Usher 8955a, Scan Speak 12M, and Peerless HDS 3-way project

[Jed]

So where are the pics?!! I've always wondered how the 8955 compares to say the RS225. have you seen any measurements Jed?

Also I wonder if the 12M is as average as the measurments show at first glance. If I remember right the FR is very smooth so linear distortion is lowered. I also wonder if the motor is more linear or otherwise better performing than we DIY can measure. Particularily at high excursion.

I'll get some pics up tonight. As for the 8955a, I'd take it over the RS225 since I've used both. Right now the 8955a is sealed too, and I can't believe the deep tight bass I'm getting. I'll try a ported alignment on Wed.

Now I say I'd use the 8955a over RS225, partially because I think it handles a LR2 slope at 300-400HZ very easily, not that the RS225 doesn't, but it has that cone breakup that has to be dealt with.

As for the 12M, the distortion measurements I've seen from HobbyHIFI, Zaph, and others all point to the same trends- average HD performance. Yes it has a smooth response, but with a little work I'll get most any driver to have a smooth response. I could be faulted for jumping to conclusions, but you have to follow what you want to do, and I'm not the first person to change his mind!

[JonMarsh]

Wow,

Can't believe it has been almost 2 months since I've posted about this speaker development.

Several things have happened with this speaker development. First, the driver selection has evolved. My previous speaker before I started working on this one included a Visaton MHT12 tweeter and Eton 4-300 mid. The Eton is excellent indeed- especially when used from 400-3k. So, the distortion for this midrange is on average .5% or less in this range. Not great, not bad. Fairly average. I still like the cone nevertheless. Here's where my idea started to change. In my cabinet now is the Usher 8955a sealed, Eton 4-300, and Visaton MHT12.

Next up would have been to try the 12M and HDS with the 8955a, but Zaph posted rather average results for the Scan Speak 12M, so I got to thinking that the 12M maybe not so much better than the Eton. Well, yes it can do LR2 as well as some other things very well, but I thought since they are new, I might as well sell the 12M and move onto something else. So I did. I really want something of an improvement over Eton 4-300. LR4 sounds just fine to me at 3K, so what other driver would work well, I thought to myself.

I've always been interested in the Accuton Mids. So, today I placed a purchase for C79. It has EXCELLENT distortion characteristics and is useable from what I have seen up to 3.5k. Depending on how hard I want to push them, down to 250HZ. Not shabby. I considered C90, but it just didn't seem to be THAT much better than C79 and it won't work if I want to use my favorite tweeter, the Visaton MHT12.

The HDS dome will have to wait for something else where I can use its excellent low end.

The Usher 8955A is playing beautifully with Eton 4-300, and MHT12 in the cabinet I have pictured in this thread. When I have time I'll snap some pictures and maybe we could change the title of this thread.

However, I can't wait to get my C79s and replace the Eton 4-300 with them. I'll have the C79s mounted and playing by the end of this week for sure.


In summary, next up is Usher 8955a, Accuton C79, and Visaton MHT12.

Jed

Interesting Jed- I'm going to be curious to see how that turns out. I usually feel like the lone wolf on the forum talking about Accuton and testing them. I'm going to use a C79 in a CC speaker this year if I can find the time, but the towers are going to use C89's- less obstruction of rear wave, and a bit more Xmax. Good luck with this new direction and keep us posted.

I've been working pretty much 7 days a week (feels more like the Beatle's song "Eight Days a week" when you keep that up for a while), so I'm totally stalled - got in the new router bits I orderd, and that's IT in the last two months. Looks to continue until the end of May, unfortunately. But then I'll be out of the repetitive hard deadlins every week or two mode for a while, and maybe I can slack off a bit... that or I'm taking out a contract on my boss.

~Jon

[JonMarsh]

I could be faulted for jumping to conclusions, but you have to follow what you want to do, and I'm not the first person to change his mind!

Change your mind? Hey, that's cheating! How can you be held accountable for anything if you go and do that? :W

[TacoD]

I usually feel like the lone wolf on the forum talking about Accuton and testing them.

