My first REW Measurement - DIY Speakers - Did I get speaker polarity wrong?

[nikb]

I would concur that this looks like a mic issue first and foremost... The exact nature of the issue is another question but @Keith_W has a good idea IMO. If you can get your hands on a UMIK or other known good mic we will make some progress.

I dont think I know anyone here with a measurement microphone - I might have to just order one (again).

 

[Keith_W]

Right now you want to diagnose the problem at the lowest possible cost. If you can't borrow a microphone, try playing pink noise and measuring the RTA with an app on your smartphone. Or better still, try using two different smartphones. Smartphones are notoriously inaccurate, but even a smartphone measurement will tell you if your speakers have that 20dB upward tilt. If two smartphones say the same thing, then you have effectively confirmed that it is your measurement setup which is at fault.

 

[Digital_Thor]

My quad capture Roland sound card has a high pass function for voice/speech, which could explain why the curve is so nicely raising with frequency - and considering, that most deviations with microphones are in the higher frequencies, that could be the cause.
So maybe some high-pass function in your sound card software is turned on, that prevents unnecessary low frequency rumble, when the microphone is specifically used for speech?

Since most microphones measure pretty ok in lower frequencies, you could try with just your phone and some freeware, and see what you get

 

[Digital_Thor]

Hmmm... seems like there is no software for the sound card like on Roland, where you can specifically control the card. The Stage line product only have a physical button that introduce a 100Hz HP.
https://www.img-stageline.com/products/audio-tools/preamplifiers/mpa-102/
So, if my previous suspicion should hold any water, it should be a windows sound setting somewhere... but increasingly unlikely, since windows do not have a specific HP function - only some sound cards have this.
Still, I would try your phone - it's quick and free. Inspiration here:
https://www.diyaudio.com/community/threads/android-apps-for-measurements-of-loudspeakers.413429/

 

[Hayabusa]

Sorry, I did not quite understand what you were asking about. Now I think I understand better: my microphone is a ECM40 FMJ. On the body it says "Stageline ECM-40 electret measuring mike". So to me it looks like an electret condenser microphone, in which case I should try to compare it with a known good mic.

On the other hand, as others have pointed out, this will probably not lead to a 20db drop (unless its totally broken). I am back suspecting the room in combination with really bad positioning of speakers and listener. Just playing with the REW room sim.

It needs phantom power...

 

[nikb]

Right now you want to diagnose the problem at the lowest possible cost. If you can't borrow a microphone, try playing pink noise and measuring the RTA with an app on your smartphone. Or better still, try using two different smartphones. Smartphones are notoriously inaccurate, but even a smartphone measurement will tell you if your speakers have that 20dB upward tilt. If two smartphones say the same thing, then you have effectively confirmed that it is your measurement setup which is at fault.

I have played some loud pink noise from both speakers using REW (the "known good" speakers measured earlier) and measured the spectrum at the seating position using my Ipad (using the built in microphone) and the free app "Audio / Spectrum Analyzer". This is the result:

This lets me hope that the upwards facing slope measured earlier was indeed an artifact.

 

[nikb]

Hmmm... seems like there is no software for the sound card like on Roland, where you can specifically control the card. The Stage line product only have a physical button that introduce a 100Hz HP.
https://www.img-stageline.com/products/audio-tools/preamplifiers/mpa-102/
So, if my previous suspicion should hold any water, it should be a windows sound setting somewhere... but increasingly unlikely, since windows do not have a specific HP function - only some sound cards have this.
Still, I would try your phone - it's quick and free. Inspiration here:
https://www.diyaudio.com/community/threads/android-apps-for-measurements-of-loudspeakers.413429/

The StageLine mic preamp has four physical buttons: phantom, phase, LP 12khz and HP 100hz.
I have only used the phantom button and therefore disabled phantom power at the interface.

I have no idea, what the phase-button is supposed to do.

 

[Keith_W]

This lets me hope that the upwards facing slope measured earlier was indeed an artifact.

Agree.

Now the question is - is it your PC, or your interface, your cabling, or your microphone?

Re: PC. As mentioned, sometimes Windows can screw your measurement up. But you're not using Windows. I don't know anything about Linux, so I can't help you there. Sorry.

Re: interface. Check, double check, and triple check that you don't have some setting turned on that shouldn't be. Some of those interfaces have on-board DSP. I don't know about yours.

Re: cabling. Sometimes XLR cables are not constructed properly. If there is a short in your cable, it can produce an odd looking frequency response. I certainly suffered from that problem before because I am such a cheapskate that I only buy rock bottom price cables. Well, as it turns out, crappy cables may have manufacturing defects. Try swapping your cable for another XLR cable. Just to be sure.

