I'd be impressed to see some modelling done by Tony's methods. But it never seems to happen. I was a bit grumpy last night at having to think about another unlikely combination of drivers. 
I would have picked this Peerless 830874 polycone:
Peerless 830874 HDS 164 PPB MidWoofer Speaker
It's just easier. Frequency response is easier to filter. Below.
A series network would be a lot of fun to try here with the paper 830875. Below might work. The 10R 10W wirewound on the tweeter is level adjust. What is neat about this filter, is it includes a 5kHz bass notch by its nature.
Bits at Wilmslow and Maplin.
Bits & Pieces
5 Way Tagstrip | Maplin
I'd use ferrite coils and 250V cheapie polypropylene caps. Some fiddling probable. Maybe a 1.5mH coil eventually. Try it. It's all good fun.
I would have picked this Peerless 830874 polycone:
Peerless 830874 HDS 164 PPB MidWoofer Speaker
It's just easier. Frequency response is easier to filter. Below.
A series network would be a lot of fun to try here with the paper 830875. Below might work. The 10R 10W wirewound on the tweeter is level adjust. What is neat about this filter, is it includes a 5kHz bass notch by its nature.
Bits at Wilmslow and Maplin.
Bits & Pieces
5 Way Tagstrip | Maplin
I'd use ferrite coils and 250V cheapie polypropylene caps. Some fiddling probable. Maybe a 1.5mH coil eventually. Try it. It's all good fun.
Attachments
hehe Steve, I actually modeled the filter you put in earlier to see how it went with these FR's The result was below. I'm not sure if the sloping (what looks like a house curve) FR was what you intended, I tend to model for a flat on axis response.
I haven't had any time to do much in the way of modeling of this driver myself. I was going for a very different approach, and the very short time I spent was ok on the tweeter side but the mid was not going very well.
here is the thread where I first tried this out http://www.diyaudio.com/forums/multi-way/229040-win-isd-pro-multiway-design-4.html#post3408537 the end result was pretty good according to Steve. I actually did the modeling for my Synergy Active crossover using the response blender as well http://www.diyaudio.com/forums/analog-line-level/164886-synergy-active-crossover-6.html#post4172537
You just haven't seen it Steve
There were a few reasons I didn't post anything here...
1. I was trying to encourage Tristan to do it, with just some pointers to get him started.
2. I've been working all day!
3. It's a damn tricky driver combo to work with! (and I'm unsure of what to use for the z offset). Note that I decided on 50mm for z offset, which I think is probably reasonable, but 10 mm either way makes quite a difference to the modeled response..
edit: and it isn't my method, it's based on the tutorial by Dave Dal Farra http://audio.claub.net/Simple%20Loudspeaker%20Design.pdf
Tony.
I haven't had any time to do much in the way of modeling of this driver myself. I was going for a very different approach, and the very short time I spent was ok on the tweeter side but the mid was not going very well.
here is the thread where I first tried this out http://www.diyaudio.com/forums/multi-way/229040-win-isd-pro-multiway-design-4.html#post3408537 the end result was pretty good according to Steve. I actually did the modeling for my Synergy Active crossover using the response blender as well http://www.diyaudio.com/forums/analog-line-level/164886-synergy-active-crossover-6.html#post4172537
You just haven't seen it Steve
There were a few reasons I didn't post anything here...
1. I was trying to encourage Tristan to do it, with just some pointers to get him started.
2. I've been working all day!
3. It's a damn tricky driver combo to work with! (and I'm unsure of what to use for the z offset). Note that I decided on 50mm for z offset, which I think is probably reasonable, but 10 mm either way makes quite a difference to the modeled response..
edit: and it isn't my method, it's based on the tutorial by Dave Dal Farra http://audio.claub.net/Simple%20Loudspeaker%20Design.pdf
Tony.
Attachments
One of these days (maybe when I retire, so no time soon) I will pick some drivers, do all the modeling using these methods, build it and then do actual measurements to see how close to reality the sims were.
I know from my sims based on actual measurements that the sim part works very well. I also think from my recent experiments with the active crossover that the response blender works very well, both in blending and in baffle step modeling.
One thing I'm pretty certain about is that it will result in a much better crossover than using text book formula's and impedance flattening circuitry.
Tony.
I know from my sims based on actual measurements that the sim part works very well. I also think from my recent experiments with the active crossover that the response blender works very well, both in blending and in baffle step modeling.
One thing I'm pretty certain about is that it will result in a much better crossover than using text book formula's and impedance flattening circuitry.
Tony.
I really didn't look too closely at levels here, Tony. Like you I have a hard working day. On examination, the D2905 tweeter seems to be about 90dB. That's OK.
As we agree, this is not the loveliest of driver combinations. That series filter seems to do the right things. Generally the simple ones work best with polycones, but this is DIY and there are no right answers and some fiddling required. You just do the best you can with it.
It's slightly better than the Acoustic Reality circuits IMO. And is easily tweaked. I have no idea if it will sound good. But it's a point of departure. Here's some modelling with similar sorts of drivers and the above series circuit. Seen worse, phase works well. And no one else has a better idea.
As we agree, this is not the loveliest of driver combinations. That series filter seems to do the right things. Generally the simple ones work best with polycones, but this is DIY and there are no right answers and some fiddling required. You just do the best you can with it.
