Jump to content

Scanspeak 22W/8857T-00 8" v 22W/8851T-00


mikes875

Recommended Posts

I am building some large 4 way DIY speakers and are having difficulty deciding between these 2 drivers. Almost identical except the 8857 is Aluminium and the 8851 is Paper. Anyone have experience with these and have any suggestions?

 

The other drivers are

32W/4878T01 (purchased)

12MU/8731T-00 (planned)

R2904/7000-09 (purchased)

 

I am planning on 2 each of the 12MU and 22W in a MTM D'apolito configuration. These will be in parallel resulting in 4 ohm each side. There will be a single 32w below on each side.

 

I am going to use an active crossover with DSP.

 

Any experience with the 2 varieties of 22W?

 

Thanks

Mike

Link to comment
Share on other sites



Was wondering that too. Thought that the 12MU will get to below 1khz easily enough so might get more advantage on the lower frequencies with the 22 than the 18. Also 22 is more capable at higher volumes with its higher sensitivity, although the 12 might struggle to keep up in the mid range in that respect

Link to comment
Share on other sites

Thanks Dave. I have been doing more research and I think you are right 18WU would be a better fit. If nothing else will mean that the drivers look like they belong together. If I didn’t have the 32w the 22’s might have made more sense. Thanks

Link to comment
Share on other sites

I have 2x 12MU 8731T in my speakers and I run them down to 230Hz with a 12dB slope. I then cross over directly to exactly that woofer too - 230Hz is where the woofer's cone cry begins so they're very accurate below that (and they have awesome bass). There is no need for a further driver between those if you use 2 midranges. I drive them with a 150W power amp and they get painfully loud without the midranges ever running out of puff. Different tweeter though.

Looks like this:

WDpsX1C.jpg

Edited by Ittaku
Link to comment
Share on other sites



Thanks. That is much lower than I was expecting the 12WU to go. After more research and advice on here I am now planning on the same as you with a single 22W 8851 above to will the gap between the 12 and 23 a bit. Sounds like it is hardly necessary, but the neither is the whole build really...

Link to comment
Share on other sites

2 hours ago, mikes875 said:

The 32w arrived today and they are the subwoofer version T00. Not that different but still do I send them all the way back to Europe...

That's very annoying. They are quite different though in a box the T00 will give you lower extension but they're also less efficient. I have the T001 because of its amazing flat response to 30Hz in free air and I'm using it in an OB. In a box the 00 would actually give better low frequency extension with its amazingly low resonant frequency of 18Hz. The T00 still has cone cry from the same frequency so it'll likely still be good for 230Hz.

Edited by Ittaku
Link to comment
Share on other sites

Yeah starting to wonder if it might be better for my application as efficiently is less of an issue as I am going to use an active crossover anyway so have plenty of control over power. It is going to be in a ported box and and can alter the design slightly to get the extra volume needed.

Looking at the response curves although they break up at the same level the 01 is much flatter. If that really matters that much at those frequencies anyway

Link to comment
Share on other sites

2 minutes ago, mikes875 said:

Yeah starting to wonder if it might be better for my application as efficiently is less of an issue as I am going to use an active crossover anyway so have plenty of control over power. It is going to be in a ported box and and can alter the design slightly to get the extra volume needed.

Looking at the response curves although they break up at the same level the 01 is much flatter. If that really matters that much at those frequencies anyway

Well as I said the flatter response is only relevant in free air, the way I use it. If it's in a box, that roll off doesn't mean anything for the real world application. In the box it will be flat down to impressive levels. If I were making a boxed speaker I'd use the 00. The marketing distinction calling on a woofer and one a subwoofer in my opinion is also about the bottom end capabilities with that lower resonant frequency. I suspect their top ends are comparable.

Edited by Ittaku
Link to comment
Share on other sites



5 minutes ago, mikes875 said:

I think you are right it is mainly marketing. The 01 has a lower moving mass 112g v 150g. Do you think that will impact transients etc?

Well yes it will have lower group delay being lower mass, but putting a driver into a box impacts group delay much more than the driver mass does, and making the box ported makes it another magnitude slower again. The 01 I have in open air is the fastest low extension woofer I've ever heard in a complete system bar none. Nonetheless the 00 will likely be excellent in a box all round. Speed isn't everything.

Edited by Ittaku
Link to comment
Share on other sites

Either are going to be excellent I am sure. I have been planning my design on the 01 and now this comes along. Seems silly sending them back. Might need to mull over it over some more I think...

