BillFitzmaurice.info

Seth wrote: ↑Thu Jul 14, 2022 11:50 am set the crossover point at 1.2k on a LR18 and see if that improves things.

I wouldn't do 18 for the slope. That'll be taking a chance on blowing the driver. I'd keep the slope at 24. It's a 1.5khz driver, but can operate lower with a steep slope at 1.2khz.

I'd tend to agree with you that it might be a phase issue.....try swapping the polarity on the driverack for the tops and see what happens.

Bruce Weldy wrote: ↑Thu Jul 14, 2022 12:44 pm

Seth wrote: ↑Thu Jul 14, 2022 11:50 am set the crossover point at 1.2k on a LR18 and see if that improves things.

I wouldn't do 18 for the slope. That'll be taking a chance on blowing the driver. I'd keep the slope at 24. It's a 1.5khz driver, but can operate lower with a steep slope at 1.2khz.

I'd tend to agree with you that it might be a phase issue.....try swapping the polarity on the driverack for the tops and see what happens.

The passive filter in the plans is a 3rd order (18dB) filter.

I'm thinking, if the OP initially had it set up on a 2k LR24, that would pretty much explain why he was in need of so much EQ in the 1k region... Signal would have been down -6dB at 2k and another -24dB by 1k, added to whatever natural roll-off the driver has in that region. Darn right there's gonna be a hole there. Trying to make up for it on the other end with the woofer doesn't sound like it was working very well.

First order of business is get the crossover freq set correctly. I'd feel okay at 18dB, wouldn't hurt to get Bill's blessing on that though. Who know's why he's spec'd 18dB. Could be that it sounds better. Could be that it's sufficient and less costly to make than a 4th order passive, yet 24 would be better if active??? Dunno. Bill?

Then verify phase, match levels, and since it's bi-amped and he has the means to test and adjust for it, phase align. Then I'd like to see where everything's at once that all gets set, prior to any EQ.

We use that same 18dB filter throughout the line with the NSD. The 1.5kHz recommendation from Eminence is with a 12dB filter. You can use 48dB if you wish, 18dB is sufficient.

Also, You should measure the woofer and compression drivers separate without any crossovers and you will see why they cannot sum at any frequency from 1-2k regardless of polarity and phase.

mj90210 wrote: ↑Thu Jul 14, 2022 1:51 pm Also, You should measure the woofer and compression drivers separate without any crossovers and you will see why they cannot sum at any frequency from 1-2k regardless of polarity and phase.

I wanna see. Post 'em up.

Bill Fitzmaurice wrote: ↑Thu Jul 14, 2022 1:39 pm We use that same 18dB filter throughout the line with the NSD. The 1.5kHz recommendation from Eminence is with a 12dB filter. You can use 48dB if you wish, 18dB is sufficient.

Then I stand corrected on the 18 vs. 24db......but, I'd still run steeper if I was bi-amping. I'm just really careful...

Seth wrote: ↑Thu Jul 14, 2022 11:11 am simply having the mid's and highs out of phase. I'm sure it's something you've looked into, but it's worth revisiting and starting with basics.

it looks as though you had your system setup incorrectly from the get-go

That's pretty irrelevant, I said during measurements I was muting the other drivers - and that I was pushing my crossover well below / above the usual point.

I am pretty sure I was pushing the crossover point well below the 1k - so it should've been responsive below 2k and instead of just fell sharply.

Next time I have the rig up I will try to get better measurements, or maybe I will try to do this in the parking lot of our storage area...

This is the measured response of my J10 Lite, loaded with NSD 2005, crossed at 1.2kHz, passive. There's nothing unusual about it, response being +/-3.5 dB within the HF passband.

snowphish wrote: ↑Fri Jul 15, 2022 8:25 am That's pretty irrelevant, I said during measurements I was muting the other drivers - and that I was pushing my crossover well below / above the usual point.

I am pretty sure I was pushing the crossover point well below the 1k - so it should've been responsive below 2k and instead of just fell sharply.

Next time I have the rig up I will try to get better measurements, or maybe I will try to do this in the parking lot of our storage area...

It's all relevant in getting the information we need to help you solve your problem remotely. I know it's frustrating. I've been there, for sure. I'm here to help.

If it were me, this is the route I'd take in getting it sorted out. If you follow these steps, it will result in the information I need to help you diagnose the problem.

