Designing filters for full range drivers

0 Members and 1 Guest are viewing this topic. Read 1803 times.

Danny Richie

  • Facilitator
  • Posts: 12117
    • http://www.gr-research.com
Designing filters for full range drivers
« on: 28 Jun 2018, 06:29 pm »
I made this offer for free crossover design work for the month of June:  https://www.audiocircle.com/index.php?topic=157533.0

I had quite a few people take me up on it and I really enjoyed it. I might just extend the offer for the rest of the summer. Here are a few interesting ones that I can share.

The Jordan Eikona.



And the popular Tang Band W8-1808.



I know some of you are thinking, filters for a full range (or wide band) driver? Yes! Almost all of them require a filter. Rarely do I ever see or hear one that doesn't need a filter to correct for a number if issues. There is no free lunch. You can't just hook a wide band driver up straight to a set of binding post and think nothing could be better than no crossover. No, they need filters. And if high quality parts are used throughout the filter can be very transparent. The problems that the filters correct are often a far greater detriment to the performance than anything a high quality filter can cause. In some cases the binding posts used are more of a detriment than the passive parts used in a filter.

Danny Richie

  • Facilitator
  • Posts: 12117
    • http://www.gr-research.com
Re: Designing filters for full range drivers
« Reply #1 on: 28 Jun 2018, 06:38 pm »
Okay the Jordan Eikona wasn't mounted in a sealed box. I was real excited about the way it was mounted, but it didn't have any effect on the response.



Here is the on axis response. It has a couple of peaked areas and the drop off below 700Hz is mostly baffle step loss. The top half of the top octave drops like a rock too. 



Here is the off axis response. It looses off axis response pretty quickly and predictably for a driver of its size.



The spectral decay shows some issues in that first humped up area.



So here it is after a correction filter. It is now much more balanced.



And by dropping the peaked area in the 700 to 1kHz range the stored energy there is much less offensive.



Here is a before and after response.


Danny Richie

  • Facilitator
  • Posts: 12117
    • http://www.gr-research.com
Re: Designing filters for full range drivers
« Reply #2 on: 28 Jun 2018, 06:52 pm »
Okay and now for the Tang Band W8-1808 driver.



The TB driver has a pretty aggressive rising response. This is one of those drivers that will sound pretty hot, and may reach fatigue levels pretty quickly. And it is compounded by the step loss below 500Hz.



This is a pretty large diameter driver so it looses off axis response pretty quickly. You'll need to be parked in front of these with them aimed right at you or you'll loose the top end quickly.



The spectral decay shows a little bit of stored energy across the board in all areas but no big issues anywhere. I had to raise it up 5db on the scale to get a good look at it down low, so the top end blew off the scale.



The impedance curve shows a resonance bump right at 900Hz. There is also a corresponding peak in the response right at that point. And there is a little stored energy in seen in the spectral decay in that area as well.



But check out the response with a correction filter. This is incredibly smooth. I left the top end up just a hair because these things drop off so sharply in the off axis. So the listen can play with towing them out a few degrees to get the top end response they want.



And the spectral decay with the filter. Nothing to get too excited about there, but not bad either.



Here's a before and after for comparison. This will be a lot easier to listen to.



Nice driver!

The huge drop off in the off axis could be supplemented by a rear mounted or up firing tweeter to add back some of the lost spacial ques to the room response. That will improve the imaging and sound stage layering a bit.

mcgsxr

Re: Designing filters for full range drivers
« Reply #3 on: 28 Jun 2018, 08:21 pm »
Interesting Danny!

I use a 5 inch TB driver (W5-2143) in a 0.2 sealed box these days as a DIY adventure.    Supported by biamped woofers in ported boxes that act as the stands for TB's.

It appears the trade off for balanced frequency response is lower efficiency.  Fair trade I'd say.

How complex are the filters?  Not asking for the recipe, just a comment on number of components?

Thanks.   :thumb:

Captainhemo

Re: Designing filters for full range drivers
« Reply #4 on: 28 Jun 2018, 08:48 pm »
Necely done  :thumb:
Will be cool if the  owner's post some impressions after  they get them  back  and do  some  listening 

jay

Danny Richie

  • Facilitator
  • Posts: 12117
    • http://www.gr-research.com
Re: Designing filters for full range drivers
« Reply #5 on: 28 Jun 2018, 09:06 pm »
Interesting Danny!

I use a 5 inch TB driver (W5-2143) in a 0.2 sealed box these days as a DIY adventure.    Supported by biamped woofers in ported boxes that act as the stands for TB's.

It appears the trade off for balanced frequency response is lower efficiency.  Fair trade I'd say.

How complex are the filters?  Not asking for the recipe, just a comment on number of components?

Thanks.   :thumb:

That driver has a little bit of a rising response as well, and with a cabinet that small you'll have some baffle step loss that will equate to a 5 or 6db drop that will start close to 900Hz. You could have a 10db spread or more from bottom to top. So some correction would help a lot. Looks like a couple of resonance issues in the impedance curve, but it is hard to say how much they are effecting the response without looking at a spectral decay.

