First time OB builder

Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.
Robert has already given me a few clues on how to use the MultiWin, and I have experimented with varying degrees of success. You mention that you leave the first line at the far left (0ms) and move the second red line to the floor bounce (start of first bump). I was under the impression that you put the first red line at the point of floor bounce and the second and third red lines to the right of that. I might give what you’re doing a go and see how that goes.

I have been giving it some thought as to how to measure, and really are we just kidding ourselves if we measure to get optimal results when we would get a different result from our listening position. Which has led me to think perhaps the truest way of measuring the speakers is to place the mic at ear height in the seated position, as that is where your ears are going to be placed while listening to the speakers. Should we really be adjusting the speakers to get the best results there?

Hi again - timezone is different up here - slowly waking up, sipping my first cup of coffee :faint:
As i understand it. The first and last line indicates the outer borders. So for now, leave them "out" there ;)
The second red line should be placed right before the first floor bounce. It works here. There are still ripples at the measurement, when gating correct. These might be from your front baffel or driver design.
If you measure from your listening seat. You are including the room to much and therefore making the sweetspot very small.
Measuring a speaker, is about making the speaker working its best, in its enclosure - without the room affecting.
Hope it helps a little :)
An externally hosted image should be here but it was not working when we last tested it.
An externally hosted image should be here but it was not working when we last tested it.
An externally hosted image should be here but it was not working when we last tested it.
 

Attachments

  • To soon.jpg
    To soon.jpg
    113.9 KB · Views: 54
  • To late.jpg
    To late.jpg
    116.8 KB · Views: 51
  • Getting close.jpg
    Getting close.jpg
    112.2 KB · Views: 54
Last edited:
Hmmm... tried to make links for pictures. Did not work. Then I found out that I could upload pictures directly. And know I cant remove the links again.
Pic 1: To much gating = low frequency missing
Pic 2: To little gating = some room reflections included
Pic 3: Getting close. Some disturbance might come from baffel and driver design.
 
The Help file in the software is basically useless and out of date (wrong version), and well under written, but what I saw in the DuoWin section lined up with what I understood Robert to say.

Still learning how to use it properly, but I put the first line at the bump, the second line I use to control the top frequencies and the third line to control the bottom frequencies.

I make the window only as big as I need it and no bigger to make it more precise. As I say I still have lots to learn and may be doing it completely wrong.
 

Attachments

  • Help File.jpg
    Help File.jpg
    563.7 KB · Views: 60
Well IF you are doing all of this manually, yes. BUT if I am understanding you correctly, the Ground Sound DSP is auto-magically performing all of the xover and EQ corrections to achieve the target functions, yes?

What does Ground Sound recommend?
Greg

NO - not auto automatic - the user still has to steer the software – no one button setup solution.

The optimizer can calculate PEQ to optimize the frequency response according to the parameters than the user has full control of and afterwards the use can sort them out or alter the result or delete some of the PEQs and even recalculate some of the result or from scratch. This tool require the skills of the user of how to and it give the user opportunity to make a top notch result – full flexibility at all time.

Well – although the software is a powerful tool, it requires correctly “analysed” data and before you get correct analysed data you have to have correct measurements. This has been a challenge this far!

We have been struggling with the hardware and connections – I assume that Paul has solved the issues according to my instructions? Especially not having phantom power from input 2 to output 2 in the reference signal path - TRS-XLR cable replaced by TRS-TRS cable I hope! I have encouraged Paul to send me some measurements after the right implementation of the hardware, so that I could help him getting the gating correct and sort out if the measurements now are near normal standard – nothing yet.

Robert GS
 
The Help file in the software is basically useless and out of date (wrong version), and well under written, but what I saw in the DuoWin section lined up with what I understood Robert to say.

Still learning how to use it properly, but I put the first line at the bump, the second line I use to control the top frequencies and the third line to control the bottom frequencies.

I make the window only as big as I need it and no bigger to make it more precise. As I say I still have lots to learn and may be doing it completely wrong.

Yes - I am working on a manual for the software - will take some time though.

Paul – you can do it like Tordenguden show it. Even better is to use the ETC visualization (button in the toolbar next to PIR) then you will see something like the attached picture.
The first red line should be placed where the signal rises above the noise floor and the second red line at the first reflection. This will give you the frequency response of the driver without room.
 

Attachments

  • MultiWin_ETC.jpg
    MultiWin_ETC.jpg
    204.6 KB · Views: 62
Hi Tordenguden I did try it your way, but I felt I had better control over both ends (low and high) the way I was doing it. I don’t know what the difference is between the MultiWin and the DuoWin as I am trying to curb my natural curiosity and not try learn too many things at once. There is an incredible amount of functionality in the software and I don’t want to skip ahead too soon and miss something vital. But if I was to hazard a guess I would say that MutliWin, AmbiWin & DuoWin would be all somehow inter connected. Guess we need to wait for Robert’s useful version of the Help to find out what each of these things do.