- Pull one cab out of storage and test it in the parking lot
- Factory reset the Driverack, save any settings you need prior
- Set the highpass on the highs LR18 @ 1.2kHz
- Set the Lowpass on the mids LR18 @ 1.2kHz

- Mute the Tweeters
- Using a dB meter and playing a 1.2kHz sine wave, trim the gain on the woofer so that you're reading as close as possible to 94.00dB wherever you're measuring from, on axis
- Make a note (written or digital) of the exact dB reading. If your meter indicates out to the hundredth, note the exact, to the hundredth reading as best you can.
- Mute the woofer, unmute the tweeter
- Trim the gain on the Tweeters so that you're reading 94.00dB at the same location you took the woofer reading from
- Make a note (written or digital) of the exact dB reading
- Unmute the woofer, so they're both playing 1.2kHz at the same time
- If the volume has gone down, invert the phase on the tweeters. What does the dB Meter read now?
- Make a note (written or digital) of the exact dB reading and whether or not the polarity needed to be inverted

- Do not adjust the volume or gains from where they are in the tests above. Use the same volume and gain settings for the rest of these tests

- With the test mic in the same location you previously took your dB reading from, use SMAART to generate a response graph. Name it, save it, screenshot it, I wanna see it.

- Phase align the woofer and tweeter
- Make a note (written or digital) of the settings made
- Use SMAART to generate a response graph. Name it, save it, screenshot it, I wanna see it.

- Mute the woofer
- Use SMAART to generate a response graph. Name it, save it, screenshot it, I wanna see it.

- Disable/remove the crossover on the Tweeter
- Use SMAART to generate a response graph. Name it, save it, screenshot it, I wanna see it.

- Mute the Tweeters, unmute the Woofer
- Use SMAART to generate a response graph. Name it, save it, screenshot it, I wanna see it.

- Disable/remove the crossover on the Woofer
- Use SMAART to generate a response graph. Name it, save it, screenshot it, I wanna see it.

- Unmute both and enable both crossovers again
- Run the Driverack's autoEQ
- Use SMAART to generate a response graph. Name it, save it, screenshot it, I wanna see it.

- If there's a way to screenshot the Driverack AutoEQ corrections, I'd like to see that too


You might wanna put a piece of masking tape or use some other method of identifying the cab you test so you're able to identify and use it as reference in testing all the other cabs for uniform polarity among the herd in the future

Seth wrote: ↑Fri Jul 15, 2022 1:08 pm
You might wanna put a piece of masking tape or use some other method of identifying the cab you test so you're able to identify and use it as reference in testing all the other cabs for uniform polarity among the herd in the future

Thanks for taking the time for that! I will try to spend some time before the next festival to do this, might be a few weeks but hopefully I'll report back

Hmm, and Hi.

In your opening post, you talked about:

In order to get a more linear response I have to EQ almost +12db around the gap

I think a good part of what you are experiencing has come from driver choice.
The impero 12a is roughly 7dB down in output compared to the 3012HO at 1.2K (and the same region in general), and the impero doesn't have the highly desirable "rising response to 2-2.5K" the 3012HO has either. Both these things combine to explain your results in part.
So, 7 of your +12dB is simply making up the deficit, in reality, you're only adding +5dB.
The above also accounts for no loss calculation at all due to the crossover slope, type, and/or frequency.

I'm also unsure if the latest plans (which I understand had last major revision in 2012) have the same information I have in an older set of plans, but some of these things may help improve your crossover point response, before EQ:

1. the red terminal on the NSD's is minus or ground, not plus or positive, and that's opposite to the woofer terminals. Get this wrong, and you're going to cause nuls between the woofer and the HF driver, worsening the problem.
2. If you are biamping only, you still have to have the 20uF cap in series with the NSD's for power on protection, and to prevent amplifier high frequency oscillation.

Passive crossover in the older plans is 1.2K
Woofer(mid) LP 2nd order (not 3rd order as mentioned before, but could be now in more modern plans), which is 12dB/octave.
HF HP is 3rd order, which is 18dB/octave.
So if you were trying to replicate that for biamp using active crossovers, you could safely try LR12 for mid LP, and LR18 for HF HP
The 12dB LP for the woofer will not cruel response so much in the crossover region, but you would definitely need to check phase response at the crossover point.

Hope this helps...