STR-DN 1080 - Calibration setting not stored

[phil_merge]

Hi,

After performing the auto speaker calibration and selecting a setting ("Full Flat" for example), this setting is not stored. 

When I switch the receiver off and on again, the speaker calibration setting is ignored, although it still says "Full Flat" in the menu. The sound is as if it was set to "Off".

I need to manually set the setting to "Off" and to "Full flat" to hear the correct speaker calibration.

Thanks for your help!

Phil

[Beeper85]

Thanks Philippe

[Beeper85]

Ok here we go with some first images, though I am sure Philippe will have better results that are more accurate and show more detail. These do at least show this is a real issue objectively.

The images are available in this Dropbox link as struggling to embed here.Images

All measurements taken from exact same place.

First image shows reading of white noise sample with receiver on after calib and in full flat. (I agree pink would be better and I'll let Philippe do that) . 

Note that around the 240 to 260Hz the signal never exceeds -20db.

Next image is after amp is powered down and back on. It shows as in 'full flat'. You would expect an exact same reading (keeping a close eye on the peaks around the 240 to 260Hz). The signal far exceeds -20db in this range. It is no way the same as it was before power cycling.

The next image shows changing the setting via the osd setting of calibration type to 'off'. Again note the out put reading. It is the same as above in image 2, almost identically. Essentially off and full flat are applying the same EQ adjustments.

The last image I switch back to full flat. If it really was in full flat when it was turned on you would expect image 2 and 4 to be the same. However, after switching to this setting the readouts are almost identical to image 1 - full flat before power off. 

We can only conclude that image 2 and 3 (full flat after a power down and off are the same) and image 1 and 4 are the same (full flat after calib and full flat after actively reselecting it after a power off/on).

This shows exactly what I am hearing. 

[dutchice]

Hi Ben,

Good hands on job! It definitely excites me to see that community members roll up their sleeves and share their findings. As I previously already mentioned, the results I posted just rules out the existence of a general fault across all devices. It does by no means rule out or rule in a fault on your device. So again, it could very well be that there is a misconfiguration issue going on on your side.

A few remarks on your testing method and results:

• White Noise is the least suitable source signal to use for such an analysis due its homogeneous spread of energy through the frequency range. Flicker noise on the other hand is a total different story and is indeed widely used in the industry to analyse a flat frequency response spectrum.
• The reason why we selected a real-life sample off Netflix for our test was deliberately to match your issue. Signal generators often will produce synthetic results.

Lets wait to see for the results by Philippe.

Cheers,

Dutchice

[Beeper85]

Thanks for checking in!

Agreed, I won't pretend this is anything but a crude test (and definitely has its shortcomings!). I'm sure Philippe will get better results.

Whilst inaccurate, it does show commonality between settings which is what I was looking to show. Will be good to demonstrate this properly once Philippe is able to.

[dutchice]

Hi,

A quick addition to my previous post, since I forgot to mention that the CT (Calibration Type) setting is simply a correction to the DCAC EX calibration in order to deliver certain sound characteristics. Therefor the first area to look at when in doubt about the correct function of the CT setting is to verify that your calibration is performed successfully. Have you ever encountered error messages after performing a sound calibration in your living room? Is there anything unusual/special about your cables/cable-connections/speakers? Have you ever tried to simplify your speaker setup to 2.0 using only the two best speakers attached to the AVR and replicate the issue?

These are all areas where I believe you should start looking for answers.

Cheers,
Dutchice

[Beeper85]

Nothing unusual. Running monitor audio radius speakers for centre sides and surrounds, and a BK gemini ii sub.

All wired normally and calibration works as expected with results showing as calibrated and saved. Accurate distance reads and level corrections that I would expect.

I'll try with just a two speaker set up.

[grolschie]

Dutchice's test doesn't conclusively rule out a problem on all devices. It's entirely possible that a bug could be triggered by a certain combination of settings or calibration values.

[Beeper85]

Valid point (and there has to be somehting causing this!).

I will note all calibration settings shown after running the auto calib (though I do know that I have had slightly different readings on these so it cannot be down to a very specific +/- db or distance reading...)

[phil_merge]

Hi.

Here are the results of my measurements.

Due to lack of time, I for now only post the measurements made with a portable system.

But these measurements are very accurate. Please only consider the upper curve of the spectrum analysis. The lower full black curve is the live measurement at the moment of the snapshot on the analyzer unit and can thus be ignored!

I used a sine sweep tone.

I measured a full scale spectrum (20-20.000 kHz) as well as only a low scale spectrum (20-1.720 kHz) for more precision in the lower frequencies, where the room correction usually is most effective and needed.

1. Sine sweep - Full Scale - Off

No room correction applied. There's a peak frequency at 906,25 Hz. Not a very linear frequency response in my room (not a recording studio).

Sine sweep - Full scale - Off

2. Sine sweep - Full scale - Full flat

One can see that the room correction is active and has corrected some bumps in the frequency response. The peak frequency is now detected at 5,9688 kHz, where there's quite a lift in the spectrum.

One can clearly hear the effect of the room correction (especially when pink noise is used instead of a sweep tone!).

Sine sweep - Full scale - Full flat

3. Sine sweep - Full scale - Full flat after reboot.

This was measured after rebooting the unit. In the menu, "Full flat" is still shown as the room correction mode, as it was left when the unit was shut down. So you would expect that when powering the unit back on, it would remain in "Full flat" mode. Well, visually it does, but not sound-wise. You can clearly see (and hear) that we get back to the same spectrum as in "Off" mode. The bump at 906,25 Hz is back. It is very much identical to the spectrum measured in "Off" mode. Now you have to manually switch the unit to "Off" and back to "Full flat" to get the audio output as intended for "Full flat".

Sine sweep - Full scale - Full flat after reboot

I made the same measurements using the "Engineer" setting, with the same result. When rebooting the unit, it still shows "Engineer" on the display, but the spectrum analysis is back to "Off" mode.

Theres a limit in picture uploads, so I will make another posts showing the measurements for the low frequency range.

Of course one could say that I could easily post any pictures I wanted. But honestly, the time I spent on this is not really worth the price of the unit. I really try to help solve this problem, which I still believe all units have.

Thanks for everyone interested in this and trying to halp solve this problem!

Philippe

[phil_merge]

I forgot to mention that only the upper line of the spectrum analysis is relevant.

The lower full black spectrum is the live measurement at the moment of the snapshot on the analyzer module and can thus be ignored.