2452H-SL in a 2-way

[Zilch]

The driver is readily available and its performance is very good for its price point. The waveguides are readily available as well. This simple fact makes both products appealing to the DIY crowd and DIY folks should be able to come up with really nice 2-way systems using these products.

Two way? That would be me. ;)


What kind of technical support? Do you want me to bolt one on to a consumer horn or waveguide, take some measurements and plug those measurements into LEAP?Yes.

2452H-SL on PT-F95HF (JBL part# 338650-001,) the combination used in VP7212MDP, i.e., "Used as intended."


Someone must have bought all these 2452H-SL's off eBay... where'd they all go? Where are all the 2-ways using this c.d.?

Two projects are built, and two more are under construction, LE14H-3 being the preferred woofer, with 2235H as alternate, both @ 1kHz. We're presently using my half-ass adaption of N3134 crossovers as I don't have LEAP yet.

We'd like a compensation filter that could be used with 24 dB/octive L/R active as well as a full passive crossover incorporating it, such that a switchable version might be constructed.

I know how to do the building them part.... :yes:

http://audioheritage.org/vbulletin/attachment.php?attachmentid=25910&stc=1&d=1182375046

[4313B]

I can't get my head around using a 15" woofer up to around 1KhzI use twelves myself these days but plenty of people still like the fourteen or fifteen.

[Robh3606]

Two projects are built, and two more are under construction, LE14H-3 being the preferred woofer, with 2235H as alternate, both @ 1kHz. We're presently using my half-ass adaption of N3134 crossovers as I don't have LEAP yet.

Post a couple of measurements on the waveguide. I can throw it into LEAP if your curious what it comes up with.

Rob:)

[Maron Horonzakz]

Is that a Haywood Wakefield cabinet between the speakers???:applaud:

[Chas]

I use twelves myself these days but plenty of people still like the fourteen or fifteen.


Dammit! Another variable to think about:banghead:

[Zilch]

Is that a Haywood Wakefield cabinet between the speakers???:applaud:Equipment stand.... ;)

[Zilch]

Post a couple of measurements on the waveguide. I can throw it into LEAP if your curious what it comes up with.Thanks, Rob. This good enough for LEAPing?

[Robh3606]

Hello Zilch

Sure they should work just fine. Do you have an impedance and phase measurement with the waveguide on?? If you do post that as well. What's the difference between the files??

Rob:)

[boputnam]

Thanks, Rob. This good enough for LEAPing?OK, now I'm confused.

Zilch - you've been listening to that confab, this way...? :blink:

Dood...

[Zilch]

Hello Zilch

Sure they should work just fine. Do you have an impedance and phase measurement with the waveguide on?? If you do post that as well. What's the difference between the files??Separate file for each of the drivers, providing SPL and phase.

I'll generate the impedance curves to go with in Sinusoidal.... :yes:


OK, now I'm confused.

Zilch - you've been listening to that confab, this way...? :blink:

Dood...No, Bo, that's the raw driver response on the waveguides, four units.

Rob's going to plug those values into LEAP along with the additional information he's asked me to generate, and LEAP will prescribe a passive filter, or several, for me to try.

I've been running them on a modified version of the 4430 crossover....

[boputnam]

No, Bo, that's the raw driver response on the waveguides, four units.

...

I've been running them on a modified version of the 4430 crossover....Ah. The sequence of your posts confused me. I thought you were showing the N3134 response. I'd like to see that.

I have a (possibly) vested interest in this - I am considering the VP7212MDP.

[Zilch]

I thought you were showing the N3134 response. I'd like to see that.Here's one of the ones in Barzilay cabs with LE14As, with the pots at "Normal" (mid) setting. Owner likes the sound of them in his listening space without additional adjustment, but he has that available, if desired.

Even adjusted, they don't get perfectly flat; I believe I can do better.... :yes:

[boputnam]

I believe I can do better.... I hope so. That is a "mountainous region" as we say in the biz...

I'm as troubled with the overall profile as I am the shoulders of some of those anomalies... :biting:

I actually don't know that I'd like to face that response, as an engineer walking in.

