Tips for laptop sound card calibration

[danillll]

Hello,

As you notice this is my first post in this forum and I would like to take the opportunity to thank you guys for all the knowledge and information you're sharing and special thanks to the author of REW John for sharing his very useful software.

I just started playing with REW and I'm going step by step, taking my time and making sure that every calibration step is as perfect as it can be.

Here are couple of challenges I encountered.

Even though I develop software for a living, I only have laptops at home, desktop can stay at work .
All my 3 laptops have no line in, only MIC in, however I noticed that all 3 docking stations have line in, so BINGO!!! first problem solved.

After hooking up the loop cable and run the sound card calibration, I didn't like the first result, you can tell there was lots of noise in the graph.
So I ran the line out from laptop to my A/V receiver and you can hear an awful static interference.

The static went away when I used the laptop stereo headphone jack, removed the docking station and unpluged the power.
But I needed to use the docking line in, googling this issue, I found out that this has to do with the ground in the power supply, pcuser.com.au/pcuser/hs2.nsf/lookup+1/931993547A4099C6CA256E5E008307B5 ,
My solution was to connect the docking station power supply plug to another plug that has no ground and voila, noise is gone and I ended up with this beautiful graph.



I hope this will be helpful for people with laptops and running into similar issue.

Next step is to calibrate my SPL meter.

Regards

[Wayne A. Pflughaupt]

Welcome to the Forum, Dan!

Wow, looks like a much-better-than-average soundcard there. I expect you'll have some subwoofer graphs up for us in no time!

Regards,
Wayne

[danillll]

wayne, thanks for your reply.

This is my first subwoofer graph, not sure how bad it is.

I noticed a peak around 66hz, is this bad? what db range should I stay in, is it +- 10db?

Measurement Settings:
start Freq = 15HZ
end Freq = 150HZ
Everything else is set to default

I am connected directly to the sub.

I am using RS SPL 33-4050 , in the SPL calibration step, I was reading +- 80db (it goes from 77db to 82 db), so I set it to 80db in REW.

My sub is PB10-ISD, Is there a room for improvement? I am trying to avoid spending $, but if adding BFD would make a big difference I would go for it.

One last question, my receiver Crossover is set to 80HZ, now I know this means that all low frequency < 80 would go to the sub only, but what about the frequency higher then 80 would this setting still send it to the subwoofer?

I hope I am not missing any configuration setting, but please let me know what you think.

Thanks

[danillll]

this is another graph that shows the spl cal graph along with the sound card after running set target




what do you guys think?
Is the idea of the graph to try and lift the drop, in my case 50HZ to 57HZ, above the target line? I think I asked it before, but what is an acceptable drop range?

Thanks

[danillll]

this graph makes more sense, this time I went through the receiver which has the crossover set to 80HZ.



The graph shows the SPL cal, the target on (80HZ cutoff and target level 82.2db), the response and the sound card.

Any input on the previous questions would be appreciated, thanks.

[brucek]

I hope I am not missing any configuration setting, but please let me know what you think.

For subwoofers, always use the standard Vertical graph axis of (45dB - 105dB) and the Horizontal graph axis of (15Hz - 200Hz) using the Graph Limits button in the top right corner of REW.

Use a start frequency of 0Hz and an end frequency of 200Hz.

Be sure to set your graphs to LOG and not LIN using the FreqAxis button in the top right corner of REW.

For full range, use the standard Vertical graph axis of (45dB - 105dB) and the Horizontal graph axis of (15Hz - upper limits you desire, i.e. 20KHZ - certainly no higher than your soundcard can extend).
For full range only, enable smoothing to eliminate the comb filtering. Use a 1/3 octave smoothing.

My solution was to connect the docking station power supply plug to another plug that has no ground

Many people use the battery when running REW on a laptop. They are notorious for noise. For REW only a temporary cheater plug works.

I was reading +- 80db (it goes from 77db to 82 db), so I set it to 80db in REW.

