Inconsistency in Measurements taken moments apart

andrewpirie

Registered
Thread Starter
Joined
Apr 29, 2018
Messages
6
hi there

i've been having a play about with rew and noticed that i can take measurements of my room only moments apart and find quite different results in the waterfall graph. one showing decay after 300ms around 50hz and then another showing all the bass frequencies around 50hz decaying well before 300ms.

what can be the reason for this? obviously getting false readings would lead to incorrect decisions being taken to what treatment is needed.

can it be noise from outside the room?

i'm using the umik-1 and have the calibration file loaded for that serial number.

thanks!
 

AudiocRaver

Senior Reviewer
Staff member
Joined
Nov 21, 2016
Messages
974
Location
North Carolina, USA
More  
Preamp, Processor or Receiver
Onkyo TX-SR705 Receiver
Main Amp
Crown XLS 1502 DriveCore-2 (x2 as monoblock)
Additional Amp
Behringer A500 Reference Power Amplifier
Front Speakers
MartinLogan Electromotion ESL Electrostatic (x2)
Center Channel Speaker
Phantom Center
Surround Speakers
NSM Audio Model 5 2-Way (x2)
Subwoofers
JBL ES150P Powered Subwoofer (x2)
Sounds like there is a compressor or fan somewhere in the house that you don't notice but REW does. When it kicks on, it can really mess up your measurements. So can footsteps on the floor upstairs.
 

Matthew J Poes

AV Addict
Joined
Oct 18, 2017
Messages
1,903
If you can take a measurement with the mic in a single location 3-4 times and upload the mdat file, we might be able to see what is going on.
 

andrewpirie

Registered
Thread Starter
Joined
Apr 29, 2018
Messages
6
thanks for getting back, that's great. i've just taken some new measurements. switched my cable box off, since that can hum now and again, and hopefully my upstairs neighbours weren't walking about. although there is a road outside with cars going past now and again.

i took four measurements one after the other and again there are very inconsistent results in the bass frequencies.

i've uploaded the data files. any pointers would be appreciated!
 

Attachments

  • 1st May 2 8_02_53.mdat
    2.5 MB · Views: 17
  • 2nd May 2 8_03_07.mdat
    2.5 MB · Views: 7
  • 3rd May 2 8_03_22.mdat
    2.5 MB · Views: 6
  • 4th May 2 8_03_38.mdat
    2.5 MB · Views: 7

John Mulcahy

REW Author
Joined
Apr 3, 2017
Messages
7,297
Pretty much everything below 70 Hz in those measurements is just noise, the level starts to roll off below about 400 Hz. You haven't selected an output device in REW so not much info about what you are connected to - what speakers are you using and how are the outputs from your PC connected to your system?
 

andrewpirie

Registered
Thread Starter
Joined
Apr 29, 2018
Messages
6
Pretty much everything below 70 Hz in those measurements is just noise, the level starts to roll off below about 400 Hz. You haven't selected an output device in REW so not much info about what you are connected to - what speakers are you using and how are the outputs from your PC connected to your system?

that makes sense that it's not right

i'm using my hp laptop and using the 3.5mm headphone output to a 2 x rca interconnect into my naim pre-amp.

thanks
 
Last edited:

John Mulcahy

REW Author
Joined
Apr 3, 2017
Messages
7,297
Couple of things to try then:
  • Connect/drive only one channel at a time (can do that in REW by selecting output channel as L, then selecting output channel as R, currently have L+R selected). If that restores more normal bass levels one channel may be inverted
  • Make sure the bass modules are connected and being driven :)
 

andrewpirie

Registered
Thread Starter
Joined
Apr 29, 2018
Messages
6
Couple of things to try then:
  • Connect/drive only one channel at a time (can do that in REW by selecting output channel as L, then selecting output channel as R, currently have L+R selected). If that restores more normal bass levels one channel may be inverted
  • Make sure the bass modules are connected and being driven :)

i'll try that. though i think all that is good. the bass cabinet it definitely connected as that's the main speaker input.

i'm pretty certain it's cabled correctly and not out of phase. will make doubly sure!!

would i be right in thinking that i should obtain very consistent results if the speakers are correctly wired and working?
 

John Mulcahy

REW Author
Joined
Apr 3, 2017
Messages
7,297
Yes, if there is signal there to be measured it should be very consistent, provided the mic is in the same place each time. The issue is the apparent lack of bass there to be measured.
 

Matthew J Poes

AV Addict
Joined
Oct 18, 2017
Messages
1,903
I think John gave you the best advice. There isn't enough bass in this measurement to suggest everything is right.

Can you give more info on the signal path? What is the bass source? A subwoofer? How is it connected? What is your source? Preamp? I'm just wondering if we are getting only the bass limited portion of your speakers and not your full system, that the sub or bass module isn't being send the signal for some reason.

where is the mic? It's a very clean impulse but there is also clearly room in that measurement.

You might try switching to a 1m FFT length to get a lower noise floor and a slightly louder volume. Maybe another 6db's or so. That might also help ensure that any noise from outside sources is minimizing its effect. As John said, below 100hz there is mostly noise in the measurements.
 

andrewpirie

Registered
Thread Starter
Joined
Apr 29, 2018
Messages
6
hey guys

thanks for taking time to get back to me on this. this is user error!! one speaker was out of phase. i changed the speaker cables and put in some second hand naim naca5 that day which had been terminated incorrectly - one of the plugs was the wrong way round and hadn't listened properly!

i was concerned about the measurements i'd taken before as i didn't really trust them, hence doing the raft of measurements on this thread at the start. i've added measurements i've taken today in this post with the treatment i have, less the two gik monster traps, and then without the two corner gik soffit traps removed.

i've also added the original measurement which is totally different, and again lacking bass, but at that point i had other speaker cable, was wired correctly and sounding fine. i was speculating that the umik-1 mic wasn't being used on the 27 april measurement. this looks similar to the previous graphs i posted with the low levels in the bass frequencies.

these new measurements have been taken at 1m. i did try at 81db, but thought that it must be clipping, due to the large amounts of decay in the waterfall. but it's the same graph at 75db.

do these new measurements seem ok? when the tone starts there are a couple of clicks and then another as it moves along the bass frequencies. i'm suspicious of the low end here, and that it's over done.

also, the mic is sat on the back of my listening couch at the point i sit in and pointing flat out in front

i've also attached screen shots of the preferences and measurement page of the measurements taken today

thanks again

andrew

ps it's a straight hifi room. there is no additional sub
pps the room is 6.2m long, 4.1m wide and 3.2m high
 

Attachments

  • with treatment May 3 11_26_02.mdat
    2.9 MB · Views: 6
  • without soffit and monster traps May 3 11_29_52.mdat
    2.5 MB · Views: 4
  • without monster traps May 3 11_29_52.mdat
    2.5 MB · Views: 2
  • Andrew Room with Room Treatment Apr 27 11_58_32.mdat
    2.5 MB · Views: 3
  • rew settings screen shots.pdf
    391.5 KB · Views: 7
Last edited:
Top Bottom