Calling John Krutke -- I think you will find this interesing

[Dylan Matlow]

I posted this elsewhere on the 27th but since I noticed there has been activity here on John's account and not there I thought I would see if I can get noticed here :T


Well hopefully John sees this because I would like to send him one for his feedback when they finally appear at my door step...

I could use some feedback here as well.









This FR is just the same cone/surround on a different lower Le, less open motor (no pole venting) in order to get a feel for the FR:



It has a stamped frame that allows slightly closer c2c spacing however that makes a home-made jig necessary for routing work instead of the handy jasper jig.

I'm looking at ~60-80W RMS in theory (which still needs bench tested) and 5.5mm of xmax (1 way). (~8-9mm xmech)

The sensitivity and BL are a bit on the low side, however it should do nicely IB or in car. They still do quite well in small sealed if a slight 1-2 dB peak is acceptable and they do rather impressively in a number of ported alignments especially with a bit of EQ. I think I will be working with .3 cuft sealed or .4 cuft @ 50hz with 2dB of cut at 100hz, Q of 1. The extra power handling with the high temp coil makes up for this lack of sensitivity.

I'm working on finding an appropriate cast frame with venting under the spider landing for a future higher end high value unit with nice thick short paths, a phase plug and hopefully a spectra fiber (or any HSHMPE High Strength High Modulus Polyethylene Fiber) cone if I can find a manufacturer that is capable of producing them. If I can't make a cone out of pure fiber in a reasonable time frame at a fair price I will work on a wood pulp mix with this fiber in it. All units are made with 1006 steel as my standard unless a super alloy (Hiperco) or pure (99.95%+Iron) is used.

I guess you could call this project extreme DIY. I would like to work some more with other people on future projects but between finals and the current projects I have on my plate I am completely swamped for a little while still. I'm pressed for time and money to keep doing such insane projects :coffee:

[jkrutke]

Hello,

I'm not as active on many forums anymore. I stop in here occasionally.

Normally, I don't test drivers that are not publicly available, though I've strayed from that rule in the past. But the bigger issue is that more recently, I don't feel an urge to test "boutique drivers" from smaller vendors. Bad results could resonate too loudly and affect sales of a company that is sensitive to bad press. Not that I care about that, but it's the part where the vendor gets defensive about their product that leaves a sour taste in my mouth. I'm not saying you're like that, but unfortunately I don't know you.

Likewise, good results could make the public want a driver they might not be able to get.

However, you seem to be working on this project from a DIY'er perspective. I'm not sure if you're potentially a vendor or just a DIY'er making drivers. Maybe we could work something out.

[ThomasW]

Realistically he's a vendor, so it would be better if communication between the two of you occurred off line.

http://www.ologyaudio.com/

http://www.alibaba.com/company/10844385.html

http://www.pesupport.com/cgi-bin/config.pl?read=348608

[Dylan Matlow]

Hello,

I'm not as active on many forums anymore. I stop in here occasionally.

Normally, I don't test drivers that are not publicly available, though I've strayed from that rule in the past.

Thanks for the reply
While I have the first sample here (which was me just inspecting the cone on a generic motor), the actual 500 units will be here in a month or 2 (or however long the boat is going to take). I will be sending you 2 of that batch. I'll measure the impedance plot of 50 units and send you the "average" unit and the unit with the lowest fs just to get some kind of gauge on quality control.

But the bigger issue is that more recently, I don't feel an urge to test "boutique drivers" from smaller vendors. Bad results could resonate too loudly and affect sales of a company that is sensitive to bad press. Not that I care about that, but it's the part where the vendor gets defensive about their product that leaves a sour taste in my mouth. I'm not saying you're like that, but unfortunately I don't know you.

Likewise, good results could make the public want a driver they might not be able to get.

If the results are bad I can just price them at cost and compete in the $8 range and it should still be a good deal unless something is so wrong... At which point I hope there are at least a few people that believe in the cause that are willing to jump in and help out.

The first sample with the correct motor I had them get the t/s on I wanted them to adjust the compliance to lower the Fs since the Qes was a bit high.

The next sample they made had a moving mass of something in the neighborhood of 25 grams at which point there was the rather interesting process of explaining to them exactly what steps they needed to take in order to drop the Fs by changing the suspension. :rofl:

The only scenario I see that could possibly make this unit unusable is if they manufactured and shipped the 25 gram version on accident, which I don't see happening. I defiantly wouldn't be complaining to you about the quality of your testing (a great service by the way which I have used many times).