I probably should point out for anyone not familiar with the software, the Help file that comes with the software is from a 3rd party manufacturer and is less than optimal. Robert has been kind enough to try bridge the knowledge gap / write clearer instructions for people like myself who are not experts in setting up and using the software. To a large degree the software manufacturer makes an awful lot of assumptions that you have knowledge of just about everything. People like myself who have never plugged in a microphone, or seen an external sound card before are left floundering without the incredible assistance of people like Robert, I can’t praise Ground Sound’s customer service enough.

Hi Robert, unfortunately no I still don’t have the cables sorted. I am still waiting on that new TRS Y piece to turn up from the US. I can see it was posted on the 8th of December, but once it leaves the US there is no tracking on it. I hope it turns up soon, as I am eager to have the cabling problem sorted properly once and for all.

I have the feeling it probably won’t be long enough anyway, and I might need to order some additional cables, as I ordered it before you explained why I needed the TRS-TRS (no 48v phantom) it was just luck that I guessed the right one to order, but going off the picture it doesn’t look long enough to loop from the front to the back, which means I may need a TRS female to TRS male extension.

I have also been practicing a bit more with the Roland 55 AU to get the SNR up. What I found was while I had high readings on the tweeters by the time I got down to the mid woofers and woofers it had fallen below an acceptable level, so I believe I now have it tuned in a way where I have good solid green readings on all of the drivers, which showed immediate improvement in the flatness of the captured response, especially in the lower frequencies. I am now also using the ASIO mode like you suggested.

I had a quick go at what you described and I think I have it right… From your picture I can’t see what button you have pressed down the bottom to get your third line. Are you using the same version of software as me?
 

Attachments

  • ETC.jpg
    ETC.jpg
    259 KB · Views: 52
I'm using version 1.17
Still looks weird - but I think it will clear up, when you get the rigth cable. SNR will allways be worse, the lower the frequency - it's the reflections and long wavelengths again :)
Why did you buy in the US? Don't you have music-gear-shops down there? It's just a 1/4" jack to jack cable, at about ½ meter. Should be extremely common in the music business.
Could you post pictures of the soundcard with cables and the buttons. Then it will be easier to figure out, if we are doing the same thing ;)
 
Are you using the same version of software as me?
I think it depends upon OS Windows XP here and Windows 7 on your laptop?

Tip of the day ;) is:
If you already know that the tweeter has much higher sensitivity than the rest of the drivers - say 8dB - then I recommend you to use the option of the DCN28 to attenuate the output of those two single channels (tweeter stereo). This is done via the encoder on the front panel menu (push 7 times) displays "Ch. Att." and turn to the tweeter channel (EG Ch1 & 5) push and attenuate to -8dB. Now you will be able to obtain better Signal Noise Ratio in the measurements of the other drivers without stressing the tweeter.

Robert GS
 
Ok the themes or styles in Windows XP would explain why your buttons look different when highlighted. Yes I am running Windows 7 on my laptop. Would I be correct in assuming that the buttons you have selected are the AmbiWin, DuoWin and Normalise?

There isn’t a great deal of efficiency difference between the drivers. The dipole tweeter is around 93dB at the front and slightly less at the rear. The 12MU8731T00 are 87.2dB wired in parallel which should give me around 93dB, and the 18WU4741T00 are 87.2dB wired in series - parallel (not exactly sure what that give me).

Is the attenuation through the DCN28 the same as the attenuation that is set through the software for each driver? i.e. do they do the exact same thing?

Tordenguden see pics for wiring. Instead of just replacing the loopback cable for another one, the Help says it is more accurate to use a Y adaptor.
 

Attachments

  • IMAG1184-1.jpg
    IMAG1184-1.jpg
    185.5 KB · Views: 66
  • Stimulus.jpg
    Stimulus.jpg
    79.7 KB · Views: 56
  • TRS female to 2 TRS male.jpg
    TRS female to 2 TRS male.jpg
    24.2 KB · Views: 48

There isn’t a great deal of efficiency difference between the drivers. The dipole tweeter is around 93dB at the front and slightly less at the rear. The 12MU8731T00 are 87.2dB wired in parallel which should give me around 93dB, and the 18WU4741T00 are 87.2dB wired in series - parallel (not exactly sure what that give me).

Well - that's the data on the paper you are referring to?
What did you measure?

