the output of the preamp (where the 2.75V DC will polarise it), and the negative to the following stage - you may ned to add a resistor from -ve to 0V, if there isn't one in the following stage (not critical, 47K would be fine).

 

[iso9001]

Thanks Nigel. I'll try that asap.

I guess what I was wondering is why didn't they add the coupling cap in the preamp ? I mean, they obviously assumed that no one would ever use anything other then thier amp. But, I wonder if thats the only reason ?

Say for instance you have a 15ft stretch of cheap twisted pair wire before you got to the amp. Could it effect the noise immunity at all by leaving the 2.75 in there ?

 

[audioguru]

Cheap twisted pair is used for balanced telephone lines, not unbalanced audio that needs shielded audio cable to avoid interference pickup. The DC level on the signal could cause distortion, not noise, in an amplifier following it if there isn't a coupling capacitor to block the DC.

 

[Nigel Goodwin]

I guess what I was wondering is why didn't they add the coupling cap in the preamp ? I mean, they obviously assumed that no one would ever use anything other then thier amp. But, I wonder if thats the only reason ?

Yes, that's the ONLY reason - and to save a few pence!.