[selmeralto]

Okie dokie. Right now I have neither the ASD harness nor the Bimmertech. So I'll put the harness on first, see what's what, and report back. If that doesn't fix the problem and if I get the amp I'll post the results as well. In either case, if we have a fix I'll then ask our technologically informed colleagues to speculate on what the cause might have been!

[simoneves]

I exchanged messages with Technic yesterday about a related issue. He told me that the ASD module used in all these cars is actually a power amp as well, and that the same one is used in some low-spec Minis and BMWs to actually drive four speakers directly. In our cars, those speaker outputs are instead connected to the higher-impedance line-level inputs of the HiFi or HK amps.

His ASD Bypass Harness physically loops the four line-level signals from the HU direct to the main amp, so that they don't go through the ASD module any more, and hence any additional noise that might otherwise be introduced by the ASD module is no longer an issue. He did point out that the ASD module has a noticeable line-level gain, so the overall level is likely to drop when the harness is installed. That would reduce both the upstream noise and the music, unless the amp gain was adjusted to compensate (which is of course only possible with a third-party amp).

That said, he also stated that he had never actually noticed any significant additional noise being introduced by the ASD unit, so the hiss that's there is likely coming from further upstream.

[selmeralto]

Originally Posted by simoneves

I exchanged messages with Technic yesterday about a related issue. He told me that the ASD module used in all these cars is actually a power amp as well, and that the same one is used in some low-spec Minis and BMWs to actually drive four speakers directly. In our cars, those speaker outputs are instead connected to the higher-impedance line-level inputs of the HiFi or HK amps.

His ASD Bypass Harness physically loops the four line-level signals from the HU direct to the main amp, so that they don't go through the ASD module any more, and hence any additional noise that might otherwise be introduced by the ASD module is no longer an issue. He did point out that the ASD module has a noticeable line-level gain, so the overall level is likely to drop when the harness is installed. That would reduce both the upstream noise and the music, unless the amp gain was adjusted to compensate (which is of course only possible with a third-party amp).

That said, he also stated that he had never actually noticed any significant additional noise being introduced by the ASD unit, so the hiss that's there is likely coming from further upstream.

Terrific information. Thanks. I'll get to the ASD harness as soon as I can.

[selmeralto]

I think I may have isolated the source of the high frequency whine.

I think it's in the ASD circuitry. I installed the Technic ASD harness and I'm pretty sure the noise has vanished. I say this provisionally because I was listening to music in the car for a while at high levels and my ears are ringing a bit. But I don't think the sound is there anymore.

There is some corroborating evidence: the high pitched whine was there independent of the volume of the audio system and the ASD operated similarly independently of the audio volume. And if I recall correctly, didn't someone say that the ASD had its own power supply? If so, perhaps the sound was coming through the circuitry so the ASD harness simultaneously eliminates the simulated engine noise and the high pitched whine.

I'll test this over the next few days when my ears are fresher.

I'll leave it to the technologically more educated folks on the forum to explain why this works - if it does - and to correct my references to the technology as necessary.

In the meantime, since I've now eliminated the messy center speaker (by disconnecting it) and the artificial engine noise and high pitched whine (by installing the ASD harness), I'm left with the pristine sound of the crummy HK system.

The journey of a thousand miles ...

[simoneves]

That doesn't surprise me. If the ASD module is really a combined sound generator, mixer, and power amp (and probably not a very good example of the last) then its lower-impedance higher-power speaker outputs aren't likely to be anywhere near as clean as if they were just low-level line outputs, so running then into the line level inputs of the good amp is bad audio engineering and just cheapskate design by BMW. They should have a different ASD module with just line-outs for cars that DO have a separate amp, but that would cost them. Bypassing the ASD completely would undoubtedly be an improvement on paper, and apparently in practice too.

[selmeralto]

Originally Posted by simoneves

That doesn't surprise me. If the ASD module is really a combined sound generator, mixer, and power amp (and probably not a very good example of the last) then its lower-impedance higher-power speaker outputs aren't likely to be anywhere near as clean as if they were just low-level line outputs, so running then into the line level inputs of the good amp is bad audio engineering and just cheapskate design by BMW. They should have a different ASD module with just line-outs for cars that DO have a separate amp, but that would cost them. Bypassing the ASD completely would undoubtedly be an improvement on paper, and apparently in practice too.

