There cannot be ground loops in the MS3 with the stock HU

[Trambounet]

This is related to the other post where I was seeking help for alternator whine...

I have been reading articles about how alternator whine is most likely caused by ground loops encompassing the HU, the amplifier, the RCA -/shield and the car chassis itself.

After dusting off my trusted multimeter, I do not believe that this can happen on the MS3, and here is why: at the HU, the audio signal reference is not tied to the +12V reference, hence the audio signals are isolated. By the way, the input jacks on the Alpine PDX5 appear to be isolated as well. As a result, there cannot be current flowing through the shield of the RCA's, even if the HU and the amplifier are not grounded at the exact same potential reference. So there is no ground loop and this is not how alternator whine is picked up.

So now the million dollar questions are:
- For those of you who have a spare stock HU, can you please verify the audio outputs and the case are isolated? Maybe my car is effed up somehow.
- Does the RCA shield need to be grounded? And where? car chassis/12V ref or HU audio signal ref?

 

[GoFast]

RFI perhaps?

 

[Trambounet]

It has got to be EMI/RFI.

I am going to ground the shield of my RCA to the car chassis later today and see what happens. I found out that my RCA cables are shielded but the shield is not grounded, so it must be acting like a 6ft antenna. Nice.

 

[Mid_Life_Crisis]

Nice theory, but a Ground Loop Isolator wouldn`t stop RFI and it will stop this noise. Part of the problem is that the Bose system uses a different kind of output/input than typical aftermarket stereos, so when you swap out the H/U or the amp, you get the horrendous alternator whine. I wish I could find the article I read but I can`t remember where I saw it right now.

 

[Trambounet]

Nice theory, but a Ground Loop Isolator wouldn`t stop RFI and it will stop this noise. Part of the problem is that the Bose system uses a different kind of output/input than typical aftermarket stereos, so when you swap out the H/U or the amp, you get the horrendous alternator whine. I wish I could find the article I read but I can`t remember where I saw it right now.

If you could find it that would be great!

 

[MikeHTally]

Unless the shield in the RCA cable is damaged, it's a sufficient ground (for that purpose). If you park very close to another car, does ITS radio exhibit a whine? You may have an issue with one or more plugs or a flakey connector on the alternator.

 

[Trambounet]

Well my theory just blew in pieces. You guys are both right!
1. Grounding the shield didn't do jack
2. I ran a wire from my amp ground to the plug where the 12V/ground for the Bose amp is, and the noise substantially was reduced. Still there but much more bearable. Hint hint.

I think if I ground the amp where the Bose amp was grounded I might be ok. But I can't find where the ground stud is. I found the HU ground stud, but couldn't find the Bose amp stud. Anybody got any suggestions?

 

[Trambounet]

Part of the problem is that the Bose system uses a different kind of output/input than typical aftermarket stereos

Do you mean that the Bose system uses balanced differential signals?

 

[MikeHTally]

Ground is ground. Doesn't matter which you use.

 

[car46999]

Ground is ground. Doesn't matter which you use.

Not 100% true. The more steel that is between components grounds, the more resistance that is created. Such that the grounds will not be the same. Steel is less conductive than copper.

 

[Mid_Life_Crisis]

Do you mean that the Bose system uses balanced differential signals?

That sounds familiar. These things work when the source (H/U) and receiver (amp) are both designed for them, but when you try to mix them with an unbalanced pair, it wreaks havoc on the signal quality and ground noise. When I first added an aftermarket amp to my car, the noise was unbearable. I put in a GLI and it was almost inaudible (I could hear it, but was never quite sure it wasn`t my imagination). I later swapped out the head unit to gain a better aux input channel and the noise was completely gone.

 

[Mid_Life_Crisis]

Not 100% true. The more steel that is between components grounds, the more resistance that is created. Such that the grounds will not be the same. Steel is less conductive than copper.

