Adding Aux input to older car stereo

[infectme247]

Hi everyone, this is my first post. I don't see a beginners' section to start a new thread, so I'll post here, as this one seems very relevant.

I recently modified my old car stereo to create an AUX audio input, by mounting a female 3.5" jack to the head unit and soldering connections from it directly to the PCB. It works, and the sound quality is great in general, but there are some issues that I need to understand and resolve, and this is where I need your help.

The concept was to be able to use my mobile phone as sound source, by connecting its headphones output (also 3.5" female jack) to the head unit, with a simple "patch" cable. The main issue is that when I first plug the cable to the phone, it correctly indicates "headphones connected", but when I plug the other end to the head unit, the phone "loses" the "headphones", and any subsequent music reproduction occurs only from the phone's speaker; no signal is sent to the car stereo (*) although both ends of the cable are connected.

Then, in order to get sound as intended (from the car's speakers) I need to unplug the cable from the headunit, while music is playing, and plug it in again. It doesn't work if I unplug it from the phone. Only from the head unit. However, this seems to be an issue with the phone, since a portable mp3 player that I tried, will work no matter which end of the cable I reconnect.

The point is that, either with the phone or the mp3 player, once the source starts playing, there is no sound from the car speakers unless you reconnect the cable at some end. If the source was already playing when I first connect the cable, then no problem!

What is even weirder (or at least, it seems to me) is that, when it's all playing fine, if I stop playback from the source for a while, the same thing happens: I need to reconnect the cable at some end, or music won't come off the speakers. Can anyone explain why this happens? I'm sure it will prove to be the "expected behavior" afterall, but I can't get my head around it.
My best guess is that, somehow, the original signal path in the car stereo takes over when there is no input from the AUX IN, (although its volume is set to minimum, therefore inaudible) but why do I need to reconnect the cable? Is it a problem with the ground?

Another minor issue is that I get a lot of static/crackling noise whenever I plug the cable. I remember this could be remedied with a capacitor in series at the input, right after the jack... correct? Anyone remember any ballpark figures for line level audio?

(*) I suppose that no signal is sent (I have not measured it with a multimeter), based on the fact that my phone cannot play audio simultaneously from both the speaker and the headphones.

Thank you in advance for any answers, I did my best to describe the problem as simply as I could, I hope it makes sense... Also, please try to keep the words small and the terminology to the minimum, english is not my native language and I don't have much knowledge about electronics... it had been quite some time since I last did any soldering... Anyway, any insight is highly appreciated. Thank you

edit: I was planning on uploading a schematic, but it's quite simple. 2 connections from the female jack's tip (T) to 2 points on the PCB (for the front left and rear left speaker), 2 connection from ring (R) to 2 points on the PCB (for front right and rear right) and 1 connection from the sleeve (S) to the chassis for ground.

[Mod Note: placed in a new thread]

 

[Externet]

Welcome.
The 'older' part in your question can be a problem to guide you.
"Older" with an integrated circuit selecting audio path inputs want to see no DC riding on the audio. If there is some DC with the source, can go into hiccup with the behavior you describe.
"Older" with rotary potentiometer volume control are more forgiving.
A key for proper insertion of an auxiliar signal is to use a 3mm cancelling jack. Jacks typically canibalized from sound cards are the ones to use. They have 5 pins. The injection point to the stereo is by the 'high' sides of the volume control, or the microcontrolled audio routing integrated output that feeds the volume control for the power amplifier.
The cancelling feature contacts disconnect the internal audio source in the stereo audio path enabling only the inserted plug as feed.
The telephone audio output stage can also have a protection circuitry muting the audio when inserting the plug as it usually short-circuits the audio briefly. The hookup location on the PCB may also have DC riding on the audio triggering protection on the telephone.
Line level audio is around 1 Volt peak; 0dBu; 0.7Vrms.

 

[infectme247]

Hi Externet and thanks for the reply and welcome.

The car stereo should be 12-14 years old (clarion rd3-01) and has no rotary controls (e.g. volume or tuning pot); all user input seems to be IC based.

I am sure I've bumped into the type of jacks you are talking about, in designs I've seen in the past. After a bit of googling, they seem to go by the name "closed circuit audio jack". The problem is, I've tried really hard to fit the existing jack neatly and snuggly on the head, and I don't know how I'll manage to fit another one on the same hole and space, and apart from that, I understand that I will have to cut tracks on the pcb, in order to "insert" the audio jack into the signal path and then back again to the PCB. I would like to avoid that if possible, as I don't have any similar experience, and I'll probably ruin the whole thing.