The really weird response 250Hz to 450Hz (++, really to 700Hz) would be best served (i.e., most elegantly) by a "MESA" algorithm, available only on the Dolby Lake processors.

Thanks for sharing the measurements. Always interesting to ponder!! :yes:

.

[Zilch]

You can't take anything below 1 kHz there seriously, Bo. You're looking at room stuff down there. That's unsmoothed in-room sinusoidal in ZilchLab @ ~1M, and the delay was not set for the gating.

Low frequency response would have to be done nearfield or groundplane.

It illustrates how the crossover is working with these drivers at the high end, mostly, and even that would be better measured MLS....

[boputnam]

You can't take anything below 1 kHz there seriously, Bo. You're looking at room stuff down there. ....I keep thinking I'm looking at measurements of your beautiful 2-ways... :o:

[Zilch]

What do we do, put them in spreadsheet and use the mean of four units?

Same with the frequency response? :dont-know

[4313B]

What do we do, put them in spreadsheet and use the mean of four units?

Same with the frequency response? :dont-knowNo, you select a matched pair and build your filter based on them, along with your woofers and your baffles.

It looks to me like you are going to have to take one or two of them apart and tweak them.

Where are the fr and imp runs of the woofer mounted with the tweeter on the baffle?

Where are the fr runs of the woofer and the tweeter together (you will need to use an L-Pad or fixed pad to get the tweeter to the same output level as the woofer for that run)?

[Earl K]

The factory curves on PWT are very much different from other 4" drivers.

Hi Zilch,

Would you please post a pic or a link to the referenced PWT curve ?

Thanks :)

[Robh3606]

Hello Zich

Do you have the ungated MLS files for the compression drivers on the horn?? Or a sine run?? Need to see what the raw unsmoothed curves look like. Also need a raw to see what lowend really looks like. Do you know what he sensitivity is 1 watt 1 meter on the wavequide?? Need to adjust the curve SPL level for that number.

Rob:)

[boputnam]

...Where are the fr runs of the woofer and the tweeter together (you will need to use an L-Pad or fixed pad to get the tweeter to the same output level as the woofer for that run)?:yes:


...Need to see what the raw unsmoothed curves look like. Also need a raw to see what lowend really looks like.:yes:

[edgewound]

I would love to see those curves deleted from this site.

They give the initial impression that JBL drivers suck, which would give plenty of potential users the means to run other way.

As Bo said....he wouldn't want to have to scale that mountain range as a sound engineer.

[boputnam]

I would love to see those curves deleted from this site.

They give the initial impression that JBL drivers suck, which would give plenty of potential users the means to run other way.Or at the least, make the graphs better attributed.

The following quote is what led me, for one, to the wrong conclusion...


Here's one of the ones in Barzilay cabs with LE14As, with the pots at "Normal" (mid) setting. Owner likes the sound of them in his listening space without additional adjustment, but he has that available, if desired.This post introduces the "mountainous response" trace and directly implies the measurement is of a full-range cabinet with LE14A's as the woof.

Easy to misconstrue, for sure.

[4313B]

I would love to see those curves deleted from this site.

They give the initial impression that JBL drivers suck, which would give plenty of potential users the means to run other way.

As Bo said....he wouldn't want to have to scale that mountain range as a sound engineer.If we did that then we would also have to delete the Q&D crap too and that isn't ever going to happen. We lost the fight to keep this a top notch site roughly four years ago and it isn't coming back. It is just another hobby site where members are allowed to post whatever they fancy (doesn't even have to be Lansing related) and other members are allowed to challenge those posts in a meaningful manner.

The curves as posted are fine but probably need explanation. People have come to the wrong conclusions because of other posted graphs and to be blunt - that's their misfortune. The professionals know how to employ the components and don't have time to read this site anyway. The people with money to buy finished product don't read this site either because they don't have time and they don't care. All that's left are ten or twenty guys who visit here daily. It's a non-issue.

This whole thread needs a wash and the DIY junk needs to be moved to it's own thread. Personally, I think Rob and Zilch need to work it out themselves and then post whatever when they think that they are done.

