Suggestions for HT sub upgrade from PB-1000

When you set the speakers to large does the AVR still redirect the bass to the sub? I'd imagine that'll just increase the overall output. Conrad probably knows more about this stuff.

How's it sound?
 
Small is the right setting for your towers, if you're using a sub.

Take a listen to it, but it'll likely sound boomy and a bit overwhelming.
You've basically flipped the problem. You were dropping to a 75dB dip from 85dB before, now you've got a 95dB peak from an 85dB base.

You'll certainly have more output!

I'd say before I had a 87dB base and a dip to 75 dB, and now I have the same 87 dB dip, but a peak to 90 dB.

It sounds MILES better. The bass sounds so much better, that I can dial it down from -4 I think.
It sounds better with the Dynamic EQ off than it sounded before with it on.

I would not say it is overbearing at all. But then I listen to a lot of EDM, and people who listen to EDM tend to be bass junkies...
 
When you set the speakers to large does the AVR still redirect the bass to the sub? I'd imagine that'll just increase the overall output. Conrad probably knows more about this stuff.

How's it sound?
Yes it does, because I set the sub to LFE+Main.
So it's getting the full signal.

I wonder if the massive dip with the fronts sent to "Small" is because of how Audyssey EQ's them?
Either that or something is really wrong with the crossover or distance settings, but nothing I did got rid of that massive drop. It feels almost like I need an EQ to ramp that drop up.

I could try to re-do the whole room calibration and plug the ports on the towers. That should reduce their low end, maybe Audyssey sets them to small and does a different calibration?
 
The dip is a phase cancellation in the sub response. It’s reflected sound interacting with direct and other reflected sound cancelling the signal.

If you move to another place in the room you’ll get a different reading.

If you can get the dip close to the crossover region you’ll be able to adjust the distance/delay/phase control and change the interaction of the subs and mains through that region, hopefully filing the dip.

This is one of the main reasons people invest in multiple subs. By putting more and differently phased signs into the room you end up with fewer cancellations, just peaks. Peaks can be EQd down, you can’t boost a null.
 
You could confirm this by measuring the sub alone on the LFE channel. If the sub is the only speaker outputting then it has to be the sub/room/seating position combination.
 
Yes it does, because I set the sub to LFE+Main.
So it's getting the full signal.

I wonder if the massive dip with the fronts sent to "Small" is because of how Audyssey EQ's them?
Either that or something is really wrong with the crossover or distance settings, but nothing I did got rid of that massive drop. It feels almost like I need an EQ to ramp that drop up.

I could try to re-do the whole room calibration and plug the ports on the towers. That should reduce their low end, maybe Audyssey sets them to small and does a different calibration?

Yes if you change something you do have to re-calibrate, that is the 'proper' way but if you like what you hear....

Speakers are now getting a full range signal which means both the sub and speakers are now contributing toward room modes (peaks) in the response. This is probably 'hiding' any potential hole that you may have perceived as a lack of bass before.

Prior to setting the speakers to large, did you try adjusting the phase on the sub? Use that as a tool to get the response flat as possible and then apply EQ at the end. Don't know how accurate the Auto EQ system is on Denons but you might find tweaking the distance on the sub after calibration provides an even better response.
 
I ran out of time today.

Yes, I tried shifting sub phase, that made the dip wider and narrower. The dip follows the crossover point exactly. So i think i have to play with the distance on the sub.

My plan is to get a usb sound card, plug the Audyssey mic into it. Verify what I am getting against the PAA3X and then I can use REW properly and post readable graphs.
 
Depending on how much you’ll pay for a USB dac you might be better off just getting an UMIK-1. If the sac is €20 then it’s a big difference. If it’s €80 then it’s probably worth getting the “proper” tool.

Phase on the sub and distance in the AVR are the same thing.
 
Okay so, the sub got sold. I had a fairly high buy it now price, but someone wanted it.
I will have a PB2000 Pro tomorrow. Bigger is always better anyway :)

The DAC/ADC was a SoundBlaster Play and it cost me 17 EUR. I also already have it here, I didn't have to wait for it. I can actually use it with my laptop on a daily basis. The integrated DAC in my laptop is horrible and does not do even my HD280 Pro's any justice. They sound so much better with the SoundBlaster DAC.

I will now go through what you wrote and get the HDMI output working from REW on my laptop to the AVR.

Unfortunately I am totally swamped with work, so I have very little time to work on this. The weekend should bring some relief.

I also want to take the time and thank you for your support, it has been invaluable. I appreciate it very much.
 
Could not get FlexAsio to work, but Asio4All works and I see all my channels.

In the mic calibration if I set the level to 75 dB, the input is at -30...
I do not have a USB mic technically, but you could say I do as the Soundcard is USB, and it's plugged into it.

I set my AVR to pure direct mode, so there is no processing at all, and I'm just testing noise.
It does seem that the mic is a lot less sensitive higher up.

What is a good way to do the calibration of the microphone in REW?
Should I do sweeps? Pink noise?
Then make a file with the difference?
 
