The last shot is typical hum pickup plus broadband noise. 2 mV is not bad under those circumstances. The power supply ripple has artifacts from rectified power. I don't think you have lots more to learn from the scope. The ultrasonic stuff would be quite visible if it were an issue. At this point you are better served with a soundcard type analyzer at a minimum if you have an issue. Is there noise other than a low hiss you get from the speakers?
Thanks.
Its actually not the usual noise hiss audible from drivers in some setup when dac/amp are turned on and nothing is playing.
In this case my drivers are dead quite even with the ear close to them.
This is actually a buzz/scratch/crackle kind of distortion which seems to modulate with the music.
it is observable on the bass drivers with other drivers turned off. Its audible on axis with the driver and till about a meter or so from it at nominal music listening levels. With mid,tweeter playing this distortion disappears into the music. I have some links to the distortion which i was able to record.
I have done some soundcard spectrum analysis of the DAC output, with no smoking guns yet. I have to recheck my calibration of the chain, but i think i read around 15uV rms noise floor on the DAC output and some THD , THD+N numbers i posted earlier.
Last edited:
Hi Demian,
The first screenshot has periodic spike in the noise hash, is that something expected? contrast on the image is poor, but its visible on a closer look. Was very clear on the scope display.
I see those types of spikes occasionally on wide band systems. Usually there is something like a phone or wifi running nearby. Could be many things. Probably not the sound you are hearing.
If you are hearing crackles in the sub at higher levels but you don't see something obvious or consistent in the signal going it it could be the driver. It takes very little stress to get a crackle or other noise from a diaphragm flexing. Or a rubbing voice coil at the extremes. If you drive the woofer at something like 5 Hz to near its excursion limits these noises can be heard easier since the 5 Hz is inaudible. If there is electronic stress it can be easier to see with the repetitive drive. Really annoying and hard to find is a bit of glue that broke loose behind the dust cap. A length of rubber tubing used as a stethoscope is really helpful in finding this stuff and air leaks.
If you are hearing crackles in the sub at higher levels but you don't see something obvious or consistent in the signal going it it could be the driver. It takes very little stress to get a crackle or other noise from a diaphragm flexing. Or a rubbing voice coil at the extremes. If you drive the woofer at something like 5 Hz to near its excursion limits these noises can be heard easier since the 5 Hz is inaudible. If there is electronic stress it can be easier to see with the repetitive drive. Really annoying and hard to find is a bit of glue that broke loose behind the dust cap. A length of rubber tubing used as a stethoscope is really helpful in finding this stuff and air leaks.
Thanks again. That was a good idea.
I tried out feeding a 5Hz tone from REW. The drivers oscillated slowly, being subsonic i could only hear the distortion. It was present.
I doubt it is mechanical, as i tried the same motion by carefully pushing the drivers in and out and couldn't hear anything. But your pointers made a lot of sense to me. These drivers have large copper phases plugs instead of dust caps and dont really have large excursions in daily usage. So some dust settling in the small gaps is a possibility. I gently pushed the cone in and out and fired some short bursts from an air can to gently clean out the exposed part of the phase plug.
While playing the 5Hz tone i put a mic close to the driver and observed spectrum on REW's RTA. I see some spikes at 60Hz and 300Hz above the noise grass.
Couldn't see anything suspicious on the amp output and DAC output fed to two channels of the scope.
I tried out feeding a 5Hz tone from REW. The drivers oscillated slowly, being subsonic i could only hear the distortion. It was present.
I doubt it is mechanical, as i tried the same motion by carefully pushing the drivers in and out and couldn't hear anything. But your pointers made a lot of sense to me. These drivers have large copper phases plugs instead of dust caps and dont really have large excursions in daily usage. So some dust settling in the small gaps is a possibility. I gently pushed the cone in and out and fired some short bursts from an air can to gently clean out the exposed part of the phase plug.
While playing the 5Hz tone i put a mic close to the driver and observed spectrum on REW's RTA. I see some spikes at 60Hz and 300Hz above the noise grass.
Couldn't see anything suspicious on the amp output and DAC output fed to two channels of the scope.
Last edited:
Thats good to know. Incidentally i noticed that the noise spikes seen on the DAC signal is exactly twice the switching residual from the ClassD amp output (400KHz).
There is a strange wiggle in the carrier sine wave from the amp as it has a zero crossing. This coincides exactly with spike in the noise floor on a quiet DAC's output signal.
I am not sure if any of this has a way to alias back into the audible range.
I havent tried a low pass filter yet. Neutrik EMC XLR connector has a LP behavior but at around 10MHz.
Havent touched the crystal, what do you expect that to do?
If the amplifier's switching frequency and the DAC's switching frequency depend on different crystals with equal or rationally related frequencies, changing the load capacitance of one of them may change the beat frequency.
I had a problem with a soft beep in an audio signal from a DAC that went into an ADC some time ago. (Of course it is unelegant to put a DAC and an ADC in cascade, but there were some practical reasons why they were connected like that.) It turned out to be DAC clock feedthrough causing aliasing in the ADC. Touching a pin of the ADC's crystal (either directly or with a small screwdriver) very obviously changed the pitch of the tone.
I had a problem with a soft beep in an audio signal from a DAC that went into an ADC some time ago. (Of course it is unelegant to put a DAC and an ADC in cascade, but there were some practical reasons why they were connected like that.) It turned out to be DAC clock feedthrough causing aliasing in the ADC. Touching a pin of the ADC's crystal (either directly or with a small screwdriver) very obviously changed the pitch of the tone.
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.