Calibrate REW to 75dBSPL, not 80dB. The SPL meter level is set during the Check Levels routine by adjusting your receiver volume and observing the SPL meter at the listening position. Then the REW input level is adjusted and finally the Calibrate SPL routine is run to match REW's internal SPL meter to the 75dB you set for the real SPL meter.

I know this means that all low frequency < 80 would go to the sub only, but what about the frequency higher then 80 would this setting still send it to the subwoofer?

Yes, and would decrease at a rate set by the order of the crossover (i.e. 24dB/octave). Always turn on the REW target line to have an example of the crossover.

this time I went through the receiver which has the crossover set to 80HZ

Always go through the receiver when testing subwoofers.

brucek

[danillll]

thanks brucek for your inpit. This is the graph with the suggested settings.



What do you think?
any tips to enhance the response?
would BFD makes a big difference?
What frequencies have bad response?


Thanks

[brucek]

any tips to enhance the response?
would BFD makes a big difference?
What frequencies have bad response?

Usually before equalization, we try positioning options for the sub to smooth the response. Often you can get away without needing any equalization.

Your response ain't bad, but the peaks at ~35Hz and ~ 65Hz will likely be noticed. See if you can position the sub to smooth it out . If not, then you could use a BFD to take care of it.
Also be sure that after you arrive at the smoothest response, you have to add the mains and examine the crossover. Any peaks or dips at that point can be removed with the subs phase control. When you measure with the mains, do the measurement exactly as you did with the sub alone (but do re-check the Check Levels routine).

brucek

[danillll]

Thanks a lot Brucek, I will move the subwoofer around and get new measurements once I get back home.
One last question, what about the drops around ~55HZ and ~89HZ can they be noticed? Is there a way to push them up? maybe using BFD?

One more question I loaded the calibration file for my RS 33-4050 spl. is the sub frequency response in the graph includes the adjustments? In other words do I need to add the cal value to it or it's already been added by rew?

Thanks again.

Danny

[brucek]

what about the drops around ~55HZ and ~89HZ can they be noticed?

The dip at 89Hz may be overwhelmed by the mains, so you would not touch it until you found that out.
The one at 55Hz could perhaps be helped with some filter gain from a BFD, but more than likely it is caused by a phase cancellation. The more you throw gain at it, the stronger the 180 degree signal canceling it becomes. Waste of headroom with no result.

is the sub frequency response in the graph includes the adjustments?

That's why you loaded the file into REW, so it compensates automatically, just as your soundcard calibration file adds automatically once it's loaded.

brucek

[danillll]

Thanks again Brucek.
The graph is subwoofer only, but I see your point, maybe adding the mains to the picture will enhance the 89hz dip and as for the 55HZ I will see how it acts with the mains.

Which reminded me, when I calibrated my AV receiver, I set my mains to 75 db and the subwoofer to 80db for extra kick (it was suggested by somebody on the avsforum), I guess I need to adjust the subwoofer back to 75db to run the test with REW? right?


Cool, can't wait to get home and tune this

You guys rock!!!

Danny

[brucek]

I guess I need to adjust the subwoofer back to 75db to run the test with REW? right?

Well, the actual SPL levels between the sub and mains is up to your listening preference. Before you run REW, you should set the levels using the SPL meter and your internal receiver test tones to your liking.

Then run REW with the mains and sub, and see what the crossover looks like. The setting of 75dB at the listening position in REW is only to facilitate the test and get a good graph.

brucek

[Wayne A. Pflughaupt]

danillll wrote:

One last question, what about the drops around ~55HZ and ~89HZ can they be noticed? Is there a way to push them up? maybe using BFD?

The graph is subwoofer only, but I see your point, maybe adding the mains to the picture will enhance the 89hz dip and as for the 55HZ I will see how it acts with the mains.

You just can’t tell about these things with any certainty until you add the mains. For instance, if the 89 Hz dip is coming from the room, then it’s highly likely your mains will show the same thing.

Re the 55 Hz dip, there’s not much chance it’s a phase cancellation since this is a “subs only” graph. Phase issues will only come to light once the mains are added (although it could show up in a subs-only graph if there are two subs in the room at different distances from the measurement mic).