You can’t calibrate that mic without a known reference. I would focus on sweeps from 20-200 which is probably where the mic is most accurate. You’ll just have to trust it. You could build a cal file manually if you could get the correct data out of the other measurement device, but we don’t know how accurate that is or how to get the data.

Run some sweeps. Do sub only on LFE first, then sub only, then mains only, then sub+l and sub+r.

Learn how to take a measurement with a timing reference. I have the instructions somewhere and I can dig them out tomorrow for you.
 
But I have a known reference. The PAA3X. I can scroll through the display bit by bit and it shows me the dB at every frequency.
I've been putting the Audyssey mic right next to it and comparing.
Even without any correction it's not far off at all, 4-5 dB hot at bottom end and 2-3 dB cold top end.

I'm gonna set the towers to large and the sub to LFE and run some fast sweeps through them, then record the result with both REW and PAA3X, and then correct the REW input based on the PAA3X.
I'll then do the same with the LFE.
 
But I have a known reference. The PAA3X. I can scroll through the display bit by bit and it shows me the dB at every frequency.
Great, that should do it then. You can just have an offset per frequency.

I've been putting the Audyssey mic right next to it and comparing.
Even without any correction it's not far off at all, 4-5 dB hot at bottom end and 2-3 dB cold top end.
For what you're dealing with that's probably good enough and can be compensated for easily.

I'm gonna set the towers to large and the sub to LFE and run some fast sweeps through them, then record the result with both REW and PAA3X, and then correct the REW input based on the PAA3X.
I'll then do the same with the LFE.
When I say mains, the measurements that I aways ask for are:
  • L with crossover engaged, with no sub
  • C with crossover engaged, with no sub
  • R with crossover engaged, with no sub

These three measurements give you the "upper" half of the response, above the crossover.

- sub with crossover engaged and no main

This gives you the "lower" half of the response, below the crossover.

With those measurements you can use REW to model the interaction of each main and the sub through the crossover, adjusting gain and delay for each to get the smoothest response. This is different from what Audyssey will do and can usually be improved on.

Lastly it's worth doing:

- sub on LFE channel

This is where the EQ should be applied. If you can get that flat to target then adjusting the crossover will just work with that response and will work out smooth when you engage the crossover.
 
Yeah, what I was saying is just to make the .cal file for the mic...
And then I need to buy the app for the phone so I can change things myself.
 
OK so all of my Q-Acoustics benefit greatly from having their ports plugged.
Especially on the towers it significantly flattens the response and gets rid of the massive peak in the 50-200hz range while leaving the highs completely unaffected.

I am in direct mode and on a sweep my entire range is between 75 and 85 dB now, mostly concentrated at the 80 dB mark. Before there was 20 dB difference.

I am going to calibrate the mic now and then post some REW screenshots.
 
Does direct mode perform bass management?

If you want to upload the file you can rename it to a txt file and do it that way. It's usually more useful to have a look at the file than screenshots.
 
Excel porn:
1615550533822.png
 
I don't know why, but REW is ignoring my calibration on the top end. Maybe it is out of range, I'll try to rebalance it at a different frequency point.

EDIT:
It worked, now the graphs look identical, but my RTA graph is exactly 15dB too low.
How do I shift the entire thing 15dB up?
 
Last edited:
Figured out, got it calibrated, so now I can just use REW and return the Phonic.

1615556707864.png

Also attached the final calibration file.
Sound Blaster Play 3 with the latest available driver and then the Audyssey mic that came with the Denon. Input set to maximum in the SB3 Play control panel. Gives -18 input on the mic.
 

Attachments

  • audyssey_sbplay3.txt
    655 bytes · Views: 40
Last edited:
That file doesn't open for me as I'm on a much later version. I can see if I can get an earlier version to run alongside the latest version.

It's worth keeping REW up to date with the latest beta (now RC6 I believe), there are never really any issues and new features get added all the time.
 
Ok, that'll do it.
If it's generally reading low I think you can add a sens line to the start of the file that describes the offset.
 
I just needed to re-do the SPL calibration, it's bang on now.
The PB2000 Pro will be here in a bit, then I can hook that up and spend Friday playing games with diagrams...

Here's the current .mdat file for the LFE channel sweep. Everything above 150hz can be disregarded. Below 20hz as well.
Of course it's not terribly useful, as the PB2000 Pro is about to arrive.

.mdat in graphical form:
1615561026959.png
 

Attachments

  • sub_lfe.zip
    91.5 KB · Views: 36
Last edited:
So the PB-2000 Pro is installed. Audyssey recalibrated with the plugs in the speakers.
I set the sub to 50% and Audyssey set it to -11.

I set the crossover to 80hz for everything and speakers to small. I can't say it sounds very good right now, I need to look and see what's going on.

However, if I set the Sub to +6 and play something at -12 I can shake everything and the sub does not get overloaded. So in case I want to do such perversion, or in movies it should be great.
Now need to fine-tune the response. This sub is definitely overkill, but the previous one was just barely not enough.
 

The latest video from AVForums

Is 4K Blu-ray Worth It?
Subscribe to our YouTube channel
Back
Top Bottom