Thanks, Simoneves. I can see that I was actually referring to you and getting it wrong: I should have said independent power amp, not separate power supply. In any case, I hope this is culprit.

[simoneves]

After further thought, the other problem with BMW's design is that is breaks the otherwise "balanced" (audio electronics term, look it up) signal chain, as speaker outputs (while still consisting of two connections) are not (usually) balanced as the negative connection is (usually) tied to ground (the higher-current return ground of the power stage, but still ground).

The balanced inputs on the stock amp (and on the usual replacements) can certainly tolerate an input derived from an upstream output normally intended to drive a speaker, in the same way that they could accept input from an upstream unbalanced line-level output (the input -ve connection being tied to the output ground), but you lose the isolation from ground.

The lack of such isolation is the #1 cause of interference in the pro audio world, especially when running signals over longer distances through electrically-hostile environments (which a car certainly is). This is why balanced connections exist in the first place (XLR cables for microphones etc.)

Using Technic's bypass harness (which, from the pictures I've found of it, does indeed consist of eight loop-back connections, as required for four independent balanced signals) presumably maintains the balanced signal from the HU output all the way into the real amp input, thus avoiding the introduction of noise.

Edit: First statement not entirely true. Speaker outputs -ve connections aren't usually tied to ground, but are usually the negative side of bi-polar output derived from the +/- power rails of the power amp (usually something like +/30V). The voltages on these are, however, usually fixed with respect to the actual ground (0V) and so it still doesn't have the interference-rejecting capability of a true unbalanced signal (where the two voltages can be anything you want, even a million volts if there's a lightning strike happening next door, as long as the *difference* between the two is the same, which is the actual signal). Either way, BMW's arrangement lets noise back in that would otherwise be rejected.

[Viffermike]

Originally Posted by simoneves

After further thought, the other problem with BMW's design is that is breaks the otherwise "balanced" (audio electronics term, look it up) signal chain, as speaker outputs (while still consisting of two connections) are not (usually) balanced as the negative connection is (usually) tied to ground (the higher-current return ground of the power stage, but still ground).

The balanced inputs on the stock amp (and on the usual replacements) can certainly tolerate an input derived from an upstream output normally intended to drive a speaker, in the same way that they could accept input from an upstream unbalanced line-level output (the input -ve connection being tied to the output ground), but you lose the isolation from ground.

The lack of such isolation is the #1 cause of interference in the pro audio world, especially when running signals over longer distances through electrically-hostile environments (which a car certainly is). This is why balanced connections exist in the first place (XLR cables for microphones etc.)

Using Technic's bypass harness (which, from the pictures I've found of it, does indeed consist of eight loop-back connections, as required for four independent balanced signals) presumably maintains the balanced signal from the HU output all the way into the real amp input, thus avoiding the introduction of noise.

Edit: First statement not entirely true. Speaker outputs -ve connections aren't usually tied to ground, but are usually the negative side of bi-polar output derived from the +/- power rails of the power amp (usually something like +/30V). The voltages on these are, however, usually fixed with respect to the actual ground (0V) and so it still doesn't have the interference-rejecting capability of a true unbalanced signal (where the two voltages can be anything you want, even a million volts if there's a lightning strike happening next door, as long as the *difference* between the two is the same, which is the actual signal). Either way, BMW's arrangement lets noise back in that would otherwise be rejected.

I think you hit the nail squarely on the head, simoneves. I suspected to semeralto in a series of PMs that the source of that noise might be power differential-related RF; I just wasn't sure where it could be coming from because, really, it could have been from ANYwhere. ASD was a primary suspect, to be sure; I just hadn't heard of anyone who, after installing the TechnicPnP harness or coding it off, said that the system whine had disappeared.

Now we know: ASD is more than artificial exhaust note. It's a damned noise source!

My question now is whether or not coding ASD off eliminates the whine. I suspect that it does not because by coding ASD 'off', one actually doesn't deactivate the module; one actually just selects an option that doesn't have a "soundtrack" (for lack of a better term). That still leaves the system noise being pumped into the system.

