[Bumpy]

Title says it all.

Particularly interested in anyone that's using DSP for just the low pass filter.

[Openly Baffled]

Hi Bumpy,
Why do you ask?
I'm working on a bi-amped three way system using separate channels for woofer and then mid/tweeter. The high end then uses a passive crossover to split mid and tweeter.
The woofer - mid crossover is a MiniDSP 2x4HD which also provides system EQ and phase alignment.
As it's open baffle, the woofer crossover is a low 150Hz and is currently a 48dB/Oct Linkwitz–Riley configuration.



Sorry for the image rotation - no idea why it only happens here.

[Bumpy]

Thank you OB

I am running 3 way open baffles. The mid and tweeter are perfect, but the mid runs full range towards the low frequencies, so the low pass on the bass, to get a good crossover, is not straightforward. Equally the open baffle bass generally needs a bit of a boost in volume. The DSP units you mention seems suitable, so I have a couple of further questions.

1. What amplification are you using for the bass.

2. Where are you getting your signal for the DSP unit

Thanks

PS I get a lot of pictures that fall over. Annoying isn't it - this is the only site that does it !!!

[JohnR]

I said before file a bug report. It would help if you uploaded an original image to dropbox and provided a link.

[Bumpy]

I said before file a bug report. It would help if you uploaded an original image to dropbox and provided a link.

Sorry John, I foolishly thought once a moderator had commented on it something would be done.  This coupled with the fact I couldn't work out how to file a bug report

[EIEIO]

I too use the MiniDSP as crossover/EQ/DAC/remote volume control. System is (very) loosely based on OBL-15 design – Alpair 12P running wide open and 15PR400 for the bottom. Amps are what I had on the shelf, for the moment - 8wpc tube for the top end and Dayton APA150 for the 15’s.

There’s lots of ideas for improvements but it sounds so good as is… upgrades will come slowly!

Input is from CD player currently but setting up my Mac Mini (headless) to plug into the 2X4HD.

[lokie]

Using a Berhinger 2496 to feed sub amps. Signal coming from the second output on preamp.
Also utilizing a PLLXO to high pass between the pre and main speaker amps.

All is still in experimental stage but sounding pretty good.

[Bumpy]

I too use the MiniDSP as crossover/EQ/DAC/remote volume control. System is (very) loosely based on OBL-15 design – Alpair 12P running wide open and 15PR400 for the bottom. Amps are what I had on the shelf, for the moment - 8wpc tube for the top end and Dayton APA150 for the 15’s.

There’s lots of ideas for improvements but it sounds so good as is… upgrades will come slowly!

Input is from CD player currently but setting up my Mac Mini (headless) to plug into the 2X4HD.

Thanks, for clarity does 100% of you signal pass through the DSP?

[Openly Baffled]

1. What amplification are you using for the bass.

2. Where are you getting your signal for the DSP unit

1) The bass drivers are 4 ohms combined and I have 250WRMS to provide a little head room for normal listening levels. The high(ish) power is required because of the bass slope applied to offset open baffle bass attenuation below the narrow baffles Fc. The slope applied is continuous -5dB/oct, below theory, but about right for my finite room environment.

200Hz = 0dB
100Hz = +5dB
50Hz = +10dB
25Hz = +15dB
Low Cut @ 16Hz / 48dB/Oct

You need a high power driver with a lot of Xmax to handle this. I am tuning the system using a Umik-1 and REW. Overall the result is very pleasing.

2) The MiniDSP 2x4HD has 3 inputs. I'm using USB (with the internal DAC) from a music server and TOSLink from a Chromecast in 192KHz (Premium) mode. The RCA Line input is connected to a CD player. It's very quiet and easy to use.

[Bumpy]

1) The bass drivers are 4 ohms combined and I have 250WRMS to provide a little head room for normal listening levels. The high(ish) power is required because of the bass slope applied to offset open baffle bass attenuation below the narrow baffles Fc. The slope applied is continuous -5dB/oct, below theory, but about right for my finite room environment.

200Hz = 0dB
100Hz = +5dB
50Hz = +10dB
25Hz = +15dB
Low Cut @ 16Hz / 48dB/Oct

You need a high power driver with a lot of Xmax to handle this. I am tuning the system using a Umik-1 and REW. Overall the result is very pleasing.

2) The MiniDSP 2x4HD has 3 inputs. I'm using USB (with the internal DAC) from a music server and TOSLink from a Chromecast in 192KHz (Premium) mode. The RCA Line input is connected to a CD player. It's very quiet and easy to use.

Thanks OB, the mini DSP 2x4 seems to do most things well, but if used just as a low pass filter is there scope to change the phase?

[gils]

Hello

example:

U-Frame 20 cm  speaker 18 " QTS 0.7 / fs 27 Hz // + 10 db at 30 Hz



U-Frame 45 cm  speaker 18 "  QTS 0.7 /fs 27 Hz // + 0 db at 30 Hz



200Hz = 0dB
100Hz = +5dB
50Hz = +10dB
25Hz = +15dB

It is not good to make such a correction, the distortion will be too great and produce a serious bad quality.

For a baffle plane, it is better to choose the right speaker and to adapt its load, the speakers of large diameter

[Bumpy]

Thanks Gils

 I am not skilled enough to know what those graphs are showing

Two of the rules I always remember with DSP are

1. Turning down peaks is better for SQ than turning up dips.

2. Match the speakers as best you can so the DSP has less to do.

I think rule 2 is what you are saying.

