Hi All,
First of all, I am a beginner in diy speaker building.
Currently, I am confused regarding my tweeter's measurement result that showing big dip in freq between 5 khz- 10 khz. This happens to both tweeters. What am I missing in this measurement? Can anyone help me, please?
Tweeters are SB Acoustics SB26STC 4 ohm mounted into speaker box with baffle of 23 cm x 60 cm. Tweeter is "almost" flush mounted due to my miscalculation of router's depth, the flange still 1mm protruding from baffle
http://www.sbacoustics.com/index.php/download_file/-/view/400/
Tools:
- Muse class D amp, M20 EX, 20 watts into 4 ohms to tweeter.
- REW software
- UMIK-1 measurement usb microphone
- a laptop to run REW, give pink noise, connect also to Umik-1.
Thank you in advance,
Pacogun
First of all, I am a beginner in diy speaker building.
Currently, I am confused regarding my tweeter's measurement result that showing big dip in freq between 5 khz- 10 khz. This happens to both tweeters. What am I missing in this measurement? Can anyone help me, please?
Tweeters are SB Acoustics SB26STC 4 ohm mounted into speaker box with baffle of 23 cm x 60 cm. Tweeter is "almost" flush mounted due to my miscalculation of router's depth, the flange still 1mm protruding from baffle
http://www.sbacoustics.com/index.php/download_file/-/view/400/
Tools:
- Muse class D amp, M20 EX, 20 watts into 4 ohms to tweeter.
- REW software
- UMIK-1 measurement usb microphone
- a laptop to run REW, give pink noise, connect also to Umik-1.
Thank you in advance,
Pacogun
Attachments
Try measuring another type tweeter and see if you get the same dip in response.If you don't then the SBs are faulty. How far away is the microphone? Try moving the mic back to around one meter. Having the mic very close to a driver will cause nulls in the response.
Also you did go here and follow the setup instructions? http://www.minidsp.com/applications/acoustic-measurements/umik-1-setup-with-rew
Speaker measurement http://www.minidsp.com/applications/acoustic-measurements/loudspeaker-measurements
Also you did go here and follow the setup instructions? http://www.minidsp.com/applications/acoustic-measurements/umik-1-setup-with-rew
Speaker measurement http://www.minidsp.com/applications/acoustic-measurements/loudspeaker-measurements
Last edited:
Thanks David,
Now I removed the tweeter's padding foam, so now it leveled and flush with baffle. Tweeter and UMIK are 1 meter distance and level. But still there is dip.
And for comparison, I did the measurement also with my vintage Technics SB5000 and ALSO shows dip on similar freq (the second image with full measurement from 20 HZ). This Technics should show flat freq response according to its manual.
Please have a look on attached images.
Thanks,
Pacogun
Now I removed the tweeter's padding foam, so now it leveled and flush with baffle. Tweeter and UMIK are 1 meter distance and level. But still there is dip.
And for comparison, I did the measurement also with my vintage Technics SB5000 and ALSO shows dip on similar freq (the second image with full measurement from 20 HZ). This Technics should show flat freq response according to its manual.
Please have a look on attached images.
Thanks,
Pacogun
Attachments
The dip could have to do with edge diffraction:
Diffraction from baffle edges
But it looks more like your measurement mic may have a dip in response at 8 kHz.
Diffraction from baffle edges
But it looks more like your measurement mic may have a dip in response at 8 kHz.
Is the mic calibrated?
What does the near-field look like?
What did the tweeter look link in a test baffle?
What does the crossover look like?
Pink noise is really only good for room turning. You need to change to a gated pulse measurement. I recommend HOLM or ARTA. ( free) You may be measuring reflections.
Edge diffraction of course.
Manuals are advertisements.
What does the near-field look like?
What did the tweeter look link in a test baffle?
What does the crossover look like?
Pink noise is really only good for room turning. You need to change to a gated pulse measurement. I recommend HOLM or ARTA. ( free) You may be measuring reflections.
Edge diffraction of course.
Manuals are advertisements.
The dip appearing in both measurements at the same frequency range, means either the mic or the calibration file you downloaded (You DID download the calibration file for your mic, using the serial number specific for your mic?) are not correct.Can you try the mic and REW on another computer to verify that the computer isn't causing the problem? I would get in touch with the company you bought it from and see what they suggest. Also you might ask on here MiniDSP UMIK-1 Microphone - Home Theater Forum and Systems - HomeTheaterShack.com
Last edited:
Thanks guys for your help! This speaker system will be active speakers using Minidsp and biamps.
Now, I tried to run REW on different PC (Acer SK30 desktop PC run on WinXP and old V 5.00 REW) and surprisingly, I guess the Dip is gone! I attached the measurement result of my Vintage Technics SB5000 (with same Umik-1 and same calibration file).
Does it mean that the fault at laptop (toshiba satelite L750D run on Win 7 and Beta V 5.01 REW)?
Now, I tried to run REW on different PC (Acer SK30 desktop PC run on WinXP and old V 5.00 REW) and surprisingly, I guess the Dip is gone! I attached the measurement result of my Vintage Technics SB5000 (with same Umik-1 and same calibration file).
Does it mean that the fault at laptop (toshiba satelite L750D run on Win 7 and Beta V 5.01 REW)?