A custom installer I used to know would never put a high power system in a car without putting a deep cycle battery in to run the stereo. He effectively isolated the entire stereo system from the primary electrical. The entire system was fed from and grounded to that battery, which was then grounded to the heaviest nearby chunk of metal he could find, preferably frame, or a reinforced section of a unibody car.

 

[car46999]

That sounds familiar. These things work when the source (H/U) and receiver (amp) are both designed for them, but when you try to mix them with an unbalanced pair, it wreaks havoc on the signal quality and ground noise. When I first added an aftermarket amp to my car, the noise was unbearable. I put in a GLI and it was almost inaudible (I could hear it, but was never quite sure it wasn`t my imagination). I later swapped out the head unit to gain a better aux input channel and the noise was completely gone.

I would argue that there have been successful additions of amps to MS3/M3 car audio systems. Be it the installation or components (to include wiring). It seams that it can be done without noise.

 

[car46999]

A custom installer I used to know would never put a high power system in a car without putting a deep cycle battery in to run the stereo. He effectively isolated the entire stereo system from the primary electrical. The entire system was fed from and grounded to that battery, which was then grounded to the heaviest nearby chunk of metal he could find, preferably frame, or a reinforced section of a unibody car.

That is one approach. Older cars with more steel would have a lower resistance, todays cars with less metal would have more resistance. With a unibody the welded body panels are what your are relying on.

 

[Trambounet]

I put in a GLI and it was almost inaudible (I could hear it, but was never quite sure it wasn`t my imagination). I later swapped out the head unit to gain a better aux input channel and the noise was completely gone.

In your case the GLI (which basically is a transformer) allows to duplicate the balanced input as an unbalanced output. All it did was shifting the ground reference of the signal. You do need to do that if the amp is not compatible with balanced signal, otherwise the amp will amplify audio + to car chassis, which is only half of the story (the other half being car chassis to audio -) and noisy as hell.

 

[Mid_Life_Crisis]

I would argue that there have been successful additions of amps to MS3/M3 car audio systems. Be it the installation or components (to include wiring). It seems that it can be done without noise.

I never said it couldn`t be done. As I said, I did it. I needed to put a GLI in the path, but I did it.

 

[Mid_Life_Crisis]

In your case the GLI (which basically is a transformer) allows to duplicate the balanced input as an unbalanced output. All it did was shifting the ground reference of the signal. You do need to do that if the amp is not compatible with balanced signal, otherwise the amp will amplify audio + to car chassis, which is only half of the story (the other half being car chassis to audio -) and noisy as hell.

I know what a GLI is. I suspect that closing the output circuit of the H/U through the transformer windings allows it to remain a balanced signal, so the H/U stays happy. The transformer coupling prevents a direct signal or ground path through the wiring between the H/U and the next component (amp, crossover, whatever) which eliminates ground reference induced noise from that path. There are, of course, other places that a ground reference differential can induce noise, but they are trivial by comparison to the signal path (the ones that, as I said before, you are never quite sure if it is just your imagination).

 

[Trambounet]

I am curious, how close to the amp was your GLI?

 

[Mid_Life_Crisis]

I am curious, how close to the amp was your GLI?

The amp at the time was under the seat. I spliced into the harness under the seat and put in RCAs. The GLI was the cable connecting those to the amp. So whatever the length of the built in cable was. Maybe 12 - 18 inches.
A GLI should be as close to one of the components as you can manage. Try both components if possible to see what works the best. In my case it was much easier to put it near the amp, so that is where it went.
Long low power runs can introduce stray capacitances that can muddy up a signal. The question is: which end is at the most risk? I really think a GLI belongs as close to the source as possible so the signal being fed through the coils is as pure as possible. It was a pain in the butt to do that though, so near the amp it went.

 

[car46999]

I never said it couldn`t be done. As I said, I did it. I needed to put a GLI in the path, but I did it.

I was implying that people have done it without ground loop isolators. But the funny part is that when I read your post I thought you were referring to a Volkswagen GLI. (cheers) guess I should not read while drinking.