The BaSSlines (was High Sensitivity Design)

[TacoD]

Wow great job! Looks very neat with the lexan baffles.

[dlneubec]

They are finally complete! Below are the final details for the BaSSlines design. I'm extremely pleased with the results, which I believe are my best effort so far.

Included below is the crossover schematic, the simulted frequency response, the simulated polar response from 500-3500 in 500hz increments, the parts list (using all PE available crossover parts) and cost estimate for the pair, not including the lexan, the hardwood, mdf or veneer and finishing supplies and a photo in our main room.

[Jim Holtz]

Wow! Those are really pretty Dan. Excellent job as always! :T

Jim

[ttan98]

Looks like the midrange(95dB spl) is heavily padded, I found the tweeter at norminal 91-92dB padded too. What is the overall efficiency of the speakers? My guess is about 88dB which is in par with most commercial speakers. I was hoping the overall effciency would more than 90dB.

I really like the system configuration, ie OB tweeter and midrange and BR woofer. Most tempted to build it myself at a later stage. I may use different midrange(cheaper), tweeter(ribbon) and woofer, and will try to extract better spl say 95dB using active and 92-3dB using passive from the driver combination.

Overall congrats to a very nice looking speakers and sounds good too I am sure.

[dlneubec]

Thanks, Jim!

TTAN98,

I haven't figured the overall sensitivity, but there are some mitigating factors making it harder to predict. I would expect it to be a little higher than that in room. The woofer is about 93db and has no more than 3db baffle step, due to the near floor mounting. The midrange (actually rated 96db) and some of the low end treble is attenuated 2-3db more than normal due to the overall summing and because of the additional energy coming from the back of the cone. If you look at Zaphs tests of the midrange, you will see that in the peak of the passband, from the 1-2khz region, it is closer to 98db. Zaphs test

The tweeter is boosted about 4db or so on the low end due to the waveguide, centered around 3-4khz, which makes the attenuation at the crossover look higher than it would with a flush mount. You can see this in the waveguide tests I posted earlier in the thread.

[ttan98]

Dan

If you have the time use a reference driver which you know the spl accurate preferably(within +/-1dB is ok), run it (MLS) using Soundeasy, one speaker after the other. The display on the MLS will quickly tell you the difference. I think accurate spl is not necessary, +/-1dB is good enough.

[Cataclysm]

Beautiful speakers; really striking to look at, great effect with the driver configuration, and I am sure they sound downright awesome. :T


Gah, you have me wanting to build a set of these just to see what they would look like with a leather wrapped baffle and some aluminum mixed with gloss black to finish it off.

... I have been inspired :rofl:

[Smokinghot]

Awesome work....makes me eager to start my own project.

FWIW..., and at this point I wouldn't even bother, but you can get rid of the ghosting you see on the cut edges of plexi/lexan with a low intensity torch. It's a very touchy process, so any thinking of trying it out should practice on scrap before being too brave.

[dlneubec]

Thanks Guys!

ttan98,

When I get a chance I will see what I can do.

Cataclysm,

There is at least one guy building them now and he is doing his own look, all wood, no lexan. I can send pdf's to anyone who wants to give the build a try. :T

Smokinghot,

I've thought about heating the edges of the baffle, but I actually like how it provides some definition to the edges. Also, it would be a major setback if I overheated slightly and got bubbling :E (I have tried it a little on a test piece), so I'm happy not to take that chance.

[TMadden]

I am in the process of building these, with some variations, not that I don't respect your design, it's awesome. Mine will be all veneered MDF , I don't have your skills or patience! Finally got the woofers, sure look nice, can't wait to hear them. Also have the tweeters. I have several mids I am going to try, PHL 1360, Audax PR170MO and Z0. I know you ruled out the Audax because of X-max, but I found them listenable down to 450 Hz or so in a previous system. I will probably try the PHL 1120 before it's over too. I will be triamping at first, have a modded DCX2496 I have grown pretty fond of. I also intend to get the B&C mids and build your crossover, I definitely want to hear your build too.

[ttan98]

I am in the process of building these, with some variations, not that I don't respect your design, it's awesome. Mine will be all veneered MDF , I don't have your skills or patience! Finally got the woofers, sure look nice, can't wait to hear them. Also have the tweeters. I have several mids I am going to try, PHL 1360, Audax PR170MO and Z0. I know you ruled out the Audax because of X-max, but I found them listenable down to 450 Hz or so in a previous system. I will probably try the PHL 1120 before it's over too. I will be triamping at first, have a modded DCX2496 I have grown pretty fond of. I also intend to get the B&C mids and build your crossover, I definitely want to hear your build too.

Hi,
It is interesting to know someone is trying so many different types of mid-rangers. I would like to know how you compare PHL1360 against the older model PHL1120? My guess the PHL1360 is better than Audax PR170MO, slightly warmer and fuller in sound, I heard from someone that the Audax has leaner sound.

I have DCX2496 original version and you have a modded version, what mods did you do? I heard the output analog stage mod made the most difference? Can you give me a brief description how you find the difference, I am thinking of doing the mod as well.

Cheers.

[TMadden]

Hi,
It is interesting to know someone is trying so many different types of mid-rangers. I would like to know how you compare PHL1360 against the older model PHL1120? My guess the PHL1360 is better than Audax PR170MO, slightly warmer and fuller in sound, I heard from someone that the Audax has leaner sound.

I have DCX2496 original version and you have a modded version, what mods did you do? I heard the output analog stage mod made the most difference? Can you give me a brief description how you find the difference, I am thinking of doing the mod as well.

Cheers.

I haven't compared all the mids directly yet. I did alternate the PR170M0 and ZO in a previous system, and prefered the ZO, sounded a bit more open and revealing to me. Others have prefered the MO. I have used the 1360, but haven't done comparisons with the Audax's yet. I had the 1360's paired up with compression drivers on waveguides, wow, what a revealing combo, I heard low level detail I never heard before. Yes the PHL 1360 do sound warmer and fuller, as they are capable of playing down to 60-70 Hz, the Audax's are only usable down to 450 Hz or so. I think the PHL 1120 will be similar to the Audax's, perhaps a bit more robust. It's optimized for midrange , where the 1360 is more of a midbass driver. I'm curious about the B&C midbass also, they look very promising. Since the AE 12" woofer is very usable up to 500Hz or even higher, I should be able to make any of them work well with it.
I put Jan Didden's active output stage in my DCX, I love it. Now I have remote control of volume, and bass and treble levels, and the input and output levels are compatible with home stereo equipment. It's hard to comment on sound quality with out an A-B comparison, but it sounds very good to me. I am thinking about doing a clock upgrade also, not sure if it's worthwhile or not. Can anyone comment on that?

[ttan98]

Hi,
It is interesting to know someone is trying so many different types of mid-rangers. I would like to know how you compare PHL1360 against the older model PHL1120? My guess the PHL1360 is better than Audax PR170MO, slightly warmer and fuller in sound, I heard from someone that the Audax has leaner sound.

I have DCX2496 original version and you have a modded version, what mods did you do? I heard the output analog stage mod made the most difference? Can you give me a brief description how you find the difference, I am thinking of doing the mod as well.

Cheers.

I haven't compared all the mids directly yet. I did alternate the PR170M0 and ZO in a previous system, and prefered the ZO, sounded a bit more open and revealing to me. Others have prefered the MO. I have used the 1360, but haven't done comparisons with the Audax's yet. I had the 1360's paired up with compression drivers on waveguides, wow, what a revealing combo, I heard low level detail I never heard before. Yes the PHL 1360 do sound warmer and fuller, as they are capable of playing down to 60-70 Hz, the Audax's are only usable down to 450 Hz or so. I think the PHL 1120 will be similar to the Audax's, perhaps a bit more robust. It's optimized for midrange , where the 1360 is more of a midbass driver. I'm curious about the B&C midbass also, they look very promising. Since the AE 12" woofer is very usable up to 500Hz or even higher, I should be able to make any of them work well with it.
I put Jan Didden's active output stage in my DCX, I love it. Now I have remote control of volume, and bass and treble levels, and the input and output levels are compatible with home stereo equipment. It's hard to comment on sound quality with out an A-B comparison, but it sounds very good to me. I am thinking about doing a clock upgrade also, not sure if it's worthwhile or not. Can anyone comment on that?

I don't want hijack this thread, last question, can you tell me where to get the Jan's mod and how much? Also where you buy the PHL1360? Thanks

[TMadden]

I don't want hijack this thread, last question, can you tell me where to get the Jan's mod and how much? Also where you buy the PHL1360? Thanks

I got the mod here: http://www.pilghamaudio.com/index.asp?pgid=51
319 euros about $430 dollars for the kit
The PHL's I bought at Zalytron, he made me a good deal on 4 of them.

[ttan98]

Dan,

Like most mid range driver mounted on an open baffle, you usually get a bump on around 600-800Hz. I was wondering what sort of bump you get from the B&C mid MD38, in your speaker.

Can you post your B&C raw FR response curve measured mounted on your acrylic frame?

Thanks.

[dlneubec]

Here is the most recent measurement of the B&C in open baffle. I believe this was after I changed from the prototype mdf baffle to the Lexan/hardwood baffle, but it may not have been. Shown are on axis, 15-30-45 degrees off axis. There is no bump like you are talking about.

[ttan98]

I use a different mid driver, my bump occurs around 800hz.

Looking at the black trace, it seems your mid bump occurs at 1.5Khz and notch (dip) at 2.3Khz and break up around 5Khz. The black trace looks like an M shape, just like mine. Just by applying the mid x-cross you fix up the M shape and flatten the response?

[dlneubec]

The black on axis that I posted above is important, but the off axis is also very important. Notice how in the off axis plots, the dip is filled in? You want to look at off axis response and not just consider on axis. You also have all the extra rear readiation going out into the room with a open baffle dipole mid and monopole woofer and tweeter, so it is often necessary to have a dip in the on axis FR with a dipole mid or the midrange will sound bright.

This was not from the final xover design, but here are some measurements taken on the mid with an earlier crossover at 0-15-30-45 degrees.

[lbstyling]

do you have a final overall graph done yet dan?

[dlneubec]

I meant to post this here, but never got around to it. I have been struggling trying the get the crossover right since December, after I showed the prototypes at the Lexington DIY. I had decided to make some physical changes in the driver layout, prompting a new set of measurements. I have just recently figured out my error. Below is what I posted on the PE forum in regards to the trouble I've gone through and my major screw up. ops: I'll post some more updated results after this post.

As many of you know, I was shocked at how bad the BaSSlines sounded to me at Dayton and at InDIYana for that matter. For those of you who also heard them at Lexington last year in their prototype form, you know how much better they can sound. Well, I've finally figured out what happened and boy was it a bonehead screw-up on my part.

At Lexington, they the XO's were preliminary but the response was not too bad. I decided to make two changes to the design after that. First I lowered the tweeter nearly 2” to improve the vertical polar response. Then I sloped the MT baffle back 5º, instead of leaving it vertical, in order to try and get a better tweeter alignment with the mid. The result of these changes was a need for all new measurements and that is where the ship veered strongly off course.

In the first measurement below you can see what happened at the crux of the problem. The red curve is the bass measurement used in my model, the black is what it should have been, both with nearfield/farfield and port contributions merged. When I did my nearfield measurement, I applied a full 6db of bsc, which I didn’t do in the prelim measurements and should not have done here, because the 12” woofer is very close to the floor, meaning little or no bsc was required. In addition, I must have made an error in merging the nearfield and farfield data, since there is obviously more than 6db of bsc shown here.



As a result, my model going into Dayton (and InDIYana) looked like the red line in the graph below, or so I thought. What the response actually looked like is the red curve to about 350hz and the black curve below that. So, instead of having about a 3db drop in the midrange for the open baffle, I had more than a 6db drop. It's no wonder they sounded so bass heavy. Still, they sounded good enough in my listening room, because of the bass null I have at the LP and also because we have an open floor plan that includes most of the bottom floor and up a two story foyer to a tall vaulted ceiling and an upstairs family room, so that I didn't pick up on the problem as I should have. Still, I have been tweaking them on and off since December, after I took the new measurements, because things were just not coming together and I was mysteriously unhappy, despite what the model told me. You can see why now.



The blue line in the graph above is what the BaSSlines response looks like now compared to the model from Dayton and InDIYana. The woofer level is down a bit. In actuality, the midrange is up by about 3db or so and the tweeter is up 4-5db. Below is what the entire response looks like now, a good 3-4db more sensitive overall as a result (the graphs are not at real world spl levels). There is about a 2.5db drop through the midrange to adjust for the additional output from the open baffle midrange. Lessons learned; if you rely on measurements too much and you make a mistake in taking those measurements or applying them, you can have a heck of a time figuring out what is causing the problems you are hearing. Don’t get me wrong, you need measurements, IMO, to do the best job possible, however, I should have picked this problem up sooner when listening, but room complications somewhat maskd it. Also, use more than one room when voicing. If I had done this, I believe I would have picked up on the problem immediately. I may have to haul them back to inDIYana and Dayton next year so those folks can hear what they should have this year.

[dlneubec]

Here are some other results. First is a simulation of the vertical polar response. These are from 500hz to 3500hz in 500hz increments.



The second graph is the measured frequency response, left speaker over right. The mesurements were taken at 101", with a gated window of 6ms (speaker was up on a stand inside the house). It has been cropped because the data below 500hz is not acurate. I'm shooting for a 2-3db dip in the midrange due to the open baffle mid, whose rear radiation is not shown in this gated graph, but is heard in the room. I had an oops with one of the first pair of HDS tweeters I purchased. This was before they went unavailable for a couple months. The replacement I ordered was from a new batch and measured a bit differently, which may explain the difference above 13khz or so, or perhaps there is a little difference between the fabrication of the DIY waveguides from the left to right speaker. I don't find it an audible difference.



I don't know how accurate this method is, but since I don't have a software like REW to measure the in room response of both speakers at the listening position, I tried to measure it using MLS in SoundEasy. I took separate measurements of the left and right speakers with a very large gated windown (500ms, IIRC) and then summed their responses in SE. This is what the summed response looked like. I'm not sure this is valid, but you can see how the midrange is filled in. Actually the tweeter is down a little bit, I'd say. Also, the bass null I've been struggling with at my listneing position and that I mentioned in the previous post, is probably the dip that shows up from about 30-80hz. I think this is part of why it took me so long in figuring out what was going awry. ops:



Here is the measured impedance and impedance phase. It dips slightly below 4ohms at 2.5khz.

[lbstyling]

sounds like you've been busy!!
ive had a couple of projects that just wern't 'right' for one reason or another- soul distroying!
- i built a fiberglass mould for casting a concrete speaker box that when i used for the first time, the caretaker guy where i worked tripped over it while locking up and cracked it in half leaving me with one hell of a problem when i arrived to look at my new cabs on monday morning- a load of workshop machinery with concrete set in it!!
i sympathise,
weve all been there!

your fr graph is looking good - does it feel right?
im tempted to build this or something similar myself.-
on the plus side - man thats a sexy looking beast, id prob be satisfied just to have that quality of cabinet work!! mmm

[lbstyling]

how do you feel its performing against your original aim of the bagby/salk speaker??
retrospectivly what do you think it was that was special about the design now youve built them??

[dlneubec]

sounds like you've been busy!!
ive had a couple of projects that just wern't 'right' for one reason or another- soul distroying!
- i built a fiberglass mould for casting a concrete speaker box that when i used for the first time, the caretaker guy where i worked tripped over it while locking up and cracked it in half leaving me with one hell of a problem when i arrived to look at my new cabs on monday morning- a load of workshop machinery with concrete set in it!!
i sympathise,
weve all been there!

your fr graph is looking good - does it feel right?
im tempted to build this or something similar myself.-
on the plus side - man thats a sexy looking beast, id prob be satisfied just to have that quality of cabinet work!! mmm

It has been a rather embarrassing, but important growth experience. I think the gated repsonse graph is solid, down to 500hz or so. The speaker was measured up on a 3' stand at about 2.5meters with no reflecting surfaces within the 6ms gated window. Thanks for the compliments!

how do you feel its performing against your original aim of the bagby/salk speaker??
retrospectivly what do you think it was that was special about the design now youve built them??

It's hard to say, since I last heard the Salk over 15 months ago, but it was probably the best I'd heard at that point. I'd say they would sound quite similar.

There are a few key differences in the designs. This is a much smaller speaker (16" to 7" tapered, 42" high versus 15" parallel sided baffle about 46" tall) and is considerably smaller visually due to the lexan baffle. My speaker has a sloped baffle to get closer to a time alignment between the drivers. The Salk is vertical.

The Salk speaker had a mid-tweeter CTC that was about 1-1/2 to 2" greater, so I think this speaker would have a larger central lobe in the vertical polar response. The boost from the shallow the waveguide, after being equalized flat in the crossover, should mean it has reduced distortion in the 2khz-6khz area as a result, and the HDS tweeter is already a steller performing to begin with as far as distortion goes. It should also have more off axis directivity above 3-4khz or so.

My woofer box is about 10 liters larger than the Salk box and tuned probably 3-4hz lower. I don't know what the woofer box construction looks like in the Salk, but mine is likely much different, with a ribbed panel reinforcement approach with cross bracing.

My MT baffle is thinner, about 1-1/8" and separated from the woofer baffle (there is 3/8"/3/16" speaker gasket between them), with the midrange driver pressed against the baffle from the rear to reduce panel vibration. The Salk is much thicker (probably 2" or more) and monolothic top to bottom. It has a fairly deep tunnel with large roundover at the rear of the open baffle mid. My mid is recessed with a roundover on the front and the rear is flush with the back of the baffle. The B&C 6MD38 actually measured better in dipole configuration mounted this way than when it was flush mounted to the front of he baffle.

I believe the crossover points and target acoustic slopes are quite similar, though the crossover design is certainly different.

As far as comparisons to other speakers, that is hard for me to do, especially without 1:1 auditioning. I'd say the soundstage is deeper and more realistic with the open baffle mid and the midrange is more open, cleaner and less colored than a boxed design, but these are entirely subjective. The Pro woofer and mid make for a much more dynamic speaker than most HiFi speakers I've heard, which make them sound more live to me. People have described them as inviting, engaging and very dynamic. I'd say the non-metal drivers are more forgiving on poorly recorded material, though poor recordings still sound bad.

[lbstyling]

sounds like a well sorted design to me- i wish i could hear it!! - alas the US is a long and increasingly expensive trip from the uk.
ive always found that dynamics (micro and macro) are a underestimated aspect of hifi speakers in general- its that aspect that for me is the emotional drive behind music- ie, the reason i listen!

out of interest- do you know the THD and IM distortion figures? i was under the impression that higher efficiency designs tend to have good figures at high playback levels no?

presumably the room accoustics play a large part in the sound of a open baffle- have you looked into what improves or changes the sound for better or worse?

sorry for all the questions!

[dlneubec]

I have not done any distortion measurements. All of these drivers have been measured independently by Zaph of Augerpro at their sites. Actually, Augerpro measured the Lambda TD12M, but it is basically the same driver as the TD12H, but optimized for midrange use. All of these drivers are quite low in distortion.

I don't know that open baffles are any more affected by room acoustics than any other speaker. They do need to be away from the front wall simply because you want to delay the first reflection sound enough. That said, due to their figure 8 shaped response, they are less affected by side wall relfections and have more placement flexibility from that standpoint. Like any speaker, they need to be played with to get the ideal location in the room. In my last crossover iteration I decided to go for the ideal on axis response, so I have them toed in and they seem to sound best that way.

[sfdoddsy]

Sorry for resuscitating a long dormant thread, but I found it too amusing to resist.

Just when I thought I was being original, the internet proves me wrong.

Drivers aside, snap!

[Evil Twin]

Yes, but you found your own way there...

And many will argue the journey is just as important as the destination.

[sfdoddsy]

Thank you ET-san.

At best my journey was quick and cheap.

Reading through this old thread, Dave-san's skills leave me in embarrassed awe.

[dlneubec]

How's that for coincidence. I haven't been on this site for years and when I do, up pops a thread from a speaker I designed and built 11 years ago!. Now, to make it even more interesting. I'm just in the process of redesigning these speakers. Same woofer and bass bin, new MT_T section, with new midrange, same brand, but B&C 6mbx44 6.5 mid to replace the old 6md38, a new front tweeter (no waveguide), the Morel ST1108 and I'm going to add a rear tweeter to maintain the dipole radiation pattern (to the degree possible) for improved balance between power and direct response. The rear tweeter has to be small and will fit just above and behind the morel. It's looking like maybe the SB26stcn-4 at the moment. Will have an all new XO, of course!

https://www.parts-express.com/bc-6mbx44-65-professional-neodymium-woofer-8-ohm--294-6014

https://www.parts-express.com/morel-st-1108-supreme-1-1-8-soft-dome-tweeter-pair-black-faceplate--277-096

https://meniscusaudio.com/product/sb-sb26stcn-4/

The reason I'm not building new is that I had a home accident in November that left me with 4 fractured ribs, which are still healing, so that derailed my new build plans. Also, I have a $250 credit from a win at the 2019 MWAF that I can use for the mids and at the same MWAF, they gave the first place guys a pair of the ST1108's, so I already own those. My out of pocket for the drivers will be about $80 give or take. The BaSSlines have been in storage for about 8 or 9 years, so I thought I might as well use what I have on hand that I can get done in the time I will have left after I heal up, not to mention that I'm less motivated to build these days anyway.

Hope all of you at HTGuide are doing well and hope to see some of you at the MWAF 2020! Here's a little motivatation. Last year, in addition to the credit and the Morel's, I also scored a major door prize, a Wavecor SW275BD02 subwoofer:

https://www.parts-express.com/wavecor-sw275bd02-10-3-4-mica-paper-cone-subwoofer-8-ohm--298-1076

Now, I don't really need another subwoofer, but this one is darn nice, so I guess I should build something with it

[wolf_teeth]

Keith got the other of those 10" subs, so maybe you 2 could collaberate...

You gonna hit up InDIYana (April 24/25) this year too, Dan?

Ben

[dlneubec]

I do plan on attending InDIYana again this year. Not yet sure about 1 or both days.

[wolf_teeth]

Oh- forgot one more thing-
Are you going to mount the tweeters the same way this time? IIRC, they were different in the previous iteration of the BaSSlines.

Later,
Wolf

[dlneubec]

Ben,
You are correct. The original design had the tweeter faceplates removed and they were mounted from behind in a 3/4" deep mini waveguide, which was not particularly effective. The Morels will be flush surface mounted as will the rear tweeters.