If the above is indeed the case, the harness -- not coding -- is the only option to really eliminate ASD. And I gotta tell ya, that makes me want to order that harness yesterday!

[selmeralto]

Just one caution: I couldn't hear the noise today but that was after lots of listening to loud music. I'll double check tomorrow after some wine and sleep tonight to clear my ears.

[simoneves]

Originally Posted by Viffermike

And I gotta tell ya, that makes me want to order that harness yesterday!

I already asked Nelson to add one to my amp order!

The other point is that if the ASD is a power-amp, that means its outputs are perfectly capable of generating far higher voltages that would normally be expected by a line-input.

Presumably the gains in the upstream system are set up such that the ASD is never provided with a strong enough input signal such that it would generate an output voltage too high for the stock amp to handle.

Maybe this means that the HU line-out level is artificially low compared with the standards expected from third-party units (I seem to remember "4V line outputs" being a selling point of such products) and that would mean that the system would have a higher overall noise floor.

One would hope that the Bimmer-Tech (and indeed the JL) have inputs which are at least as robust, but not having the inputs of one's new expensive amp being driven by something of indeterminate function, quality, and tolerance, which might be capable of creating signals of damaging levels, is surely yet another advantage of the harness?

[simoneves]

I'm wondering if there's any mileage in creating an alternative harness or device that would take the four (or perhaps two is enough) speaker level outputs from the ASD box, convert them properly to a safe line-level, and then route them into two more of the regular inputs on the Bimmer-Tech amp, separately from the music signals which were bypassed to it directly?

This would allow the ASD sound to be mixed back in at a sensible (and user-controllable!) level.

Of course, by default this would add the hiss back too, but you'd have the full DSP power of the amp to fix that with a low-pass filter on those channels.

This would probably require the ASD bypass harness to be combined with the adapter harness for the amp, to route the two (or four) balanced connections (four or eight wires) from the ASD outputs to inputs 5-6 (or 5-8) of the BimmerTech. The four balanced audio inputs of the ASD would obviously also need to be tied together (and to ground?) to ensure that the output of the ASD is *only* the engine sound and not more noise from its otherwise-floating inputs.

Technic ?
Nelson ?

[simoneves]

Originally Posted by simoneves

One would hope that the Bimmer-Tech (and indeed the JL) have inputs which are at least as robust...

I looked at the Match 82DSP spec, and it describes the main eight inputs as being "high-level", which I take to mean that they are perhaps designed to handle speaker-level signals without damage.

However, just because you can, doesn't mean you should.

[Zooks527]

Originally Posted by Viffermike

I just hadn't heard of anyone who, after installing the TechnicPnP harness or coding it off, said that the system whine had disappeared.

Mine did. Problem was, I wasn't looking for it at the time, and by the time I thought to look it was too far past the time I made the changes for me to determinately say so.

FWIW, I spent part of last night's ride home trying to find the hiss. Not a sign of it, and it was clearly there last year before I did the harness and amp.

[BWT]

I have ASD coded off. I noise that I get is quite faint (but more noticeable than the OEM HK amp) and can only be heard with the car parked and volume turned right down. The noise is present even with the ignition switched off which leads me to think it is the noise floor of the amp.

[Viffermike]

Originally Posted by simoneves

I'm wondering if there's any mileage in creating an alternative harness or device that would take the four (or perhaps two is enough) speaker level outputs from the ASD box, convert them properly to a safe line-level, and then route them into two more of the regular inputs on the Bimmer-Tech amp, separately from the music signals which were bypassed to it directly?

This would allow the ASD sound to be mixed back in at a sensible (and user-controllable!) level.

Of course, by default this would add the hiss back too, but you'd have the full DSP power of the amp to fix that with a low-pass filter on those channels.

This would probably require the ASD bypass harness to be combined with the adapter harness for the amp, to route the two (or four) balanced connections (four or eight wires) from the ASD outputs to inputs 5-6 (or 5-8) of the BimmerTech. The four balanced audio inputs of the ASD would obviously also need to be tied together (and to ground?) to ensure that the output of the ASD is *only* the engine sound and not more noise from its otherwise-floating inputs.

Technic ?
?

