Center Channel 'Oris' Build

[Rudy81]

Now in room measurements. First, a measurement in the room of the original P. Audio setup compared to the new horn. Done, by simply swapping horns. This shows the crossover roughly around 600Hz from the bass wave guides. I just wanted to make a relative comparison by just swapping horns. I can easily see the P. Audio 2" driver is much more efficient than the Tang Band driver.

I do, however note two things. One is the hump around 1500hz as noted above, and two is the better HF response as compared to the readings I took outside. The impulse window has been shortened to 10ms. for clarity in the HF range.

[Rudy81]

In this graph, I used REW's offset to move the new horn's response to visually match the majority of the P. Audio response. This should simulate increasing my amp output to compensate for less efficient driver. The P. Audio reportedly has an efficiency around 110dB while the Tang Band 1808 shows an efficiency of 93dB.

The Tang Band and this DIY horn clearly produce more bass from 600Hz on up, which is where the crossover sits IIRC.

The good news, is that it is better in the high frequency area in terms of performance and output relative to the smaller driver. This seems to confirm my initial listening impressions that it did not lack the high end compared to my Heresy speakers in the shop.

I will have to either use my Ashly crossover to tame that peak around 1.5kHz or rely solely on Audyssey Pro to handle the problem.

Feel free to make suggestions.....I can use all the help I can get.

[CECAA850]

I wouldn't think the horn would affect the response at 1500Hz, must be the driver. Is that the only driver like that you have? Have you seen a manufacturers frequency curve of that particular driver?

[CECAA850]

HERE

Shouldn't be the driver then.

[Rudy81]

HERE

Shouldn't be the driver then.

I don't think it is the driver. The other, identical drivers on my mains don't exhibit that anomaly. I suspect the horn is creating that peak. Not a huge deal. I can flatten that out with the Ashly crossover. I am excited to really hear the higher end in the center without having to give it a big boost as I did with the 2" driver.

Now the tedious task of phase aligning the HF with the LF drivers.

At least the larger size makes a better aesthetic fit with the large sized mains.

[Rudy81]

Here's the initial setup. No EQ, XO set at 204Hz to match what I have set in my mains. As a starting point, the horn and driver combination seems promising. With a little EQ, time alignment and a final setup with Audyssey Pro, I should be able to make this work.

It looks like that pesky peak at 1.5kHz is the first thing I need to fix. I'm fairly certain that it is due to the horn shape and size.

I would have gotten to this point earlier, but my Win 8.1 machine refused to load the drivers for the Ashly. I finally decided to use my trusty Win XP media PC to run the Ashly program.

Edited November 26, 2013 by Rudy81

[NBPK402]

Looking great!

[Rudy81]

Looking great!

Thank you. This seems to be working out great and well worth the effort. One sheet of BB and a few days work.

Tomorrow I'll run Audyssey Pro and then watch a couple of good movies to get the full effect. I'm really looking forward to evaluating the dialog channel coming from primarily a single driver.

[Guest David H]

Rudy, outstanding work.

I am sure you are correct about the 1500hz bump, I assume its the flat dimensions causing that bump.

EQ it, and forget about it.

Dave

[Rudy81]

Rudy, outstanding work.

I am sure you are correct about the 1500hz bump, I assume its the flat dimensions causing that bump.

EQ it, and forget about it.

Dave

That's the plan. Working on final XO checks before I run Audyssey Pro, which is a major PITA!....but worth it IMHO.

[CECAA850]

That's an outstanding looking LCR you have now. I'm sure the sound will be of the same caliber.

[The Dude]

Do you use any form of auto eq, to eq the peak. Or trial and error.

[Rudy81]

Do you use any form of auto eq, to eq the peak. Or trial and error.

In my room measurements, the peak is not as bad once you take room effects into consideration. I will simply rely on Audyssey Pro to handle any issues. If the software in my Integra can't handle the issue, it will let me know and I can then go back and make the correction manually in the Ashly.

[Rudy81]

That's an outstanding looking LCR you have now. I'm sure the sound will be of the same caliber.

I hope so, but you are correct that it looks much better than it did with the small P. Audio horn.

[Rudy81]

I was fortunate enough to have all day to work on setting this horn up in the media room. Set the XO at 204Hz to match my mains. Decided to not add any EQ, just checked the XO point to ensure I had good phase correlation with the bass bins. Then, ran Audyssey Pro for 14 positions in the room. Audyssey did not have any problem taming the 1500hz hump and showed that this driver has much better high frequency response than my P. Audio setup.

After Audyssey was done I sat for about a four hour listening session focusing on center channel performance.

Initial impression is that the new setup sounds quite a bit different than the P. Audio setup. The new horn does not have the vertical dispersion the P. Audio horn has. Not a problem if you are seated in the room since the driver is right at ear height for those seated. Once you get much above or below the horn, you start to lose that clear sound. Horizontally, however, the DIY horn is superior. For center channel dialog, there is not bad seat in the house. I tried all 7 seats and could barely tell any difference in movie dialog. I was very surprised the horn does that well off axis.

In general comparison to the Oris horns, it still is not quite as good. The full size Oris just has such a full bodied, clean, transparent sound. The DIY center has a more 'constrained' sound...hard to describe, but you can hear a 'qualitative' difference between the Oris and this build. Overall it will be more than adequate for center channel duty however.

Now for the subjective impressions.

First concert I popped in, was Joe Bonamassa at Prince Albert Hall in Blu Ray. The very first chapter is an interview with Joe and my initial impression was that his voice sounded rather 'thin' or higher in frequency than I remember. That kind of scared me and I stared to think I had made a huge mistake.

