DIY speaker design.. Aluminum coned monitors

So I got one of these together. I still need to do some woodworking and finish the other box.

So far, it sounds a bit bright. I'm not sure if it's bass light, or treble heavy. I placed a series 5R resistor on the tweeter to pad it down a little and the speaker sounds more in line with what I expected. It still seems a little shouty around 600-700hz.

I also think the port might be too restrictive. I removed the port itself and the bottom end came up in level pretty noticeably and made it much more full. I'm not sure if this is just preference, or if there is something wrong with the box calculations.

I also think that since the woofer is a bit on the small side, I might have to overcompensate the low end a bit.

So I tweaked the crossover some, to reduce a little in the 2K range and tilt a little more in favor of the 15K range.

I like it a lot better. The smallish woofer is still light on bass, so I tried a few things. First I completely removed the port, and the low end bloomed heavily. For listening to CDs or something, this would be OK as it doesn't sound too bad, but it definitely makes the low end much heavier. Next, I added 3R in series with the tweeter to bring it down a little in level. This seems to sound much more balanced but almost a little too light on the high end. I might try 2R instead and see how the balance is.

Now, I find the woofer sound much tighter than a paper cone. it's also a tiny bit tubby sounding, even out of the box. I think it's the lack of cone flex that would normally mask such tone.

Anyway, now I hear the slight bump at 3K which is inherent in the tweeter response it seems. it's not annoying like the 2K bump was, but it's definitely noticable. it might actually be a good thing because it really brings out the definition on the guitars and vocals without tearing your face off like an NS-10 does.

I think I'm almost done with these after a few more tweaks.

So I messed with these again over the long weekend. I'm kinda on the fence about calling these close to the most awesome monitor ever..

They still sound like they have a slight push in the midrange, probably due to having such a low crossover and allowing the midrange to be untouched.

At first I thought they were still a little midrange heavy compared to my other monitors, but I've found that they translate very well once you listen without judgement. Every little midrange detail comes through perfectly, which can be jarring when you aren't used to it.

I really want someone else to hear these things!

I also think I might build another set of boxes simply to try out the passive radiator. Some folks love them, some hate them. I've only heard really cheap speakers with passive radiators, so I'm not sure what a good one sounds like.

I'm currently looking for info on properly sizing passive radiators for box response.

Well, considering it took me about a month to plan, design and build a single set, I'd say that my "hourly" rates would certainly push these into the multi-thousand dollar range if I were building them specially for someone.

That being said, these aren't nearly as physically nice as something like Amphion, which has a much nicer finish, etc.  

The internal parts could be roughly similar, or they could have used much higher dollar components as mine.  The jury is still out whether going with super-high dollar components actually adds much to the outcome.  As always, there is diminishing returns on price/performance.

These cost me roughly 400$ in materials alone, granted I used well-liked but generic speaker drivers, and medium-quality components for the crossover.  If this was a production level design, unit costs for the parts would decrease, but the overhead for labor, stocking, and finishing would rise higher than what you save by buying bulk.

I want to build these, too!
I'm not in a hurry, so will wait for the passive radiator experiment, Svart. Also interested in if you are going to pursue the waveguide, since I am a believer in time alignment.
Are you willing to publish the final crossover design and/or do a run of PCBs? Same question for the cabinet design, I think, although I can butcher my own wood.

So here's the expected response of the new crossover. I still found the top end to be a slight bit "hot", so I added a 2R series resistor to the tweeter to knock it down a couple dB.




Here's the old one for reference:

Svart and Tony,
Did either of you finish your respective speakers yet? I'm dying to know how they turned out.

I finished mine.  I like them a lot.  They sound a bit different than most 2 way monitors, with a very "there" midrange.  It's hard to explain though, they don't sound mid-forward, but yet the mids are very clear, much more than other monitors without midrange drivers.  


I attribute this to the very low crossover point and the aluminum cones, since they are very low distortion.

These also have a heavier low end and sloping frequency response to help them seem like they have more low end.  The low frequency drivers don't have a lot of low end extension, and I needed to tune the box by hand via the port length in order to get more OOOMPH out of the speakers.

Overall they are pretty nice, and it didn't take more than a few minutes to get mixes up and translating well. 

One thing though, with the much more detailed midrange, problem frequencies pop out really well, like they do with NS-10's and auratones.

I settled on 1.5" port length.  I'd say the room is a consideration in any tuning.  I know my room, and I have my main monitor system set up with a sub that gives me good translation to other systems, so I tried to match the lower extension as much as I could with these speakers.  

I would say that absolute flatness wasn't a goal, but a monitor that translates perfectly was the goal.  So far i think i got really close after the second tuning.

So we shot these out against Johnkenn 's Amphions while ericn and I listened.  Actually I'm pretty surprised that they held up rather nicely.  A little different throughout, but nonetheless performed well.

A few things I noticed, the Amphions had less extended highs than mine.  The aluminum tweeters have a sharp point in the very high end, Amphion must have removed that.  I softened mine, but didn't remove it completely.

Also, mine still have some mild midrange bump going on.  Not quite NS-10 like, but more than the amphions.

Mine definitely have more low end than the 18's.

I'm pretty happy except for the midrange.  it's useful in the same way NS-10's are, things that have any midrange problems get noticed really quick, although I think I want to "fix" the bump just a little more.

I think Eric and I agreed that the Amphion waveguides seem to smooth out a lot of issues that folks generally have in small rooms, which might be part of the reason why folks love them so much.