If it is high quality the only reason I see this driver being discontinued is if at the same price I can offer something of higher quality, and a version 2 5.25" ultra budget unit is way down my list of priorities.

However, you seem to be working on this project from a DIY'er perspective. I'm not sure if you're potentially a vendor or just a DIY'er making drivers. Maybe we could work something out.

I'd like to see a DIY driver design community develop -- where there is a good tutorial on how to design a motor/speaker, where collectively we work to design speakers and vote on what designs to move forward with next.

Granted this is still a business model that hopefully has room for some profit margin or free drivers for me, as well as free/reduced drivers for leading participants all depending on how well things work out, it is just a very relaxed and open model. If this is just a great line on a resume, that’s fine with me. If I can manage to actually make some money to help pay my tuition, even better. In the end I'm sure it will help to ebb on innovation in the field or at the very least let places like seas know exactly what we want.

I guess I should outline my ideology as well.

Honesty and accountability are key, both for if things are going well, and for if things take a turn for the worse. The more accountable the people working on the project are (the more reward they get for doing well), the more motivated they are likely to be. The happier the people are, the more productive they are likely to be. The open source format makes it easy to gauge how productive people are as well as has the potential to push innovation full throttle, but difficult to reward them adequately historically (most people can only participate for fun, it is has been difficult to even make a secondary income even if you are doing great things). If we can properly execute an open project that actually rewards people for their contribution, and if we can actually keep the group of people centralized around some business structure I believe that we could do something great. If this format goes well for the DIY community directly it could easily be expanded to commercial audio and I’m sure other people will adopt a format that evolves off this in other industries.

It is entirely conceivable that a wiki could evolve for each of the industries and these wikis would be one stop shopping (you may need some engineering/physics text books but hopefully http://en.wikibooks.org/ will pick up, or textbook publishing companies will use a similar model in due time) for everything you need to know to operate at a professional level inside the industry, if you have the passion to learn about it.

Just like usual, radical change almost always comes from outsiders. I don’t pretend to have all of the answers, or to know everything on the topic of transducer design but the best way to learn is to jump in, do it and take a risk.

[Dylan Matlow]

Realistically he's a vendor, so it would be better if communication between the two of you occurred off line.

http://www.ologyaudio.com/

http://www.alibaba.com/company/10844385.html

http://www.pesupport.com/cgi-bin/config.pl?read=348608

A bit of a diffrent venture there, I was working with Matt but he has gone MIA for the most part (girlfriend and lawschool prep), mostly car audio.

I'm still working to get those projects completed this summer but its a bit hard doing something like that on your own.

John, if you can drop me an email to matlowd3 *a_t * hotmail * d_ot* com that would be great.

[cotdt]

Dylan, you can send me one to play with

[Dylan Matlow]

Dylan, you can send me one to play with

https://www.htguide.com/forum/showthread.php?t=19203

Anything like that going to happen to it? :lol:

If anyone wants to pull them apart I'll let em go at cost, unless you just feel like donating to the cause. :T

You expressed wanting to learn the subtlties of driver design.

It looks like it is a T pole, they should have made the upper part of the pole be able to be seperate and just slide into a hole so that they could have added a short path under the T pole instead of inside what looks like the magnet... shouldn't impact the price unless that short path didnt need a tooling and one under the T did... Nice thick path though.

[Raptor550]

Its a nice looking driver . For convenience to the DIY community a round frame would be easier to use/flush mount. I can think of about a half dozen threds with people gripeing about the Vifa D26NC55 beacause of that.

[Dylan Matlow]

Its a nice looking driver . For convenience to the DIY community a round frame would be easier to use/flush mount. I can think of about a half dozen threds with people gripeing about the Vifa D26NC55 beacause of that.

I wonder how bad the diffraction would be though given that it isn't a round frame just to surface mount given that it isn't a tweeter, or make a circle cutout and fill in the gaps with material/felt? The flange is 4mm in height.

Zaph has a great write-up on surface mounting effects that can be found here:

http://www.zaphaudio.com/mtg-surface.html

If it poses too big of a problem for people I can see about getting some flush mount wood circles CNC machined into some .75" thick material, or ideally just into some front baffles if a really good design arises. Regardless cosmetics might be the largest concern. Given that the frame is stamped though the flange edge isn't completely squared off, it does have a slight round over to it.