So, popped in two favorites that I have watched more than any other movies. Gladiator and Top Gun.

Gladiator was the perfect movie for me to use to evaluate the center. The first chapter war scene has some great right to left and back panning of various sound effects. This gave me a great idea of how well this horn is timbre matched to the mains. Just no difference in sound as panning sounds moved around. That was my major goal and a huge plus.

Listening to various scenes with good, solid dialog exclusive to the center channel put my fears to rest. Russel Crowe's voice was right on, nice and deep, with what I would consider his normal pitch. I started to realize that the improved hf response in the driver is different enough that it seemed odd with higher pitched voices....such as Joe's in the concert interview.

Top gun center voices sounded good and the flying scenes confirmed good timbre matching.

So, thus far I'm pleased with the effort. I still need to spend quite a bit more time watching movies in a more relaxed atmosphere and not as an evaluation. When evaluating I often times get too involved in trying to analyse things and lose some perception by trying too hard.

Bottom line, so far so good. It's certainly not perfect, but an improvement for what matters in a center channel for watching movies.

[Rudy81]

Ok horn experts. I need your help. I watched Red 2 last night trying not to focus on the center channel.

Sound effects and sound panning was, as I mentioned, excellent.

However, I'm convinced there is some issue with some, not all, voices. In many dialog only scenes the voices sound good, intelligible, and what I would expect from a center channel. However, and this is a big however, in some scenes the voices sound 'honky' and you get the impression that they are speaking with literally a horn in front of their mouths.

This is hard to explain, but sometime it sounds fine, and other times it sounds 'odd'. I'm beginning to really understand the challenges of building a suitable horn.

I suspect the vertical dimension limits of the horn are causing the sound waves to get constricted. I checked my flare profile and it does match the Oris. Of course the Oris is about 3 times larger in the vertical.

This is really perplexing since music and sound effects sound really good. Somehow, voices are effected....but not all the time and not in all scenes. Just goes to prove that frequency sweeps definitely don't tell the whole story.

Later today I'm going to take the horn apart and try the driver alone in an open baffle configuration. I need to compare the sound with and without the horn. Fortunately I had the foresight to make the motorboard removable from the actual horn.

On a separate note, in researching open baffle designs with this driver, several DIY'ers note the 1500Hz hump with the driver. Although not on the manufacturer specs, these drivers clearly show that anomaly. It will be interesting to see how it performs without a horn.

Edited November 27, 2013 by Rudy81

[mikebse2a3]

Rudy very interesting project...!

It will be very interesting to see your open baffle results.

Have you thought about using some rigid foam (or something easy to shape) placed in the corners of this horn (using double sided tape temporarly) to experiment with making a smoother throat area transition to the driver and see how that measures and sounds especially in the 1.5kHz region?

While I was at it I might would even consider trying some acoustical foam in these corners and run a curve and listen for any changes just as an experiment.

It would be very interesting to see some polar plots in the 1kHz and above range on an open baffle for this driver.

Just some thoughts here. We know a horn with a 2" exit driver (having a throat design without a narrowing diffraction slot such as the K510 or K402 for examples) begins to show that it's polars control and frequency response begins to transition from the horn and driver combined influences to mostly the driver/phase plug influences at approximately 6.78kHz by calculation. If you do the same calculation using 8" as an exit value this calculates to 1.695kHz transition point and warrants consideration.

Of course full range drivers are very complex systems in themselves and mating them to a horn with predictable results has to be an extreme challenge. I'm very much enjoying this thread Rudy and your efforts to make this project successfull..!

miketn

Edited November 27, 2013 by mikebse2a3

[CECAA850]

Most of the energy of the human voice is from 300 to 3000Hz which is coincidentally the area of the hump and correction. I can't help but think that this is contributing to your issue.

[Rudy81]

First, my apologies to Joe Bonamassa for inferring his voice was high pitched in the opening interview of his Blu Ray......Wrong! The horn I built is creating that effect, although not easily discernible on a FR plot.

I took two readings this morning. One with the horn as I set it up yesterday. Same results, same 1.5kHz hump. I then removed the horn and set up the driver with only its baffle, no enclosure. BTW, this driver is well regarded as very good for open baffle use.

As you can see by the comparison plot, the hump is gone! The driver actually shows a flatter response by removing the horn overall. A little weak between 200-300Hz, but that could easily be a phase alignment issue with the LF bass. I'll work on that once I decide what to do next.

In listening to Gladiator, Red 2, and Bonamassa, the honkiness I mentioned earlier is gone.

Kind of a bummer since I spent so much time with the horn build. However, the goal is to get a good timbre matched center to my mains. That's all that matters right now.

The driver is definitely better than the P. Audio gear, but the Tang Band with that horn is unacceptable for center channel dialog duty.

So.....back to the shop to build a proper baffle and start over.

I had been giving some thought to Mike's suggestion of making a better transition from the driver at the throat. I just don't have the skill to build something with those kind of dimensions. I didn't think of foam or something like that. At this stage though, it would likely be wasted effort based on what I'm hearing from the open baffle.

[Rudy81]

Wow! The more I listen, the more I'm surprised by the effect of an improper horn design on a driver. I wish it would have occurred to me to start out trying an open baffle. Would have saved a bunch of time and a sheet of BB....but it was fun building it.

I wonder if my wife could use the horn as a flower pot?

This Tang Band 1808 is just a great driver for $200.