Haha, I'm feeling the same with practical every driver I'm using. E.g. the audiotechnology drivers or the Philips rsp8q planar. Botom line is that if you aren't using dayton rs180 you are on your own. Which is fine with me ....

Jed: did you consider the Visaton TI100 or the al130m, which could be the natural match with your MHT 12. Somehow I remember some posts of you about these drivers.

[Jed]

Here is the current speaker with MHT12, Eton 4-300, and Usher 8955a. A nice speaker with a relaxed sound, but offers enough resolution to get interesting. I hope the C79 offers more than the Eton, as it definitely has a lot of potential, as Jon's and other's measurements would indicate. Unfortunately, madisound does not have many accuton drivers in stock. No C79 or C90. They placed their order in September and are still waiting! I was lucky to find some C79 drivers from soniccraft.com. Hopefully they will arrive soon and meet my expectations.


The front baffle is removeable so I can change drivers out on a whim. :B

When I'm done, baffles will probably end up getting painted black. I tried the solid maple corners and messed up. Dropped one and damaged the corner, then cut it too small. Oh well.

[Jed]

Jed: did you consider the Visaton TI100 or the al130m, which could be the natural match with your MHT 12. Somehow I remember some posts of you about these drivers.

Those Visatons look nice as well but the C79 has the low Q resonance peak. I've used Excel W15CH, which I think is probably close to the Visaton units in sound character.

I have heard Avalon speakers and liked them very much, so I'm ready for the C79 challenge. Luckily, many have used this driver so there are a lot of crossover topologies to chose from to model and design the crossover network after I take my own measurements.

It pleases me that Jon is working on a system with C79s. I may borrow his bandpass design when it becomes available, or at least compare it to my own developments.

Jed

[Jed]

Change your mind? Hey, that's cheating! How can you be held accountable for anything if you go and do that? :W

Lucky for me nobody is depending on it

[MarcE]

The number of degrees is calculated by the depth of offset difference between driver A's and driver B's acoustic center and the relative distance from driver A's to driver B's center. Basically just picture a triangle- then with some trig. you can calculate the angle.

Jed:

Hi! I found this post via Google and I was wondering if you could take a few seconds and perhaps post a quick diagram of the calculation you described above? I am in the initial stages of considering a similar design as what you posted at the start of this thread and would appreciate yours (or anyone else's) help!

[Jed]

Jed:

Hi! I found this post via Google and I was wondering if you could take a few seconds and perhaps post a quick diagram of the calculation you described above? I am in the initial stages of considering a similar design as what you posted at the start of this thread and would appreciate yours (or anyone else's) help!

Roman says it better than I and includes a calculator for the task. Welcome to the forum! Let us know what you plan to build.

Jed

http://www.rjbaudio.com/Audiofiles/driver%20offset%20calculator.html

[SQdude]

[MarcE]

Roman says it better than I and includes a calculator for the task.

Jed,

It seems that the calculator gets me about 80% of what I'm looking for. Next, I will use these values to set the angle of the front baffle to the value that reduces the midrange acoustic offset and the woofer acoustic offset to zero, thus finding the angle which forces a, b, and c (from the diagram) to be equivalent and also equal to the listener's ear when seated at the primary listening position. Perhaps I'm oversimplifying some part of this? I realize that once the angle has been set and thus the approximate location of the drivers on the front baffle, I will still have to deal with the baffle diffraction... that's the next step. Am I on track at all?

Welcome to the forum! Let us know what you plan to build.

I will! Thank you for the warm welcome.

[Jed]

Sounds like you are on track, and much of this may not have to be 100% perfect because you can account for phase delay in the crossover circuit, that is if you are using shallow slopes crossed over high in frequency. Otherwise, steeper slopes, such as 4th order, are not as critical to get the acoustic centers aligned.

I'm assuming you want a LR2 design. Have you decided on drivers yet?

Jed

[MarcE]

Jed:

well, partially :P I am looking at the Hiquphon OW4 and the Audio Technology C-Quenze 15 H 52 06 13 SD and haven't decided on the third driver. I'm thinking a 10" or 12" depending on a few things. Of course, any recommendations would be greatly appreciated and duly considered... I honestly have been reading and calculating based on the aforementioned drivers, which has kept me quite busy!

Also, much thanks to David Gatti who has gotten me thusfar.

I do have the ZMA/FRD data for these two drivers, but I'm trying to figure out how to subtract the previous baffle configurations and volumes using the Excel tool FRC, with little to no luck as I don't have a copy of Excel 2000 handy Is there another tool that might be able to help me with this? It seems to "quasi-work" with the newest version of Excel and perhaps later this week I'll have more time to sit down and play with it.

So, I'm been going through some more material and learned that group delay is a function of Qts, with minimal group delay at Qts = 0.5. Seeing that I'm taking great strides to minimize this other places, should I be shooting for a Qts <= 0.5? Or perhaps Qts around 0.6 when the cavity is empty, which would allow me to hit the 0.5 mark after 100% stuffing of some sort?

edit:

Jed, do you happen to have ZMA/FRD data for the Scanspeak 25W 8565-01 that I could play around with?

Thanks in advance!

[Jed]

Jed:

well, partially :P I am looking at the Hiquphon OW4 and the Audio Technology C-Quenze 15 H 52 06 13 SD and haven't decided on the third driver. I'm thinking a 10" or 12" depending on a few things. Of course, any recommendations would be greatly appreciated and duly considered... I honestly have been reading and calculating based on the aforementioned drivers, which has kept me quite busy!

Also, much thanks to David Gatti who has gotten me thusfar.

I do have the ZMA/FRD data for these two drivers, but I'm trying to figure out how to subtract the previous baffle configurations and volumes using the Excel tool FRC, with little to no luck as I don't have a copy of Excel 2000 handy Is there another tool that might be able to help me with this? It seems to "quasi-work" with the newest version of Excel and perhaps later this week I'll have more time to sit down and play with it.

So, I'm been going through some more material and learned that group delay is a function of Qts, with minimal group delay at Qts = 0.5. Seeing that I'm taking great strides to minimize this other places, should I be shooting for a Qts <= 0.5? Or perhaps Qts around 0.6 when the cavity is empty, which would allow me to hit the 0.5 mark after 100% stuffing of some sort?

edit:

Jed, do you happen to have ZMA/FRD data for the Scanspeak 25W 8565-01 that I could play around with?

Thanks in advance!

Yes, I do. PM me.

Jed

[Jed]

Jed:

: Is there another tool that might be able to help me with this? edit:

Not that I know of. How different of a baffle will you be using versus the original anyway?

Also, download Unibox to experiment with different box configurations as well. You will see how you can tune various conditions to satisfactory results using somewhat nontextbook goals.

[MarcE]

Not that I know of. How different of a baffle will you be using versus the original anyway?

well, a rectangular two-way monitor on a stand versus a trapezoidal three-way floor stander...

[David G]

Marc
If the cabinet width is roughly the same - i.e about 24-27cm, you should not need to make any adjustments to the FRD plot. If the baffle edges are not curved as in mine, you will however get more ripple in the response.
Note : Thats assuming a 3 way. If it's a stand-mount 2-way, the differences will be much more significant.
Regards
David

[MarcE]

David:

Really? I figured for the OW4 tweeter that the baffle diffraction would change considerably in going from a baffle that looks like this:

http://www.rjbaudio.com/Asterion/asterion.html

which is where I got the OW4 data from (unfortunately, the wasn't any impedance data included, so I'm trying to either get that or generate it myself somehow)

to a trapezoidal prism similar to your Delta cabinet shape?

[David G]

Sorry for my lack of clarity. Yes you can expect significant difference in response between those 2 baffles, especially since the width of a trapezoid baffle around the tweeter is much less than in the rectangular one. Also the distance from the tweeter to the top and bottom is different.

[Jed]

If you guys don't mind, I think we should start a new thread that accurately depicts what Marc hopes to accomplish in his new 3-way. The title in this thread was a project that never materialized, and now we're talking about 1st order crossovers and completely different drivers. It might be less confusing, that's all.

Jed