[jdybnis]

How about the peerless style truncated circle? That gives you a close c-t-c and is still pretty easy to route with a circle jig and an edge guide.

[Dylan Matlow]

How about the peerless style truncated circle? That gives you a close c-t-c and is still pretty easy to route with a circle jig and an edge guide.

I couldn't find that style of cast frame sadly. I have a feeling it is a custom tooling, or they have each and every single unit machined??? A cast frame and machining each frame would have pushed the price up another $4/ea or so if you want my guess, and if I’m going to use a cast frame it needs to have venting under the spider landing or else I don’t think it is actually worth it. I don’t think they would like to machine each unit given the tool time it takes and I’d likely have to modify one of their tools in the end regardless…

I find it quite likely that I'm going to need to tool some mounting locations such as the flange, spider landing, motor mounting, washer that just use aluminum/zinc bars/pins and have fastener locations from the side in order to allow fairly painless recons (or painless cone swaps)... They may not be the prettiest but they will accommodate the largest spider landings possible, have very open frames under the landing, as well as eliminate all future tooling costs since the depth of any section (clearance under the spider landing, or the effective cone height) can simply be adjusted with different lengths of aluminum.

A washer will attach to the basket and screw in from the sides, and that washer will be custom for any given motor top plate OD and simply screw into the sides of the top plate.

It also seems possible that the area directly under the spider landing could be sealed up and radial cross drill locations could be implemented in the washer causing quite a good deal of air to blow directly across the voice coil if there was a few millimeter thick aluminum spacer implemented here. Not to mention this aluminum washer would be electrically coupled both to any short path on the ID of the top plate and the aluminum frame and be closely hugging the voice coil... Each of the radial cross drilled holes would terminate into a right angle, vertical groove in order to keep noise down. Naturally pro audio and subwoofers only for the venting setup…

Hopefully you can visualize this – I’m busy studying for a final right now or I’d draw something for you

I'm pretty sure there’s no prior art on any of that but I'll be finding out shortly. The basket is comparable to some other flexible designs out there, although it is a little more flexible and not worth $10k on a patent, I just want to make sure I'm not infringing... The venting on the other hand is actually quite useful.

-- Dylan Matlow

[cotdt]

any way to get rid of the small peaks at 2.5kHz and 6kHz? if you send me a pair, i can analyze the standing waves and try to smooth out the response by selectively applying dampening coating at the right places. and wasn't this woofer supposed to have 32mm of one-way xmax?

[Dylan Matlow]

any way to get rid of the small peaks at 2.5kHz and 6kHz? if you send me a pair, i can analyze the standing waves and try to smooth out the response by selectively applying dampening coating at the right places. and wasn't this woofer supposed to have 32mm of one-way xmax?

There is a ~20mm of xmax 6.5" underhung design that has a case of the hiccups, this isn't that...

I'll send you a pair to play with if you want to tackle moding them. Just post the FR and CSD results of what you do.

It is possible that some of that is baffle diffraction... I guess we will see though shortly. I'm under the impression this was measured in a cabinet though.

[Dylan Matlow]

Zaph?


Quite the pile... Augerpro is a bit busy at the moment so I'll get some distortion measurements whenever someone has a chance.

I feel this stands up well against seas PL units in listening tests at least, although I haven't heard the reed drivers yet... I do have a bias though, and I don't mind a bit of peak and I love the extra low frequency output this is capable of. I would be interested to see a properly executed TL with this unit. The flat frequency response ends in the 3khz range though due to increased inductance vs the graph posted earlier, but it still seems easy to use in terms of xover design.


Here is some info on a 6.5" but its not done yet (generation 4):
Pe: ~180-200W RMS
Xmax: ~12mm





Generation 3 proto 6.5":





Generation 2 proto 6.5":











I'm working on a car audio version as well that has a little less mounting depth and xmax. Here are the estimated t/s:
Qes: .75
Qms: 3.12
Qts: .604
Fs: 49hz
mms: 13.7g
cms: .77mm/N
Re: 1.82 ohms
BL: ~3.2 Tm
Le: ~0.07-0.1mH
Sd 128 cm^2
Xmax 7mm
RMS Pe: ~125W

[Piotr]

Interesting

The basket in the last post seem to be identical to the Accuton midrange C-90-T6 aka C-173-N...