Send it to me and I'll design it for you for free if you buy the parts from us. Shipping cost would be pretty low too for a small one like that.

Number of parts varies depending on the response and if there are response peaks or resonance issues to deal with.

The filter used in our LGK 1.0 kit is just three parts. http://gr-research.com/lgk10kit.aspx

The TB W8-1808 needed six parts.

The Jordan also needed six parts.

WC

Re: Designing filters for full range drivers
« Reply #6 on: 29 Jun 2018, 02:45 am »
Interesting. Would the filters depend on what sort enclosure you have the driver in?

I have the Eikona 2 drivers and was planning to put them in a MLTL enclosure. Would the filters still work, or would there need to be some tweaking done.

Were those the Eikona 1 drivers that you tested?

HT cOz

  • Industry Participant
  • Posts: 1182
Re: Designing filters for full range drivers
« Reply #7 on: 29 Jun 2018, 08:08 am »
Danny,

Given that the W8 1808 is made to work on open baffle, what about the V2 Bass section, plus W8 1808, plus the Neo3 all in open baffle?   I almost contacted you about doing this build...

Danny Richie

  • Facilitator
  • Posts: 12117
    • http://www.gr-research.com
Re: Designing filters for full range drivers
« Reply #8 on: 29 Jun 2018, 12:41 pm »
Interesting. Would the filters depend on what sort enclosure you have the driver in?

I have the Eikona 2 drivers and was planning to put them in a MLTL enclosure. Would the filters still work, or would there need to be some tweaking done.

Were those the Eikona 1 drivers that you tested?

Change the baffle size and shape and the values needed for the compensation filter also change.

I am not sure about the model of the driver.

Danny Richie

  • Facilitator
  • Posts: 12117
    • http://www.gr-research.com
Re: Designing filters for full range drivers
« Reply #9 on: 29 Jun 2018, 12:47 pm »
Danny,

Given that the W8 1808 is made to work on open baffle, what about the V2 Bass section, plus W8 1808, plus the Neo3 all in open baffle?   I almost contacted you about doing this build...

If you are going to use a Neo 3 with it then you'll want to let the Neo 3 do what it does best and let it play down low. But then if you are going to let the Neo 3 play down low then no need for a wide band driver to mate it with. A M165NQ would be a better choice.

But if someone wanted to add a couple of the 12" woofers used in the V2 with the TB driver then I have plenty of those 12's left in stock. That would be a good combo.  http://gr-research.com/v-2.aspx

jparkhur

Re: Designing filters for full range drivers
« Reply #10 on: 29 Jun 2018, 12:53 pm »
Danny,

Given that the W8 1808 is made to work on open baffle, what about the V2 Bass section, plus W8 1808, plus the Neo3 all in open baffle?   I almost contacted you about doing this build...

I have always liked the driver but it does weird things at higher volumes and at the top end.  I have worked with it in many forms and I NEVER can get it to sound right no matter the application when comparing it to a neo3 combination.  You IMO are better off just going straight to a Neo 3 combination with one of GR's applications.  I have failed many times and know the pitfalls of this driver.  A similar driver in the PE Dayton 208 PS has similar issues as it breaks up in the upper range easily and is hard to listen to.

As for the older V2 use of the Eminence Acoustinatior N2012 8 ohm drivers, I can say that I just recently helped a friend  with a pair of Super V tops and since he was short on money, we put in four N2012 drivers in the bottoms.  Are they servo NO, same NO, but for his application and about 250 watts of power per side, they sound really good.  If someone was looking for a poor mans version, this would work well, as they did in the V2, just not angled in the Super V.  I think Danny had said previously that he has a good set of these in stock and selling for 75 dollars per woofer.  Would be easy to look at or call him. 

JP


I have made every incarnation of the 1808 and never gotten it to work above avg.

Tyson

Re: Designing filters for full range drivers
« Reply #11 on: 29 Jun 2018, 02:34 pm »
I have always liked the driver but it does weird things at higher volumes and at the top end.  I have worked with it in many forms and I NEVER can get it to sound right no matter the application when comparing it to a neo3 combination.  You IMO are better off just going straight to a Neo 3 combination with one of GR's applications.  I have failed many times and know the pitfalls of this driver.  A similar driver in the PE Dayton 208 PS has similar issues as it breaks up in the upper range easily and is hard to listen to.

As for the older V2 use of the Eminence Acoustinatior N2012 8 ohm drivers, I can say that I just recently helped a friend  with a pair of Super V tops and since he was short on money, we put in four N2012 drivers in the bottoms.  Are they servo NO, same NO, but for his application and about 250 watts of power per side, they sound really good.  If someone was looking for a poor mans version, this would work well, as they did in the V2, just not angled in the Super V.  I think Danny had said previously that he has a good set of these in stock and selling for 75 dollars per woofer.  Would be easy to look at or call him. 