Link to comment
Share on other sites

Decided to keep the 32w/4878T00 and have ordered a 22W/8851T00 for each side. The supplier gave me a good deal because of the mix up. Look like they will compliment each other nicely as 22w go pretty low on there own. Thanks everyone for your input

Link to comment
Share on other sites



  • 5 weeks later...
I have 2x 12MU 8731T in my speakers and I run them down to 230Hz with a 12dB slope. I then cross over directly to exactly that woofer too - 230Hz is where the woofer's cone cry begins so they're very accurate below that (and they have awesome bass). There is no need for a further driver between those if you use 2 midranges. I drive them with a 150W power amp and they get painfully loud without the midranges ever running out of puff. Different tweeter though.
Looks like this:
WDpsX1C.jpg

How is your project going? I got one of mine going last weekend with the miniDSP crossover.

I ended up with the 32w 00 sub version + 22w 8851, the same 2x 12MU as you and D2908/7140 Beryllium Dome Tweeter. Crossed 170, 900, 3000 (still experimenting with cross freq). Sounds awesome, can’t wait to get the second one ready to test. The 22w response in a sealed chamber is unbelievably flat, like a ruler.

I tried crossing at 230 as you did without the 22w and it was quite noticeable that it was missing. Just thought I would let you know in case you wanted to consider another driver to fill the gap. Yours is the 01 model so will do a better job of the higher end of the range so probably works better
Link to comment
Share on other sites

2 hours ago, mikes875 said:


How is your project going? I got one of mine going last weekend with the miniDSP crossover.

I ended up with the 32w 00 sub version + 22w 8851, the same 2x 12MU as you and D2908/7140 Beryllium Dome Tweeter. Crossed 170, 900, 3000 (still experimenting with cross freq). Sounds awesome, can’t wait to get the second one ready to test. The 22w response in a sealed chamber is unbelievably flat, like a ruler.

I tried crossing at 230 as you did without the 22w and it was quite noticeable that it was missing. Just thought I would let you know in case you wanted to consider another driver to fill the gap. Yours is the 01 model so will do a better job of the higher end of the range so probably works better

My project is all but complete in the driver and crossover design. The only thing I haven't finished is doing an absurd aluminium baffle but I'm still waiting on my engineer to do the finite element modelling to get the resonant frequencies where I want them though I'm in no hurry since it sounds so damn good at the moment.

 

Mine uses passive crossovers and the two midranges are different - both cross over at 230 Hz but one rolls off 6dB at 900 Hz while the other 12 dB at 2500 to meet the tweeter. I'm not sure what you mean by "noticeable that it was missing" unless you say it sounded better with it in, as the sweeps confirm it is flat across the 230 Hz crossover point entirely. There are also massive issues with using the quality components I'm using to get the crossover down to 170Hz and the band pass filters passive create a lot more excitement of the frequencies they allow through when the cross over points are close together, making balancing them more difficult. The DSP I'm using is for room correction only and not for active crossing over except for the woofer to subwoofer at 30Hz. To get 230Hz, the current woofer uses 10 gauge 6.2mH air coils which are only .32 ohms and look like this, (pen for size comparison):

IMG_20180412_100050.thumb.jpg.01d864878b18f4eb67581a9a01c74481.jpg

 

Getting it down to 170Hz would make for a monstrous inductance requirement and increase the resistance which will somewhat raise the Qt further - though the drivers are so low Qt already that they have plenty to play with but one thing I love is the absurdly low group delay the woofers provide thanks to that low Qt. Another issue is the massive capacitance required for 170Hz for the upper midrange - It would require an absurd number of parallel caps if I use the quality that I desire in the signal path as the largest metallised film caps that multicaps makes are 30uF and I already need 80 odd uF for the midranges in parallel.

 

Everything's a compromise even in what apparently is a no-compromise approach, and mine revolves around a passive design. Yes you're probably thinking I could avoid these issues with an active crossover but I'm so wedded to the use of my audio research SE 150 that to do it active I'd have to buy more of these babies... maybe 3 more of them for a 4+1 way design (my current design is effectively 3.5+1) and while I might consider it for a cost-no-object design, I don't have the power and cooling in my listening room to support that... lol. In a nutshell a (high quality) passive bandpass for 170~900Hz would introduce far more complexity and I'm not really seeing or hearing any limitations with my woofers between 170 and 230Hz. It would also make for much more excitement in the midranges if their cross over points are closer together from 900 to 2500 Hz (also it would make one of the midranges redundant since 900 is the HIGH pass point currently.) I also noted the 22w you're using shows cone cry beginning from 750Hz.

 

As you also noted, mine is considered a woofer by scanspeak and the T00 is considered a subwoofer (whatever that means in practice), and you're using a box while I'm not which changes the sound quite substantially so they're not really comparable since our projects are quite different. If only we were in the same city we could audition each others' to see how we feel.