/Peter

[Dylan Matlow]

http://www.mainstreet-audio.com/Merchant2/merchant.mvc?Screen=PROD&Product_Code=ES-06plus&Category_Code=CDTMIDS

The ES-06+ GOLD uses it too. It is quite a substantial basket and seems to have been traditionally used on quite expensive units since it does come at a bit of a premium even among cast frames... I guess this unit will be an exception to that trend though.

The accuton on the other hand kind of reminds me a bit of the omega driver motor, except with a bit of a thicker top plate and lower sensitivity with a ton of xmax... There will be variants with less xmax and higher sensitivity though well suited for 3-4 ways since a change in the voice coil and top plate is pretty simple... Thats quite another topic though

[Drew]

Hi Dylan,

This seems pretty cool - definitely taking DIY to the next level. If your hope is to make these available to the DIY community I have a suggestion: a wideband, low distortion midrange driver. These are pretty rare and what ones do exist are big $$ - look at the Accuton that Jon is using. More than $400 each and a range of 250Hz to 2.5kHz.

I don't know what you constraints are, but if you could cook up something that would be clean from say 150Hz to about 4kHz, I'm sure you'd have a lot of very interested people. Even at $50-70 each. Something like the Peerless 830883 but without the FR bobble @1k.

I hope I don't open the floodgates for everybody suggesting their pet idea here and thereby overwhelm you. But this particular application seems to be undeserved.

Whoops I just noticed that the last word there should read under-served, not undeserved!

[Drew]

Hell, as long as I'm here I may as well ask: why the (seeming) fascination with phase plugs? I'm not a driver designer by any stretch, and this is an honest question.

In your first post you mention that you'll hopefully be able to add a phase plug, and your second driver already has one. Why? Really, I'm just curious...

One more question - I've seen it mentioned before that an inverted dome (I'm assuming a spherical section but that could be wrong) is perhaps the ideal cone profile. Like some of the HiVi and Revel drivers. What are your thoughts on this - is this something that you could source? I'm imaging a HiVi M8 with an improved motor...Mmmm, tasty.

[Mazeroth]

Augerpro is a bit busy at the moment so I'll get some distortion measurements whenever someone has a chance.

If I was serious about starting a speaker company I would probably learn to take my own distortion measurements.

[Dylan Matlow]

Hell, as long as I'm here I may as well ask: why the (seeming) fascination with phase plugs? I'm not a driver designer by any stretch, and this is an honest question.

In your first post you mention that you'll hopefully be able to add a phase plug, and your second driver already has one. Why? Really, I'm just curious...

One more question - I've seen it mentioned before that an inverted dome (I'm assuming a spherical section but that could be wrong) is perhaps the ideal cone profile. Like some of the HiVi and Revel drivers. What are your thoughts on this - is this something that you could source? I'm imaging a HiVi M8 with an improved motor...Mmmm, tasty.

Feel free to ask any questions this thread is random enough anyways

You *can* (not always) see a more extended response with a phase plug.

http://www.seas.no/seas_line/midrange/H422.PDF

http://www.seas.no/seas_line/midrange/H522.PDF

The above is a good comparison where the only change is the addition of a phase plug.
A properly designed cone profile with a properly designed non-traditional phase plug, or 3, or 4 part sandwich cone can be far superior to a unit with just a phase plug, so long as there is ample % open area in the former to relieve resonance and compression.
Using a phase plug to help remove heat from the assembly is about as effective as a traditional pole venting structure however the hole required to mount the phase plug is much smaller than a pole vent meaning that when fully saturated the pole can carry more magnetic energy than a unit with a pole vent. Less standard venting structures can be more effective at cooling the structure than a phase plug.
The mms goes down, but the Sd goes down with the addition of a phase plug.

http://www.loudsoft.com/my_documents...per%202003.pdf That paper can explain cone profiles quite well and makes me want to get that software… Shame it costs too much for me right now… possibly if I sell enough of these I can get that toy to play with

[Dylan Matlow]

If I was serious about starting a speaker company I would probably learn to take my own distortion measurements.

I trust John's measurements more than a manufacturer's. I have a Behringer microphone and preamp I just need to put a quality jig back together (the speaker workshop jig got lost in the move from Ohio somehow) and get a good external soundcard for the laptop and I can head to some nice open spaces for measurement.