[Zilch]

Would you please post a pic or a link to the referenced PWT curve ?We'll have to wait for Techbot to post the engineering specs. I don't have permission to disclose it.... :(

[boputnam]

...the DIY junk needs to be moved to it's own thread. Done.

Now, I'll go back and deal with the junk and poor attribution.

Thanks for the good idea, "4313B"...;)

[Earl K]

We'll have to wait for Techbot to post the engineering specs. I don't have permission to disclose it....

Okay, Cool . ( I thought maybe it was buried in some SRX / VRX product profile that I hadn't read .)

:)

[Zilch]

It looks to me like you are going to have to take one or two of them apart and tweak them.Yes, same was true of the 2435HPLs purchased on eBay, which I documented earlier here. Approximately 1/3 of them needed reconditioning to perform as well as the others.

The same also needs doing with "Legacy" drivers -- measure and refurb as required. It's surprising how uniformly old LE85s of varying and indeterminate provenence perform, for example, but it's still a crap shoot.... :yes:


Okay, Cool . ( I thought maybe it was buried in some SRX / VRX product profile that I hadn't read .)The "In-product" performance curves are shown there, and the requisite compensation to make them perform that way may be in "Tunings," but that's not real helpful in coming up with a slick passive filter to do it, other than serving as evidence that it can be done. :yes:

To the best of my knowledge, there's only one product using this particular driver/waveguide combination:

http://www.jblpro.com/installedsound/vpseries/specsheets/JBL_VP7212MDP.v5.pdf

[Zilch]

Do you have the ungated MLS files for the compression drivers on the horn?? Or a sine run?? Need to see what the raw unsmoothed curves look like. Also need a raw to see what lowend really looks like. Do you know what he sensitivity is 1 watt 1 meter on the wavequide?? Need to adjust the curve SPL level for that number.As suggested, I've chosen #7 & #8 as representative samples, and remeasured MLS. I'm not certain but that the data files aren't unsmoothed anyway; in either case, I'm attaching max res (2048 points,) unsmoothed here.

There is no woofer or cabinet at this point; looking rather simply for a HF compensation filter and topology similar to that earlier provided for 435Be on H4338, and thus attenuation is not an issue....

[4313B]

Rob will need one frequency response curve and one impedance curve for each driver.

#7 and #8 impedance curves are above in post 16?

Once he gets them he can plug them into LEAP and generate a compensation circuit.

[Zilch]

#7 and #8 impedance curves are above in post 16?Yes, @ #16 above. Looks like #8, the red curves, is the one to use. :yes:

How would you handle the notch and spike in the VHF, roll everything above 16 kHz off, maybe?

I guess we'll see if LEAP figures out something clever to do up there.... :dont-know

[Robh3606]

Here's a first crack at it. I was messing with it last night. Loaded the impedance file this morning. That's driver file #5. Basically I am driving this. I have tried using the Optimizer on compensations and gotten some really odd outcomes. I use Greg's basic filter and go from there. OK if you want crossover at 1.5K not going to work at 1K. I was hoping there would be a bit more loading before the crash. I may be able to get it to work @ 1K but the attenuation would be a problem.

The red curve is 24db L/R at 1.5K

Can you post the raw ungated MLS??

Rob:)

[Robh3606]

Here's a second shot that is OK out to 1K but there is no real network to roll off the driver just compensation. I could really use either a sine sweep of the ungated MLS. Will try the other curves later.

Rob:)

[4313B]

I may be able to get it to work @ 1K but the attenuation would be a problem.Probably :dont-know

[Robh3606]

Hello JBL4313B

I don't know what the 1 meter SPL is. I am using the Quide Curve for the driver. So the attenuation isn't that bad yet figure about 15db @ the peak so maybe??? What surprises me is the roll off. I have a 12db passive on top of the horns roll off to get the 24db slope at 1.5K on the first plot.

With the second one I don't think I can get it to work real well @ 1K with a proper crossover in place??

Any ideas??