That would take more than another harness design, I think - it would take a line-level converter separate from the Match/Bimmertech amp to convert the speaker-level signal to line-level before it goes into the amp.

I think we've discovered inadvertently why BMW uses differential balancing in its audio system: To keep audio sources that are not controlled by the volume control of the HU -- and, likely, the HU itself -- independent. This would not only include ASD, but iDrive and system sounds such as the door ajar/seat belt 'gong'. By using a hotter speaker-level signal for those sources and feeding it into the signal as essentially a sidechain source, the relative volume would stay constant no matter how much power is applied via the volume-control gain.

[selmeralto]

Just to confirm. Tried again to hear the noise after the harness install. I couldn't detect anything. As far as I'm concerned, this is a fix.

[simoneves]

Originally Posted by Viffermike

That would take more than another harness design, I think - it would take a line-level converter separate from the Match/Bimmertech amp to convert the speaker-level signal to line-level before it goes into the amp.

Forgive me, but unless I'm misunderstanding what Technic told me, and my interpretation of the pictures I've seen of the harness, no additional converter is required because without the harness, the speaker-level outputs from the ASD are *already* connected to the line-level inputs of the amp.

The Bimmer-Tech amp is documented as having "high-level" inputs which are capable of this sort of connection (even though, as I said before, it's bad audio engineering) so if it can handle those inputs on 1-4 without the harness it can surely handle them on 5-8 with it, with 1-4 then fed directly from the HU.

My only concern would be some kind of ground level issue with two different sources of inputs, but hopefully if the amp inputs are truly differential then that is not a problem.

Surely all it takes is four (or eight) more connections, from the (now empty) "output" pins on the bypass harness connector that now goes into the ASD unit, to the input 5-8 pins on the amp end of the adaptor harness, and something to tie the "input" pins on that same connector to ground to stop the ASD inputs floating.

I suspect that it would require all four channels (eight connections) to retain the full ASD sound, since while ASD sound predominantly comes from the rear speakers, it also has a bass component which surely has to come from the subs, which are in parallel with the front speakers, so it must be mixing low-frequency sound into the front outputs and mid-frequencies into the rear.

Obviously I don't have any of this hardware myself yet, but like I said, unless I am misunderstanding what I've been told and have seen, I am pretty damn sure this is workable. I'm hoping that Technic and will chime in with their thoughts.

Originally Posted by Viffermike

I think we've discovered inadvertently why BMW uses differential balancing in its audio system: To keep audio sources that are not controlled by the volume control of the HU -- and, likely, the HU itself -- independent. This would not only include ASD, but iDrive and system sounds such as the door ajar/seat belt 'gong'. By using a hotter speaker-level signal for those sources and feeding it into the signal as essentially a sidechain source, the relative volume would stay constant no matter how much power is applied via the volume-control gain.

Again, forgive me, but I don't think that differential connections are a prerequisite for how it works.

The volume and fader controls on the HU control the level of the music component. The HU then generates and mixes in the gong/PDC sounds at a fixed level (after the volume/fader) before those four line-level signals are sent to the trunk. The ASD is then mixed in also at a fixed level. The amp then powers the speakers with a fixed gain. Hence, the gongs and ASD come out of the speakers at a consistent level, while the music level varies with volume/fader.

The only reason for differential connections is to make them more immune to interference, especially in the long signal path from the HU to the trunk. It's just a shame they throw that away again by reverting to non-differential (or quasi-differential) for the connection between the ASD and the amp, even if that's only about three inches!

[simoneves]

Originally Posted by BWT

I have ASD coded off. I noise that I get is quite faint (but more noticeable than the OEM HK amp) and can only be heard with the car parked and volume turned right down. The noise is present even with the ignition switched off which leads me to think it is the noise floor of the amp.

Unless the car is completely asleep (several minutes after turning the ignition "off") the ASD and the amp will still be powered up. Coding ASD off only stops the ASD module from mixing in the sound. It doesn't change the electrical noise apparently induced by the funky signal path upstream of the amp.

I'd be extremely surprised if the Bimmer-Tech amp had any noticeable output noise of its own, and this is corroborated by selmeralto's claim of the hiss having gone away with the harness.

[Viffermike]

Originally Posted by simoneves

Forgive me, but unless I'm misunderstanding what Technic told me, and my interpretation of the pictures I've seen of the harness, no additional converter is required because without the harness, the speaker-level outputs from the ASD are *already* connected to the line-level inputs of the amp.

But doesn't the ASD module enter the signal path after the main amp? That's my understanding ... in other words:

HU ---> main amp preamp (line level) ---> main amp power amp ---> output (speaker level) ---> ASD (speaker level) ---> speakers

If the ASD enters the signal path before the main amp power stage, then differential balancing would theoretically eliminate any noise it emits. Since it plainly doesn't, I've got to think that the ASD enters after the main amp power stage, which would make total sense because ASD is outputting a speaker-level signal.

The problem with what you propose is that the Match amp would need to deal with line-level signals for the first four-six channels (depending on how the subs are powered by the amp), then a speaker-level signal from the ASD on any other open channel. If the Match amp has that capability, great -- but a hotter signal from ASD before the power stage could cause all sorts of mixing and fidelity issues.

Originally Posted by simoneves

The Bimmer-Tech amp is documented as having "high-level" inputs which are capable of this sort of connection (even though, as I said before, it's bad audio engineering)

Yep; it is, but if all inputs are carrying the same voltage and resistance, it shouldn't cause issues. It's when they start to differ (see above) that problems can arise.

Originally Posted by simoneves

My only concern would be some kind of ground level issue with two different sources of inputs, but hopefully if the amp inputs are truly differential then that is not a problem.

Completely agree. Let's see if Technic or anyone from Bimmertech will chime in on this.

[//M235ixD]

Delete!

[simoneves]

Originally Posted by Viffermike

But doesn't the ASD module enter the signal path after the main amp? That's my understanding ... in other words:

Like I said, I don't have either hardware myself (yet), so I can't confirm the topology of the wiring, but it would make absolutely no sense for it to be how you describe.

It is simply not possible to make a electronic device that can be inserted between a power-amp and a speaker to "mix in" another audio signal without completely replacing the effect of the upstream amp. The electrical rules of impedance, and how inputs match with outputs simply preclude it.

It could be made to function, and it would make sound, but then replacing the amp would have absolutely no effect (at least not in terms of sound quality and power capability), since its output would simply end up converted back to line-level at the input of the ASD box, and the ASD box would still always be the thing actually driving the speakers.

Originally Posted by Viffermike

The problem with what you propose is that the Match amp would need to deal with line-level signals for the first four-six channels (depending on how the subs are powered by the amp), then a speaker-level signal from the ASD on any other open channel. If the Match amp has that capability, great -- but a hotter signal from ASD before the power stage could cause all sorts of mixing and fidelity issues.

The inputs of the Match amp are perfectly capable of accepting a line-level signal (balanced or unbalanced) or a speaker-level signal (bipolar), and there should be no limitation on the combination of those types applied to the eight inputs, as long as done correctly to avoid ground cross-talk.

It would be perfectly legit to connect the four balanced line-level signals from the HU to inputs 1-4 (as the bypass harness already does) and also connect the four speaker-level outputs from the ASD box into inputs 5-8 (or some subset thereof).

I haven't played with the DSP control software much, but I believe that the amp contains eight independent signal paths, each with gain, multi-band EQ, and delay processing, and then those eight signals can be mixed together in various combinations into the eight power amps.

To emulate the stock set-up, basically you need to make four mixes which are the sum of the signal from the HU for each corner with the corresponding signal from the ASD, with the ASD component having a low-pass to kill any hiss introduced by the ASD, then send each to its corresponding power amp and output, and the input to the sub power amp(s) from a mix of the four corners (or the two left and two right).

I don't know how to explain it any more clearly than that... :\

[Viffermike]

OK; I think I get it now, and what you're saying does make sense, simoneves . You're totally right about anything downstream after the power stage causing havoc. I guess what was throwing me regarding the signal path was the ASD module's function as a separate power amp (or, at a minimum, a booster amp). It just doesn't make sense to me to mix line-level and speaker-level inputs in the same channel in the OEM amp -- particularly when the speaker-level signal is the one 'barely there'. How is that even possible ...

If you do try this, it'll be very interesting from a fidelitic and processing standpoint.