Lamont Sanford
Lamont Sanford's picture
Offline
Last seen: 1 year 1 month ago
Joined: Mar 31 2006 - 8:32pm
Post your room frequency response
ncdrawl
ncdrawl's picture
Offline
Last seen: 12 years 11 months ago
Joined: Oct 18 2008 - 9:18am

can you outline, in detail(as much as possible) the method that you used?

ive been wanting to do this myself, but am a little confused as to how it is done.

Lamont Sanford
Lamont Sanford's picture
Offline
Last seen: 1 year 1 month ago
Joined: Mar 31 2006 - 8:32pm

Set up the Radio Shack SPL meter at your listening position. Get a test tone CD like I mentioned above or download a set of tones from various sites. Run the full spectrum pink noise track at 80dB by adjusting the volume on your system. Then just go through each track and write down the dB measurement. Create a graph using Excel or OpenOffice.

Some people have software that will do this as well. The SPL meter has an output jack to run to their computer. Of course, there are other ways to do this with different equipment and so forth. But this is the most economical way and is why I used it for posting on this forum.

As you can see I was only 1 dB off the 80 dB volume setting of the full spectrum pink noise track. The average dB for the entire spectrum measured using the SPL meter was 81 dB.

Here is my data:

    Hz dB
    20 67
    25 70
    32 75
    40 78
    50 83
    62 85
    80 86
    100 86
    125 86
    160 86
    200 85
    250 85
    315 85
    400 84
    500 84
    630 84
    800 85
    1k 85
    1.3k 85
    1.6k 84
    2k 84
    2.5k 84
    3.2k 84
    4k 83
    5k 82
    6.4k 81
    8k 79
    10k 75
    13k 67
    16k 64

    Mean 81.03333333

    StDev 6.247577308

I should add that the Sansui speakers have been modified. The capacitors on the crossover have been replaced with same value new capacitors. The outboard set of loudspeakers had the OEM woofers replaced with a pair of Pioneer A30IR50-51F for lower frequency response. I also plugged the ports on the outboard boxes as well. I disabled a single midrange and tweeter in each box to switch them from 5 speakers per box to 3 speakers per box. Having 8 midrange speakers and 8 tweeters was a little fatiguing to the ear in my room. The outboard midrange speaker working is the opposite of the inboard midrange speaker working since both midrange in each box is mounted at a slight angle facing each other. This is to accomplish the same effect as one full Sansui SP2500 where the midrange speakers are designed. In other words, to cross each others' sound travel to create a wider listening area.

ncdrawl
ncdrawl's picture
Offline
Last seen: 12 years 11 months ago
Joined: Oct 18 2008 - 9:18am

correction adjustments for the RS meter(rumoured to be off a bit)

http://www.danmarx.org/audioinnovation/rsmeter.html

In order to correct the low frequency roll-off, you can do the following modifications to your meter. This will make the meter FAR more sensitive to low frequencies and allow measurements with very good accuracy to well below 20 Hz. Due to the increased sensitivity at very low frequencies, it is possible for low frequency "thumps" to slam the meter if using a very low SPL setting. In order to prevent meter movement damage, take precaution not to peg the meter off scale on a regular basis. If this is occuring regularly, you either need to move up to a higher SPL range on the meter or take more caution as to how your performing your tests.

All of the following capacitors must be rated for AT LEAST 15 volts or so. Size does matter, so try to use the smallest package possible.
C1 & C2 are changed from 1 uF to 10 uF
C3 & C4 are changed from 1 uF to 47 uF
C7* is changed from 10 uF to 220 uF
C8 is changed from 100 uF to 470 uF
C9 is changed from 22 uF to 220 uF
C15 is changed from 100 uF 220 uF

Lamont Sanford
Lamont Sanford's picture
Offline
Last seen: 1 year 1 month ago
Joined: Mar 31 2006 - 8:32pm

I believe those are for the analog version of the RS meter. There is a lot of information out there about what model RS meter you are using and what are the correction values and so forth. My digital meter has a correction value of no more than 3 dB at the lower ranges. So, no need for corrections or adjustments. I'm not saying that it is 100% correct or within the standard -/+ 3dB error rate. Just reminding everybody that there is a lot of stuff on the Internet about these meters all which have some sort of different answer. Also, the lowest test frequency I used was 20Hz so not sure why anyone would want to measure below that for a room frequency response. Most manufacture graphs will be anechoic chamber measurements. Sansui's SP2500 published frequency response for anechoic chamber is 40-20khz and listening room is 25-20khz.

http://www.classicsansui.net/images/Literature/Speakers/SP2500%202.jpg

Keep in mind that my Sansui's were modified and altered a little for overall room response. See above post #72037
...