Rob:)

[4313B]

I don't know what the 1 meter SPL is.It's really loud.

So the attenuation isn't that bad yet figure about 15db @ the peak so maybe???Yeah, I think there's enough there.

Any ideas??Just what I posted above. I'd have to actually build it and try it. It probably needs an inductor to ground to pull that leg around 500 Hz down. Since the driver/horn is dropping at 12 dB/octave you should only need a 12 dB/octave electrical filter. I used a 6 dB/octave above. As you can see it is 3 dB down at 1 kHz. I have no idea how it would actually sound since it naturally wants to start attenuating around 1.5 kHz. It might sound strained. But who cares, JBL's done the same thing with some of the smaller Consumer horns to get them down into the 700 to 900 Hz range and those are $18,000 loudspeakers that no one is complaining about. I'm anxious to hear what Bo thinks about the VP7212MDP. :yes:

*****

Well bloody hell! I didn't save off the schematic properly so I could post it for you. :banghead: I really hate the way LEAP does the whole file save thing.

[Robh3606]

Well we must have been thinking the same thing here it is with a big cap and an adjustment to the "trim" inductor. It rolls off below 500Hz. Yeah I agree the save sucks and you can't really undo. So I have a single pole instead of 2 pole.

Rob:)

[boputnam]

...As you can see it is 3 dB down at 1 kHz. I have no idea how it would actually sound since it naturally wants to start attenuating around 1.5 kHz. It might sound strained. Do we know the crossover point JBLPro uses in the VP7212MDP? I've looked and can't see that you guys have discussed that. Granted, that is a 2-way with an incredible 12-in woofer, but some of the smartest guys I know are doing 12-in 2-ways... ;) Maybe JBL take the 2452H-SL up to 1.5kHz...?


...I'm anxious to hear what Bo thinks about the VP7212MDP. :yes:Yo tambien. I have uniformly positive reports from a universally tough crowd. I will report soon as I can.

[Earl K]

Some nice looking curves there guys ! Bravo !


I have no idea how it would actually sound since it naturally wants to start attenuating around 1.5 kHz. It might sound strained.

,,,here it is with a big cap and an adjustment to the "trim" inductor. It rolls off below 500Hz.

- I do something similar in that I use a "big cap / small inductor to give a higher "Q" curve " to buttress up the drive voltage of my impossibly small horn setup . It has worked well for HiFi purposes .
- It's not something I'd practice in my SR work .

- This technique seems to be easier to successfully implement ( sonically ) when the driver has a lowish Fs ( which is why I was interested in the PWT pic for the 2452 ).
- The Altec 288-8K ( which I use ) has a pretty low Fs ( lower than 500 hz ) .
- I'd think that forcing EQ ( LF, near cutoff ) onto the 2431 in a likewise fashion ( when it's fixed onto a overly tiny horn with no loading ) would dictate the use of some other form of damping filter ( such as a series LCR ) to stop audible weirdness at 900 hz.

:)

[Robh3606]

Hello Bo

As far as I know it's an active system. I will see if there is anything on the spec sheet. With DSP they can do whatever they need.

Here's a schematic for the graphs posted. They are not converted to standard values and the attenuation is unknown at this point so they are for reference only. Gregs filter is a real champ!!

Rob:)

[boputnam]

...As far as I know it's an active system. I will see if there is anything on the spec sheet. With DSP they can do whatever they need....Yes indeed, it is active, and there is an onboard DSP.

But the spec sheet (http://www.jblpro.com/installedsound/vpseries/specsheets/JBL_VP7212MDP.v5.pdf) gives few details on the engineering... I cannot find anything on the JBLPro site about the 2452H-SL and recommended crossover points.

However, there are two ways to skin this cat. :idea:

The VP7212MDP uses the 2262FF woofer, a not-too-distant brethren of the 2262H (http://www.jblpro.com/pub/technote/JBL_TN%201-33%20rev3.pdf). The 2262H is very linear to ~1.2kHz (cf. Figure 5).