It's slightly better than the Acoustic Reality circuits IMO. And is easily tweaked. I have no idea if it will sound good. But it's a point of departure. Here's some modelling with similar sorts of drivers and the above series circuit. Seen worse, phase works well. And no one else has a better idea.
Attachments
Tony, many thanks for all the date. That should help no end I've finally managed to get PCD running in excel, although the macros don't appear to be working. Could this be a compatability issue? I've been doing a lot more studying the last couple of nights and feel that I'm slowly getting my head around things.
Steve, thanks for the suggestions also. A series filter could indeed be fun to try. Can't hurt anyway
I've finally managed to get PCD running in excel, although the macros don't appear to be working. Could this be a compatability issue? I've been doing a lot more studying the last couple of nights and feel that I'm slowly getting my head around things.Steve, thanks for the suggestions also. A series filter could indeed be fun to try. Can't hurt anyway
Hmmm I wonder if it uses any windows dll files... I seem to recall it should pop up a message saying do you want to enable macros, and maybe also another one, which you need to answer yes to. Typical MS security stuff. However I just did a search and this post from Jeff Bagby (the writer) doesn't sound promising Passive Crossover Designer on a Macintosh
Sorry I have been leading you down a path of disapointment!! I'll see if I can find out what some mac users use..
Tony.
Sorry I have been leading you down a path of disapointment!! I'll see if I can find out what some mac users use..
Tony.
It's starting to sound like you need to go down the path of running a windows VM... This site has some info Download Windows XP For Free and Legally, Straight From Microsoft I'm not sure how legit it is though because the download link for the VM is not microsoft.
If you get an XP VM running on virtualbox or vmware player then you can more than likely run something like xsim http://www.diyaudio.com/forums/multi-way/259865-xsim-free-crossover-designer.html Boxsim may be an option as well (I've never played with it, but I think it is what Steve uses).
Sorry to have got you enthused only to run into a brick wall. I thought that excell for mac would work just as well as excell for windows, but apparently it doesn't support visual basic (which PCD uses).
Tony.
If you get an XP VM running on virtualbox or vmware player then you can more than likely run something like xsim http://www.diyaudio.com/forums/multi-way/259865-xsim-free-crossover-designer.html Boxsim may be an option as well (I've never played with it, but I think it is what Steve uses).
Sorry to have got you enthused only to run into a brick wall. I thought that excell for mac would work just as well as excell for windows, but apparently it doesn't support visual basic (which PCD uses).
Tony.
I have an old copy of XP plus license kicking about, so will try that. I've also just come across the MiniDSP kit, which has got me thinking even more!. I have a spare power amp too......
I've got enough bits to try Steve's series suggestion, so will no doubt bash that together over the next few days.
I've got enough bits to try Steve's series suggestion, so will no doubt bash that together over the next few days.
Even better! Virtual box should do the trick and is free. I use vmware player myself but I have tried virtual box as well and it works fine (I was using on a windows XP machine to run windows 7 64bit) which was a pretty impressive feat. Work laptop was XP but I wanted to run some windows 7 stuff).
Tony.
Tony.
Tristan, Good news. I've dug up a reflex 830875 design for you!
Veeper TM Monitor
Louis Coraggio was a top guy, IMO. His whole website is interesting. Once the bass is about right, it's not hard to tie in a tweeter to match. 1.5mH and 10uF looks OK. It's not far removed from the series filter on bass response, as it goes.
Veeper TM Monitor
Louis Coraggio was a top guy, IMO. His whole website is interesting. Once the bass is about right, it's not hard to tie in a tweeter to match. 1.5mH and 10uF looks OK. It's not far removed from the series filter on bass response, as it goes.
Last edited:
Ah, thanks Steve, will have a good look at that
Even after spending many hours over the last two days, setting up a virtual PC, then getting it to bridge network connections, I now realise that I need a copy of excel for said VPC! My copy is for mac, which obviously doesn't work... You'd think there'd be tools for mac available, but try as I might I've drawn a complete blank! It's now also dawning on me that I should've gone for the 830874 that AVi used, rather than thinking "more expensive, must = better" doh!
Even after spending many hours over the last two days, setting up a virtual PC, then getting it to bridge network connections, I now realise that I need a copy of excel for said VPC! My copy is for mac, which obviously doesn't work... You'd think there'd be tools for mac available, but try as I might I've drawn a complete blank! It's now also dawning on me that I should've gone for the 830874 that AVi used, rather than thinking "more expensive, must = better" doh!
Sorry I've set you on this path of frustration Tristan!! That veeper link that Steve posted, the woofer xover will be fine (it is approx 2.8Khz 4th order acoustic response) and will take care of the baffle step, but the tweeter is way off)
I was doing something similar with the woofer (only a bit lower 2.5Khz 4th order bessel) I'll have a bit more of a play and post something.
Tony.
I was doing something similar with the woofer (only a bit lower 2.5Khz 4th order bessel) I'll have a bit more of a play and post something.
Tony.
Hi Tristan, I thought I was getting somewhere decent, and then Speaker Workshop crashed on me and I hadn't saved the current iteration. Have been a bit busy and not been able to get back in the zone.
I think that this crossover of yours is cursed I'll try and get something in the next day or two.
Tony.
I think that this crossover of yours is cursed
I'll try and get something in the next day or two. Tony.
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.