Frustration! Help me please (1 Viewer)

[Brian_J]

I recently picked up the BFD to even out the response in my room. I thought i would make it easy on myself by using my radio shack spl meter with ETF or Spectraplus software but I cant seem to figure out either. The frustration is also high due to some software glitches that i had to hurdle. The searches i retrieved on this site were not very helpful. Can anyone offer detailed help on how to do this stuff, please! Thus far it seems it would have just been easier to plot everything out manually.
Brian
------------------
Zed's Dead Baby...

 

[Brian_J]

Anyone?
------------------
Zed's Dead Baby...

 

[Julian Data]

Ok.. hwg:
From what I recollect from memory for the Spectra.
Since you are EQ'n bass:
-Sampling rate 11025Hz
-Decimation 6
-FFT: 8192
-Use Smoothing Window
Be sure to create a RS SPL Cal file and load and use the "FLAT" setting in the mic settings.
Put your mic at the listenining position.
Of course, you know that when you EQ, it will be just for this position. You also might have to adjust the height/position of the mic if you have some serious modes or nulls encountered.
------------------

 

[Phil Olson]

Brian,
You've certainly bitten off a chunk! The FBD is certainly a handfull but if you carefully read through the threads on it, you'll get it.
On the ETF, follow the directions carefully. They first ask you to verify that the card is full duplex, (most are now days), then short input to output to calibrate the card. Have you done this yet.
If not.... If so, you will then need to know which measurement you want to do.
If you want to smooth out the bass, use the low freq mode when you run the data capture. Then set the gate time to something long, (1500 ms.). This will allow you to display down to 2 Hz if you want. On the display menu, you can choose different frequency ranges, I'd use 2-200 Hz. that will tell you everything you need to know.
If you read their white paper and follow the on-line case history, you should get the hang of what they're doing.
If not, come back and ask detailed questions of how you're stuck.
Hope this helps,
Phil
P.S. I think help will be easier to find if you go through every step in the instructions and ask a very detailed question when you get stuck. Open ended 'what do I do' type questions are much harder to answer.

 

[Brian_J]

Thanks for the reply.

Use Smoothing Window

Which? Hanning, Uniform...etc.
Brian
------------------
Zed's Dead Baby...

 

[Julian Data]

Uniform.. I think it's the last option..
------------------

 

[Brian_J]

What setting should you put the radio shack spl meter on while using it as a microphone?
Brian
------------------
Zed's Dead Baby...

 

[Julian Data]

I recall I left the mic value alone .. Just keep in mind to tell the software to be "FLAT" check that box.
------------------

 

[Brian_J]

Thats not what I meant. I mean what value do you set the actual meter itself at, 60, 70, 80, 90...?
I have yet to get either software to work where i feel totally comfortable with it. Manual readings seam much more accurate. Also, when I first ran some tests I thought i was measuring my sub output but then noticed that I could take the microphone off the cord and the graph would not move! Obviously, I was not measuring my sub but my laptop! It works now but I am just not as confident of the readings yet as I am when I do it manually. If both were the same I would feel better.
Brian
------------------
Zed's Dead Baby...

 

[Julian Data]

do you want the graph to read what the meter says?
If so, you have to play with the setting. Say set your meter to the 90dB scale and run a 40Hz tone to 90dB and see if it correlates in the software. If not, you need to do some calibrating.
jd
------------------

 

[Brian_J]

Julian,
Thanks, I just didnt know if the RS meter recorded at different levels or just 'on'. My whole frustration comes down to the fact that i had a bad rca cable and was only recording electrical hum from my computer. I changed that and now am actually recording the room response.
Earlier you mentioned setting the following in spectra plus:
-Sampling rate 11025Hz
-Decimation 6
-FFT: 8192
Could you explain why and what these things mean in terms of low frequency?
Next, should you set the peak hold on or not and do you set the rta for infinite average?

Be sure to create a RS SPL Cal file and load and use the "FLAT" setting in the mic settings.

I think I know already but just to make sure can you tell me how this is done.
Finally, do you just use pink noise or a sweep?
Thanks for all your help
Brian
------------------
Zed's Dead Baby...