Edited by Ittaku
Link to comment
Share on other sites

Looking great, very serious crossover you are designing there, I am cheating going active I am afraid. I see your point on the amp front, I am now looking at a 8 channel NCore amp as my existing ones are not really up to the job.

 

When I said missing, yes I did mean it sounded better with, the 22w produces very clean bass sealed, well damped and crossed reasonably high at 170 Hz. You are right, there are so many differences between our projects so my observations don't necessarily correlate to yours (e.g. T01 v T00, passive v active, ported v open etc.). Just thought I would let you know my results. as with the active it is very easy to try different settings and see how it compares without having to rebuild the crossover.

 

Sorry you probably could have done without the seed of doubt if you are already fixed on your design. When I tried with out the 22w I didn't spend much time tuning the response curve (as I did with it) so it is probably not really a fair comparison anyway the more I think about it.

 

If you are ever it Sydney you are always welcome to come and have a listen

  

Edited by mikes875
  • Like 1
Link to comment
Share on other sites

23 hours ago, mikes875 said:

... and D2908/7140 Beryllium Dome Tweeter. Crossed 170, 900, 3000 (still experimenting with cross freq). Sounds awesome, can’t wait to get the second one ready to test. The 22w response in a sealed chamber is unbelievably flat, like a ruler.
 

How are you getting on with this tweeter? I'm curious about the baffle you're using and how it measures/sounds.

 

Is the 22W behaving OK up to 900 Hz?

 

Some pics of your progress would be nice.

Link to comment
Share on other sites

Still very much in the testing phase with lots of sub optimal areas. Drivers are not recessed, only one speaker is built etc however already sound awesome, pretty confident the end results will be worth it. Will do a separate thread with build progress and test results on the 22w. Have been playing with a cross between 700-1000.

Very happy with the D2908/7140. Well worth the high cost IMO. My wife doesn’t normally pay any attention to my speaker projects although this time she said that if she closed her eyes she felt like the singer was sitting in the room. The Mid & Tweeter is to thank for that I think. Only issue is the physical dimensions are large 120mm. The same size as the 12MU mid. Makes the MTM a bit more spread out than I would have liked

Link to comment
Share on other sites



That (wife bit) would be considered my many as the ultimate recommendation! That D2908 and 12MU is surely a winning combination.

 

What struck me on reading your first post this morning was that you'd purchased the R2904/7000 tweeters, which I've been wanting to try, but at $400 each, am a bit hesitant. They, like your Be ones, have a healthy natural roll-off at the lower end, which could potentially make my life easier. Everything else seems to have that step in FR, which gives too much output at and below 1 kHz.

 

But even with the SBA tweeter, whose datasheet shows a similar good roll-off, I get that step, so either the tweeters are non-typical, or it's my baffle.

 

Pics of a work-in-progress are pretty interesting to some of us.

  • Like 1
Link to comment
Share on other sites

I got a good deal on the Be domes so decided to keep the r2904/7000 which I bought second hand for my next project.

I also have a set of ScanSpeak 23W/4557T subs that go down to 33 Hz in a 27 litre enclosure and can probably cross up to 400 Hz . Next plan will be to find a suitable mid and make a 3 way out of them with the R2904. All drivers sealed so low volume/size. Perhaps with Hypex Fusion plate amps when they come out. Got to get these ones done and then spend some time with the family first[emoji3]

  • Like 1
Link to comment
Share on other sites

4 hours ago, BioBrian said:

Is the 22W behaving OK up to 900 Hz?

Results, UMIK mic about 30cm from front of 22W 8851. Blue is with no filters, purple is with filters applied to flatten. One full freq sweep, the other is zoomed into relevant section. This is in my bedroom with speaker only about a meter from the bed so far from ideal testing situation. Although it does show that it is not having any problem within the range I need

 

22W.png.033803a34e411dbab86ab2516828805b.png

5ad01fcb62185_22W100-1000.png.e012516dc034941646719fa81fe10d47.png

 

 

Link to comment
Share on other sites

And the D2908/7140. It has a 25uF capacitor attached to protect it, looks like it is impacting the lower frequencies as expected, and well out of the range that I need it for which is good. Could probably get away with a much smaller capacitorD2908.png.e6a6c86ec3a9130831ccbf10ba0b8e08.png

Link to comment
Share on other sites

Yes, 10 uF might be safer, and still show what you need. A parallel coil as well (guessing around 0.47 mH, or bigger to start with) might deter that alarming 380 Hz peak. (But don't ask me, I'm notoriously passive).

 

Curious about that 10 dB drop above 10 kHz, but I take it this isn't your final baffle.

 

With the 22W, I guess the stuff above 2 k will be well out of the picture, with the final filter.

Link to comment
Share on other sites



  • Recently Browsing   0 members

    • No registered users viewing this page.
×
×
  • Create New...
To Top