Buddha
Buddha's picture
Offline
Last seen: 12 years 5 months ago
Joined: Sep 8 2005 - 10:24am

Fantastic!

I gotta learn more.

Thank you for posting that!

RGibran
RGibran's picture
Offline
Last seen: 2 years 4 months ago
Joined: Oct 11 2005 - 5:50pm

I can see Ethan, shaking his head and rolling his eyes.

Check out the free program REW. Then come back and show us what your room response really looks like. It ain't pretty!

RG

Kal Rubinson
Kal Rubinson's picture
Offline
Last seen: 1 day 20 hours ago
Joined: Sep 1 2005 - 9:34am


Quote:
can you outline, in detail(as much as possible) the method that you used?

ive been wanting to do this myself, but am a little confused as to how it is done.

There are problems using this technique because the RS meter (or any other meter) cannot distinguish delayed responses due to room acoustics from the immediate response. It is better to use a quickly swept source such as with RoomEq Wizard (freeware) with your meter and a PC.

It is also tedious, so you can save time and effort with an RTA such as on REW or TrueRTA and there is a free version of the latter, as well, although the pay versions have better resolution.

Finally, let me put in a plug for the XTZ Room Analyzer (~$300 or so). It is a complete and easy package (software + USB mic) that will do both of the above as well as give you a nice T/F view of the whole response.

Kal

Lamont Sanford
Lamont Sanford's picture
Offline
Last seen: 1 year 1 month ago
Joined: Mar 31 2006 - 8:32pm

I have REW and plan to post something as well soon. Soon as I learn how to use the damn thing. Nevertheless, the meter technique above is fine for ease of mind. I mean, the system sounds damn good and the simple test shows a flat line to support that claim. This just reinforces it for me. Its somewhere between subjective and objective. Better than nothing. I also don't want to see members recommending this and that without showing something tangible in response. This is an entry level forum. Take the time and show your graphs using the stuff you are recommending for your own system. Check the title of the thread. I'm not trying to stifle people. Just throw something tangible in the hat rather than the same old tired "spend your money and/or time on this/that rhetoric."

Right now I'm going to the trap club and shoot some birds and to then to Sam's Club to get cat litter. I hope to have something from REW by the end of next week. God willing and the creeks don't rise.

Monty
Monty's picture
Offline
Last seen: Never ago
Joined: Sep 16 2005 - 6:55pm

That seems way too unbelievable to me.

Lamont Sanford
Lamont Sanford's picture
Offline
Last seen: 1 year 1 month ago
Joined: Mar 31 2006 - 8:32pm


Quote:
That seems way too unbelievable to me.

I can't help it. I find it weird myself. I was listening to it one day after making some changes and thought it sounded very good (for once). So, I got the test CD and the meter and had my graph in 20 minutes. I wrote up what I did to the loudspeakers as far as the changes go. It wasn't anything planned to achieve a flat response. More accidental and an after thought. Is that a byproduct of experimenting?

Anyway, in starting this thread it's my intention to see other graphs of members rooms. In other words, I'm not 100% confident in my measurements. Just thought it was weird and wanted to see other graphs. This also a good topic for entry level measuring.

ncdrawl
ncdrawl's picture
Offline
Last seen: 12 years 11 months ago
Joined: Oct 18 2008 - 9:18am


Quote:

Quote:
That seems way too unbelievable to me.

I can't help it. I find it weird myself. I was listening to it one day after making some changes and thought it sounded very good (for once). So, I got the test CD and the meter and had my graph in 20 minutes. I wrote up what I did to the loudspeakers as far as the changes go. It wasn't anything planned to achieve a flat response. More accidental and an after thought. Is that a byproduct of experimenting?