With respect to any DC offset in the source, isn't it weird that it happens with 2 devices? both the phone and the mp3 player? This is what you mean here, right?

If there is some DC with the source, can go into hiccup with the behavior you describe.

If I insert a couple of these circuits (http://goo.gl/acGvKH) between the Tip to PCB and Ring to PCB, should it be enough?

Line level audio is around 1 Volt peak; 0dBu; 0.7Vrms.

I'm sorry I wasn't clear enough, I meant what capacitor values to use in series with the jack, in order to prevent the pop/crackling sound when I plug or unplug the cable. (if it works in the first place)

Unfortunately, the whole troubleshooting process is quite "black box" to me, and I can't make much use of the multimeter, since I have to put the stereo back together and take it to the car, to see if it works. I thought that it wouldn't work outside the car, with only the proper voltage supplied (without the iso connectors from the cars) and furthermore, I don't have any appropriate speakers to test it with.

Thank you very much, again.

 

[infectme247]

I'm giving the "closed circuit jack" approach a second thought, and it really seems like the only way to go in my case... However, I'm afraid I won't be able to pull it off, so I've started reconsidering my original approach, which was to use the stereo's CD changer input, and "stream" the audio through it. This had obvious advantages, but unfortunately it didn't work, that's why I ended up with the connections on the PCB .

After reading a few articles on the web and looking at the pinout for the ISO connectors (see connector C here: https://en.wikipedia.org/wiki/Connectors_for_car_audio) I connected the pins 13-15 with a resistor in between (this was supposed to trick the stereo into detecting a cd changer) and also connected the external audio signal and ground to pins 18-20 according to the diagram. I tried resistance values ranging from 5KΩ down to 45Ω but the stereo never "detected" a cd changer, so I quickly abandoned it and opted for the connections on the pcb instead.

I read at some car audio forums that I had to activate the CD changer port through the car's ECU first, for this to work, but when I took it to the local dealer/workshop, the technician insisted that this was not possible or necessary for my car model, thus the feature was not "locked" in any way, and I had obviously done something wrong since it doesn't work...

Any thoughts on that? anyone?

 

[Externet]

Hi.
I consider a car stereo with a microcontrolled audio path integrated a 'newer' model.
My 2000 LandRover received the surgery for an external audio source, only after understanding the data sheet of its audio path management integrated circuit. Better if you start there learning how it works in order to harness it properly.

What I did is eliminate the cassette selection and fed the audio management IC input pins corresponding to cassette, as auxiliar jack input.
The cancelling jack is then not necessary. Pressing the cassette button enables now the auxiliar jack.
A plain 1μF capacitor in series with the jack audio may suffice if any path wants to see no DC.
Clicks and pops can also be from other reasons.
Guessing resistors is guessing that will hardly yield any result. The CD changer usually wants to see a serial data handshaking with the head unit. The same serial data that displays which CD slots are loaded and their titles and tracks.

 

[BobK]

I think the problem might be that the phone is not sensing headphones because the impedance of the input to your car radio is too high. Headphones are typically 32Ω. A line input is typically 10KΩ. If this is the problem, you can correct it by placing a resistor from each channel output to ground. A 32Ω would certainly fool it. A higher resistance, like 100Ω would likely work as well.

Bob

 

[infectme247]

Hi.
I consider a car stereo with a microcontrolled audio path integrated a 'newer' model.

I'm sorry if the title was misleading, I should have mentioned the specific model from the start. The connections I've made are almost at the end of the signal path to the speakers, so I have completely bypassed any source selection from the head. (radio, CD or CD changer) I guess I'll need the datasheet for the IC to see what options I have, if I try to hijack any specific input channels instead...
The problem is I don't want to lose the radio completely, the CD player has been removed long ago (it didn't work) so I don't know if the IC will even bother reading the corresponding input pins anymore, and the CD changer doesn't seem to work without any data/protocol transmission... I guess I'm running out of options!

Also, I didn't guess the resistor values, I read posts by people who claimed they got it to work. Some said 1KΩ worked for them, others as low as 60Ω. I used a potentiometer and started with a (safer) high enough value, and then worked my way down, until the CD changer would be detected... No luck unfortunately, and it makes sense that there should be some "handshaking" going on first... I think the CD player works this way, too.
For now, I'll try the 1μF capacitors, and see if that helps.