[Nemophyle]

i will keep an eye on the devellopements of your drivers, the 6.5 looks sexy

any info on motor design ?

[Dylan Matlow]

i will keep an eye on the devellopements of your drivers, the 6.5 looks sexy

any info on motor design ?

Overhung, Extended pole. The pole piece is driven into saturation which I have a hunch might cut down on distortion a bit, cut down on Le and modulations in the field since the voice coil itself will have a harder time influencing the soft magnetic structure (and therefore harder to reach the hard magnetic structure as well) due to the decreased relative permittivity of the soft magnetic structure as the steel is pushed to the top of its B/H curve throughout a considerable length of the pole. If I'm wrong on this I'd like some more info on this myself, the relation between permittivity and distortion is still a little cloudy to me but this makes sense to me.

Using a T style pole would have likely been a little more effective in terms of decreasing low frequency distortion since it can help make the plateau of flat BL around X=0 larger, but since the pole is saturated I felt that the added useful BL was worth the tradeoff. With the saturated pole, any significant reduction in the OD of the pole would limit the total B across the gap and reduce the BL of the system. If I used a T pole I would also have had (well I wouldn't have to but it is nice) to get a copper sleeve tooled for placement under the "T" and a copper cap that would be coupled to the phase plug thus pushing the price upwards.

Super heavy 6mm external (outside of the voice coil OD) short piece inside of the magnet structure helps cut down on distortion some more and the stacked magnets allow plenty of voice coil clearance. I also suspect that the large "H" inside the magnet structure due to the stacking has some impact on distortion as well. I was lucky to find this tube size already so no tooling was needed, keeping prices down.

1006 steel is used throughout which may be excessive but it is available and still a fraction of the price of Hiperco-50 which I am going to use for some other designs.

Top plate is 7mm thick, voice coil is 31mm tall (CCAW), 82% BL range is +- 12mm. (0.9448 inches p-p)
:T

[Hank]

Dylan, I admire your industriousness - you are obviously serious about your endeavour. I will agree with Drew's suggestion for a quality midrange wholheartedly. The landscape is full of 4" - 6.5" drivers and some are real performers. As Jon is pointing out to us, there are only the TB and Dayton mids that are affordable. Can your manufacturer do a good, wide-range mid?
I'm going to meet with my projector manufacturer in Suzhou, China and Taiwan in 2 weeks. They are one of the largest driver ODM's and I'll see what they've got in midrange capability.

[Drew]

As Jon is pointing out to us, there are only the TB and Dayton mids that are affordable. Can your manufacturer do a good, wide-range mid?

And the Dayton mid, nice as it is, shouldn't be used below about 700Hz. So you still need a woofer that's clean till about 1.5k to mate it with. A nice wideband mid would be great!

Again - just as a suggestion - I'd go ahead and use a T-shaped pole piece and put the copper sleeve in there if the driver would benefit from it. The world is awash in cheap drivers that are poor to mediocre performers. I'm not saying that your units are bad, just that performance is what drives people to buy. Especially on this board. if I'm going to spend $300 on crossover components, what do I care if the midrange costs $50 instead of $40?

At the risk of speaking for others, I think that most on this board would pay a few more bucks for better performance. Even if you're talking about the difference between good and great performance. I sure would.

[Dylan Matlow]

The BL would drop to about 5.7 with a T pole with the same voice coil and a thick pole sleeve. I would have to move to a 2" dia coil, start over, make the plant irritated and increase the price a bit more than $10/ea.

As it is with the extended pole the BL will drop to the 95% point at +- 7.5mm
This BL point will be comparable with a T pole, however what we will see a real change in is the 98% BL point.

The 6mm thick short path that is present is substantial enough for its placement to do a great job.

Having 3 sets of short paths would indeed reduce distortion more but I am a bit limited in terms of funding as a college student so I'm doing my best given the current situation.


The cone supplier is the limiting factor for me getting the level of performance I need. I've been searching for the right supplier for quite some time now... Also I could use access to a cone profile FEA program but that toy will have to come later...

I'd love to make a line of high efficiency high quality speakers for a nice 3-4way unit but I'm just getting my feet wet with these first few budget oriented units.

[Spanky Ham]

Just like usual, radical change almost always comes from outsiders. I don’t pretend to have all of the answers, or to know everything on the topic of transducer design but the best way to learn is to jump in, do it and take a risk.