Anyway, in starting this thread it's my intention to see other graphs of members rooms. In other words, I'm not 100% confident in my measurements. Just thought it was weird and wanted to see other graphs. This also a good topic for entry level measuring.

gotta love the amen corner.... hell with em, lamont.

mjalazard
mjalazard's picture
Offline
Last seen: 2 years 11 months ago
Joined: Sep 8 2005 - 4:38pm

I've done the same with an old Audio Control C101 equalizer. I hook up the pink noise generator into a CD input on the pre-amp with a long pair of interconnects. I then jury-rig the microphone to approximate my sweet-spot location on the couch. I can then watch, in real-time, any changes in the frequency response as I muddle about with my stereo and acoustic treatment.

geoffkait
geoffkait's picture
Offline
Last seen: 3 months 1 week ago
Joined: Apr 29 2008 - 5:10am

Love the Rat Shack Meter! One thing to try if you haven't done so already is measure the peaks down in the room corners, and upper room corners, for that matter. Set the test tone for around 300 Hz or so. With the test tone at moderately-high average SPL, you should notice very high SPLs in the room corners, maybe as high as 9 dB (!) above the average. If you have fairly conventional speakers you should also notice very high levels directly over each speaker and in the area between the speakers about a foot above the plane of the tops of the speakers.

Cheers

Lamont Sanford
Lamont Sanford's picture
Offline
Last seen: 1 year 1 month ago
Joined: Mar 31 2006 - 8:32pm

Thanks. There really isn't anything unbelievable about it. I mean, the measuring instrument (meter) was set at 80dB before testing each tone using the full spectrum pink noise track each individual track is based upon. I did the whole 30 spectrum of tones that the pink noise was based on. The mean should come to about 80dB. At the very least the meter is working fine.

I didn't even realize the results were controversial. It just made sense to me. I did do a 12 band set of measurements. The results were more radical and less uniform. These 12 tones and the related pink noise were downloaded from Bink's site and converted from WAV to CDA format when burned to a CD-R. That is why I got a good commercial test CD from testaudio.com and did a full 1/3 octave set of measurements. The Bink's tones just didn't sound right and were home made by the time I downloaded them and converted during the CD burning process. The end result was I used a good commercial test tone CD and a good meter and results posted are what they are. I can live with them because they translate to what I hear listening to music.

If I was a speaker manufacturer I would be interested in what I did and investigate more; i.e. the SP2500 specifications and what I changed may result in a good set of new speaker boxes.

Here is a link to the CD I chose to use...

http://www.testaudio.com/testaudio/audio_test_3.asp

I don't think I need an equalizer with my HK receiver...

... That was what I was trying to determine anyway.

I don't spend all my time drinking at The Open Bar - spitting and cussing.

Lamont Sanford
Lamont Sanford's picture
Offline
Last seen: 1 year 1 month ago
Joined: Mar 31 2006 - 8:32pm

Okay, I've calibrated my SPL meter to the REW software. All I can tell it is doing is displaying on my screen the same measurement on my meter. I don't see how this is going to change anything to my manual measurements. Do I need an output cable from my computer to the receiver to use REW to make a graph on a REW generated sweep tone? It looks like I can make all sorts of graphs that look like the ones on Stereophile. You guys are overpaid.

What type of RCA adaptor will I need? My spare computer cable is male on one end and female on the other end. I would like a splitter as well to get mono to both channels on the receiver. Radio Shack qaulity. None of the $200 adaptor recommendations. Thanks....

Something like this?

http://www.amazon.com/Belkin-Audio-Cable-Splitter-1-Mini/dp/B00004Z5CP

RGibran
RGibran's picture
Offline
Last seen: 2 years 4 months ago
Joined: Oct 11 2005 - 5:50pm

I recommend this website partly owned by the creator of REW.

THIS page in particular. See "Guides/REW cabling and Connection Basics" amongst many other helpful topics which you should read in full.

If you should decide to participate in the forums be advised to check your ego at the door. No warnings. One outburst and your toast.

RG

Lamont Sanford
Lamont Sanford's picture
Offline
Last seen: 1 year 1 month ago
Joined: Mar 31 2006 - 8:32pm

Uh, to get the software I needed to register starting with that page. Did you read it all in full?

RGibran
RGibran's picture
Offline
Last seen: 2 years 4 months ago
Joined: Oct 11 2005 - 5:50pm

Just trying to help. If you had even cracked the help files you would not have asked the connection question. Attempting to use the the program without following the help very carefully is a recipe for failure.

Yes I read it all, several times and made several mistakes. But with the help of the fine folk at that forum was able to equalize my sub with excellent results but that was almost two years ago. The software has been improved. Things have changed. Dive in. Be patient. Your not going to get this done in twenty minutes. I have my doubts you'll ever get it done so prove me wrong.

