Shin:
Thanks, really appreciate all the effort to allow people to learn from your experience. I'd be interested in the CD file as well as the LEAP files. I don't currently have LEAP, but I'll see if I can remedy that. Either way if you could take some screenshots of the modelings you've done I'm sure it would be of interest to a lot of people. If you can't already tell I'm toying with the idea of copying your project if it turns out to be a success for you. The cabinet construction seems complex enough that I'll probably have to seek professional help if I decide to go down that road. My equipment and woodworking experience probably isn't up to the task. Best of luck and thanks!
Thanks, really appreciate all the effort to allow people to learn from your experience. I'd be interested in the CD file as well as the LEAP files. I don't currently have LEAP, but I'll see if I can remedy that. Either way if you could take some screenshots of the modelings you've done I'm sure it would be of interest to a lot of people. If you can't already tell I'm toying with the idea of copying your project if it turns out to be a success for you. The cabinet construction seems complex enough that I'll probably have to seek professional help if I decide to go down that road. My equipment and woodworking experience probably isn't up to the task. Best of luck and thanks!
I also have no idea how Ant manages to do all those angle cuts... my table saw never makes then straight enough to fit together perfectly!
You could always do the same design using 'normal' boxes though. I doubt the angles really make that much difference beyond baffle diffraction, do they? That can be done with a big chamfer tool.
You could always do the same design using 'normal' boxes though. I doubt the angles really make that much difference beyond baffle diffraction, do they? That can be done with a big chamfer tool.
Tenson said:
Simon, what you really need is a good crosspull double bevel mitre saw, forget the table saw for this kind of work. I was banging on about these during my work on the Perceives. Put simply; without one I wouldn't have been able to do them. I rate it as more important than my table saw or router put together. A really good one will set you back around £350-500 but I've seen nice ones around £250, avoid the cheaper stuff as you'll stuggle to get the full range of features as well as decent and accurate cuts will be tough no matter how much care you put into setting things up before making a cut. Things to look out for are:
- The depth of cut (important if your doing large slices in laminated MDF to create angles such as those I do on my baffles). If your slicing through 4 layers of 18mm MDF then you'll need a minimum of 80mm.
- The sliding range of the mitre, 30cm+ is best and for most cuts that's more than enough, however if your careful and take your time its possible to double that length to 60cm by simply turning the work around and come at it from the other end, most times though I find 30cm is enough to avoid that and all but a few cuts.
- If your spending good money then absolutely ensure that you get a double bevel type as this is a god send that will save you much headache and time. Basically this will allow you the most flexibility when making cuts and you won't have to compromise or find work arounds that you'd experience on a single bevel type.
- Look for ones that offer 45degree+ in both the angle and the bevel of the cut in both directions.
- Also get yourself a really nice blade for it, these cost around £50 upwards. I immediately noticed a difference when I swapped the freebie for a decent one - smoother and quicker cut action and didn't damage the wood/MDF around where the cut had been done.
One that meets all these requirements would be the Bosch GCM 10SD or even better the GCM 12SD which has a 12" blade instead of 10".
Here's a quick pic of the GCM 10SD which I use:
These have come down a lot in price compared to when I bought ours for nearly £500, best price I could find was here:
http://www.axminster.co.uk/product.asp?pf_id=460297&src=froogle
That also comes with a stand for £380 too. Best money you'll ever spend trust me, it opens up a whole world of new cabinet ideas that can be realised.
Enough with the pimping of Mitre saws though
Another trick to getting complicated angles done is cheating. I do this all the time. What it normally entails is cutting only one piece of work with the angle on it and then sanding the excess of the ajoining panel and finally covering up your sin with a thin 9mm MDF 'skin'. That wasn't a very good explaination, I can see it my mind but its difficult putting it into words. A picture would be better
Undoubtedly the most complicated part of the Tarantism is the mid-treble enclosure which is a real nasty piece of work on face value. If you disect each part, come up with a strategy(usually using the cheats I mentioned above) to tackle it and carefully plan before cutting then you'll have something that looks very good but was imminently approachable. Providing you know what and how to tackle each step then you'll have a clear picture about where your headed. This is essential to success and the final quality IMO.
Motion was interested in the steps required and a cutting list for this design (I'm honoured someone is actually thinking about building something I've designed BTW, so thanks!) but what I'm about to post will also help illustrate the steps, some of the cheats and the general sentiments of the paragraph above this one. You can get complicated shapes that look like what the proffessionals do with their CNC routers it just takes a bit of effort and good tools and careful planning.
Here is a step by step illustration of what is involved during the construction of the mid-treble baffle, this is perhaps the hardest thing to build in the entire speaker so if you can do this then the rest will be equally forthcoming.
An externally hosted image should be here but it was not working when we last tested it.