Re: microphone. If all the above check out, then by process of exclusion, it's your microphone. Time to hit Amazon.

 

[Hayabusa]

The StageLine mic preamp has four physical buttons: phantom, phase, LP 12khz and HP 100hz.
I have only used the phantom button and therefore disabled phantom power at the interface.

I have no idea, what the phase-button is supposed to do.

did you see my post on phantom power?

 

[nikb]

Agree.

Now the question is - is it your PC, or your interface, your cabling, or your microphone?

One question: my understanding is that my calibration file is for my combination of microphone and pre. I am suspicious of the preamp.
I could connect the mic stgraigt to the interface and use phantom power from there. Would that be a problem?

 

[JayGilb]

Since the motu has mic level input and phantom power why even have preamp in the signal chain ?

Have you tried directly connecting the mic to the motu ?

The mic appears to have a standard xlr connector.

 

[nikb]

did you see my post on phantom power?

Yes, maybe I misunderstood: the preamp is powered and has its own power supply. When I connect the motu to the pre, I do not get a signal unless I push the phantom button on the pre.

 

[Keith_W]

One question: my understanding is that my calibration file is for my combination of microphone and pre. I am suspicious of the preamp.
I could connect the mic stgraigt to the interface and use phantom power from there. Would that be a problem?

Sure, it's worth a try.

 

[JayGilb]

One question: my understanding is that my calibration file is for my combination of microphone and pre. I am suspicious of the preamp.
I could connect the mic stgraigt to the interface and use phantom power from there. Would that be a problem?

If the mic uses 48 vdc power, I see no reason it shouldn’t work connected to the motu.
If you can verify that mic requires 48v and not 24 or 12, you should be able to skip the preamp in your signal chain.

 

[nikb]

I have made a new measurement with yesterdays "known good 2 ways", this time without pre (phantom power on motu was turned on instead).
Look at this:

For this measurement, the calibration file for the mic/pre combo was used. What do you think: will this be good enough for future measuerments?

 

[Juhazi]

Now it's ok! Just use higher spl. Always use exactly same positioning of speaker(s) and mic in later measurements.

Please use freq scale 20-20 000Hz and dB net 50dB (eg. 30-80dB) Room noise level varies typically between 30-50dB. In your graph everything below 30Hz is room noise (ambient)

If you measure at spot and L/R/L+R be prepared to see hugely different response curves and impulse etc.

It is a different scenario to measure quasi-anechoic,nearfield or room response. Outdoors measurement would be nice to see, study what the speaker's anechoic output is vs. the room effects.

 

[kemmler3D]

I have made a new measurement with yesterdays "known good 2 ways", this time without pre (phantom power on motu was turned on instead).
Look at this:

View attachment 487082
For this measurement, the calibration file for the mic/pre combo was used. What do you think: will this be good enough for future measuerments?

Yep looks like something was funny with your mic pre, I guess! Maybe the phantom power wasn't working. This and the app measurement are both basically normal looking.

Now I would say you can apply the earlier advice on measuring driver phase and start diagnosing your DIY build!

 

[DVDdoug]

Off topic, but...

Older electret microphones can lose the charge and may not remain linear. Most of us use omnidirectional condenser (non-electret) microphones for taking audio measurements.

I have an old Radio Shack SPL Meter (probably 40 years old) which has an electret mic. Awhile back I was going to do some measurements and I was thinking the mic may have "deteriorated". But instead of buying a new SPL meter I bought a calibration gizmo. To my surprise the SPL meter was still perfect (as accurately as I can read the analog meter).

 

[nikb]

So now I am returning to my diy speakers.
I was suspicious of the left speaker lacking bass so I switched them around. Still think this speaker is lacking bass - actually I am sure. So I pulled the crossover and double checked it, but found nothing. I even reflowed some bad looking solder joints.
Put it back together, but no change. That speaker sounds anaemic, with practically no bass.

The bass filter of the crossover only has three parts (LRC). Can a fault in one of the other filters lead to a lack of bass by diverting the signal in some way?

I will have to pull both crossovers and compare them I guess.

 

[JayGilb]

So now I am returning to my diy speakers.
I was suspicious of the left speaker lacking bass so I switched them around. Still think this speaker is lacking bass - actually I am sure. So I pulled the crossover and double checked it, but found nothing. I even reflowed some bad looking solder joints.
Put it back together, but no change. That speaker sounds anaemic, with practically no bass.

The bass filter of the crossover only has three parts (LRC). Can a fault in one of the other filters lead to a lack of bass by diverting the signal in some way?

I will have to pull both crossovers and compare them I guess.

Please post a response plot of your diy speakers with the specs posted by @Juhazi now that you have the mic problem figured out.