• WANTED: Happy members who like to discuss audio and other topics related to our interest. Desire to learn and share knowledge of science required as is 20 years of participation in forums (not all true). Come here to have fun, be ready to be teased and not take online life too seriously. We now measure and review equipment for free! Click here for details.

SB Acoustics SB13PFCR25-4 COAX 5" Paper Cone Coaxial

headshake

Active Member
Joined
Apr 20, 2019
Messages
297
Likes
234
I am looking into making a low-cost coaxial with this driver. My aim is $200-250USD per speaker.

Google found this post:

The measurements look like shit. After a little VirtuixCAD tinkering, I could see most of the mess was the person's baffle.


Top lines are a baffle sim made with manufacturer data in vcad. The bottom two lines are traced from the german guy's real measurements.

german_real_vs_sim.png

the orange line is a 44mm edge radius baffle - it improves things a bit!

mini Diffraction.png

A fuzzy look at the directivity with a 44mm edge radius. I had to use the manufacturer's 20 degree to generate all the off-axis data. XO is around 2khz. Again this is fake and fuzzy but looks nice!

mini Directivity (hor).png

I guesstimate that the baffle will cost $45-65USD to produce using laser-cut sheets.

PE 0.23 cu.ft. mini bookshelf $26.98

44mm edge: $50 (if you have a menard's store you can get it delivered there for free)

Sb coax $66.20

I have not decided if I am going to go cheap active or passive:

Dayton KAB-250 $45.00

or

Passive XO parts $150 - $50 ... I priced this on PE and madisound and came up with a range
The XO parts are here: https://www.madisoundspeakerstore.com/pdf/SBAcoustic_PFC_Coaxial_Crossover_Designs.pdf
I have not simulated what the SBA XO is doing.

I am still debating if I want to make a 3way, but I think the same box and baffle can be used to make a bass box to go underneath.

That's all I have for now. I have not purchased anything yet but wanted a ballpark look at the costs and what it could do.
 

bigjacko

Addicted to Fun and Learning
Joined
Sep 18, 2019
Messages
639
Likes
293
44mm edge: $50 (if you have a menard's store you can get it delivered there for free)
Looking forward to hear more. :) How are you going to join the polyurethen, baffle and box? Where did you get the laser cut baffle price?
 
OP
H

headshake

Active Member
Joined
Apr 20, 2019
Messages
297
Likes
234
Looking forward to hear more. :) How are you going to join the polyurethen, baffle and box? Where did you get the laser cut baffle price?
Hey!

Baffle to box will use titebond II wood glue. The poly edge to the box will use "loctite power grab". I used power grab to glue the mdf to aluminum for my plate amp boxes and I am a believer.

For the laser cutting service for the baffle I'll use ponoko.com. I did a range of prices because I don't know what materials I want to use yet. My first speaker used mdf + bamboo + mdf for the baffle along with some other stuff. Either way, it will take 4-5 layers to make the baffle. Or if I don't want it flush mounted it can be just 3 layers. The mdf is like $8USD a layer for the baffle if I order a few. Bamboo is $18USD a layer.

The quote was for the simple shape attached as "mini.txt". Rename to "mini.svg" and that can be used for a quote. The parts size should be 177.8mm X 165.1mm. It is not a real part. The baffle will be slightly larger. The point of this part is to replace the bottom section of the PE box. This will take (3) 6.5mm layers.

One thing I hope to try with this baffle is having guide holes to make things easier to line up when gluing. The holes will alternate positions every 2 layers.

I did order the PE mini bookshelf box and the driver yesterday. Next up is to measure it all with the calipers to confirm the sizes for the baffle design and try it out with a foam baffle.
300-7062-parts-express-technical-diagram.jpg
 

Attachments

  • mini.txt
    701 bytes · Views: 18
Last edited:

voodooless

Major Contributor
Forum Donor
Joined
Jun 16, 2020
Messages
2,894
Likes
4,420
Location
Netherlands
Great little project :) Interesting to see what kind of response you get with a radius on the baffle.