I think that is more the exception than the rule. I remember reading an article in Audioxpress (I think) called "Confessions of a Speaker Designer" or something like that. The author worked for Jensen designing OEM car speakers. According to him, it is mostly science. I would guess computers have had the biggest impact on speaker design and evolution over the past ten years.


Having said that I agree with Hank on admiring your industriousness. Along with Drew and Hank, I would agree that a good performing mid would be a good niche to go after.

[JoshK]

Agree with the mid suggestions.

[Dylan Matlow]

I think I can get about a maximum of 93dB/w out of a 5" midwoofer with about 5mm of underhung xmax (should have more than enough of a flat BL plateau), at least with any reasonable f3. I could get a better f3 out of them in a ported enclosure with a higher tuning but that’s not the right way to go I believe... So unless I can trim off more mass than I think I can without paying a price somewhere else, that’s as high a sensitivity unit I can manage in a 5" mid... I wouldn't use it under 400hz or so myself for a no compromise system which is well within the range of a nice dedicated 10" woofer to pick up the slack.

On the other hand I can get a bit more extension out of a 7" with comparable sensitivity allowing a good union with a dedicated 12" woofer, I need more time to play with this though and I'm still trying to figure out which of these driver pairs would be best to start with... My time is rather divided between so many things at the moment.

I'm doing my best to make these at least a bit dipole friendly but the 5" version as it stands starts a bit of an early rolloff, the 7" looks OK.

On one hand a dipole favors the 7", on the other hand the 5" is better suited for higher frequency playing and can still mate quite well with a 10", or an 8" dipole.
The 7" could be tailored with 90 or 91dB/w and still mate well with a subwoofer if need be, whereas a 5" is always going to push that envelope a bit in terms of lower frequency THD almost no matter what you do...

Input is always great, I’m still a bit up in the air on the midranges on the matter of 1.25-1.5” voice coils with a phase plug, or 2” voice coils with as much venting as can be managed given the magnetic strength required and plenty of perforations in the former in order to keep the resonance down. I'm leaning towards as small a diameter as I can manage though that can still keep the parameters I need...

A nice dome mid has crossed my mind but I’m not sure that’s the best way to go.

"I would guess computers have had the biggest impact on speaker design and evolution over the past ten years." -- Without computers I would have never run across DIY audio in the first place :T
The processing that can be done using computers is definitely essential though for me to be able to handle this on this sort of budget...

I'll go a bit more into motor design when I get the chance to write something substantial between all the projects I'm working on

Don't mind me if something here is a bit off, it has been a LONG night! (6.5" midwoofer)

[JonMarsh]

What you're discussing is a bit like the tradeoffs between a C79 and C90-T6 Accuton, it's not really possible to be all things to everyone. Drivers like the Tangband W4-1337S cone unit, or the RS52 dome mide, which are quite clean with regards to energy storage up to 8 kHz or so are very nice and give more options for crossover topology, but have other limitations. The 7" mids require a different approach in the system design, because even with exotic cone materials there are much lower limits to how high you can push them, and what CTC distances are practical. Unlikely to have a one size fits all, but you're certainly exploring the right issues with your "thought experiments".

And i DO like the looks of those 6.5" Proto's you're working on... you should definitely have something interesting there....

~Jon

[Drew]

Input is always great, I’m still a bit up in the air on the midranges on the matter of 1.25-1.5” voice coils with a phase plug, or 2” voice coils with as much venting as can be managed given the magnetic strength required and plenty of perforations in the former in order to keep the resonance down. I'm leaning towards as small a diameter as I can manage though that can still keep the parameters I need...

Well, here's another bit of input:

If I were in your shoes, I wouldn't worry about the sensitivity too much. Just because the Accuton is 93dB doesn't mean that you have to strive for that. Look at how many people use the Dayton RS series with sensitivities in the mid to upper 80s.

For me personally - wideband is important. Wideband makes crossover design so much easier; you have some breathing room rather than trying to use a unit up (and down) to the ragged edge. I'd like to be able to crossover at about 200Hz on the low end and about 2kHz on the high end. This implies good behaviour from 100Hz to about 4k.

A 5" unit is a nice size since if you get any bigger you start to have serious C-To-C issues with the tweeter. I'd avoid a dome mid for no other reason than the RS52 is great and is already pretty cheap. It could use a smaller flange.