Attachments
Sorry, I just realize that I use the different Technics speaker (now it is the right speaker, instead the left speaker). So to be fair, I attach also the right speaker FR measurement using Laptop.
And thanks David, I will try to ask also in that HTS forum.
When I got chance, I will try to measure the SB Tweeter with the Acer Desktop.
And thanks David, I will try to ask also in that HTS forum.
When I got chance, I will try to measure the SB Tweeter with the Acer Desktop.
Attachments
Hi there all -
RE. Tweeter 'dips & peaks'
I'll give you a little trade secret when it comes to DIY speaker design.
Make yourself a 'long lead' phase reversal switch.
This takes 4 wires and a DPDT switch.
With this switch, you can audibly hear the phase orientation of any two drivers
over and covering the region of the crossover frequency.
What I am getting at is: are your drivers 'in phase' or 'out of phase' ?
Some unknowing speaker builders just need to reverse the connection of 2 wires
to get their speakers sounding great. This 'phase thing' is very clear around 5 Kz.
IF USING PASSIVE CROSSOVERS :
If you can find a tweeter that you like that is "louder" than the mid & bass -
Run it through an L-pad.
Another little 'trade secret' = An L-pad actually dampens impedance peaks of
HF drivers, thereby aiding crossover design.
Good luck to you all,
Mark
RE. Tweeter 'dips & peaks'
I'll give you a little trade secret when it comes to DIY speaker design.
Make yourself a 'long lead' phase reversal switch.
This takes 4 wires and a DPDT switch.
With this switch, you can audibly hear the phase orientation of any two drivers
over and covering the region of the crossover frequency.
What I am getting at is: are your drivers 'in phase' or 'out of phase' ?
Some unknowing speaker builders just need to reverse the connection of 2 wires
to get their speakers sounding great. This 'phase thing' is very clear around 5 Kz.
IF USING PASSIVE CROSSOVERS :
If you can find a tweeter that you like that is "louder" than the mid & bass -
Run it through an L-pad.
Another little 'trade secret' = An L-pad actually dampens impedance peaks of
HF drivers, thereby aiding crossover design.
Good luck to you all,
Mark
This is a puzzler no doubt. Try measuring the SB tweeter without moving the mic or tweeter and just switch between the Acer and Toshiba laptops. Moving the mic even a very small amount can effect the high frequencies when measuring.If they are the same, then you need to look at the mic/calibration file being the culprit. If different, then of course the Toshiba would be suspect. Keep the decibel range and the smoothing the same between measurements , I noticed in some of the plots you posted that the range is 20 dB between markings and 10 dB in others. It just makes it easier to compare results.Yes contact people on that HTS link, maybe they have run into the same problem.As a last resort, find someone that has a calibrated mic to make the same measurements also, it's just possible what you are getting is correct.
Last edited:
REW does allow gated measurements , please see the link I already posted Loudspeaker measurement with UMIK-1 and REW | MiniDSP
Ups!
Forgive me TvrGeek and guys for giving INaccurate info. I do Not use pink noise when running Measurement Sweep. Pink noise is used only when setting db level in REW. After that, in measurement sweep mode, REW making "Whoooo-oop" sound for in room measurement.
UMIK-1 setup with REW | MiniDSP
@DavidL,
I will try your suggestion for measuring in the same condition. Because what did happen, I have to move the mic from my workshop into my living room where my desktop is. It is a pain to move around with desktop pc, so I move the mic instead.
Next step, I will try to use that laptop to measure SB tweeter in the living room, same spot where I did measurement to the same tweeter using desktop. I will post the result, guys.
Last edited:
So just to make sure I have this straight, you are going to remeasure using the desktop and also using the laptop without moving the tweeter or mic? The settings on the decibel and smoothing scale will also be the same? Otherwise we will be back to comparing apples and oranges again.
Last edited:
Yes! I will do my best to create exact same condition for fair comparison. I'll keep posting.
I tried to create exact same condition to test SB Acoustic 26STC Tweeter, using Toshiba L750D Laptop VS Acer Desktop. I do Not moving the mic and tweeter even a millimeter, only swap the mic plug and audio output plug. The Mic is UMIK-1 calibrated mic, set on 0 degree on axis.
Both desktop and laptop audio output are set to 100% and mic input volume to 65%.
However, bear with me everybody! Before I reveal the result, I found that to produce the same DB level:
- Muse D class amp's Volume has to be cranked up to 90% with Laptop, while
- Only 50% volume with Desktop
I am not too sure why like this. With my current knowledge (I am a noob), probably the laptop produce less voltage than the desktop, so laptop's audio output needs more amplification.
Now, the result for Laptop is in attached image: (still have dip)
Both desktop and laptop audio output are set to 100% and mic input volume to 65%.
However, bear with me everybody! Before I reveal the result, I found that to produce the same DB level:
- Muse D class amp's Volume has to be cranked up to 90% with Laptop, while
- Only 50% volume with Desktop
I am not too sure why like this. With my current knowledge (I am a noob), probably the laptop produce less voltage than the desktop, so laptop's audio output needs more amplification.
Now, the result for Laptop is in attached image: (still have dip)
Attachments
Last edited:
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.