I think the problem might be that the phone is not sensing headphones because the impedance of the input to your car radio is too high. Headphones are typically 32Ω. A line input is typically 10KΩ. If this is the problem, you can correct it by placing a resistor from each channel output to ground. A 32Ω would certainly fool it. A higher resistance, like 100Ω would likely work as well.

Hi Bob, just a little clarification if you don't mind. When you say "from each channel output" you mean at the source, right? (the resistors should be right after the jack's "T" and "R"?), or at the PCB output stage? Sorry if this sounds dumb! I suppose the latter would definitely require the use of a closed-circuit jack, somehow.


To be honest, I only got into this "mod", because the "tutorial" I found made it seem like a 5 minute job. I see there is a lot more to it, afterall... I'm going on vacation in a couple of days, so I'll leave it as is for now, but I will definitely try your suggestions as soon as possible.
Thank you both for your input.

 

[Externet]

BobK observation is highly valid. We do not know if your sound source mutes if there is no load equivalent to headphones. Every brand and model can be different, and the help we try to provide is partially from blind information.
A load to the audio output can well be 100Ω resistors as stated. Try and you may avoid deeper surgery. That is two resistors from L,R 'lines' to ground.

 

[infectme247]

We do not know if your sound source mutes if there is no load equivalent to headphones

This sounds quite likely, because I had noticed that the phone detects headphones, even if the other end of the patch cable is unplugged. So at first I thought that it doesn't have any sort of "mechanism" to detect actual load, and that it only relied on having a closed circuit in its headphones jack... But then, the situation which I described in the previous post happens, where plugging it to the head, is equivalent to unplugging it from the phone, so the phone seems to drop the headphones although it has a cable plugged in.

It also explains why the other source I tried (the mp3 player) played noticeably lower in volume and (slightly low-pass filtered, now that I'm thinking about it...)

I will give it a try as soon as possible and let you know. Thank you for your time.

 

[BobK]

If I understand your question correctly, it does not matter which end of the cable the resistors are on. The corresponding lines at both ends are, after all, shorted together (hopefully)

Bob

 

[infectme247]

it does not matter which end of the cable the resistors are on.

Yes, this is obvious. I just wanted to make absolutely sure, that you didn't mean the output to the speakers on the PCB, when you wrote "from each channel output to ground".


Anyway, thanks again, I'll be back with feedback when I get a chance to try both your suggestions.

 

[infectme247]

Good news. I chopped up a spare cable, added the 2 resistors as BobK described, and it seems to do the trick just fine. So far I've tried 2 mobile phones and a tablet, which didn't work with the previous one (they wouldn't sense headphones when both ends were connected) and now they all stay in headphones mode no matter what, for as long as the cable is plugged.

The sound is a bit weird though, but I guess I must have shorted some channel by accident. It sounds to me as if all "monophonic" audio is lost, and I only hear the "Side" channel (for those of you familiar with Mid-Side audio encoding) but it's true that I did a very sloppy job, since I'm away from home and have access to merely a knife and duct tape! Multimeter to check continuity is out of question for now!! Anyway, the idea was to see if this works in principle, until I have the chance to make a proper one, so I think I'm on the right track.

Funny thing is, I thought this problem was because of me doing something wrong, and never got to look it up. However, after a bit of googling these days, it seems that a lot of people had this problem, even with factory-fitted AUX inputs, and especially with Samsung android devices (like mine) which obviously have some kind of "intelligent" sensing/protection mechanism. Some people noticed that, adding an extra cable with level control seemed to fix the problem, while others had more "ingenious" solutions, such as using a Y-splitter with a useless pair of headphones permanently plugged, and using the other output instead. I think they all boil down to adjusting the impedance with a little extra resistance in any form, right?

Anyway, it's amazing the kind of functionality you can get out of a (almost broken) car stereo with just a few minutes of soldering and practically no money at all. Music, GPS narration, accepting calls, all on the car speakers for practically no money, and while keeping the original head unit. A lot more than what I asked for!

The thread turned out to not have anything to do with the actual car stereo modification afterall, so if the moderators could rename it to something more appropriate, I think it would be more useful to others in the future.

Thank you all for your help