Which Components Should I Use?

[Ian Mackenzie]

Me, I was only Yankin (get it) your chain.

I take time out to play with this stuff between brain farts!

When I get a mini I was screw with Sound Easy and see what it says when you have a baffle like yours.

Ian

[johnaec]

To bad that Johneac sounds like he is losing interest, I know how you feel John, it sounds pretty complicated but don't give up on your dream system, you might never forgive yourself. It's not as complicated as it sounds, even if you don't do it perfect, I don't think people will notice or measure it.

It's not that I've lost interest - I actually have every driver, crossover, EQs, amps, and most of the cabinet material right here at hand for the complete system. Frankly, it's about the most interesting thing to me right now! Withough going into detail, the problem is that I'm dealing with some very serious medical issues right now, and sometimes it's hard to even function. Sometimes, I feel like the interaction I see on this forum is one of the things still helping me get by. So please don't let my input influence any enthusiam - I'm dealing with things the best that I can, but sometimes my health just gets in the way...hopefully, the Lord willing, that may change...

John

[grumpy]

Hi John, Hope your health takes a turn for the better. You've got this motley
crew behind you

[Baron030]

Hi Ian

I now understand what you are saying about baffles. They can have some pretty dramatic changes to the response. And I don’t mean to argue, but there one thing that I am certain about. There is a something strange going on between 800Hz and 1 KHz with this driver. And it even appears on the JBL 2012H driver’s spec sheet. Admittedly, every measurement that I have ever made does show this anomaly to be larger then JBL’s published response by a few DB. So, maybe you are right about my baffle is making that peak worse. Now, I have tried plugging the ports and the results are posted here. Other then an a few db increase in the roll off below 350 Hz, I don’t see much difference with plugging the ports.

Before starting my 15” 4-way project, I really debated a long time between “closed vs. ported” and even between “Dog Box vs. Sonotube”. Well, right or wrong I did go with a ported rectangular dog box, with an angled back panel, and a heavy fill. For a box this small, 1” to 2” of fiber glass is about the most that can be packed in without covering the ports and a small vent hole on the back of the driver. So, I guess that constitutes a heavy fill. And I don’t think adding more fiber glass fill to the doggy box is going to change much.

During, this debate I ran tests using smaller test boxes, with no baffle at all, other then the box itself. Now, this was a time when I did not have CLIO. So, the response curves were not as accurate as they are now. But, in all cases there still was a peak at about 1 kHz. And it was very objectionable during listen tests, which necessitated adding a passive EQ circuit to an otherwise all active system.

The only real point that I am trying to make here is that the 1 KHz. peak is not an error in my testing and I don’t think it is a baffle related anomaly.

Now, my first attempt at an EQ circuit was a 1.5mH coil in series with the driver. Surprisingly, it did work well enough to prevent me from selling the 2012 drivers for a big loss. But, it did not mate well with my 2382/2446h driver. There was a dip between 1 KHz and 1.2 KHz and I would think that other HF horns with a deeper low end then the 2382 would not have this problem. So, you are probably in the ball park with a 1.38Mh and a 49uF crossover. Maybe, not at home plate yet, but definitely in the ball park.

Johnaec, I am sad to here about the medical issues, I hope you get better soon.

Baron030

[Doc Mark]

That combination of drivers looks OK, but you may have problems getting passive crossovers lined up easily. What I'd do is use a passive only for the VHF 2405, around 7K or 8K, and use an active 3-way crossover for the rest, each range with its own amp. Of course, you'll need some sort of EQ with that, looking at the different response characteristics of the different drivers. But I think you could get it sounding fairly good, and capable of prodigious, (sp?), output if desired.

John

Hello, John,

I hope you're feeling better, and that your medical misadventure is on the way to being history, instead of current news. In any case, I was perusing this thread, and realized that when I thanked Baron030 for his suggestion of using a passive crossover for my 2405's, and active for everything else, you had already suggested that exact same thing, and I completely forgot to thank you, too, Sir!! Please accept my humble apology, and also my thanks for that outstanding idea!! It is a simple, but absolutely perfect solution to my quandry about making an actual 5-way system, including a subwoofer, and I very much appreciate your having suggested it!!!

Also, please don't give up on creating your dream system, John!! As one who gathered JBL components, then carried them around, and stored them in various homes in which we lived, for more than 20 years, I know that it sometimes takes quite a while for things to come together! But, I have every condidence that for you, just as it happened for me, time will eventually present itself, and your health will improve, and then you can finally realize your long-awaited dream system!! It will have been well worth the wait, my Friend!! Hang in there, and please keep us posted as to your progress in getting better. Take care, and God Bless!

Every Good Wish,
Doc

[Ian Mackenzie]

Thanks,

Well its definately not diy user friendly if this is typical of production samples.

Some peaks are okay and some arn't as in of they are well damped and narrow.

If you can hear this one and is does not show on the impediance curve its weird.

The JBL data sheets are probably smoothed and I note your response curves are un smoothed but what the heck its there.

[Baron030]

Hi Ian

Yes, I would have to agree with you about the 2012H as not being very user friendly.
But, the 1 KHz peak does respond well to notch filtering. And if people can accept using this driver over a relatively narrow range of say about 400 Hz to 1200 Hz then my LCR solution for active crossover networks does work fairly well. Pictured below is the exact same data that I posted earlier. But, now it has 1/3 octave smoothing and I “db shifted" the left and right drivers outputs a little bit. So, that they are now normalized at 800Hz instead of the 1 KHz like I had posted earlier. Other then smoothing and db shifting, no other processing was done. With my 24db/octave electronic crossover set at 400Hz and 1200 Hz, you can see how I have gotten this driver to perform well in my system.

Now, the real challenge is to find a passive solution for someone like Kevin who wants to use an electronic crossover to handle the lower crossover point and then have a passive crossover network handle the upper crossover point.

Well, maybe I am stubborn or just plan stupid. But, I still think that a 1 KHz series notch filter combined 2nd or 3rd order low pass filter could be a solution to Kevin's problems. And of the two different orders, I like your idea of using a 2nd order as an 18db/octive low pass filter. From my experiments with parallel LCR filters, I remember that I was always playing the 1 KHz peak against a LCR filter induced dip at 800 Hz and the higher the LCR "Q" the better the results that I got. And I would think that without a series notch filter, a 2nd order low pass network would have fairly flat response over a 400 Hz to 1000 Hz range, but not without a big dip at 800Hz. or it could have a really flat between 400 and 800Hz and still have a 1 KHz peak. Of the two flavors of 2nd order filter that I just described, the one with the 1 KHz peak would be the easier to fix with a series notch filter. But, I am not sure how the overall system impedance would be affected. It could really mess things up with the high pass section of the crossover or then again maybe not? :dont-know

There is one thing that I can say for certain. If someone can get past this hurtle, then I think there will be a lot more successful DIY projects out there using the 2012H driver.

Baron030

[4313B]

Aren't 2122H's and 2123H's alot cheaper?

[Robh3606]

Well here's a 2122 in a 4344 baffle enclosure.

[Ian Mackenzie]

Like the scale Rob,

I wonder what its is like where you sit!