RG

Lamont Sanford
Lamont Sanford's picture
Offline
Last seen: 1 year 1 month ago
Joined: Mar 31 2006 - 8:32pm

You're right. It also states the Radio Shack meter is not good above 3,000Hz. Everybody has an informed opinion on the RS meter. They want you to buy/use some other meter without explaining why. I get the impression that REW is mostly used for subwoofer measurements.

Kal Rubinson
Kal Rubinson's picture
Offline
Last seen: 1 day 20 hours ago
Joined: Sep 1 2005 - 9:34am


Quote:
You're right. It also states the Radio Shack meter is not good above 3,000Hz. Everybody has an informed opinion on the RS meter. They want you to buy/use some other meter without explaining why. I get the impression that REW is mostly used for subwoofer measurements.

Probably. REW is equally useful for the low end of any speaker as well as the rest of the spectrum. However, fixing mid- to high-frequency problems can be accomplished with reasonable acoustic treatments but those become much more massive and obtrusive as the frequency decreases. As a result, measurement and EQ of the low range is more critical.

Kal

Lamont Sanford
Lamont Sanford's picture
Offline
Last seen: 1 year 1 month ago
Joined: Mar 31 2006 - 8:32pm

I'm just waiting on a splitter. Like somebody posted, "It ain't going to be pretty".

RGibran
RGibran's picture
Offline
Last seen: 2 years 4 months ago
Joined: Oct 11 2005 - 5:50pm


Quote:
It also states the Radio Shack meter is not good above 3,000Hz. Everybody has an informed opinion on the RS meter. They want you to buy/use some other meter without explaining why.

From the above referenced Guide/Connections link:

"Both the Radio Shack and Galaxy CM-140 output a mono line-level signal. (note, we recommend the Galaxy meter for full range measurements. The Radio Shack meter is not suitable for full range, although it can be used up to ~3000Hz with no problems. HERE'S a fairly good article on the Radio Shack meters)."

Despite its shortcomings, it will give you a good indication of just how scary the top end can be in a typical room.

RG

Lamont Sanford
Lamont Sanford's picture
Offline
Last seen: 1 year 1 month ago
Joined: Mar 31 2006 - 8:32pm

I read that write-up by D. Koya. I don't understand how my simple test of the full spectrum pink noise was set at 80dB and after measuring each individual tone in the spectrum (30) the average dB was 81dB. My room isn't that large but there is a small foyer to the left and an extended area to my right that leads to the fireplace and the center of the room has a 10' vaulted ceiling running from the system wall to my listening area wall to add volume to the room (space not sound per se) and I sit right below the pinnacle of the ceiling. In fact, my favorite listening position is 11' from the center of the system at a slight angle to to the right facing the system. I actually set up the RS meter on small tripod and set it on the coffee table in front of my listening position. So it is approx. 8' from the right channel speakers with the meter turned slight to the left facing the center of the system. I realize there is a lot of information about calibration for these meters but my little test doesn't show that need as you can see from my data and the graph. I don't get it. My test CD is advertised to be used with the RS meter. Could the maker of the CD actually create a set of tones that will create these results? That is the only explanation I can come up with for such a flat graph.

Is one of Stereophile Test CDs one with the same pink noise and individual tones like I used so I can use a different media. I don't like downloading tones and then converting them to an audio CD. They always end up sounding like crap. Anyway, I need a different 1/3 octave pink noise CD to compare with the one I used so I can repeat the same test.

ncdrawl
ncdrawl's picture
Offline
Last seen: 12 years 11 months ago
Joined: Oct 18 2008 - 9:18am

EVerything you need, Lamont

I Can make you up some custom tones in Sequoia if youd like.

Lamont Sanford
Lamont Sanford's picture
Offline
Last seen: 1 year 1 month ago
Joined: Mar 31 2006 - 8:32pm

Well, thank you sir.

KBK
KBK's picture
Offline
Last seen: Never ago
Joined: Sep 30 2007 - 12:30pm

I believe I heard that an RTA program is available for my iPhone 3g. Anyone? It's quite the little computer, so I can't see why not. Apparently it has a VERY good microphone,and the hardware for i/o at an excellent quality level is there so it makes sense to me to design an RTA program for it.