My 15" drivers (without DSP) give enough bass for me in the OB when driven with my 1.8W triode amp. I will be using the DSP to divorce the bass drivers from the midrange, to introduce a low pass filter for the bass driver and perhaps turn up the bass just a little

I have tested (without DSP) the Rotel amp that is to drive the bass drivers. It has the right characteristics.

[gils]

hello

you have to reduce the distortion rate as much as possible, make unnecessary gains only accentuate the distortion and produce a bad low quality, it also generates unnecessary high power to balance the bands.

best regards Gilles

[Openly Baffled]

you have to reduce the distortion rate as much as possible, make unnecessary gains only accentuate the distortion and produce a bad low quality, it also generates unnecessary high power to balance the bands.

Hi gils, in principle I agree but at very low frequencies I have two listening observations:

1) The bass, even with such high boost, is tight, transparent and effortless. Being free of resonance it still has that amazing open baffle feel. Low double bass notes sound like the instrument is in front of you.
2) Distortion below 50Hz is hard to hear. Even synthetically created distortion to 10% is much harder for me to discern that at higher frequencies.

I have A/B listened with various filters and slopes and like the higher boost more.

What I do observe is that the poles and zeros of the bass slope correction filters should be outside the bass speakers functional frequency band to reduce the phase impact.

Below are REW measurements that show the distortion effect of the greater boost. I can hear the impact of the flatter response - I can't hear the extra distortion - mostly listening to jazz and rock.

Bass Boost Distortion +15dB max (15-200Hz)


Bass Boost Distortion +10dB max (15-200Hz)

[gils]

hello Openly Baffled,

To reduce this huge distortion, you tried to lengthen the U-Frame?
What is the reference of your bass speaker?

now is good



best regards Gilles

[Openly Baffled]

Hi Giles,

I actually dislike the sound generated by a U baffle. I have tried several designs and I can always perceive a "box style" resonance that detracts from the clean open baffle sound.

Pictured below is one of my test set-ups from last year where I experimented with a variable U frame using piano hinges and sides trimmed to various depths from 250mm to 50mm. None of these combinations appealed. The effect of the sides was audible and the main measurement impacted was impulse response.

As I can't hear the distortion, I might repeat my measurements outside in case the "huge" distortion is an artifact of the room and it's contents. There is a lot of vibration in the room during the frequency sweeps.

[gils]

I understand better, nice job. 

the idea of only doing bass in U-Frame would be good from 20 hz to 150 Hz, then tuning with low medium or wide band from 150 Hz up to 20 Khz, the short acoustic circuit is less important in Flat- frame.

the agreement of the 18 inches are at 100 Hz see less depending on the speakers used.
For tests, I have tuned my JBL 4311b on the box 2x18 inches U-Frame at 45 Hz, the quality of bass is remarkable in U-Frame.

this is a JBL 4311b recording with a U-Frame subwoofer 45 Hz.

https://drive.google.com/file/d/1ZRb54_dasZrVGI_1l2VWz0Dkrj0-A8b_/view?usp=sharing

best regards Gilles

[Bumpy]

I too used temporary side baffles to create a U frame. With the baffle over the full height of the speaker, about 8"deep. I did not like it as some of that lovely OB sound was gone. Now I run with a deeper U baffle but only behind the bass driver. Its 18" (45cm) deep. It has no top but I may experiment with that later.

Here it is during my experimental lash up. NOTE, I abandoned the lid

[Openly Baffled]

Hi Bumpy,

Your experience and observations are maybe similar to mine. I also experimented with the lower extended baffle to avoid the midrange coloration.

In the end, using comparison to an actively EQ'd 22" flat baffle, I decided to ditch the extended baffle all together. This included an extended flat baffle that still seemed to cause the loss of purity in the bass (I tried 22" 26" 30" 36" widths at the lower woofer mid-point).

My conclusion, was that any mechanical bass loading to extend the lower end (below 40Hz) was resulting in the loss of sound purity in live double-bass listening. Electronic EQ with the same gain seemed to me to have less impact - so I went with active EQ.

However I agree that it's a continuum and based on your drivers and baffle configuration there may well be excellent mechanical solutions - I just didn't have the patience to find one. I suppose active EQ and Watts were cheaper than time!

Thanks for the "work in progress" photo.

[Bumpy]

Hi Bumpy,

Your experience and observations are maybe similar to mine. I also experimented with the lower extended baffle to avoid the midrange coloration.

In the end, using comparison to an actively EQ'd 22" flat baffle, I decided to ditch the extended baffle all together. This included an extended flat baffle that still seemed to cause the loss of purity in the bass (I tried 22" 26" 30" 36" widths at the lower woofer mid-point).

My conclusion, was that any mechanical bass loading to extend the lower end (below 40Hz) was resulting in the loss of sound purity in live double-bass listening. Electronic EQ with the same gain seemed to me to have less impact - so I went with active EQ.

However I agree that it's a continuum and based on your drivers and baffle configuration there may well be excellent mechanical solutions - I just didn't have the patience to find one. I suppose active EQ and Watts were cheaper than time!

Thanks for the "work in progress" photo.

Thanks OB

I am hugely encouraged by the runaway success of my midrange, so I press on.

Equally I note what you say about the extended baffle. You certainly explored the options:)  I think once you have discovered the open baffle sound its very recognisable when it is compromised.

Out of interest do you think you speakers get below 40hz now. Mine don't at present. This doesn't concern me but its nice to know what others are getting.

BTW, lovely looking speakers. I see that your bass and midrange drivers are closer to each other than mine, so there will be a difference there.

In case you missed it elsewhere here is a slightly more refines version of te side baffles. I have also pulled the speakers further out from the wall.