The detail is very difficult to see in that diagram but I didn't want to ruin the frame size in your web browser byposting a huge image. So here is link to a large image that shows the measurements and more detail:
http://usera.imagecave.com/zeroex_15/midbaffle.jpg
Hope this helps to inspire others into trying something other than a simple box. There are also very real acoustic benefits to creating non parallel walls and irregular shaped baffles, so its far from just eye candy!
Finally Motion: These diagrams aren't finished yet. It would be better if I'd provide a short explaination that goes with each pic in the series, they're also missing details on the angles needed. I'll finish it off though and will likely document the rest of the section of the speaker in much the same way if there's the demand. Might be nice idea to corelate the photo's of the build progress with these diagrams to provide an even better idea of what's going on.
Wow, thanks very much. Really excellent drawings. I think I understand what you mean now be "cheating". You kind of veneer the rough surface with MDF to make it really clean.
Side thought:
Something I've always kind of wondered about doing was building cabinets either out of or with a layer of Corian. Ever think about this?
Side thought:
Something I've always kind of wondered about doing was building cabinets either out of or with a layer of Corian. Ever think about this?
Yep, that's pretty much it. Its purpose here is to hide all the joints from the layers of laminated MDF, I left the bare on the baffle of the Perceives and it gave me a real head ache in the end. Tried all sort to stop them creeping back through after time including wood hardeners and copious amounts of high build primer but could never completely eliminate them although it was greatly reduced.
This time around I want rid of that problem so came up with this which allows for a single visable joint but this follows the line of the baffle shape and will simply be made a detail after recessing it by a mm.
There's all sorts of situations where you can use this to hide joints and sins. The added bonus is that it also adds to the wall rigidity and will help slighty with resonances since 9mm MDF has a different resonance signature to 18mm. You could also use different materials to extend this effect further.
TBH I've never looked at it. MDF is cheap and I'd always assumed such stuff to be expensive. Its tough to get hold of from my local DIY centers, actually I've never seen any outside of kitchen places and the stuff they sell there isn't suited to loudspeaker DIY unfortunately. I'm guessing you've got access to tons more stuff than us over in the states.
I'm almost certain Corian would be superior to MDF though for this application.
I did some modeling of the Seas Excel 8" in a 40l box tuned to 32Hz. It looks like you run out of excursion pretty quickly, 15W will cause this driver to reach it's max excursion at 20Hz. Much lower power will cause it to reach excursion at lower frequencies. If you implement a high pass LR4 filter centered at 28Hz you can dump 120W (max power) into the driver and just BARELY reach excursion at 50Hz, but then it drops down again for all lower frequencies. If you implement the high pass LR4@28Hz this gives you an F6 of about 34Hz.
Did you have any intention of implementing a high pass protection filter for the Seas woofers?
Did you have any intention of implementing a high pass protection filter for the Seas woofers?
m0tion said:
20hz is a not far off an octave below tuning frequency though. Its natural you'll have virtually zero power handling there. Its not something I consider a real problem that would be encountered often enough in daily use. If you really want to protect the drivers then like you say, its possible to use a HP filter or go sealed instead.
Another note is that you've got to consider is that the design is essentially 2x W22's in an 80ltr enclosure tuned to around 30hz.
I played around with a few choices but nothing extravagant as I was limited to a strict 40ltrs for each of the two woofers. This meant no 12 or 15's and non of the really nice pro-drivers. Instead I modelled most of the 8" woofers from the likes of Scan, Seas, Eton and Volt. One that did look interesting and that would also possibly negate the need for subs was the Scan 23W but this driver would likely not work well crossing at about 300hz. Four of these would be pretty good on the bottom end though, and extend deep in a small sealed enclosure without the need to start using any EQ. I guess if you swapped the Scan 12M mids out for the 15W's and cross lower at say 100hz then that might just work. I didn't fancy going in that direction though.
The rest modelled very similarly which is unsurprising given that all the TS specs pretty much blended together.
I'm guessing the Scan 22W woofers will likely have less distortion than the W22 low down but overall I think its the winner when used up higher.
Just thinking out loud; I'd have loved to go with the ATC 9" driver again but four of them would have run to £1000! They worked brilliantly in a sealed 50ltr enclosure. I recently picked up some TL plans for these drivers that looks awesome but sadly the box is pretty large.
Hi, good to see you back in the right spirit
Impressive project, and one thing I know for sure; you will get it done
I hope you make a simple bass box test
Could this bass principle really make wonders as they claim it does (look at bottom of page) - to me it looks like a simple and effective matrix with BR
http://www3.ocn.ne.jp/~hanbei/eng-angular.html
BTW... for what its worth....this little driver moddels quite impressive
http://www.rawacoustics.com/item__RA1658_,805.html
Good luck!
