Hello all, finally finished up the speakers I've been eager to share. They are a three way speaker utilizing all Dayton audio drivers. The woofer is a an SD315, Sig 180 on mid, and the ND25FW for the tweeter. I find these drivers offer excellent performance and allow for easy integration. At the time of writing the total cost in drivers is about $300+ tax. I feel this is an excellent value for a speaker that offers such response and dynamics. I find myself thoroughly enjoying their representation and I have no complaints with them which is pretty rare for me, If I could change one thing, and this may come to be a change I revert, is the tweeter could be wider. I have a few I can try out to see.
The total cost of these is well, I'd rather not know at this point
The are finished in cherry veneer and clear shellac, I put a lot of time and effort into the finish and I'm very happy with the result. One must really take their time if they want to get things perfect. I wasn't able to get that, in the form of poor pattern matching but the veneer was expensive and I was already cutting it short with an 8x10 sheet. I've heard cherry will darken over time so I'm pretty excited for that. I wanted these to be something I could live for for a long time.
They are currently being run active via EQ APO and the results sound great and appear to be so in the data as well. I will eventually have to muster the curage to move them outside and try to get some good measurements on them. I tried a Dayton DSP408 and was sad to find it didn't satisfy my noise requitements, would've had been neat to add even more Dayton stuff to show it's value but I think $160 is just a little too much to ask for such a device in my setting. I'm eager to develop passive filtering for the speakres for ease of use and to being the noise down, and because I see it as a necessary knowledge set for me knock out. The current filters are implimented via onboard motherboard sound, and my computer thinks its surround sound, so movies have no dialogue (open to a fix for that if anyone knows of one, can't video player that will force stereo playback).
Just 2nd order at 300,2500 and a few filters for the mid bass and low end to knock down some room issues.
The total cost of these is well, I'd rather not know at this point
The are finished in cherry veneer and clear shellac, I put a lot of time and effort into the finish and I'm very happy with the result. One must really take their time if they want to get things perfect. I wasn't able to get that, in the form of poor pattern matching but the veneer was expensive and I was already cutting it short with an 8x10 sheet. I've heard cherry will darken over time so I'm pretty excited for that. I wanted these to be something I could live for for a long time.They are currently being run active via EQ APO and the results sound great and appear to be so in the data as well. I will eventually have to muster the curage to move them outside and try to get some good measurements on them. I tried a Dayton DSP408 and was sad to find it didn't satisfy my noise requitements, would've had been neat to add even more Dayton stuff to show it's value but I think $160 is just a little too much to ask for such a device in my setting. I'm eager to develop passive filtering for the speakres for ease of use and to being the noise down, and because I see it as a necessary knowledge set for me knock out. The current filters are implimented via onboard motherboard sound, and my computer thinks its surround sound, so movies have no dialogue (open to a fix for that if anyone knows of one, can't video player that will force stereo playback).
Just 2nd order at 300,2500 and a few filters for the mid bass and low end to knock down some room issues.
All orders of filters and other tweeters were tested, not much of a fan of the results with first order and low fs tweeter. The nd25fw sounds better than any other tweeter I've tried so I'd like to stick with it.
Poor things lasted an hour before I knocked a tweeter off. Was able to hide the damage completely but I need to figure out how to secure these boxes. The mid stays in place, but the tweeters are on ice. Considered rubber feet but would probably get staining from plasticizer migration. Someone suggested museum putty. I have a few types of antislip mats but those too might just melt into the wood eventually.
I'm finding the initial overall look a little plain. I came up with the idea of some wood inlays just near the bottom of the woofer box. Something that catches the eye but doesn't really distract from the drivers.
Attachments
I've used these on oak kitchen cabinets and have not seen any staining.
3M Scotch SJ5201 Series Bumpon Self-Adhesive Bumpers: 0.433 in. x 0.125 in. (Light Brown)
https://www.amazon.com/3M-Scotch-SJ5201-Self-Adhesive-Bumpers/dp/B007IB11SO
3M Scotch SJ5201 Series Bumpon Self-Adhesive Bumpers: 0.433 in. x 0.125 in. (Light Brown)
https://www.amazon.com/3M-Scotch-SJ5201-Self-Adhesive-Bumpers/dp/B007IB11SO
Very cool project. The ND25FW tweeter is indeed a gem, a real bargain... A lot of performance for $25.