The 2262H is used in the SRX712M (http://www.jblpro.com/srx700/PDF/JBL.SRX712M.pdf), a passive/biamp wedge I know very well. The recommended crossover point for bi-amping is 1.2kHz. Grist for the mill... :hmm:

[Zilch]

Note: All FRs done with 47 uF protection capacitor.

[Zilch]

But the spec sheet (http://www.jblpro.com/installedsound/vpseries/specsheets/JBL_VP7212MDP.v5.pdf) gives few details on the engineering... I cannot find anything on the JBLPro site about the 2452H-SL and recommended crossover points.Look for my FR curves in theirs. I see it rolling off at 1.5 kHz.

Also see the DI curve. The slope changes to flat at 1.5 kHz, where the waveguide takes control, looks like.... :yes:

[Zilch]

O.K., got close, unit #8, 9 dB of comp:

2 uF || 8 Ohms, then two series notches:

1.5 mH, 5.6 uF, 5 Ohms, and
0.13 mH, 0.47 uF, 4 Ohms.

Can LEAP refine and optimize this topology?

[4313B]

1.5 mH, 5.6 uF, 5 Ohms -> 1.5 mH, 6.6 uF, 6 ohms (1.6 kHz notch)

0.13 mH, 0.47 uF, 4 Ohms -> remove

You might want to try something like a 2.0 mH, 2.2 uF, 20 ohm or 3.0 mH, 1.5 uF, 20 ohm as well. (2.4 kHz notch)

Your two big problems are at ~ 1.6 k and ~ 2.4 k and the dip at ~ 3 k is a real bummer.

Maybe Rob has a bit more time and other ideas to try.

[boputnam]

Look for my FR curves in theirs. I see it rolling off at 1.5 kHz.

Also see the DI curve. The slope changes to flat at 1.5 kHz, where the waveguide takes control, looks like.... :yes:Dunno if you're "seeing" it right.

In the SRX712M, JBL takes that 12-in 2262H up to only 1.2kHz. That same woofer is mated to the 2452H-SL in the newer VP7212MDP. I'd guess that cabinet crosses at the same point.

As well, regarding a 15-in 2-way, the VP-series uses the 2265H mated to the 2452H. While JBL again don't disclose the crossover for the VP-series, the VRX-series uses that same woofer, the 2265H and again, the crossover point is 1.2kHz.

I think you might find that you need to run the 2452H-SL down to 1.2kHz to get a good crossover to the woofer - otherwise you are running the woofer up too high. JMO...


Your two big problems are at ~ 1.6 k and ~ 2.4 k and the dip at ~ 3 k is a real bummer...Yup. I particularly don't like that dip at 3k "coming out of" the hump at ~2.4k. Once you treat the ~2.4k anomaly, the one at 3k might worsen...

[Zilch]

I think you might find that you need to run the 2452H-SL down to 1.2kHz to get a good crossover to the woofer - otherwise you are running the woofer up too high. JMO....We're not doing 1.5 kHz, rather trying to get to 1 kHz, Bo, for 2235H as one option. A major difficulty in accomplishing that passively is the driver/waveguide rolloff below 1.5 kHz.... :yes:


Two projects are built, and two more are under construction, LE14H-3 being the preferred woofer, with 2235H as alternate, both @ 1kHz. We're presently using my half-ass adaption of N3134 crossovers as I don't have LEAP yet.

We'd like a compensation filter that could be used with 24 dB/octive L/R active as well as a full passive crossover incorporating it, such that a switchable version might be constructed.What JBL does in that monitor is of some interest, but they are using a different woofer toward a specific, and also different, purpose.

[4313B]

Well I got it to 1kHz as shown in post 33 using a series LCR in parallel with a series resistor and added in two notch filters but the file save didn't do what I intended so I lost the schematic. I was too tired to bother with it further last night.

[Zilch]

Well I got it to 1kHz as shown in post 33 using a series LCR in parallel with a series resistor and added in two notch filters but the file save didn't do what I intended so I lost the schematic. I was too tired to bother with it further last night.Wouldn't a high Q two-pole highpass as Earl suggests get it at 1 kHz from what I've done? Not sure I consider 60 uF a very practical approach, tho.