Yep. About $30US, overall. You might be able to justify a iPhone on the back of this one, as a business expense.

http://www.clickapps.com/moreinfo.htm?pid=19930&section=IPHONE

The iPhone is capable of doing 44khz/16 data I/0, and all that. combined with the internal processor, this is enough to do FFT analysis like TRUE waterfall plots and Cepstral (see the liberty audio suite guys) analysis. This would become a Real and actual fully blown 'pocket loudspeaker design' package.

uhm.

Only if it has 2 channel analog input, though. I'd have to check on that.

ncdrawl
ncdrawl's picture
Offline
Last seen: 12 years 11 months ago
Joined: Oct 18 2008 - 9:18am


Quote:

Cepstral (

You mean SPECTRAL?

Kal Rubinson
Kal Rubinson's picture
Offline
Last seen: 1 day 20 hours ago
Joined: Sep 1 2005 - 9:34am


Quote:
Cepstral



Quote:
You mean SPECTRAL?

Nope. See: http://en.wikipedia.org/wiki/Cepstrum

Kal

bertdw
bertdw's picture
Offline
Last seen: 2 months 2 weeks ago
Joined: Sep 18 2007 - 5:41am


Quote:
EVerything you need, Lamont

I Can make you up some custom tones in Sequoia if youd like.

NC, is there some sort of track listing that goes with this disc, or instructions, or an explanation?

Lamont Sanford
Lamont Sanford's picture
Offline
Last seen: 1 year 1 month ago
Joined: Mar 31 2006 - 8:32pm

It looks like the Bink Test CD.

http://binkster.net/extras.shtml

bertdw
bertdw's picture
Offline
Last seen: 2 months 2 weeks ago
Joined: Sep 18 2007 - 5:41am

Thank you very much, Lamont. I like to play with these things.

ncdrawl
ncdrawl's picture
Offline
Last seen: 12 years 11 months ago
Joined: Oct 18 2008 - 9:18am


Quote:

Quote:
EVerything you need, Lamont

I Can make you up some custom tones in Sequoia if youd like.

NC, is there some sort of track listing that goes with this disc, or instructions, or an explanation?

yes, as lamont said..it is the bink test cd

1. Left Right
2. Left Right Center Surround
3. 700/1000Hz Dual Tone
4. 700Hz
5. 1000Hz
6. Log Sweep
7. Linear Sweep
8. 80Hz warble
9. 100Hz warble
10. 120Hz warble
11. 16Hz
12. 20Hz
13. 25Hz
14. 31.5Hz
15. 40Hz
16. 50Hz
17. 60Hz
18. 63Hz
19. 70Hz
20. 80Hz
21. 90Hz
22. 100Hz
23. 125Hz
24. 160Hz
25. 200Hz
26. 250Hz
27. 315Hz
28. 400Hz
29. 500Hz
30. 630Hz
31. 800Hz
32. 1250Hz
33. 1600Hz
34. 2000Hz
35. 2500Hz
36. 3150Hz
37. 4000Hz
38. 5000Hz
39. 6300Hz
40. 8kHz
41. 10kHz
42. 12.5kHz
43. 16kHz
44. 20kHz
45. Piano A 440
46. stereo piano solo
47. stereo drum solo
48. 120bpm metronome
49. Crest Wave
50. DC offset 10%
51. DC offset 100%
52. 1kHz inverted polarity
53. 1kHz -60dBFS
54. 1khz Square -2dBFS
55. 1kHz Square 0dBFS
56. White Noise
57. 440Hz 20 minutes
58. Pink Noise 20 minutes
59. Digital Silence

KBK
KBK's picture
Offline
Last seen: Never ago
Joined: Sep 30 2007 - 12:30pm

Cepstral analysis in loudspeaker design is very handy for evaluating the distortion that is generated in the output signal by internal box pressure modes, nodes, etc. It gives you the character of the distortion -in time- that lies behind the signal output. Therefore , it can clearly indicate the overall S/N ratio of the given box execution and design.

When writers in magazines talk about seeing into the signal noise floor, inner detail, snap, wavery and blubbery distortions, etc...floppy bass, midrange mud, 'goes to crap at high levels' this measurement will show that aspect comparatively, between differing designs. It separates box execution from crossover execution.

Attention John Atkinson, you should be playing with this particular measurement and see if it applies to any published loudspeaker measurements, when it comes to correlating to observed and heard qualitative levels of the given specific design..