It looks like the unit in the first post already has a lot of potential for a mid application. I'm wary of a paper cone from a distortion standpoint. A nicer frame would be nice as well. But all good things in time.

I'm really trying to be helpful here - not critical.

So, all that said, how can I get my hands on two of the units that you have right now? I'd like to mess around with them a bit. If you don't feel like posting publicly then go ahead and PM me.

I'll be keeping an eye on your 6.5" as well...

[Tommythecat]

My $0.02:

Start talking to a manufacturer who can do Neodymium and shoot for the 95db point.

And think bigger. Who wants to mate a serious midrange to a 10" woofer? Im thinking more of a 12-15" woofer that can be used in up to 3 cubes tuned to 35Hz. Something with a bit more sensitivity so we can keep up a total system average of 94-95db 1W from 40-16kHz+. A no BS three-way fullrange speaker.

[Dylan Matlow]

If I were in your shoes, I wouldn't worry about the sensitivity too much. Just because the Accuton is 93dB doesn't mean that you have to strive for that. Look at how many people use the Dayton RS series with sensitivities in the mid to upper 80s.

For me personally - wideband is important. Wideband makes crossover design so much easier; you have some breathing room rather than trying to use a unit up (and down) to the ragged edge. I'd like to be able to crossover at about 200Hz on the low end and about 2kHz on the high end. This implies good behaviour from 100Hz to about 4k.

Given that the motor is liberally underhung, assuming I can get everything else right the low frequency distortion should be great, it is just the f3 point that will be a bit higher with a high BL design. It is however very easy to remove a bit of BL and gain back a lower f3 in exchange for more or less 3dB sensitivity. The gap can be made wider and more copper can find its way there, the gap can be made taller for more Xmag and a larger flat BL plateau for less distortion at higher levels, or the magnetic structure can be reduced and the price falls…. There are more options but this outlines a few.

I'm still working with motors though and trying to trim the mass of the coil with thinner edgewound wire (so as not to sacrifice much in the way of power handling, yet still increase the box size requirements). I’m kind of leaning towards kapton for the former material but I’d love to see peoples opinion of this.

I’m also seeing how light I can go for the cone and am planning on trimming mass from the surround with Arnitel from dsm (~0.2mm) thick for the surround should be more resilient than a thicker rubber surround and I'm exploring various spider changes but that may not be for this unit...

It sure would be easier to do this if I wasn't doing this by hand and the tedious method and could optimize the coil quickly with a program...

In regards to the top end of the frequency range, I'll do my best... This will go even better if someone out there has Loudsoft Finecone and wants to assist. (Please?) ;x(

A 5" unit is a nice size since if you get any bigger you start to have serious C-To-C issues with the tweeter. I'd avoid a dome mid for no other reason than the RS52 is great and is already pretty cheap. It could use a smaller flange.

It looks like the unit in the first post already has a lot of potential for a mid application. I'm wary of a paper cone from a distortion standpoint. A nicer frame would be nice as well. But all good things in time.

I'm really trying to be helpful here - not critical.

So, all that said, how can I get my hands on two of the units that you have right now? I'd like to mess around with them a bit. If you don't feel like posting publicly then go ahead and PM me.

I'll be keeping an eye on your 6.5" as well...

The RS52 is a great unit and I'm sure I'll have something along these lines eventually but not now.

The current SL13641 unit is a fairly high excursion unit for a 136 mm unit with a bit lower sensitivity, with no short path, and quite the value unit... If I had to guess just from listening I'd put the xover at about 2.5khz. I just ran an impedance sweep of it now that I have an impedance jig handy and it looks like there’s a slight bobble at 1.5khz but hopefully it doesn't pose much of an issue...





Here is the closeup in as close a scale to the ones we are used to seeing from zaph:



If you look at the ancient frequency response from an early sample back in a prior post there is a slight dip at 1.5khz that corresponds but hopefully this won't introduce too much in the way of distortion. That frequency response is from a unit where I was just testing the cone since it looked interesting on a motor with a Le of .1mH, 6 ohms, 1mm of xmax, and no pole venting so the current version rolls off sooner.

The only way to get the 13641 right now is through me, I'm frequently available through AIM: OlogyAudio. Something tells me this isn't the place to discuss pricing?

Needless to say I don't have a salary and everything goes back in to fund future projects.

