2. digital
  3. 163272-135704

High Frequency Oscillation hurt resistors in Magico A5 crossovers

Hi,I am looking for help understanding how HFO could blow crossover resistors in Magico A5 speakers. Equipment playing at the time of the incident is as follows...
Magico A5 speakersHegel H30 mono blocksHegel P30 preamp (lt and rt inputs on the Aux inputs from the laptop and Dragonfly Cobalt)
Dell laptopAudioquest Dragonfly CobaltAmazon Hi-res streaming
I was listening to Amazon music at a very moderate level when the application said there was an update available.I instinctively clicked accept while music was paying. It was only off for a few seconds to update and when complete and I restarted music the A5 tweeters were not on. These happened simultaneously and there was no audible noise what so ever indicating a potential issue. After multiple source tests and tweeter test, I was certain the tweeters were fine and it was an electronic issue. The cross overs were removed and sent to Magico and that was when I was told that HFO blew the resistors.
Can anyone explain how/why this happened so I can ideally avoid it happening again?
Thank you!
128x128howaanders8
kellyp's avatar
kellyp
104 posts
 
I think the LF > HFO etc has been covered adequately.  A schematic of your crossover would help. Anyway, with updates, components are reset, turned off, turned on.  It is reasonable that the +5V PS voltage to your USB DAC was also reset.  The switch mode PS used are 200-300 kHz.  The voltage has to settle.  And the ripple, particularly with Dell products is already quite bad.  The dragonfly has some PS filtration, but it is not great.  Being as this happened during a reset, and that is the way resets work, it is my best guess.  I would not expect your speaker engineers to shed and light on something like this. 

This may happen again for no fault of your own.  How can you avoid this?  Get an outboard PS/USB and forget the laptops poor voltage source.  Not only will you add protection, you should experience better fidelity.

kennyc
1,488 posts
 
Magico is technologically driven so it’s likely that it’s not the speaker’s faulty design but rather the HFO caused by an external source - also likely caused by running the update while having the speakers engaged.

Seems like the OP has the solution - don’t do software updates while the speakers are engaged - likely a good idea/lesson for us all. But unless the OP has a deep technological curiosity, seems like a potential fruitless rabbit hole to further track down the source of the HFO. I’d expect minimal/no incentive for component manufacturers to spend time/effort to assist you in this endeavor.  
erik_squires
16,090 posts
 
I want to be clear:

Oscillation burning out tweeter resistors or tweeters is not new to Magico.  It is very rare, but it does happen and when it happens it is the electronics which are at fault. If you try to protect the speaker from this, you might as well start trying to introduce surge protectors into the speaker itself.

We can't even get people to use tone controls, can you imagine protection circuitry like this??  :-)

A bigger concern is that honestly not all electronic makers are as careful about protecting their amps from oscillation, and may eschew making the amp stable regardless of load for the sake of high bandwidth.
73max
46 posts
 
What kind of speaker cables are you using? Mega capacitance cables can cause high frequency oscillations with some amplifiers. Even more likely if you biwire. HF Oscillations are self-sustaining with ever increasing level once started due to the signal “reflecting” in the cable. With digital audio and high bandwidth amplifiers the chance to trigger this increases. While amplifiers  are often designed to mitigate this, I am not a fan of very high capacitance speaker cables. The effect is similar to a feedback loop you hear when a microphone picks up it’s own speaker but well above audible frequencies.
howaanders8's avatar
howaanders8 OP
23 posts
 
Wireworld Eclipse 7 Copper I believe. No bi-wires.
More to discover