Some thoughts on securing the tweeter cabinet:
I think you are wise to avoid any rubber or polymer non-skid material. Most clear finishes are susceptible to staining/softening by prolonged contact with polymers, at least for the first 6 months after cure. Some 2-part finishes (2 part polyurethanes and 2-part catalyzed lacquers) are less sensitive, while some slow cure finishes like tung oil and boiled linseed oil remain sensitive for years.
You could install 2 or 3 short studs in the top of the midrange cabinet, and drill matching holes in the bottom of the tweeter cabinet. The tweeter cabinet could be lifted off, but it would not slide off. Another idea is a couple of small (1/4") neodymium disc magnets.
j.
Some thoughts on securing the tweeter cabinet:
I think you are wise to avoid any rubber or polymer non-skid material. Most clear finishes are susceptible to staining/softening by prolonged contact with polymers, at least for the first 6 months after cure. Some 2-part finishes (2 part polyurethanes and 2-part catalyzed lacquers) are less sensitive, while some slow cure finishes like tung oil and boiled linseed oil remain sensitive for years.
You could install 2 or 3 short studs in the top of the midrange cabinet, and drill matching holes in the bottom of the tweeter cabinet. The tweeter cabinet could be lifted off, but it would not slide off. Another idea is a couple of small (1/4") neodymium disc magnets.
j.
Magnets are a neat idea for sure. I ended up doing the studs or dowels idea. I did buy a few little rubber (or whatever material they were) pads, different brands and such. Put a few on and checked them often, only took a day for them to start to leave marks.
Here are some pictures of how I secure the cabinets. I made four 1" holes with a forstner bit with depth stop. Cut some 1" red oak discs from a dowel and sanded until they were smooth enough to sit tight but also slide out with a pull. Took awhile to sand down 16 little discs. A bit annoying to drill over the finished box but a drill guide with a large DIY'd base made it super easy.
I'm not really smart enough to figure out how to mate two holes perfectly on different surfaces, so I just plopped the circles in the holes, applied glue to the tweeter and placed it where I wanted it.
The end result looks great. The fit is snug enough to avoid any annoying rattling or vibrations and loose enough to taken off if need be. Quite proud of this little work around. The gap seems to have made for a much needed visual separation of the boxes.
My plan is to develop passive filtering based on the active filter tasks. Active certainly has plenty of benefits but I'd really rather just run one cable to the speakers. I haven't really been satisfied with the hardware solutions that have been within my budget and I'd like to bring the noise down to a minimum. The woofer boxes have two nl4 outs near the top to pass signal to the mid and tweet while storing it all the components in the woofer box.
While I hvaen't taken any good off axis data yet, I speculate that the speaker would benefit from a tweeter capable of a lower crossover point. While testing a few tweeters I noticed and improvement in coherency when using tweeters lower than 2500hz. I was wondering it would be worthwhile to try something that can reach into the 1.5khz territory, which I can only think of a few. I may try a Dayton RST, though it lacks a waveguide and overall dispersion looks kinda messy, never seen a speaker response with it that was neutral. Seas DXT is another consideration if I win the lottery. Both would fit my current cutout but it can also be expanded.
edit - Actually I think I just had my tweeter a bit bright, mono listening made it obvious, padding it down helped a ton.
While I hvaen't taken any good off axis data yet, I speculate that the speaker would benefit from a tweeter capable of a lower crossover point. While testing a few tweeters I noticed and improvement in coherency when using tweeters lower than 2500hz. I was wondering it would be worthwhile to try something that can reach into the 1.5khz territory, which I can only think of a few. I may try a Dayton RST, though it lacks a waveguide and overall dispersion looks kinda messy, never seen a speaker response with it that was neutral. Seas DXT is another consideration if I win the lottery. Both would fit my current cutout but it can also be expanded.
edit - Actually I think I just had my tweeter a bit bright, mono listening made it obvious, padding it down helped a ton.
Last edited:
I have a question for you generous folks. I can definitely confirm that the lower I take the mid to tweeter crossover point, the better things sound. The ND25FW however is recommended for 2500hz xover point and has a high FS. What I understand is that it's important to avoid crossing too low and getting close to the FS to avoid a peak in response(?), and to avoid tweeter distortion. I decided to take the ND25FW down to 2000hz and the results seem to be fine so far. At the max volume I will use these at, distortion appears to be fine and still much lower than the rest of the spectrum, and my filter task seems to be sloping properly even though it's much closer to FS. If it sounds good and looks good, I suppose it's good. I wonder if it's worth investing in scan speak Hd2606 for a lower crossover point, looks like Heismann has run that to 1800hz.
Looking at the specs for the ND25FW it seems that it is ideally suited for 3k-up. Below that, distortion starts to rise. But there are other things at work here besides distortion.