As far as I can recall, your original stored data (quasi-anechoic fft data) can be utilized for the mathematical calculations and subsequent plotting, so you already have what you need to evaluate it's usefulness over the bulk of the history of Stereophile's Loudspeaker measurements.

Lamont Sanford
Lamont Sanford's picture
Offline
Last seen: 1 year 1 month ago
Joined: Mar 31 2006 - 8:32pm

Graph: Binks Audio Test CD Frequency Response Graph
Receiver: HK 3380
Speakers (x4): Sansui SP2500
CD Player: Pioneer DVR 220
Media: Binks Audio Test CD

Mean across entire spectrum: 76.36 dB
Standard deviation across entire spectrum: 10.31 dB

Measured from listening position with Radio Shack digital SPL meter on tripod with 1khz sine track set at 80 dB.

The Binks CD tracks are a drastic difference than my orginal graph using the testaudio.com audio test CD.

I have the stuff now to run the sound tracks straight to the receiver so I will post some more without using the CD player as time allows. If my computer, software, cables, and system all work together...

RGibran
RGibran's picture
Offline
Last seen: 2 years 4 months ago
Joined: Oct 11 2005 - 5:50pm

Nice Lamont.

Looking forward to seeing the sweeps.

No smoothing.

RG

Lamont Sanford
Lamont Sanford's picture
Offline
Last seen: 1 year 1 month ago
Joined: Mar 31 2006 - 8:32pm

The testaudio.com test cd is made up of pink noise tracks that are filtered. The Binks tracks are sine wave set at specific frequencies. Thus, the material difference between the two graphs. I would like to get a copy of the Rives test cd 2 and get a third and fourth set of measurements through the CD player and then get an average for the three CD tests (Rives has two different sets of audio sound tracks; one adjusted for the Radio Shack analog meter).

I'll probably end up just putting together a web page with results after everything is said and done.

Also, I'm going to try and make my own set of filtered pink noise. Something funny going on there. I want to test that too.

Jim Tavegia
Jim Tavegia's picture
Offline
Last seen: Never ago
Joined: Sep 1 2005 - 4:27pm

Lamont,

Thanks for the link. A great find and an even greater tool.

Jim

Lamont Sanford
Lamont Sanford's picture
Offline
Last seen: 1 year 1 month ago
Joined: Mar 31 2006 - 8:32pm

Graph: Room EQ Wizard Graph
Receiver: HK 3380
Speakers (x4): Sansui SP2500

Media: REW Measurement test. 1mb sweep taking 24 seconds.

Measured from listening position with Radio Shack digital SPL meter and no calibration file used. Also, no smoothing.

RGibran
RGibran's picture
Offline
Last seen: 2 years 4 months ago
Joined: Oct 11 2005 - 5:50pm

You're using a vertical axis of -60 to +130dB (190dB spread). This would make most speakers look great. You should switch to a vertical axis of 45

Lamont Sanford
Lamont Sanford's picture
Offline
Last seen: 1 year 1 month ago
Joined: Mar 31 2006 - 8:32pm

No problem, I'll get it done. I just went with the defaults and made sure there was no smoothing checked off anywhere. I only have a slight idea of what I'm looking at with this new graph using REW. It was literally my first try at it.

All these test tones are making me crazy. So, I listened to an original 20 year old cassette of Chris Rea's, The Road to Hell. It sounds great.

Jeff0000
Jeff0000's picture
Offline
Last seen: 1 year 2 months ago
Joined: Mar 30 2009 - 8:28pm


Quote:
I listened to an original 20 year old cassette of Chris Rea's, The Road to Hell. It sounds great.

Lamont ... Chris Rea, wow, that brings back some neat memories. I haven't heard anything about him for ..gosh, at least 20 years. I guess I was a fool to think his time is over

Monty
Monty's picture
Offline
Last seen: Never ago
Joined: Sep 16 2005 - 6:55pm

Chris Rea has always been one of my favorites. The guy has a catalog that is mind boggling. If I ever blow my speakers, it will be listening to, "It's All Gone" from his On The Beach album. I swear, the guy has at least 4 versions of every song he's ever done and they all wind up on an album somewhere down the line.

