Box port or crossover dippity do

[8ntractor]

nearfield is not to bad. I can measure each woofer and the port at 12" or 6 "

ground plane I could try that I would have to read up some more on it. it looks like you need a larger space. I can point the speaker at the ground and put the mic on the ground. let me do some more homework on it.

 

[AllenB]

Just to clarify, an internal box sim plus an external box sim plus a filter transfer function sim would give near the same end result as an anechoic measurement (maybe more useable too). Maybe this gives a clue as to how confident I feel about answering your earlier question at this time?

 

[Lojzek]

I am not quite sure what is the issue that we are trying to solve.

I'd be curious how the individual driver FR with/without crossover would look like on a single graph, their intersection in the xo region. I would put the mic 1 meter away, between the woofer and tweeter axis, straight in front of it and choose the gating for woofer about 200 Hz and 800 Hz for the tweeter. I recall doing a simulation some time ago and the tweeter did not need padding and I removed the R3 and C4 to boost the hf as much as possible.

There is also the possibility to prepare the frd's and zma's as described in White Paper "Accurate In-Room Frequency Response to 10Hz" by J.Bagby, and work out crossover in simulator first.

 

[AllenB]

Thanks Lojzek, getting good data to start with can be challenging, and when I can't see it or touch it.. well, it's like playing chess and changing sides.

 

[8ntractor]

Thank you both and thank you for your patience. I know I wasted your time with the poor measurements and the spl meter.


I read the white paper and some information on the ground plane I don't see how I can do the ground plane and I do not have excel. the programs do not work correctly in open office. The white paper may be beyond my ability.

the reason for this thread is to make what ever changes necessary to the cabinet, port or crossover to try to find a way to use the Nextel drivers in the Thor as a replacement for the Magnesium drivers that corrode in some environments. I am woefully unqualified to do this but I cannot see replacing the drivers every few years. and I know that there are others with the same problem. The Nextel are not a slouch as a driver they are used in the pass and montana speakers and others and in some ways may be as good or better than the mag drivers. I am willing to build new cabinets if needed. I would like to use as much of the crossover as possible and the speakers and measurements that I am getting look, to me anyway, like the crossover is close.

I also wanted to learn more about speaker building and have tried to read and follow Allens excellent sticky on speaker design without measurements. and several other websites and even a book on the subject.

here are some measurements from 1/4" from the drivers. (they would make great headphones just need the tube amp hat from the old stereophile ads) with the cross and a measurement of the tweeter and woofer pair without the crossover. I have loaded these into xsim.

Again I am sorry for my lack of measurement skill and knowledge.

 

[AllenB]

You know, if you had comparable generic manufacturer response plots from each driver, with impedances, the crossover changes could be simulated.

 

[8ntractor]

Yes I do have those. Here is the original thor crossover with the sim Nextel drivers and millennium tweeter.

attached is also a proposed crossover with a resistor to flatten the impedance and the second and third caps and the two resistors removed from the tweeter side. the coil is increased on the woofer side from 1 to 1.15
adding a 1 cap in parallel to the 8.2 cap on the woofer side

not sure about the resistor without a cap to lower the impedance.

not sure about removing so many components on the tweeter side I currently have removed the resistor and cap that is in parallel with the tweeter.

because the measurements do not really match xsim I am reluctant to do the changes. attached also in 18inchs with and without the star felt.

 

[8ntractor]

for reference here is the original thor crossover with the magnesium woofer simed

 

[Lojzek]

Actually we could use some spl measurements from 1 meter away for a sim. I think this would not be difficult for you to execute. I propose mic in front of the tweeter ( this is 0 deg horizontally and vertically, dead center). The gating for the tweeter 800 Hz, for the woofers 100 Hz. Woofers need to be connected in parallel, no filter attached, for the tweeter neither. Raw measurements without any smoothing applied, for the XSim will perform 1/24 smoothing anyway. It is important that amplifier voltage remains constant at all times. You would provide a separate woofers.frd and a separate tweeter.frd that holm lets you export as textual files. We will deal with zma's later.

 

[AllenB]

zma's later

Yes, unless you have impedance without crossover.

 

[8ntractor]

thank you -- I will get those measurements in the morning I have them with the spl meter but wuld trust the unik to be more accurate. The amp unfortunately is a Marantz 1060 as my primare is being fixed. not sure if that amp will maintain constant voltage. I do have the tweeter and woofer pair at 32" that I ran yesterday. it is some ugly gated at those levels post 165 has the same measurements with slightly different gating. I did enter them into xsim. as frd files. is the one meter critical? the room is going to get worse. I attached the xsim these are the raw drivers. thank you these are not gated the chart is gated woof and tweet 32 inch centered on tweeter no cross woof pair

 

[AllenB]

Yes, unless you have impedance without crossover.

I meant downloaded from the manufacturers website.

 

[8ntractor]

I do have the zma file from lojzek I think they are from seas.

 

[Lojzek]

The amp unfortunately is a Marantz 1060 ... not sure if that amp will maintain constant voltage...is the one meter critical?

If you don't touch volume controls (both amp and sound card) it should remain at a constant level, assuming wire connections are ok. 1m is just a convenient distance to capture the far field sound.

 

[8ntractor]

Thank you the gated cross below is the result of the gated raw woofer and the gated raw tweeter at 32" centered on the tweeter. I have the frd files but cannot attach them.

 

[Lojzek]

Forum does not accept .frd extensions. Either rename to something.txt or zip the files. There is also a possibility to attach the XSim file which will pass through ok. I prefer separate files from holm.

 

[8ntractor]

Thank you it loaded the txt files. I will include the orginal crossover

 

[8ntractor]

here is a simple cross with the gated numbers the woofers stays the same the tweet looses all but C2 when I run this with the manufacture frd the L3 also stays to get the best fr.. Lots of speaker ripple but everything over 400 is within +/- 2db

 

[8ntractor]

maybe I am over thinking this. here is a chart with and without felt gated to 400 with the current crossover so let me ask this question. if you were building these speakers what would you do at this point ?

 

[Lojzek]

I suggest trying to record the spl at about 1m with original crossover, drivers in phase and out of phase to observe what is going on with reverse null at xo point, gating 100 Hz. If felt makes the sound smoother, then why not.