 

[Julian Data]

Changing the sample rate moves the area or the BW (Bandwidth) to a lower level since you won't be concerned with upper frequencies.
Majority of the info is in the Help file that comes with the Spectra.
You could use a higher sampling rate. Decimation is the breakdown of the sample rate on how it's measured.
I used peak hold when running and then saved it an overlay. Then did some changes on the EQ and noted the differences. It's faster that way.
You would either do individual warble tones or run a sine that ranges from 20-100Hz. but depending on the XO on your proc/rec, I wouldn't even bother going past it.
As for the mic compensation file, I think TV helped you on this (As he did with me when first stated with this. )
Open up notepad and input the following:
10-20.0
12.5-16.5
16-11.5
20-7.5
25-5.0
31.5-3.0
40-2.5
50-1.5
63-1.5
80-1.5
100-2.0
125-0.5
1600.5
2000.5
250-0.5
3150.5
4000.0
5000.5
6300.0
8000.0
10000.0
12500.0
16000.5
20001.5
25001.5
31501.5
40002.0
50002.0
63002.0
80002.0
100001.0
12500-0.5
160000
20000-1.0
As for the sine, pink noise, or even warble, I in fact used them all and since you have a BFD, you can save these settings and see for yourself on which you like. Here's what I did:
First attempt: indvidual sines from Stryke CD.
Second Attempt: used the pinknoise from the proc.
Third Attempt: used individual warble tones from the Stryke CD.
Fourth Attempt: Use a sine wave from 10-100Hz from the Stryke CD.
That's for different ways and resulting in different atttenutations.
Which one do I prefer? Either the sine wave - not individual- or warble tone as these types tones sort of replicate on what the actions of the sub would be doing room, IMHO. When you watch a movie or listen to music, the sub isn't locked on to one frequency during play as it just moves when told to do so .. naturally.
Make sense?
------------------

 

[Brian_J]

Julian, Thanks for all your help. Ive got the hang of this now, ever since I got a good rca cable (rggh! that cost me some time). I moved the mic around the room and interestingly enough the best position for the sub (in terms of no major dips and more flat) is in a corner that I thought was too far away. As it is Ive cut down on some major peaks where I have it now and threw in some additional power to the bottom end of the sub and this has made a lot of difference. The neat thing with the bfd is that i can set a few programs for different purposes. Close to flat being for music and a bump from 30 to 45 being for movies. Cool.
Question-Say I have a chart that looks like a wave, low at 20 rising to a peak at 50 and back down to 80 (with some up and down movement in between of course). Should I cut everything from say 30 to 70 to a level closer to what the sub is at 20 and 80 or do I just try and cut the hump a bit so that it is flatter and accept the falls towards 20 and 80? This is not a crossover issue by the way as the receiver is currently out of the loop.
Thanks,
Brian
------------------
Zed's Dead Baby...

 

[Brian_J]

Also, shouldnt those be plus signs for the compensation file at the low frequency range?
Brian
------------------
Zed's Dead Baby...

 

[Julian Data]

If you like the "flat" sound, then by all means cut the peaks. I like the flat sound for most audio.
------------------

 

[Patrick Sun]

Yes, in the compensation chart above, add those correction values to your measurements for the low end.
Sidenote: you can safely ignore SPL readings over 10KHz with the RS SPL meter (those correction values for over 10KHz was very inadequate).
Be sure to wear earplugs when you are doing your measurements.
Using individual sine waves will allow you to find the nulls in your room. Warbles are more suited to full range speaker testing. But take a composite of the 2 sets of measurement if that's what it takes to make you happy!
------------------

 

[Brian_J]

How far do you "flatten" the chart to? As noted above, if you have a chart that looks like a wave, low at 20 rising to a peak at 50 and back down to 80 (with some up and down movement in between of course) how flat do you try to make it. Should I cut everything from say 30 to 70 to a level closer to what the sub is at 20 and 80 or do I just try and cut the hump a bit so that it is flatter and accept the falls towards 20 and 80?
In other words, do you flatten to the lowest lines on the chart?
Brian
------------------
Zed's Dead Baby...

 

[Patrick Sun]

Umm...maybe flatten to within 6dB of the low end around 20hz and 80Hz.
------------------