Possible 3 Way Build - Akai SW-130 Resurrection

I have a schematic for the first crossover. I have parameters for all the drivers, I believe. Not sure how to handle the two woofers in series.

What would be the preferred crossover simulator to run on these circuits? Would I be able to reasonably predict performance on a simulator? It has been a very long time, so this stuff is new to me.

I also have a couple of decent Shure mics and a two channel A/D mic controller with USB. These are my son's and have been stored here. PreSonus Audiobox USB, if that helps. I may be able to take measurements. What would be the recommended software?

I have a couple of laptops, one of which is fairly capable.

Thoughts and opinions?
 
I loaded the Vituixcad and largely figured it out. I have a similar program we use for work that is based on Autodesk. Hydraulic circuits. I created the schematic as supplied and modeled the drivers as well as I could. Dayton had response and impedance files for the woofers. I was able to use the trace tool to create the response for the midrange from its PDF. I don't see a trace tool in Vituixcad for the impedance. Left it flat. I created a trace for the frequency of the tweeter. Its impedance is basically flat.

What I am seeing in the traces I am not following. If this were a motion control system, it looks like machine went unstable and blew itself up.

My head hurts.
 
I don't know how accurately these simulations will reflect the actual results. I have some components on the way from PE that tweak the curves nicely, to the extent that the models are accurate. I am confident in the input data now. Per the earlier comments, as-is these crossovers had excessive SPL in the 150-500 Hz range, then a hole after that. I figured that would sound like a Bose WaveRadio.

I am going to leave one crossover as-is, and modify the other. Side by side listening comparison. I want to verify the computer models. Maybe try to get crude measurements with the mics and USB device I already have. Have to see if I can do that.
 
Did you include the baffle step model and diffraction model in your design if you were using manufacturer data. Typically manufacture data is based on measurement on a very large baffle, as the baffle gets smaller for the woofer, for frequencies below approx 300-500hz the volume from as much as 6dB there this has to be compensated for by padding down the mid/tweeter or using a second woofer. This assumes the speaker will be used away from a wall.
 
Thanks for the input. I will have to look into that model. Clearly, I am using a small baffle, as this is essentially a restored antique and I am stuffing as much as I can into the space available. I did notice somewhere that the manufacturer specs noted a large baffle test. So, the original curves may in real life be compensated. I have to hear that for myself. This design is probably an outlier, an extreme case of small baffle. All the more reason for a careful listen. I'll have the parts for changes anyway.

That was a good point. Thanks.
 
Was not working on this for a while, but got some good work in the past days...

This is the back panel, sanded and with a thin coat of lacquer.

Rear Finished 1.jpg


The original input was 1/4 jack only. The bi-amp plate more that covered the cut out, so that was the choice with little routing. I kept the original nameplate just because it is cool vintage and says Made In Japan.

Front view of the dry fit.

Dry Fit Front.jpg


The black is a primer and dust fingerprints cling to it. Wipes clean with microfiber.

The dry fit from inside. Not a lot of spare room in there. Glad I detailed all that pileup in CAD so there were no unpleasant surprises.

Dry Fit Rear.jpg


Woofers and mids are thru bolt with washer and nut. These nuts are removable. The real installation ill be nylock nuts. I have two platform rails for the two crossover(s) I will be using. They are premade, and I will start out there. I have specs for one of them and have modeled to the best of my ability. The other is a better quality (???) but I cannot find any definitive information on it. The big question is what this tiny baffle will do with the mid-lower frequencies. The model says I will have a 5 to 6 dB hump there. Maybe this mount will attenuate that. No way to know, so I will be trying some basic SPL tests as soon as I get the first one complete.
 
  • Like
Reactions: 1 users
So I fired up Number One last night. I have one laptop dedicated as tone generator / music source through a spare amplifier. I have the other laptop running ProSonus and a decent Shure microphone. All this stuff is working.

Some initial observations:

After playing some test tones from the laptop, I went to music, via YouTube, you know, just to have something to play. Man, the audio stream from YouTube is HORRIBLE. There were so many odd digital artifacts and errors coming through, I honestly thought something was broken. Nope. That is what YouTube actually sends. These speakers are very sensitive and projected that crap clearly. A great sign.

I hooked up my Fiio digital player on line level output. A good digital source with a bunch of clean recordings I know well. The sound was very detailed. Brilliant. I dropped the mid crossover by -3 dB since it has a selector. Between Stevie Nicks and Geddy Lee, I decided it needed it. Both sound better now. Stevie actually sounds good, lol. These are very good drivers. Sensitive. Fast. Clean. Even with one speaker in mono, I can tell the soundstage will be open. The drivers are that clean. I would call the woofers "mid-bass", as that is what they seem to be. I can hear right away that these will require a subwoofer or two. The bass clearly drops off fast. Fortunately, a have a couple subs right nearby.

What I am struggling with right now is that I am basically making judgements based on the recordings I am listening to. Some harsh. Some subdued. What stands out is how the tracks were recorded and mixed. Any track from Bridge Over Troubled Water sounds subdued. Nice, but subdued. That is probably the recording, which I know was from the 80s. I know the Born to Run tracks came from a remaster released 15 years ago. Bruce and the Band jump right out at you. They never did that on vinyl or on the original CDs. Anything Fleetwood Mac sounds about right, but hose are older CDs also. ANYTHING with Mark Knopfler sounds incredible. Of course.

Some tests I always run: Intro to Wish You Were Here. Can you clearly hear the clicking of what must be a cassette recorder playing the first guitar part. Also, who I assume is David Gilmour moving his guitar and shifting around. Breathe in. All there. I usually get those on headphones. The Vandersteens get them, but do not project them so much. Rush 2112 intro. Can the speakers distinguish all those synth sounds into what they must have been intended, or is the result a wall of mush. The reverb. The punch. Again, these sound more like my Sennheisers than speakers. The Vandersteens would pick this action out, but subtly, not kick you in the teeth with them. These are "whoa...".

I am going to run some basic sound level tests later...
 
  • Like
Reactions: 1 user
"Testing" was not very successful. I was, however, able to test my ears. I can hear a tone up to 13K. Better that I would have thought. I was also able to test my room, which has a lot of reflections. I was surprised how much the mic was able to pick up when I was standing too close to the speaker. Big dB jump.

I clearly need a better room, or maybe go outside somehow.
 
I got these finished around Thanksgiving.

I had been running these on a Kenwood KA9100 all along. Sounded great. Adding subs to the Kenwood was going to be tricky, so I integrated with my home theater receiver and its subs. The speakers sounded horrible now! I was shocked what running the digital processing did to the sound quality. Pure Direct sounded correct in level and sound quality. Anything else sucked. So, I brought down my son's Pioneer, which he would ultimately be using, and hooked it up. As far as the home theater, I turned on only the sub amplifiers. Sounded great.

These drivers were able to glaringly show the limitations of the AVR. Whoa. I would say they are sensitive and revealing. No bass to this layout, however. I think they look really cool, too.

Way too much effort. LOL.

1 20231017_212416.jpg
1 20231018_230107.jpg
1 20231023_205226.jpg
1 20231115_210226.jpg
 
  • Like
Reactions: 1 user