Print Page | Close Window

REW & Measurment mic, confirmation

Printed From: Speakerplans.com
Category: General
Forum Name: Newbie Discussion
Forum Description: Newbie Discussion/Questions. Look less stupid here...
URL: https://forum.speakerplans.com/forum_posts.asp?TID=107670
Printed Date: 30 September 2023 at 7:55pm
Software Version: Web Wiz Forums 12.06 - https://www.webwizforums.com


Topic: REW & Measurment mic, confirmation
Posted By: Jack1991
Subject: REW & Measurment mic, confirmation
Date Posted: 30 January 2023 at 8:31pm
The last few months i have been learning to measure and work with rew for ground plane measurments to set up my speaker. I have been practicing on my diy portable speaker which is a tweeter and woofer, as its easy to move around to get used to it all, before i use it on my bigger system.

Anyway, i just would like some confirmation my results are correct as there is alot to rew i dont understand or even know if its applicable to what i am doing. All i used rew for was to see the frequency response and to e.q my portable speaker. The response you can see in pic 1 is how i got it now which to my ears it sounds really good, i cant imagine it sounding any better.

Pic1

The other tabs on rew; distortion, impulse, filtered ir etc etc. I do not understand even after reading up on them. I am assuming some of them are for room response which is not what im doing. But id just like to know if there is anything i have clearly missed or is obviously wrong and i could do better.

Let me know if there is other screenshots that i should post up for you to see. I just dont know which others are relevant. Theres only so much i can learn from googling.

I used winisd to design the woofer enclosure so if there are any screenshots youd like to see from that let me know. Air port velocity is just below 14ms & cone excursion is below its limit at the max wattage the amp will give.

Cheers 👍🏼



Replies:
Posted By: Norseman
Date Posted: 30 January 2023 at 11:27pm
hiya, you are correct with the all SPL and EQ.

i am in a similar position trying to learn this software.  i do not understand the choices for crossovers regarding delay or phase, i am simply using the all SPL for EQ/tuning and the IR tab for sub/kick/midtop delay to obtain data to inform speaker delay so they play in time, i have not yet figured out how phase is affected despite countless hours of reading.


Posted By: Elliot Thompson
Date Posted: 31 January 2023 at 11:51am
You do not need to have the Mic Calibration check box enabled as it only shows the frequency response of the Microphone on the chart. Room Eq Wizard has a forum if you get lost along the way. https://www.avnirvana.com/forums/official-rew-room-eq-wizard-support-forum.10/" rel="nofollow - Room EQ Wizard Forum

If your Microphone is MiniDSP, there is a forum as well that offers many using Room EQ Wizard for their personal needs. https://www.minidsp.com/support/forum" rel="nofollow - MiniDSP Forum

Best Regards, 


-------------
Elliot Thompson


Posted By: Jack1991
Date Posted: 31 January 2023 at 1:15pm
Cool ok ill toggle that off when i get home. That is litterally just that black line though isnt it?

Thanks for the links too, i have already scoured the rew forum but not the minidsp one and i do use a umik1 so ill have a look at that. Im not using rew to e.q a room. Just using it in order to see a frequency response graph, using ground plane measurments.

As Norseman said though, we are in a similar place with no matter how much more i read i still dont know if the outcome is correct. Ideally i need clarification from someone expierienced that what i am doing is correct, and if not what to change or do. 


Posted By: Elliot Thompson
Date Posted: 31 January 2023 at 3:21pm
Yes, the black line is the Microphone Calibrated Frequency Response
Since you are using the UMK-1 you can pose the questions on the MiniDSP forum for there are lots of UMK-1 users measuring with Room EQ Wizard. Pending on your level of REW,
https://avgadgets.com/ultimate-rew-tutorial-taking-measurements/" rel="nofollow - This Link  may help you along the way.

I am using the UMK-2 btw...

Best Regards, 


-------------
Elliot Thompson


Posted By: Jack1991
Date Posted: 31 January 2023 at 3:53pm
Cool just checked out that link, i understand how to use rew and i know that the e.q'ing i done is correct and what im putting into my dsp from rew is working as im expecting. I think you might be missunderstanding what im asking though, or are you saying its not suitable to ask this on this forum?

Im not looking to e.q a room. I am only using rew to see the response of my speaker and to e.q it. That part is fine. So all im asking really is the other tabs in rew; phase, IR, distortion etc etc. Are they relevant to what im doing now. 

Basically hoping someone could check it over, i would post screenshots of each section someone would need to see and theyd say "all looks good" or "that doesnt seem right".

I just want to make sure im doing it all correctly before moving onto my system. Where as this small 2way speaker is my 'practice' piece basically. 