Aside from that, depressions from phase cancellation or nulls will typically be fairly narrow and symmetrical. This one is neither. True nulls can’t be boosted with equalization, but they seem to be pretty rare; only occasionally have I seen someone unable to raise a depression with equalization.

Depressions caused by crossover-induced phase issues (i.e., once the mains are added) can often be addressed with equalization. The reason why is pretty easy to grasp: Equalizer filters alter phase. This is a well know fact. In fact, it’s one reason equalizer detractors don’t like them.

Like I said, you just can’t tell for sure about these things until you try...

Which reminded me, when I calibrated my AV receiver, I set my mains to 75 db and the subwoofer to 80db for extra kick (it was suggested by somebody on the avsforum), I guess I need to adjust the subwoofer back to 75db to run the test with REW? right?

No reason to change your subwoofer setting. This is one reason why we recommend running REW through a receiver instead of sub-direct: REW’s SPL calibration can be accomplished with the volume control, leaving your relative sub/mains settings intact. Of course, you’ll probably need to re-adjust that after equalization…

Regards,
Wayne

[brucek]

Phase issues will only come to light once the mains are added

Well, you know how I hate to disagree with you Wayne, but I'm afraid I have to in this case.

Yes, there is a lot of cancellation around the crossover between the mains and sub, but why would you think the situation is any different with the sub alone?

Dips are caused by phase cancellations, where the sound reaching the measurement point is a combination of an original direct soundwave and a reflected sound that is 180 degrees out of phase at the dip (one half wavelength). This can, and often does come from the sub alone as the sole culprit.

It's quite easy to identify the surface that causes the dip by examining the frequency where the dip occurs. If the dip is at 89Hz for example, then the wavelength is ~12.6 feet long, and so a half wavelength is 6.3 feet away for the offending reflective surface that caused the dip. You can move the mic and see which way the dip moves to verify your suspicion of the surface causing it.....

brucek

[Wayne A. Pflughaupt]

brucek wrote:

Well, you know how I hate to disagree with you Wayne, but I'm afraid I have to in this case.

Yes, there is a lot of cancellation around the crossover between the mains and sub, but why would you think the situation is any different with the sub alone?

Sorry ‘bout that, I meant any that would appear once the mains were added... Thanks for clarifying.

Regards,
Wayne

[danillll]

you guys are real professionals in this subject, seriously, audio is a science on its own and I love when I see two experts like you arguing it's a good opportunity to learn

I am going to leave the sub in the same place and run the tests with the mains and see if anything would change, I will post all my graphs with different phase settings shortly.

[danillll]

COOLIO ....

I added the mains to the picture
level check and SPL cal was reading ~75 db
ran the sweep from 0HZ to 200HZ
Crossover is set to 80HZ on the receiver
The sub is in the same place as it was when the last graph was taken

Test 1:
phase is set to 0 and I got this



Test 2:
phase is set to 90 and I got this


Test3:
Phase is set to 180 and I got this



From my "newbie" point of view, I think test 2, phase set to 90, is the biggest winner.

do you guys agree?

It's amazing how this little knob called phase makes such a big difference , seriously wow, even if I turn it 5 degrees I can see big differences in the graph.. wow...

So my questions for you (some of them are obvious but want to double check)

if leaving the subwoofer in its current position, should I stick to phase 90, right?

The only issue is the one peak at 70HZ right? can I notice it while watching movies, in other words is buying a BFD worth the $ to fix this peak?

What about 35 to 40 is this considered peak? it is +5 db above the target line.

Finally any dips I am not aware of that I need to pay attention to.

I guess my next step tomorrow is to move the subwoofer around and see if I can smooth the peak at 70HZ

[brucek]

if leaving the subwoofer in its current position, should I stick to phase 90, right?

Correct.

The only issue is the one peak at 70HZ right?

It would be nice if the peak at 40Hz and 70Hz were lowered.

Try moving it around some more to get it smoother.

An easier method to finding better potential spots is to place your sub in the listening position and move the mic around to different spots and measure.

Even easier is to use the RTA feature of REW since it saves successive measures. For setup see here.

brucek