The attenuation via the front panel menu is done in the analogue domain AFTER the DAC, so no bit reduction in the digital domain = better.
If you attenuate in the software setup you will attenuate in the digital domain.
Conclusion - if there is a big difference between driver sensitivity AND gain of used power amplifier channels it is advisable to attenuate the drivers via the analogue path. I think you use a multi-channel amplifier at the moment and the gain of these channels will be the same whereas if you where using different power amplifier from different manufacturers you would also have to take this into account and this is simply by comparing the measurements of single drivers/power amplifier combination relative levels.

Robert GS
 
Yes that is based on data sheets but currently I have no gain added to any of the drivers, and they all seem to produce a fairly even SPL, but as you say 5 out of the 7 channels are currently coming from the same amplifier. I will keep this in mind for down the track when I upgrade the 5 channel amp for something better.
 
You could also (for the group of drivers under test) run a pink noise signal through each channel separately ... align the levels with Roberts analog stage, or your amps. I'd listen to some music program then just to be sure the tweeter isn't too low in level. (Might sound fine to you ... might want another 2-3db of gain)

Then run your tests.
 
Why did you buy in the US? Don't you have music-gear-shops down there? It's just a 1/4" jack to jack cable, at about ½ meter. Should be extremely common in the music business.
Could you post pictures of the soundcard with cables and the buttons. Then it will be easier to figure out, if we are doing the same thing ;)

Because most Australian retailers have no concept of the word moderation. What costs $5 in the US cost $20 here, so I can buy from the US pay freight on small objects and still cost much less.

The other problem is range... Anything that isn't mainstream isn't kept and needs to be ordered in for an even more ridiculous price. The Y adaptor is not a mainstream item.
 
You could also (for the group of drivers under test) run a pink noise signal through each channel separately ... align the levels with Roberts analog stage, or your amps. I'd listen to some music program then just to be sure the tweeter isn't too low in level. (Might sound fine to you ... might want another 2-3db of gain)

Then run your tests.

You have just given me an idea. At the moment I have only worked out how to create a xover that is ganged to the first channel. Which in turn provides no flexibility to the second speaker crossover i.e. Separate delays etc (I am hoping this is just me and not a software limitation). Using the individual attenuation in the DCN28 I could at least do a pseudo balance if the room dictates that one speaker must be closer.
 
You could also (for the group of drivers under test) run a pink noise signal through each channel separately ... align the levels with Roberts analog stage, or your amps. I'd listen to some music program then just to be sure the tweeter isn't too low in level. (Might sound fine to you ... might want another 2-3db of gain)

Then run your tests.

Just looking at the full range graphs in posts no's 356 & 360 and it looks like I might be able to drop out two of the 18WUs, as it is a few dB higher, as an alternative to attenuating the channel.
 
Been doing some more measurements this morning. An interesting observation when measuring each of the drivers in its “optimal” mic position (on axis ~6” away)… each driver is a lot flatter then when it is in a not so optimal position (measuring another driver).

Which has me thinking again while it is great for speaker manufactures to measure their driver in the ultimate light that makes them look fantastic. Unless all of the drivers are clumped in one spot to some degree you have to be listening to some of the drivers in a vertical off axis position.

Which makes me further think any correction should also be done when measured from the listening position. I won’t get it as flat as if I was measuring near field, but I don’t plan to listen to the speaker from 6” in front of the tweeter, I plan to sit down to listen to them.

Do I want to know what I really have in my seated position? or have a dead flat response to say it looks good? I listen with my ears not my eyes, so I am thinking what is right is from where I am sitting.
 

Attachments

  • More Measurements.jpg
    More Measurements.jpg
    378.9 KB · Views: 139
I'm a noob when it comes to this kinda of stuff, however I'm a believer in listening from the intended listening position. What good is a freq response curve that looks great from say 1m, when at 2-3m off axis the whole picture falls apart.

I do believe that near-field measurements are great for getting crossovers correct, measuring at the intended listening position and of course your ears should be the final say so.

I know for a fact that my setup isn't ideal, however it sounds great so far with very minimal "tuning" and I'd be hard pressed to make any more changes to it until I have much more time on my hands. Also - the wife approves :D

EDIT: The only downside as far as I can see about measuring at the listening position is discovering how "bad" a room can be to accurate frequency response...treatments and speaker position could be vital here more so than adjusting the audio source-to-speaker chain itself.
 
Last edited:
Yes I tend to think the answer lays in there somewhere...

I can't swing a cat in here now so there is no room for accoustic improvements, but that is not to say it isn't on my list of things to do in the place I will be getting.

I am just trying to convince myself it is a good idea because most of the people I think know what is going on seems to measure closer to the speaker.

When I try to do individual speaker measurements, then corrections, the corrections sound like crap, and the base crossover sounds better.

I have had some improvements with the corrections but I think that was when I was closer to the speaker, so it looks like more trail and error is required.

I also can't wait to get my cables sorted out because I have no idea how much that is contributing to measurement corruption.
 
Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.