It sounds very good on 24 dB L/R active presently, but I want to tweak it using your suggestions above and see if it can be improved, then retest with several drivers here.... :yes:

[4313B]

I didn't use a 60 uf capacitor anywhere.

If I have time I will look at it further.

[Robh3606]

didn't use a 60 uf capacitor anywhere.

No but I did. Used it as a single pole, in front of the compensation circuit.

That red quide line is for a 1K 24db L/R crossover.

You can get it out to 1K but the slope is screwed up.

Zilch you have a 50uf in front of the compression driver for all the measurements??

You need the real SPL at 1 meter on the wavequide to do the crossover correctly. Anyone know what it is????

Rob:)

[Zilch]

Zilch you have a 50uf in front of the compression driver for all the measurements??Yes, a 47 uF protection cap.


You need the real SPL at 1 meter on the wavequide to do the crossover correctly.Do you know that of 2435HPL on PT-H1010HF or some other driver/horn combo I have?

If so, I can easily measure their relative SPLs and we can come up with something close.... :yes:


If I have time I will look at it further.Thank you. As always, I very much appreciate your assistance with this.

There are apparently others interested in using this driver, as well, and they'll be confronting similar issues....

[rs237]

Hello,

Thank rob and zilch for the posts and measurements with the 2452HSL. One question. I am planning a two way speakers with 2234 and 2252HSL, crossover by 800-900Hz. I want a (DIY) oblate waveguide horn. I see a start attenuating around already at 1,5kHZ. Is it the horn or the driver?
It is perhaps better to use the 2435HPL?

regards
juergen

[Robh3606]

Those measurements are specific to the waveguide plus a series capacitor. Really need them with no series cap to see what the loading is.

Rob:)

[4313B]

We could use the .mls and .sini files from CLIO

*****

Here's what I came up with using the data from #8

[Zilch]

We could use the .mls and .sini files from CLIOThat's a "Do over," and I can certainly do that, but I need to know what, specifically. There's no smoothing or windowing reflected in those, right, it's just the raw acquisition data, so my settings don't matter?

Rob's implying he wants without the protection cap, also. Yes?

[4313B]

That's a "Do over," and I can certainly do that, but I need to know what, specifically. There's no smoothing or windowing reflected in those, right, it's just the raw acquisition data, so my settings don't matter?Nevermind, I played with it as much as I wanted to using the available data and posted the results above (http://audioheritage.org/vbulletin/showpost.php?p=204761&postcount=54).
I have no idea how it will sound or if you will have enough gain.

*****

If that 20 kHz bump really bothers you then throw in another notch: 0.39 mH, 0.16 uF, 5.6 ohm

[Robh3606]

Rob's implying he wants without the protection cap, also. Yes?

Wondering what it's doing. It may not have a significant effect. If you are concerened about protecting the drivers don't do it or just double them up. Just try one if there is no change then it doesn't matter.

That looks good why not just try it and drop in the attenuation later.

Zilch you don't have any 2425's or 26's laying around??

Rob:)

[edgewound]

Do we know the crossover point JBLPro uses in the VP7212MDP? I've looked and can't see that you guys have discussed that. Granted, that is a 2-way with an incredible 12-in woofer, but some of the smartest guys I know are doing 12-in 2-ways... ;) Maybe JBL take the 2452H-SL up to 1.5kHz...?



If it helps at all, the 2452H is used passively in the later versions of SRX722 and SRX725

Crossover is listed as 1.2KHz on the spec sheet at 75 watts continuous pink noise.

http://www.jblpro.com/srx700/PDF/JBL.SRX725.pdf

[Zilch]

Here's what I came up with using the data from #8Thank you again. I believe I have sufficient parts to prototype this. :thmbsup:


Zilch you don't have any 2425's or 26's laying around??Yes. Do we know the SPL on 2370A or 2344A? I guess so, and will look it up.


Wondering what it's doing.Mr. Widget taught me their value to not blowing up drivers. ;)

My Sonic Impact HF testing amp has a bit of DC, though I've never measured how much.


That looks good why not just try it and drop in the attenuation later.Yes, the crossover version will certainly require attenuation. I'll probably use a fixed and an adjustable L-pad to do that; not sure where to put them yet, is all. Adjustable at the driver output, probably. Still running just the comp version active here....

[Robh3606]

OK 2426 is about 110db on a 2370

http://www.lansingheritage.org/images/jbl/specs/pro-comp/2426/page2.jpg

Rob:)

[Zilch]

I see a start attenuating around already at 1,5kHZ. Is it the horn or the driver?It's the combination, and you'll have to measure yours. You need at least minimal measurement capability like RTA to do this stuff. SPL meter and signal generator, maybe.



It is perhaps better to use the 2435HPL?Not with respect to lower cutoff, I wouldn't think. 2452H-SL is a 4" large format driver, and 2435HPL 3" medium format. I've probably posted measurements of both using 2352 horns here somewhere.... :dont-know

[4313B]

OK 2426 is about 110db on a 2370

http://www.lansingheritage.org/images/jbl/specs/pro-comp/2426/page2.jpg

Rob:)Put a milliwatt into it and measure it at 1m and then add 30 dB to get the SPL 1W, 1m

It is perhaps better to use the 2435HPL?No. It exhibits an even tougher curve to work with.

[rs237]

Thanks to all of you for the answers.

regards
juergen

[Zilch]

Didn't have parts for the spike killer, but:

[rs237]

Not with respect to lower cutoff, I wouldn't think. 2452H-SL is a 4" large format driver, and 2435HPL 3" medium format. I've probably posted measurements of both using 2352 horns here somewhere.... :dont-know



thanks zilch ,
I have found it.

regards
juergen

[4313B]

Didn't have parts for the spike killer, but:Both the 2435 and 2452 exhibit that little dip at ~ 3 kHz. You can affect it somewhat with a broader notch but I'm not real sure it's necessary. You can do alot if you throw enough parts at something.

You might want to run plots of all four of your drivers through that filter to see what happens. If you save files you can easily compare the plots with each other, for instance 64 with 43. You can compare your "hard way" with just one solution from LEAP - you can probably sit around forever and play with the various notch filters, adjusting breadth, depth and location. Obviously what you really have to do is listen to it all and become familiar with how various changes sound and stick with the sound you like best.

It is really tough to get the 2435 to 1 kHz on that waveguide but I did it with a broad notch and nearly 20 dB of loss (the 2452 only loses something like 15 dB). I'll actually listen to it someday when I have time and see what I think of it.

What I'd like to see is a plot of the TAD bolted onto that waveguide... I would want to see if that 3 kHz dip went away.

[Zilch]

Good to 875 Hz, perhaps a bit lower, with LE14A on 24 dB/Octave L/R active XO, CX3400:

[4313B]

Cool!

I can't imagine wasting that nice, expensive top end with an old LE14A though...

[Zilch]

Cool!

I can't imagine wasting that nice, expensive top end with an old LE14A though...I'll wheel out some LE14H-3s next, but first, I pursued how low I could play this combination. It seems to be behaving purely as a waveguide, i.e., no horn coloration down to the natural rolloff, which shocked me the frist time I dialed the frequency around. I'm accustomed to horns sounding nasty at some point as I dial it down, which is not happening here.

Tell me how to measure distortion, please. I set the Auto Delay @ 1 kHz. The SPL is my normal listening volume here (bottom). Is the the -35 dB 2nd harmonic from the 2452H-SL and PT waveguide terrible between 725 Hz and 1.5 kHz there? I sure don't hear it.... :dont-know

[Robh3606]

The lower you go the more diaphram excursion, more excursion higher numbers. If you look at Geedes Distortion Coefficient low order distortion in compression drivers is simply not a concern up to max power. That his findings. Check out his site. www.gedlee.com (http://www.gedlee.com) I still would not run them too low there is no reason too. I would think 800hz would be fine. Measure each driver seperately to see what the compression driver is doing alone.

Rob:)