It is possible that at the lower crossover, the power response and directivity performance of the system is improved... or perhaps simply changed in a way which works better in your room. This benefit may overcome any increase in distortion that comes from using the tweeter below 3k.
When you switch to a passive crossover, you will find it can be challenging to design a circuit which works well with a low crossover. If you are within an octave of the tweeter Fs, the changing impedance of the tweeter can upset the crossover, and it can be hard to design around this impedance hump. The active crossover you are using now is immune from impedance effects.
I have used both the SB26CDC and SB26STAC as low as 1.8k in passive designs. I found that a 3rd order electrical filter could produce a nice 4th order response.
It is possible that at the lower crossover, the power response and directivity performance of the system is improved... or perhaps simply changed in a way which works better in your room. This benefit may overcome any increase in distortion that comes from using the tweeter below 3k.
When you switch to a passive crossover, you will find it can be challenging to design a circuit which works well with a low crossover. If you are within an octave of the tweeter Fs, the changing impedance of the tweeter can upset the crossover, and it can be hard to design around this impedance hump. The active crossover you are using now is immune from impedance effects.
I have used both the SB26CDC and SB26STAC as low as 1.8k in passive designs. I found that a 3rd order electrical filter could produce a nice 4th order response.
I suspected this would be the case. Can one simply alleviate/compensate the impedance spike with a passive network?
I suppose I won't know why things improved until I get them measured, but I'm taking a much needed break from thinking about these things for the week. Just gonna listen.
As for distortion, this is what I get when measuring at the loudest SPL I'll ever take the speakers. I will be borrowing my partners iphone so I can get an actual idea of the SPL I'm hitting, but it's very loud and honestly not a level I'd ever listen at.
Last edited:
Apologies for frequent posts but I wanted to share as much as I can in case anyone can get inspiration from my ideas.
I was getting some rattling that I somehow didn't hear before. Went and grabbed some felt sheets and cut little circles out to put in the holes, and boy did it make things look fancy and stopped the rattling.
I would agree here, I did start to notice a graininess that was hard to put my finger on but I speculate it was distortion from the tweeter. It cleaned up as I raised the xover point to 3k. I was concerned with that being too high for the sig180 but it looks ok, at least a lot better than the flat tweeters I tried at lower xover.
These are just quick and dirty in room measurements of a d27 at 2k and an nd25fw at 3k. Looks how one would expect with a sound to match. I'm finding I very much don't care for what I'd describe as nebulous at best imaging with the d27, dx25 was similar. This info is probably nothing new for most people.
I was getting some rattling that I somehow didn't hear before. Went and grabbed some felt sheets and cut little circles out to put in the holes, and boy did it make things look fancy and stopped the rattling.
I would agree here, I did start to notice a graininess that was hard to put my finger on but I speculate it was distortion from the tweeter. It cleaned up as I raised the xover point to 3k. I was concerned with that being too high for the sig180 but it looks ok, at least a lot better than the flat tweeters I tried at lower xover.
These are just quick and dirty in room measurements of a d27 at 2k and an nd25fw at 3k. Looks how one would expect with a sound to match. I'm finding I very much don't care for what I'd describe as nebulous at best imaging with the d27, dx25 was similar. This info is probably nothing new for most people.
You could also use a series notch filter across the tweeter, if you don't mind the complexity. Or if your circuit needs a little more help than wolf_teeth's suggestion provides. How the circuit is performing, cost, and user preference come into play in deciding what's appropriate for each case. The ferrofluid in your tweeter limits the impedance rise at resonance, so the peak is smaller than in some others.
This is one I used on a Vifa OX20SC00-04 (just because it was handy - values would need adjustment for your tweeter).
This is one I used on a Vifa OX20SC00-04 (just because it was handy - values would need adjustment for your tweeter).
Last edited:
Well shoot, still getting some rattling at pretty high SPL. I am going to try some rubber washers instead of the felt padding, or combine, or just knock the feet of the bottom and replace with rubber feet. I made need to also consider some sort of dampening or bracing for the woofer top panel. I have plenty of horizontal bracing but kinda forgot about the top. Not sure if it would be better to add bracing or apply some sort of dampening like the woofer cutouts joined with some elastomeric glue?
The SD315 needs more screw holes so I'm going to add a few of those. The 5 it has are simply not securing the woofer well enough.
The SD315 needs more screw holes so I'm going to add a few of those. The 5 it has are simply not securing the woofer well enough.
Last edited: