Audio Mystery

Hello,

I'm reaching out to the technical community for help in resolving an audio issue that has persisted over the course of several years in spite of countless hours spent troubleshooting. This may get a bit lengthy as I want to provide as much background and system data as possible.

Symptom: Higher frequencies sputter, breakup, and sometimes drop out for ten or more seconds.
- occurs in either or both stereo channels
- occurs most often with stops rich in harmonics such as mixtures and reeds
- occurs intermittently, sometimes after hours of trouble-free operation
- can occur with as little as one stop pulled so not a polyphony issue
-occurs with all virtual instruments, small and large
Hauptwerk instrumentation well in the green below 50% (CPU, Mem, Polyphony, Audio Level)
OSX Activity Monitor indicates low and steady CPU loading – no large spikes

System:
Apple iMac (mid 2011) OSX High Sierra 13.6
CPU – 2.8 GHz i7 Quad Core
Memory – 32 GB
Audio interface – RME Fireface 800 (drivers and firmware are up-to-date)
MIDI – MOTU 8 x 8
Furman Power Conditioner
WiFi is ON for iLok Cloud
Bluetooth keyboard and mouse
Computer is dedicated to Hauptwerk – no other apps or utilities running except RME TotalMix
Hauptwerk v6.1

Things tried:
checked cables (Firewire, speaker)
audio buffer increase beyond single 1024
RME Sync set to Master
Tried different output channels (5/6)
Today I substituted a Focusrite Scarlet 2-channel USB audio interface and results were the same - no better, no worse.

The last experiment would seem to eliminate the audio interface as the bad actor and points back at the Mac computer. As mentioned earlier however, all indicators suggest this thing is coasting, not breaking a sweat.

The only half-baked hypothesis I can formulate is that something (what?) occasionally hammers the Mac's internal I/O bus and chokes down bandwidth to the Firewire and USB controllers. This seems iffy given that the problem can pop up with relatively light registrations.

As you all can imagine, the months of frustration has compromised my enjoyment of this wonderful platform, not to mention the distraction, when, as a piano player learning organ, my head is already about to explode trying to master the pedals.

Any help would be most welcome!!

Sincerely
James

[mdyde]

Hello James,

I don't think I've ever heard of anyone having those particular symptoms before, but my thoughts would be:

- Although it won't be relevant to this particular problem, if you're currently running v6.0.1, then do first upgrade to the current version (v6.0.2) anyway. It can be downloaded here: https://www.hauptwerk.com/osx/

- Make sure no other audio/MIDI software is running at the same time as Hauptwerk, in case it's trying to play sound (e.g. at a different sample rate) through the audio interface/output that Hauptwerk is using.

- When troubleshooting, stick to using just the St. Anne's sample set, since it's simple and small, and we all know it very well.

- Make sure your mobile 'phone isn't anywhere near any audio/MIDI/computer hardware or cabling! They often interfere with electrical/audio signals when they connect to the mobile network periodically to check for messages, etc.

- If you're using 96 kHz in Hauptwerk, then try temporarily changing it to 48 kHz. Also verify that that the sample rate gets set accordingly on the audio interface itself (e.g. on its hardware display, if applicable) when you load the organ, and that it stays set (e.g. doesn't change unexpectedly when the problems occur, which might indicate some other software trying to play sounds through the interface at a different sample rate).

- Just as a test try using just the Mac's built-in audio output with headphones plugged into it, since that would eliminate anything related to external audio interlaces (FireWire, USB, etc.) and amplifiers, speakers, cabling, etc..

What's after the audio interface? It sounds like the treble amp on your speakers are being overdriven or overheating.
Have you tried different speakers?
It could be RF, in which case I'd try taking the Fourman out of the equation and see if things improve.

Martin - Thanks for the great suggestions. I have a few things to try this morning.

I'm often guilty of having my mobile phone sitting on the console table while practicing and I live in a weak signal area so the phone is transmitting max power when it accesses the network.

I'll load the St. Anne's organ and run through the Mac's internal audio with headphones.

I've kept a close eye on sample rate as the symptoms could be attributed to a sync or aliasing issue. Something to monitor for sure,

murph - thanks for your suggestions as well. I'm running 3 stereo pairs of powered monitors and a sub off the interface. Overheating of the treble amp could very well be the culprit here although I play at moderate volume. Easy enough to swap speaker pairs and see what happens. Martin's suggestion of using headphones would also help flush out this type of issue.

I'll obviously need to be careful and change one variable at a time so I have a few hours of work ahead of me. On the plus side, I'll be logging some solid practice. Stay tuned...

James

[mdyde]

Thanks, James. You're very welcome. Hope you manage to track it down.

murph - I think you nailed it. Managed to isolate the problem to the main speakers. Dang it - probably the most expensive component in my rig, an older but high-end pair of Dynaudio monitors.

Interesting that both speakers exhibit the problem which is why I ranked it low in root cause probability. Also interesting is that higher notes on the keyboard and/or 4', 2', 1' ranks don't bring it out as much as lower notes with rich harmonics. As you suggested, the treble amps are probably cooked or I'm thinking it could be a problem in the crossover.

I'm going to yank the speakers out and run some forensics with a Moog synthesizer serving as a function generator but it looks like I'm probably in the market for new speakers.

Thanks again for your insight and suggestions!

James

Which model Dynaudio monitor?

I've run into (built-in) EQ capacitors causing HF distortion when they age and dry out. For the most part, any caps past maybe 8 - 10 years are past their prime. Mostly aren't difficult to change but sometimes hard to locate an exact replacement.

Rgds,
Ed

Hi Ed,

These are Dynaudio BM15A purchased sometime around 2009.

Good point regarding the caps. I'm facing a monumental task replacing the caps on my 1961 vintage Hammond B3.

I just ran a test with the synth function generator and couldn't get it to act up no matter what I threw at it. The synth is monophonic so I may need to switch to something polyphonic in case the problem manifests itself as some sort of inter-modulation distortion.

James

Generally speaker EQ caps are more accessible (than those in a B3) - after pulling the (appropriate) speaker. Compared to replacing the complete speaker, caps aren't much of an issue.

Rgds,
Ed

If you can't get things to act up on a different set-up, try the HW rig without the power conditioner. Weird RF things can happen when these are used, which can cause the HF amps to overheat.
It could also be a PSU problem in the mac that's injecting RF into things. (I once found a faulty PC monitor causing all sorts of audio problems at a driving theory centre....).

If you can't solve the problem this way, a cheap solution is to use a cheapish audio transformer (hum isolator) on the input to the speakers. These normally only have an upper frequency response of 25k or so, which provides effective filtering.

All attempts to reproduce the problem outside of the rig environment have failed.

murph - this morning I'm following up on your suggestion regarding potential RF pollution. I'll bypass the Furman and route the main speaker AC power cables to a separate wall outlet, ideally on a different circuit if I can stretch that far.

The BM15A has both thermal protection and an opto-electronic limiter on the HF. As you suggest, not hard to imagine a large RF input triggering one or both of these safeguards.

There's also a trim control that can cut up to 3 dB above 10 KHz which may help.

Take everything off the power conditioner!!!!
(You can put it back later if it doesn't help).

Take everything off the power conditioner!!!!
(You can put it back later if it doesn't help).

I guess it really depends on your situation. I lived somewhere with dirty power and the power conditioners helped A TON. Far from hindering my sound, they cleaned it up considerably and halved the amount of hiss my speakers were producing.

Hi folks,

Wanted to provide an update to close out this issue. After weeks of careful troubleshooting, swapping components in and out, isolating components on separate power outlets etc., I have isolated the problem to the Dynaudio BM15A speakers themselves. Different monitors and headphones (off the RME Fireface) all work fine.

What iced it for me was the following comment I stumbled upon while searching for problems with Dynaudio speakers:

https://gearspace.com/board/geekzone/47 ... g-out.html

Title: "Dynaudio BM15A tweeter dropping out... "

One reply: "It's a known issue with Dynaudio's BM series. Just crank it really loud, and give it a few minutes and it will start working again."

Seems to work but what a distraction. Probably will switch to a new set of monitors. (recommendations?)

BTW in the course of my investigation, I discovered that my old Sonos Connect throws off a huge RF field about every three seconds. It's not related to the tweeter dropouts which persist after removing the Sonos, but still not a good thing.

Thanks again for insights and suggestions. Now if I can keep the reed relays working on my pedal board, I may actually learn to play at some point.

Seems to work but what a distraction. Probably will switch to a new set of monitors. (recommendations?)

Behringer 2031A speaker are a popular choice.

Rgds,
Ed