Posted By: Norseman
Date Posted: 31 January 2023 at 10:14pm
The impulse tab is for time alignment of speakers to one another.
It will show you the timing of the impulse response captured by the mic relative to a timing reference (loopback, or, when using a USB mic like umik, an acoustic timing reference from a separate full range speaker on channel 1/L

It sends a chirp to the fullrange speaker followed by a sweep to the main stack (for instance with kicks and subs soloed, and midtops muted

In this way you can see two impulses, one for sub to which you will ascribe a new t=0, and therefore see the delay needed to be applied to the kick to bring it in time.
Then you would mute the kick and sweep sub and midtops, set t=0 at sub again, and find the delay between subs and midtops, and apply the difference as delay to the midtops, bringing everything in line with the slowest cab (sub)
Now if when using the full system a musical element that played from say 50hz to 1000hz as one burst comes, the cabs all play at same timing for the listener.

The phase component of this I do not fully understand how to use.
Some say leave off the crossovers when testing , and select appropriate crossover type later at desired points in order to maintain phase coherence.
Others have said that for timing alignment, test with crossovers in place.
I have been unable to get a jargon free answer in which should be done, let alone how it should be done, so I shall follow along here keenly.


Posted By: vertx
Date Posted: 01 February 2023 at 4:50am
1. Individually sweep drivers. Flatten response of individual components to taste
2. Gain match drivers
3. Select electronic crossover frequency. Use distortion graphs, and simulations/measurements at high output to assist crossover frequency selection.
4. Match crossover slopes of each drive unit at the crossover frequency to obtain summation appropriate for both on and off axis response. i.e do you want a reduction of gain, 0db gain match, or a summation boost at this frequency. Determining this assists the shape of the overall polar response and power response of the speaker.
5. Match up the start of the impulse responses of each driver.
6. Apply crossovers and align phase at the acoustic crossover point of both drivers via Overlay tab.
7. Measure both units together and confirm intended summation has occurred both on an off axis at crossover frequency.
8. Listen and shape input EQ to taste.
9. Adjust individual driver units gain/balance to taste.
10. Repeat steps 3-10 until satisfied with end result.

I think maybe steps 5 and 6 could be interchanged in order.

You can use the Waterfall and Spectrogram tabs to look for Time Domain issues, either linked to driver construction, enclosure construction, or room reflections.

Remember to donate to John, the author of REW and share your knowledge with others Smile


Posted By: Jack1991
Date Posted: 01 February 2023 at 9:09am
Perfect cheers guys! I can see theres a lot more i can be looking into, and not as easy as someone checking my graphs now.

Thats a real good step by step too thanks, it will be much easier for me now to break down each section and try each bit and then research each bit if i get stuck.




Posted By: smitske96
Date Posted: 01 February 2023 at 9:43am
Look at the Merlijn van Veen videos/articles. He has multiples on allignment, cardioid etc.
A good start:
http://www.merlijnvanveen.nl/en/nl/studiezaal/166-subwoofer-alignment-the-foolproof-relative-absolute-method" rel="nofollow - http://www.merlijnvanveen.nl/en/nl/studiezaal/166-subwoofer-alignment-the-foolproof-relative-absolute-method

Also look at open sound meter, would make it a lot faster and is free.


Posted By: Norseman
Date Posted: 03 February 2023 at 12:47am
[QUOTE=vertx]1. Individually sweep drivers. Flatten response of individual components to taste

this is to be done without crossovers in place?


Posted By: vertx
Date Posted: 03 February 2023 at 2:11am
Originally posted by Norseman Norseman wrote:

[QUOTE=vertx]1. Individually sweep drivers. Flatten response of individual components to taste

this is to be done without crossovers in place?

Both usually, but I like to start without filter in place at low volume, making sure you do not put too much low frequency thru a comp driver, for example.

The idea being the natural response of the driver without crossover filters will sometimes assist in picking a crossover point.

For instance, in a standard 2-way 12" woofer + 1" comp, I'd want to know the natural acoustic rolloff that the woofer has in the top end, and the comp/horn combo has in the low end, prior to adding filters. Picking then, ideally, a point where both have low distortion, equal on/off axis response that would only be improved by electrical crossover filters.

Sometimes seeing the natural driver rolloff of 6db or 10-12db per octave, for example, will determine the electrical filter one could use, i.e to get a 0db sum at crossover, with 24db per octave acoustic slopes, would require only a 12db electrical filter, which might combine with the natural driver rolloff, to form the 24db acoustic crossover slope if that's the intended target.

Intended application may also play into the above, for example using the same pair of 12" + 1" combo for HiFi listening levels vs PA listening levels, one might be able to crossover lower in the HiFi setting with lower levels, thus gaining closer centre to centre spacing between the drive units.