Impressive project, and one thing I know for sure; you will get it done
I hope you make a simple bass box test
Could this bass principle really make wonders as they claim it does (look at bottom of page) - to me it looks like a simple and effective matrix with BR
http://www3.ocn.ne.jp/~hanbei/eng-angular.html
BTW... for what its worth....this little driver moddels quite impressive
http://www.rawacoustics.com/item__RA1658_,805.html
Good luck!
Redesign time. I've been looking at Mark K's Scanspeak 12M midrange tests and I've got to say that they're disappointing, I was hoping to cross at around 200-300hz but even at 400hz these would be struggling. I was also chanced upon a guy from the Netherlands who said he didn't rate them and said they have a particular character that's pleasing but not neutral. I've also heard this from Troels about the 18W sliced paper and I've used that one before but never really thought it was any other than a good driver.
So looking back I decided to stick with a driver that looks interesting and that's the Accuton C90-T6 midrange. This driver looks perfect to 200hz with good extended FR to well over 2 octaves past my intended upper XO point. The only problem is MTM configuration and the larger 6.5" Accutons, I'll gladly take a knock here and no its not the end of the world though but having said that maybe a 3.5way but thats not something I want to consider until really needed.
I wanted to keep the same form factor and had to juggle the design a bit to squeeze in the Accutons. The overall width of the cabinet has clearly increased by a good amount and so has the height. My goal of reasonably small size is now lost and I'm going to have to live with that too. The upshot is that the 8" Seas Excel drivers will be replaced by 12" pro-drivers (yay!) that will provide a big increase in bass performance. I was thinking along the lines of the 12" Beyma LX60 or maybe the excellent B&C 12TBX100 which I loved for its massive dynamics in the 35hz and up range.
Overall the design will benefit in that I'll be looking at atleast 93dB efficiency and more likely 95dB which would really suit 50w class A amps for the mid and treble. The overall dynamic range should be pretty fierce in comparison to the other. Its a more rounded design.
Here's the redone version. Just a bit of FYI: the drawing was completely done by hand, no fancy 3d rendering here and to think I got a G grade in Design and Technology at School, though TBH I was much more interested in other less academic activities at that time.
If size turns out to be a real issue a cut down version could be made whereby the top bass section is omitted. It could always be added later when room allows. Personally I think I'd want the big version though because the foot print is the same and vertical space is less of an issue.
So looking back I decided to stick with a driver that looks interesting and that's the Accuton C90-T6 midrange. This driver looks perfect to 200hz with good extended FR to well over 2 octaves past my intended upper XO point. The only problem is MTM configuration and the larger 6.5" Accutons, I'll gladly take a knock here and no its not the end of the world though but having said that maybe a 3.5way but thats not something I want to consider until really needed.
I wanted to keep the same form factor and had to juggle the design a bit to squeeze in the Accutons. The overall width of the cabinet has clearly increased by a good amount and so has the height. My goal of reasonably small size is now lost and I'm going to have to live with that too. The upshot is that the 8" Seas Excel drivers will be replaced by 12" pro-drivers (yay!) that will provide a big increase in bass performance. I was thinking along the lines of the 12" Beyma LX60 or maybe the excellent B&C 12TBX100 which I loved for its massive dynamics in the 35hz and up range.
Overall the design will benefit in that I'll be looking at atleast 93dB efficiency and more likely 95dB which would really suit 50w class A amps for the mid and treble. The overall dynamic range should be pretty fierce in comparison to the other. Its a more rounded design.
Here's the redone version. Just a bit of FYI: the drawing was completely done by hand, no fancy 3d rendering here and to think I got a G grade in Design and Technology at School, though TBH I was much more interested in other less academic activities at that time.
If size turns out to be a real issue a cut down version could be made whereby the top bass section is omitted. It could always be added later when room allows. Personally I think I'd want the big version though because the foot print is the same and vertical space is less of an issue.
tinitus said:
The data I have is straight from Accuton, I couldn't find the driver on their website so emailed them and they provided a datasheet with detailed measurements. I'm not sure what other measurements your refering to but these are definitely Accuton originated.
I've also been in touch with Tony Gee who is currently using the C90-T6 mids in a project he's working on. He used the word 'amazing' which says a bit considering all the drivers and designs he's been through.
There's also the Evolution Acoustic MM3 commercial speakers that were recently introduced and use the C90-T6's. Mike Lavigne who previously used the Von Schweikert VR9 and Karma Exquisite Midi with diamond tweets said the MM3's were one of the best he's owned.
Hard to believe that the driver measures awful from the research I've done. At this point its hearsay for me because until I get a driver, test it and get it in a working loudspeaker system then its impossible to do anything else.
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.