JBL Master Reference Monitor

[Cappelen]

Yes I have compared a SEOS 18 (1.5" version) and the M2 WG with the D2430k, and to me the M2 WG is a better device.

In short:
Lower coloration (subjective), smoother more consistent response especially in the 0-30 deg area, alot more constant DI vertically, cleaner impulse response (generally less ripple/diffraction effects in the FR).

Also the horn walls are better damped than the SEOS aswell.

But the M2 is also much wider dispersion! 120x100deg. Vs Seos 90x50

This is a difference that will not allways be positve IMO.

Sebackman (forum member) has tried the M2 wg with the 2451BE. I've seen the measurements and it looked pretty excellent (smoother than D2430K) but a little less output in the UHF. He also had some extra WGs some weeks ago.

[srm51555]

After viewing the JBL M2 Schematic I built the best sort of brace matrix I could. I was unable to remove the particle board braces already in place from my last post so they stayed in. The three main braces are baltic birch that were glued and screwed into place. Tomorrow is the last warm day here this year so I plan to prime the insides. Has anyone purchased the sound dampening material POS recommended? My plan is to have the outside professionally painted, but want to save putting the sound dampening material until after this is done. I'd like to know if this is possible or is the sound dampening material rigid?

[Cappelen]

To all owners of the M2 wg and D2430K:

It would be interesting to see measurements from others using the EQ/crossover values in POS google docs document. See https://goo.gl/GdfQ7N

I have measured some excess energy in the 7-10kHz area using a hypex plate amp, but also when using rephase to generate this filter and running them in the convolver in JRiver. (http://www.audioheritage.org/vbullet...l=1#post381367)
I'm interested to know if these deviations are unique to my setup or if other also see them.

It puzzles med that from a subjective standpoint I don't really hear the excess energy. Padding this area down with PEQ gives a slightly more "dull" presentation especially at low volumes.

Maybe I need to try a different MIC to be sure that is not a problem with the cal file I'm using for my UMM-6....

[bubbleboy76]

To all owners of the M2 wg and D2430K:

It would be interesting to see measurements from others using the EQ/crossover values in POS google docs document.

I have measured some excess energy in the 7-10kHz area both using a hypex plate amp, but also when using rephase to generate this filters and running them in the convolver in JRiver.
I'm interested to know if these deviations are unique to my setup or if other also see them.

It puzzles med that from a subjective standpoint I don't really hear the excess energy. Padding this area down with PEQ gives a slightly more "dull" presentation especially at low volumes.

Are you measuring in an anechoic chamber with calibrated measurement system?

[Cappelen]

Are you measuring in an anechoic chamber with calibrated measurement system?

That is a good point.

I should check with a different MIC.

I don't think an anechoic chambers or outdoors is necessary to measure this area precisely as I can use gating to eliminate reflections from the room.
The deviation is always there no matter distance, angle or position in the room.

[pos]

[Cappelen]

Good observation POS.

Here is a comparison between my Dayton UMM-6 using the individual calfile provided by Dayton, and my Behringer ECM-8000 using a generic cal file (based on the data you just showed)


I find it hard to believe they are both off, about equally much in this area, I should have seen this repeat itself with other drivers if it was the case.

Both drivers show the same thing.
Both running the filter in Jriver through convolution and Hypex plate amp shows the same.

Would be nice to see someone else measuring to see if this is unique to my setup...

[pos]

I would not be surprised if the calibration file provided by Dayton was off.
It was certainly the case with the one I send to cross spectrum for calibration (the XLR version).

I should be able to measure mine this Friday, at last!

[srm51555]

I will try to take some pictures of the SAM LF box, but it looks like this basically:
Attachment 67555
http://www2.owenscorning.com/around/...coublanket.asp

Has anyone found either a local or internet source for this? I went through the Owens Corning site to find a supplier and received a very prompt response from a local dealer, but after telling him I'd only need one roll if possible, I haven't heard back. Thanks, Scott

[pos]

/!\ Warning /!\

I measured my D2/H2 combo yesterday, and I think there is a strong possibility that the cap values in the passive network schematic from the tech manual are wrong. A this point I think that there is a possibility that C2 was misprinted 33.3uF instead of 3.3uF.
That would bring the 1st order high pass filter from 209Hz up to 1kHz (!), which is of course a completely different scenario, going from a mere protection filter to a real component of the actual crossover.
And of course in this case the lpad cannot as easily be omitted as it acts on the impedance peak of the driver near resonance...

Note that at some point (a few pages back in this thread) the 3.3uF value appeared on some networks photos, but I thought it was another revision of the network and replaced C1 (originally 4.7uF, not a big deal in that case), but it might be that it was in fact C2, C1 being and staying 4.7uF...

To be sure we would need someone with the actual network to tell us the real values, but at this point it is safer to either wait for a definitive answer, or replace C2 with a 3.3uF cap.

Here is a simple measurement with the passive network and DSP crossover settings.
Red curve is the acoustical target, black curve is the measurement with the passive network as described in the manual, green curve is with the addition of another 1st order high pass at around 1kHz.
3.3uF+4.7uf would give a 8uF value, which gets you an exact 1kHz fc for the 20ohm load presented by the lpad...

That is too close to text book values to be a coincidence, and that would also make more sense than a protection cap and 1st order electrical high pass. But nothing is sure at that point, until someone tells us what the actual values are.

[pos]

Cappelen, as you can see I measured similar phenomenons as you around 8kHz, albeit with a different perspective.
Let's wait for the high pass problem to be solved before addressing that one.

[Cappelen]

Cappelen, as you can see I measured similar phenomenons as you around 8kHz, albeit with a different perspective.
Let's wait for the high pass problem to be solved before addressing that one.

Interesting POS. Thanks for puting it up.

Can you tell a little bit about your measuring routine; What distance and what kind of gating is used?

To measure the crosover properly you need a semi-long time window. For me this is hard to do indoors without introducing reflections.

[pos]

I measure indoor, but do multiple measurements at different distances and positions, and then play with gating.
It is nearly impossible to publish a single meaningful measurement in these conditions, but it is nonetheless possible to analyze them (at least for the person that actually took the measurements and knows the exact conditions), and draw conclusions from them.

[bubbleboy76]

C1 and C2 looks very similar in size...
33uf would be much bigger, wouldn't it?

[pos]

You are right indeed, as both are supposed to be of the 250V type.
Why didn't you tell us before?!
The size does vary from one network "version" to another though.

Some context:
http://www.audioheritage.org/vbullet...l=1#post375535
http://www.audioheritage.org/vbullet...l=1#post375538
http://www.audioheritage.org/vbullet...l=1#post376946

So now it appears it might not be 4.7+33.3 or 3.3+33.3, but actually 4.7+3.3...

It also seems intuitively logical to have the caps values in descending order (C1 > C2, C1A > C1B, etc.), at least this is how it has been done in most previous JBL networks.