Posted By: Norseman
Date Posted: 03 February 2023 at 2:18am
for a four scoop, 4 kick system, is it possible to test only one scoop, one kick, one midtop for time alignment/IR, and apply this to all?

i understand this is not going to be accurate for the EQ etc as the coupling of drivers makes a propagation of waves all summed together at listening position, but would it be reasonable to do for IR timing only?


Posted By: vertx
Date Posted: 03 February 2023 at 2:39am
I've not worked on such a system so I can only speculate.

What your saying I've done successfully for direct radiating ported/vented enclosures - Do an alignment for one half of the left stack, then applied thee settings to the full left stack, then the right stack, and it's been OK (read: better than nothing/gets you in the ballpack)

For horn loaded system, the extension of the horn path and radiating area you will see by adding more enclosures may have an effect on low frequency extension, which may impact your lower crossover point selection, which will impact your alignment settings. How much would depend really but you have nothing to lose by giving it a go and then reviewing before the gig with the full stack. 


Posted By: Norseman
Date Posted: 03 February 2023 at 4:08am
Originally posted by vertx vertx wrote:

Originally posted by Norseman Norseman wrote:

[QUOTE=vertx]1. Individually sweep drivers. Flatten response of individual components to taste

this is to be done without crossovers in place?

Both usually, but I like to start without filter in place at low volume, making sure you do not put too much low frequency thru a comp driver, for example.

The idea being the natural response of the driver without crossover filters will sometimes assist in picking a crossover point.

For instance, in a standard 2-way 12" woofer + 1" comp, I'd want to know the natural acoustic rolloff that the woofer has in the top end, and the comp/horn combo has in the low end, prior to adding filters. Picking then, ideally, a point where both have low distortion, equal on/off axis response that would only be improved by electrical crossover filters.

Sometimes seeing the natural driver rolloff of 6db or 10-12db per octave, for example, will determine the electrical filter one could use, i.e to get a 0db sum at crossover, with 24db per octave acoustic slopes, would require only a 12db electrical filter, which might combine with the natural driver rolloff, to form the 24db acoustic crossover slope if that's the intended target.

Intended application may also play into the above, for example using the same pair of 12" + 1" combo for HiFi listening levels vs PA listening levels, one might be able to crossover lower in the HiFi setting with lower levels, thus gaining closer centre to centre spacing between the drive units.





this is a good point.  i know that the 12s in my midtops can comfortably go down to about 80hz if needed, with authority, but i have kicks in between.  i will try and measure the natural roll off of each cab with no filters applied, and use REW frequency range feature with some leway for the sweeps to avoid any damage, and of course, use low volume.


Posted By: Norseman
Date Posted: 06 February 2023 at 8:44am
as i understand it, REW can allow for you to input the sweep range for each sweep, so i imagine setting, say 900hz --> 20,000 hz may be an acceptable range for a comp driver whos spec sheet says 1000hz is the lower limit, without it breaking...

so similarly with woofers, maybe set the range to just outside its passband when running sweeps...?

this fellow in his presentation seems to think doing the timing this way then no longer is affected by choice of filters, and one can then choose the best filter for flat response summation...

i have no idea how phase factors into this however???  since filter choice affects phase would one then choose different filters based on what it does to the phase trace?



Posted By: bitSmasher
Date Posted: 06 February 2023 at 9:00am
I believe REW starts sweeps at _half_ the specified measurement frequency - that is something to be very wary of

-------------
https://www.instagram.com/batteryacidsoundsystem/


Posted By: Norseman
Date Posted: 06 February 2023 at 9:28am
meaning 40hz = 20hz?


Posted By: smitske96
Date Posted: 06 February 2023 at 10:54am
Whatch out with alligning pure on IR with subwoofers:

http://www.merlijnvanveen.nl/en/study-hall/120-why-the-impulse-response-won-t-work-for-subwoofers" rel="nofollow - http://www.merlijnvanveen.nl/en/study-hall/120-why-the-impulse-response-won-t-work-for-subwoofers


Posted By: snowflake
Date Posted: 06 February 2023 at 11:44am
Originally posted by Norseman Norseman wrote:



this fellow in his presentation seems to think doing the timing this way then no longer is affected by choice of filters, and one can then choose the best filter for flat response summation...

i have no idea how phase factors into this however???  since filter choice affects phase would one then choose different filters based on what it does to the phase trace?



this is the way that makes sense to me. And it seems to work. But the consensus amongst people more experienced than myself seems to be that alignment should be done after filters have been applied.


Posted By: Norseman
Date Posted: 07 February 2023 at 6:26am
round and round we go. lmao




Print Page | Close Window

Forum Software by Web Wiz Forums® version 12.06 - https://www.webwizforums.com
Copyright ©2001-2023 Web Wiz Ltd. - https://www.webwiz.net