Lamont Sanford
Lamont Sanford's picture
Offline
Last seen: 1 year 1 month ago
Joined: Mar 31 2006 - 8:32pm

Graph 1: Room EQ Wizard Graph (without calibration file for SPL meter)

Graph 2: Room EQ Wizard Graph (with calibration file for SPL meter)

Receiver: HK 3380
Speakers (x4): Sansui SP2500

Media: REW Measurement test. 1mb sweep taking 24 seconds.

Measured from listening position with Radio Shack digital SPL meter with and without calibration file used.

From what I can tell on the second graph using the Radio Shack SPL meter calibration file (black line) the response is flat like my orginal graph using the testaudio.com audio test CD. Is this correct or just wishful logic on my part...

Monty
Monty's picture
Offline
Last seen: Never ago
Joined: Sep 16 2005 - 6:55pm

Graph 1 is more what I would expect from a typical listening room that hasn't been treated to minimize peaks and nulls and without substantial speaker positioning to achieve the best you can do in a given room. It's certainly not a terrible looking graph.

There are almost always peaks that can be addressed, but the nulls are a lot more difficult. By treating a peak at say 80 hz, you can usually raise a null point at say 160 hz if you have the room to move speakers around.

The radio shack meter does an adequate job for most people who don't want to get out in the weeds with this stuff and can help confirm things you already suspect from your own listening experience.

Lamont Sanford
Lamont Sanford's picture
Offline
Last seen: 1 year 1 month ago
Joined: Mar 31 2006 - 8:32pm

Thanks. I think that black line in the second graph is just showing the calibration file is in use. There doesn't seem to be any material need for the calibration file above 40Hz.

RGibran
RGibran's picture
Offline
Last seen: 2 years 4 months ago
Joined: Oct 11 2005 - 5:50pm

Nice Lamont!

The black line is the mic/meter calibration file display, from the looks of which you have calibrated perhaps as well as the Rat Shack will calibrate. I

Lamont Sanford
Lamont Sanford's picture
Offline
Last seen: 1 year 1 month ago
Joined: Mar 31 2006 - 8:32pm

The dotted red line is the corrected response using the calibration file. It was actually the same as the solid red line. I left it dotted to indicate it was a different sweep than the first graph. I'm cool and gang with it. That one huge peak at just below 80Hz sounds like it is also vibrating the cassette and CD cases stored close by among other things. I certainly don't get that during normal playback with music.

Oh, I ordered the Rives Test CD. Maybe this weekend or so.

Lamont Sanford
Lamont Sanford's picture
Offline
Last seen: 1 year 1 month ago
Joined: Mar 31 2006 - 8:32pm

Concerning the two different CD I have used thus far:

The Binks CD is fine compared to the REW results. The testaudio.com CD may be a ripoff of sort. It is made up of filtered pink noise test tones. In hindsight, of course the average is going to be the same as the full spectrum pink noise. Unless, the room response and the equipment are really jacked up. I used the Filtered Noise Generator software to confirm my suspicions. It is more of a good test for the SPL meter than room response in general. In fact, it is a misleading product. The site advertises that is made for the RS meters. Duh. We would all get a nice pretty flat response using the testaudio.com CD and the RS meter. Its designed to make you happy with the response, the meter, and the CD itself.

Lamont Sanford
Lamont Sanford's picture
Offline
Last seen: 1 year 1 month ago
Joined: Mar 31 2006 - 8:32pm

I calmed down the lower end by plugging the remaining open ports on the inboard boxes. Now all ports are plugged in all four loudspeakers. Also, the roll off at high frequency is mostly the meter's shortcoming. Nevertheless, sound at these higher levels are not fatiguing and quite clear, which is important to me. The lattice grills over the speakers make a big difference. Sansui's claim that their curved "Kamiko" fretwork grilles diffuses sound has some merit. The best measured responses are with the grilles attached.

Graph is smoothed at 1/3 octave. 1/1 octave is really nice looking but not practical for this forum. I now see a 31 band equalizer being effective for a nice smooth measurement. But I don't need it for my own listening pleasure. BTW, there are no adjustments to the treble and bass controls. The HK 3380 receiver is not a bad budget receiver. I got this rebuilt unit through HK. The receiver is almost as ugly as the speaker cabinets...

Buddha
Buddha's picture
Offline
Last seen: 12 years 5 months ago
Joined: Sep 8 2005 - 10:24am

More fascinating stuff, Lamont!

What do your ears tell you about the treble response? Is the treble harder to measure properly?

Pages

Log in or register to post comments
-->
  • X