Instead of MDF, you might opt for plywood/multiplex. It's generally a better material to use and will be closer to the bamboo in looks. I have no idea about the pricing though ;)
 
OP
H

headshake

Active Member
Joined
Apr 20, 2019
Messages
297
Likes
234
Great little project :) Interesting to see what kind of response you get with a radius on the baffle.
Hi! Thanks! I hope to test out the radius on two sides (Grimm style), vs 3, 4 and see what happens. My guess is anything but 4 will be a mess. I wonder what can be heard.

Instead of MDF, you might opt for plywood/multiplex.
What is multiplex? Here is what is on the menu, How would you make your sandwich? We need (3) 6.5mm layers for the coaxial to screw into. Another 6-7mm in layers to make it flush.


I do have data that backs up the superiority of bamboo to ply. I have never compared it with a speaker though so I can't justify that it is even needed.

babmboo_vs_ply.jpg

The sheets ponoko uses are made from three bamboo layers with the middle crossing the top and bottom. With a knock test on bamboo vs mdf I can hear less highs in the bamboo.
 
Last edited:

voodooless

Major Contributor
Forum Donor
Joined
Jun 16, 2020
Messages
2,894
Likes
4,420
Location
Netherlands
What is multiplex? Here is what is on the menu, How would you make your sandwich? We need (3) 6.5mm layers for the baffle to attach to. Another 6-7mm in layers to make it flush.
Multiplex is what we call plywood ;)
The sheets ponoko uses are made from three bamboo layers with the middle crossing the top and bottom. With a knock test on bamboo vs mdf I can hear less highs in the bamboo.
That is precisely one of the reasons not to use MDF. If you're cutting it yourself: it's also not very healthy for you. Plywood is much better in that regard as well. I guess bamboo plywood will also be fine. If normal plywood is about the same cost as MDF, I'd go for that.
 

D!sco

Active Member
Joined
Jun 16, 2021
Messages
127
Likes
57
It's good to see someone taking a shot at these drivers. I'm really interested in the results. SB Acoustics has some incredible woofers, it would be nice if they could produce a high quality small coaxial.
 
OP
H

headshake

Active Member
Joined
Apr 20, 2019
Messages
297
Likes
234
It's good to see someone taking a shot at these drivers. I'm really interested in the results. SB Acoustics has some incredible woofers, it would be nice if they could produce a high quality small coaxial.
I see you are eyeballing coaxials too!

This is made for a budget for sure. It has a plastic frame. It is not the fabled satori coaxial that does not exist. But, it does have a paper cone and from what I can tell a silk dome tweeter. The specs for the tweeter are missing on 1/2 the places online. On solen, they call it a textile dome. This could explain why it drops like a rock past 10khz. My ears only work just past there so I am ok with that.
 
Last edited:
OP
H

headshake

Active Member
Joined
Apr 20, 2019
Messages
297
Likes
234
foam_baff.jpg

Madisound and PE both got things out the door fast!

I mounted the driver a different way as planned just to give it an early listen. First impression: They are fun! Not super detailed but really nice tone and decay. I think calling them forgiving is a better way to describe the sound.

Now onto trying to make a smaller foam baffle and trying the rounding.

The speaker is held together with tape and the drive popped out 3/4 screws while listening! The screws were really small and now that I am using larger ones it has not happened again. It was crazy to hear it pop out and open up like a can of spinach! I could not push the drive back until the song with deep bass was over. The speaker as-is needs a sub though. I did get a 1" port to mess with but don't expect much.
 
Last edited:
OP
H

headshake

Active Member
Joined
Apr 20, 2019
Messages
297
Likes
234
ok, foam baffle v2. I do not want to commit to cuts for the edge until it is all glued up so I used some scrap and fudged things a little for now. There is a sheet of melamine behind the speaker to help show it.
rough_edge_foam.jpg

We have real measurements! It was late and I did not calibrate the SPL so the levels are off. My concern was what was happening with the XO and to get setup for the edge radius test. A LR8 at 2000hz is in there now and I only had to do a little to level match. There are no other filters on the tweeter. The measurement is on-axis about 2ft and with a 2ms gate.

Next up, I will calibrate things and start removing the edge to see what happens.

The same measurement with different smoothing and one uses a FDW.

800hz_1500hz_lr8_at2k.jpg800hz_1500hz_lr8_at2k_FDW_PSY.jpg800hz_1500hz_lr8_at2k_PSY.jpg
 
OP
H

headshake

Active Member
Joined
Apr 20, 2019
Messages
297
Likes
234
Some more measurements... I still have not done the test where I remove the 44mm edge. There were some things in the data I wanted to understand first.

Like, this little hump and dip around 1,2khz:
fb_44mmALL__no_melamine_waterfall.jpg

The inside of the box had no stuffing/melamine in it for this measurement. I wanted to see if there would be a mess and I found it. What is cool is I sorta expected this here since this coincides with some of the box's dimensions.

fb_44mmALL__2inAir_2inMelamine_waterfall.jpg

This is what 2" of melamine 2" from the back of the box can do. I immediately heard the loss of bass putting it in there, It was not a good trade for the better on-axis. This makes me wonder about a tuned plate just for the problem freqs. and skipping the melamine. I will try a few other ways to line it too.

The last set is about 1m (I can't find my laser distance reader ATM..there is also a typo in the name, both are w/ melamine) and I calibrated the SPL to my meter. I think the phase looks good? I have no delay on the drivers. It has been really cool to play with a device that can play 100hz-ish-10,000hz with no delay and no vertical issues. It has been a nice reference point for my ears. ;P

fb_44mmALL_76db_no_melamine_1m_300msGate.jpgfb_44mmALL__2inAir_2inMelamine_dist.jpg

I had a set of filters that reached 100hz but it had lots of distortion.
 
Last edited:

alex-z

Senior Member
Joined
Feb 19, 2021
Messages
310
Likes
463
Location
Canada
Tweeter resonant frequency is 1300Hz, and given there is usually 10% tolerance, that 1200Hz hump in the spectral decay makes sense.

I would recommend trying some 2" mineral wool cut in half and staggered at 33 and 66% of the box depth. Anecedotally, I find it damps most internal standing waves without harming bass response. You may need to make the cabinet slightly larger to compensate, but usually only by 10-15%.
 
Last edited:

bigjacko

Addicted to Fun and Learning
Joined
Sep 18, 2019
Messages
639
Likes
293
Maybe you can try some acoustic fills too, those can even add effective volume to the box. How dense is the melamine and why did you choose it?
 
OP
H

headshake

Active Member
Joined
Apr 20, 2019
Messages
297
Likes
234
Tweeter resonant frequency is 1300Hz, and given there is usually 10% tolerance, that 1200Hz hump in the spectral decay makes sense.

I would recommend trying some 2" mineral wool cut in half and staggered at 33 and 66% of the box depth. Anecedotally, I find it damps most internal standing waves without harming bass response. You may need to make the cabinet slightly larger to compensate, but usually only by 10-15%.
Good point about tolerance. That is something I have not thought about making XO's. I did think about the spec FS of the tweeter and tried to put it -24db down in the XO. This is one reason I used an LR8 XO.

The question, if it is the FS, why would a sealed tweeter stop making noise 1m in front of it with a block of foam 4" behind it? This is only my 2nd speaker so there is that.

The first thing that made me think it is due to the dimensions is when I typed the dip freq. into a WL calc.. 1115hz = 12.005225804173582"

It could just be a coincidence, but that number seemed too perfect. The inside of the box is 10", the baffle is 1", and the distance to the cone is a little hard to guess (but it is prob 1" :p). Maybe this is the cone's surround messing with the back of the box? The cone is only paper so it is not stopping much.

I made a couple more images. The blue is w/ the melamine and is the smallest rise time REW can do. The remaining noise around 1200hz is all made with the signal. It is 1:1 which makes me think this is all happening at 1/2 a WL. The range of mess exactly matches the rubber surround at 1/2 the WL (my eyes see an inverted rubber surround shape in the decay but this could be newbie vision). The speaker is not flush mounted so maybe some of this will clean up in the real build.

fb_44mmALL__2inAir_2inMelamine_waterfall2.jpg
Another look at w/o melamine. The SPL's are not calibrated.
fb_44mmALL__no_melamine_waterfall2.jpg


Maybe you can try some acoustic fills too, those can even add effective volume to the box. How dense is the melamine and why did you choose it?
I have 1" & 2" melamine, a few .lbs of Acousta-Stuf, and some audio grade felt to try out.

I picked the melamine first because it crushes problems at 1200hz. I might have used too much though. Even the manufacturer graph shows the loss of bass.

mel_sw04.gif


This is the brand I am using: https://www.acousticalsurfaces.com/flat_faced_melamine_foam.html
 
Last edited:

bigjacko

Addicted to Fun and Learning
Joined
Sep 18, 2019
Messages
639
Likes
293
The first thing that made me think it is due to the dimensions is when I typed the dip freq. into a WL calc.. 1115hz = 12.005225804173582"

It could just be a coincidence, but that number seemed too perfect. The inside of the box is 10", the baffle is 1", and the distance to the cone is a little hard to guess (but it is prob 1" :p). Maybe this is the cone's surround messing with the back of the box? The cone is only paper so it is not stopping much.

I made a couple more images. The blue is w/ the melamine and is the smallest rise time REW can do. The remaining noise around 1200hz is all made with the signal. It is 1:1 which makes me think this is all happening at 1/2 a WL. The range of mess exactly matches the rubber surround at 1/2 the WL (my eyes see an inverted rubber surround shape in the decay but this could be newbie vision). The speaker is not flush mounted so maybe some of this will clean up in the real build.

fb_44mmALL__2inAir_2inMelamine_waterfall2.jpg
Another look at w/o melamine. The SPL's are not calibrated.
fb_44mmALL__no_melamine_waterfall2.jpg
Not sure what dimension and half wavelength and surround you are talking, maybe it is my problem because I am not very experienced with the graph.

I have 1" & 2" melamine, a few .lbs of Acousta-Stuf, and some audio grade felt to try out.

I picked the melamine first because it crushes problems at 1200hz. I might have used too much though. Even the manufacturer graph shows the loss of bass.

mel_sw04.gif
If when you don't damp the speaker and speaker have more bass, I think the bass coming from inside the box is not good. I think generally we don't want rear wave to come out.
 
OP
H

headshake

Active Member
Joined
Apr 20, 2019
Messages
297
Likes
234
I went through this effort because without an edge, a small baffle does not sound right to my ears. Small-speaker-itis. There can be a sibilance in some songs and a hi-hat can sound wrong.

The TLDR is that I did not know a negative group delay was a bad thing. I also did not know that a flat edge = more on-axis power and less off-axis power right where ears can locate things accurately and are sensitive to decay.

The listening test contained four different signals: the unit impulse, the pink impulse, a castanet recording, and a synthetic hi-hat cymbal sound. The two former signals are known to be the most critical for group-delay audibility, whereas the latter two resemble real-life musical signals and thus help to generalize the results better. Group-delay audibility was tested at the frequencies 500 Hz, 1 kHz, 2 kHz, 3 kHz, and 4 kHz. The results indicate that the audibility thresholds for local group-delay variation are less than ±1 ms for the most critical signals, and approximately 1.5 ms to 4.5 ms for a local positive group-delay peak and between −1.0 ms and −2.3 ms for a local negative group-delay peak for real-life signals.

liski11-3087969-large.gif

Audibility of Group-Delay Equalization​


These are the most recent measurements. I need to fix things below 700hz a bit more. The polar was done by shooting a laser distance meter at the tweeter and moving the mic around. It is 0-about 80 degrees. I could no longer shoot the laser at the tweeter at the angle I was at but it was not 90 degrees. I think it averages up nice.
44mmEdge_4sides_1meter_rough_polar.jpg44mmEdge_4sides_1meter_rough_polar3_30msGate.jpg44mmEdge_4sides_1meter_rough_polar4_4msGate.jpg44mmEdge_4sides_1meter_rough_polar4_30msGate.jpg


The XO did like a little delay. The top two lines are the XO w/ and w/O the delay.
lr8_invert_delays.jpg

The next set is all the w/ and w/o the edge radius. There is no delay and eq for these. Just the XO.

No edge = a flat baffle no edge radius
2 sides = edge radius on the left and right sides.
3 sides = edge radius on the left, right, and top.
4 = ...

44mmEdge_4_to_0_sides_15degrees_1meter_1.jpg44mmEdge_4_to_0_sides_15degrees_1meter_2.jpg44mmEdge_4_to_0_sides_70degrees_1mete_2msGate_5.jpg44mmEdge_4_to_0_sides_70degrees_1mete_2msGate_6.jpg44mmEdge_15degrees_1meter_decay_0sides.jpg44mmEdge_15degrees_1meter_decay_2sides.jpg44mmEdge_15degrees_1meter_decay_3sides.jpg44mmEdge_15degrees_1meter_decay_4sides.jpg44mmEdge_15degrees_1meter_GD_0sides.jpg44mmEdge_15degrees_1meter_GD_2sides.jpg44mmEdge_15degrees_1meter_GD_3sides.jpg44mmEdge_15degrees_1meter_GD_4sides.jpg44mmEdge_15degrees_1meter_rt60_0sides.jpg44mmEdge_15degrees_1meter_rt60_2sides.jpg44mmEdge_15degrees_1meter_rt60_3sides.jpg44mmEdge_15degrees_1meter_rt60_4sides.jpg44mmEdge_waterfall_0sides_2.jpg44mmEdge_waterfall_2sides_2.jpg44mmEdge_waterfall_3sides_2.jpg44mmEdge_waterfall_4sides_2.jpg

I think the bass coming from inside the box is not good.
Bad things should show up in the measurements. I'll post along the way.
 

bigjacko

Addicted to Fun and Learning
Joined
Sep 18, 2019
Messages
639
Likes
293
You said small baffle without round off does not sound right to you. The frequency response does show some baffle step of the baffle with less round off.

You alo provided group delay graphs. The more round off the more minimum group delay dip into negative. Although I don't know whether we should watch minimum group delay or excess group delay here. From the study it shows group delay is audible, can you hear any bad impact on the group delay?
 
OP
H

headshake

Active Member
Joined
Apr 20, 2019
Messages
297
Likes
234
From the study it shows group delay is audible, can you hear any bad impact on the group delay?
This is my current guess. At first I thought is was impossible for me to hear such small timescales and then I found that paper. I think a gradual shift in GD is ok, just not a big blip. For me, the shift in the EDT shows what I think bothers me. The EDT is only -5db in decay. I am listening at 3.2 meters so lots of the room is in the mix.

Spotify has a mono "The Girl From Ipanema" by stan Getz, ... and "What's going on" by Marvin Gaye. These are some of the songs I use to hear issues. If the decay from the T's and S's don't sound buttery, my ears focus on the T's and S's.

Some gotcha's with how I compared:
  • The baffle shape I am using w/o an edge is almost square which is as bad as you can get
  • The eq was meant for the 44mm edge on all sides configuration
  • I only have one speaker and cannot do fast swaps (but I can hear a diff the second I pull off the edges)
  • The LR8 XO might be causing some of the issues- need to compare XO's more
  • newbie thinking

Maybe I can try rephrase and see if I can make the 44mm edge radius baffle sound the same by adding some delay in the same spot?
 
Last edited:
OP
H

headshake

Active Member
Joined
Apr 20, 2019
Messages
297
Likes
234
I went back to the sim and can see the LR8 is adding some group delay. A second XO was made with a priority on the group delay.

Here are real measurements of each XO's group delay:
2ndO_GD.jpglr8GD.jpg

The new XO has a really clean EDT:
2ndO_EDT.jpg

I'm gonna give my ears a few days of listening to it before I mess around again.


edit:

2 more images. I think what happens at 60ms is what bothers me. There is like +3db @ 2khz @ 60ms with the LR8 vs a couple of second-order filters. Deceptively both are flat, both have nice-looking phase- my ears only like one.

2ndO_decay.jpgLR8_decay.jpg
 
Last edited:
OP
H

headshake

Active Member
Joined
Apr 20, 2019
Messages
297
Likes
234
I thought I had 1" melamine, but it is 0.5". Anyways, here is a comparison with the mic a foot or so away from the center of the cone.

2" melamine vs acousta-stuf vs 0.5" melamine vs nothing

They all do a decent job. The staggered 33/66 like @alex-z described did have the most bass at 200hz.

..

staggered at 33 and 66% of the box depth. Anecedotally, I find it damps most internal standing waves without harming bass response.
My little test backs this up too! Thanks for the idea!!!



0_5in_melamine__two_sheets_30__60.jpg0_5in_melamine_one_third_way.jpg2in_melamine_half_way.jpg2in_melamine_half_way_and_acousta_stuf.jpghalf_bag_of_acousta_stuf.jpgnothing.jpg
 
Last edited:
Top Bottom