2105H or 104H-2?

[johnaec]

What are the advantages/disadvantages of using a 2105H or 104H-2 with a 2405H for HF, (ala 4315B or L65)? Looking at other threads, it appears the 104H-2 might be smoother but the 2105H might be more efficient. And while the frames look the same, the magnet structures are obviously different. These would probably be crossed over around 1K, with a 2012H below that, (and Sub1500 below that). Eventually they might be replaced with a 2430H/2332 horn combination.

I've got both the 2105H and 104H-2, but someone wants the 2105H's, and I'm not sure about using the 104H-2's.

Thanks for any comments.

John

[4313B]

2105H, LE5H, LE5-9, LE5-11

[4313B]

104H-2, 104H-3

[johnaec]

Wow, Giskard - thanks for those plots! It does look like the 104H-2 is flatter, though not by much. The 4K - 7K looks better on the 104H-2, but I still haven't decided on where to cross over to the 2405H. Ribbon vs. round wire - hmmm... Also, is that depth listed the depth of the windings on the voice coil? If so, I'm surprised there's such a difference - it looks like the 104H-2 might be capable of more linear high output, correct? I noticed the magnet structure on the 104H-2 looks to be about the same amount thicker than the one on the 2105H.

I think I'm just going to have to hang on to both until I can compare them in the system. Unless anyone else has any objective/subjective opinion, it looks pretty close to call...

Thanks again!

John

[DavidF]

Hey John. Yes these graphs help a lot if you need some guidance on how to put the xover together. I would not take the rising response of the 2105 as a negative deciding factor. In fact it may be easier to work a band pass response filter with this rising response.

If you intend to put together a high output system, which sounds likely if you need to use the 2405 and 2012, you should look at the 2105. The threshold for the low end response of the 2405 is high, making it at better match with mid horns that fall off in the 8-12kHz range. A decision then is whether to push on the lower end of the 2405 or the upper response of the 2105. I think the latter idea is wins over so again this is a reason to look at the 2105. The response starts to peak and dip over 7kHz (damped paper cone and lighter voice coil may be why we see a smoother high end on the 104h) but there is more response above this to work with than with the 104h.

Unless you have some ideas on crossover, using a derivation of the 4315 design to match the 2105 and the 2405 on the high end will provide a starting point.

David F

[johnaec]

Yeah, David - I was thinking of looking at the 4315 crossovers, and also comparing those to parts of some L100T3 crossovers some guy in San Jose sent me, ( ;) ), though I'd obviously have to put in some HF padding with the L100T3 xovers. I also thought I'd take a look at what JBL does in the L65 crossovers.

To start with, though, I'll either make some little test boxes for the 2105H/104H-2 and put them through their paces, or maybe swap them into L65A's to compare them, or even test them in L150's. But I agree - running the 2105H/104H-2 higher definitely makes more sense than the 2405H lower. That's one area where I thought the "smoothness" of the 104H-2 might outweigh the 2105H; I'm not sure how easy it would be compensating for the peak in the 2105H.

I've also got a couple active crossovers to work with - the JBL DSC260A and the DBX Driverack 260. With those I have control over crossover points, slopes, EQ, and even delay. I do wish they were stereo 4 way instead of 3 way - then it'd be fully no-compromise, (other than analog/digital...). Maybe go passive LF and 3 way active above that... In the end, though, I'd prefer 4 way passive, with an option to biamp the LF.

John

[DavidF]

Originally posted by johnaec

I've also got a couple active crossovers to work with - the JBL DSC260A and the DBX Driverack 260. With those I have control over crossover points, slopes, EQ, and even delay. I do wish they were stereo 4 way instead of 3 way - then it'd be fully no-compromise, (other than analog/digital...). Maybe go passive LF and 3 way active above that... In the end, though, I'd prefer 4 way passive, with an option to biamp the LF.

John

You are probably better off going passive on mid-to-upper-mid/upper-mid-to-high. You can not rely on the active filters to compensate for the acoustic response characteristics of the drivers. To do so you have to customize the active filters, so may as well go passive. The fairly even response and resistance of the lower mid and woofer around an appropriate xover point (100-300Hz) gives you a better shot that the standard curve designed in the active circuit will work out at this xover point.

David F

[4313B]

4315 FR #1

[4313B]

4315 FR #2

[4313B]

Originally posted by johnaec
I was thinking of looking at the 4315 crossovers3114 (http://www.jblproservice.com/pdf/Network%20Schematics/3114%20Network.pdf)

3114A (http://www.jblproservice.com/pdf/Network%20Schematics/3114A%20Network.pdf)

[4313B]

Originally posted by johnaec
It does look like the 104H-2 is flatter, though not by much.I think the 3 to 4 dB decrease in the rising response is a decent amount of difference. One can read the 2105 (http://lansingheritage.org/html/jbl/specs/pro-comp/2105.htm) and 2105H (http://www.jblpro.com/pages/pub/obsolete/2105h.pdf) product sheets for explanations on what JBL was going for.

[johnaec]

Originally posted by Giskard
I think the 3 to 4 dB decrease in the rising response is a decent amount of difference. One can read the...product sheets for explanations on what JBL was going for. You're right - I didn't mean to minimize the difference.

It's long been apparent to me JBL designed the rising response in the 2405/H to add a little clarity to the voice range, kinda' like turning the treble up a little. To me, this seems like standalone voive reproduction may have been its initial intended purpose, and that maybe later it was pressed into service as a mid/highrange transducer in full-range systems. It's also obvious they've compensated for this fairly successfully in the crossovers they use, as is apparent in the 4315 graphs you posted.

My question is whether the 104H-2 might me a better "core" mid/highrange transducer, that has been modified from the 2105H parameters to be better suited for mid/high when incoprorated in a full range system?

I'll probably stick with the 2105H, though, since it's such a "known" when paired with the 2405H.

And I really want to thank you again for those response graphs, especially of the drivers! They're a LOT more detailed than JBL's published ones, (and first I've seen of a 104H-2).

John