JP


I have made every incarnation of the 1808 and never gotten it to work above avg.

What about using the 1808 with a Neo 3?  I find many "full range" drivers work exceptionally well when used as "midrange only" drivers. 

HT cOz

  • Industry Participant
  • Posts: 1182
Re: Designing filters for full range drivers
« Reply #12 on: 29 Jun 2018, 02:55 pm »
Wow good thing I didn’t waste time on that idea!!!!  I’ve also decided to go Hypex Fusion 253, so I’m now really far from where I grew up RS750 DR redo and Neo 3 towers.

jparkhur

Re: Designing filters for full range drivers
« Reply #13 on: 29 Jun 2018, 02:57 pm »
Wow good thing I didn’t waste time on that idea!!!!
HT OZ. you can go back and look at all my failures with in AC.  Just use my user name.



HT cOz

  • Industry Participant
  • Posts: 1182
Re: Designing filters for full range drivers
« Reply #14 on: 29 Jun 2018, 03:00 pm »
I’m leaning into the SB Accoustics camp, but had an itch. I trust you and have seen you posting for what 10 years now...

jparkhur

Re: Designing filters for full range drivers
« Reply #15 on: 29 Jun 2018, 03:01 pm »
What about using the 1808 with a Neo 3?  I find many "full range" drivers work exceptionally well when used as "midrange only" drivers. 

Tyson

You are right.  I just don't want to pay 200 per 1808 driver when economically and most likely performance wise I could use a M165NQ driver or similar.   I agree that that driver sounds good to my ears until the top end, but also, it is expensive on all parts. 

I love the simplicity of FR drivers and what they offer, but for the last 12 years I keep going back to the Neo 3 and an accomplice driver.  Rich Hollis is doing some interesting things and only if you could match with a good passive xo the Neo 3 and Neo 10 or 8.  I think someone with spend some time on it and I do believe that it is currently making its rounds and will get to the public soon.   It would be half of the Super 7 upper, and with flat packs, will be a great top end.  Inexpensive for the most part too

jparkhur

Re: Designing filters for full range drivers
« Reply #16 on: 29 Jun 2018, 03:04 pm »
I’m leaning into the SB Accoustics camp, but had an itch. I trust you and have seen you posting for what 10 years now...

Either way, the Neo 3 or its equal is the way to start at the top.  We all have personal perf, cost objectives and significant others to deal with.  Or kids who smash your tube amps and think they can just vacuum it up and it will still work with wires hanging out of it.  Hey, lets plug that in and see what happens.   Smoke fire...death..   Oh, sorry, I have digressed in to berating my children... Errrrrrr

HT cOz

  • Industry Participant
  • Posts: 1182
Re: Designing filters for full range drivers
« Reply #17 on: 29 Jun 2018, 03:24 pm »
Either way, the Neo 3 or its equal is the way to start at the top.  We all have personal perf, cost objectives and significant others to deal with.  Or kids who smash your tube amps and think they can just vacuum it up and it will still work with wires hanging out of it.  Hey, lets plug that in and see what happens.   Smoke fire...death..   Oh, sorry, I have digressed in to berating my children... Errrrrrr

Wow never had that happen but had to put all my stuff up for a few years and focus on work and young kids. For the next year I’m living in India without my family so it’s full on audio hobby to pass the time. Music helps keep my sanity and I’ll learn as much as I can.  I’d take broken amp and kids vs leaving them in US and working abroad, even with lots of audio time.

I love the super 7 idea but jeeze it gets expensive for just the drivers. My uper limit for drivers is about $1,500. Fusion covers six channels of Ncore amp, dsp and dac for $1k.

mlundy57

Re: Designing filters for full range drivers
« Reply #18 on: 29 Jun 2018, 03:29 pm »
Either way, the Neo 3 or its equal is the way to start at the top.  We all have personal perf, cost objectives and significant others to deal with.  Or kids who smash your tube amps and think they can just vacuum it up and it will still work with wires hanging out of it.  Hey, lets plug that in and see what happens.   Smoke fire...death..   Oh, sorry, I have digressed in to berating my children... Errrrrrr

They can also do nasty things to turntables.  One time I went to play a record and when I touched the turntable it literally fell apart. Turned out my son thought he would play a record (or just check out the turntable, who knows) and knocked it off the stand. When it hit the floor it self-disassembled. He carefully picked up all the pieces, put them back on the rack then left the room as if nothing happened.

I wasn't too happy about that. Then my parents reminded me of the time I was banished from my father's workshop for taking is watch apart to see how it worked  :dunno:

HT cOz

  • Industry Participant
  • Posts: 1182
Re: Designing filters for full range drivers
« Reply #19 on: 29 Jun 2018, 03:31 pm »
Danny we high jacked your thread. What would you do to solve this problem?

Wide band driver, with support at top and bottom?  No crossover in the critical regions with good phase coherence?  Perhaps V2 bass as all agree it works and is economical.