Looking back on the 13641, it may have been best to find a catenary dustcap and go with polypropylene or possibly even a thin white small weave fiberglass cone... Price would go up a tad but considering how much shipping contributes to the price along the way it would have added value.

Hindsight is 20/20, and if I can get rid of these 420 A stock(80 managed to become B/C stock through shipping damage) items quickly we can move on :

I was up all night last night dealing with the 6.5” but everything seems to be moving back on track more or less…
Does anyone have a close up picture of the voice coil former/cone joint, both for the rear and front of the cone for a seas excel and/or AA Poly units?
To keep quality control high in terms of defects it is best to have a glue joint on the front of this 6.5” cone… With a front glue joint however the glue will not be 100% identical from units being made by one person to the next which in terms of quality is only the slightest change in mms between units, but in terms of cosmetics can be an issue. I’m having a translucent glue used instead of the black glue of the prior prototypes which I think will look better as well as specifying a smaller glue joint, however it is possible to ad a cosmetic ring where the cone and former meet. It will increase the mms slightly but it will also create a better adhesion between the cone and former as well as make the joint more rigid which should have a slight positive impact on the frequency response as well.

Any ring there would require a tooling and would only appear on a production unit and in such a small order would likely increase the cost per unit by about $1.



The ring is to be much like the ring there, except black and a bit thinner.
Black semi gloss coating on expanded polypropylene foam (EEP) would be my first choice.

The main question revolves around: is such a small mostly cosmetic issue worth about $1/ea?

I am getting the sample made though with woven voice coil leads, one out one side of the spider, and one out the other due to the benefits of symmetry.

What you're discussing is a bit like the tradeoffs between a C79 and C90-T6 Accuton, it's not really possible to be all things to everyone. Drivers like the Tangband W4-1337S cone unit, or the RS52 dome mide, which are quite clean with regards to energy storage up to 8 kHz or so are very nice and give more options for crossover topology, but have other limitations. The 7" mids require a different approach in the system design, because even with exotic cone materials there are much lower limits to how high you can push them, and what CTC distances are practical. Unlikely to have a one size fits all, but you're certainly exploring the right issues with your "thought experiments".

And i DO like the looks of those 6.5" Proto's you're working on... you should definitely have something interesting there....

~Jon

Thanks for chiming in Jon, I'd be glad to aim for any pointers you can throw my way. That Tangband W4-1337S is an impressive cone design, I can't remember seeing any other profile that so skillfully spread out the breakup mode in a fairly linear fashion in order to keep any high Q harmonic peaks from creeping in. What a perfect example of how usefull a program such as FineCone can be. I'd be curious to see how distributing the modes in a lighter, more exotic material would perform...

My $0.02:

Start talking to a manufacturer who can do Neodymium and shoot for the 95db point.

And think bigger. Who wants to mate a serious midrange to a 10" woofer? Im thinking more of a 12-15" woofer that can be used in up to 3 cubes tuned to 35Hz. Something with a bit more sensitivity so we can keep up a total system average of 94-95db 1W from 40-16kHz+. A no BS three-way fullrange speaker.

I can hit 94 without much of an issue in a 7" form midrange, I'm not sure how it would work out in a 5" though considering the problems with even 93 out of a 5”, there’s only so much mass that can be trimmed and given the sd, there’s only so much sensitivity to realistically pull out of it.

A high quality underhung 10" is a good solution and will mate well with a 5" as well as any subwoofer even at unreasonable amplitudes.
A 12/15" 7" 2-3" (dome) 22-25mm dome 4 way would be great but what I have in mind for the midrange/tweeter coincident, concentric combination is a bit tricky (Read as expensive and time consuming).

The tweeter’s sensitivity goes up, the f3 goes up, the xover point goes up therefore making integration between a high sensitivity 7" and said dome tweeter an issue due to c2c constraints.

As it stands a nice tweeter, 5”, 10” is still a compromise but at this stage can be handled by me with less of an issue, where ultimately the tradeoff is in output, of which there should still be plenty of. You may be able to mate a lone 94-95dB/w 7” with a rather expensive ribbon currently and some nice 12/15s but other issues come up, including cost.

Thanks, I will defiantly keep a no compromise 4 way in my mind for the future.

P.S. I love sensitivity too, less power compression and more dynamic potential = good, so long as there aren’t any nasty tradeoffs... :T

BTW I just had to chime in on... the shipping weight on the 6.5" is a whopping 5kg!!!
I'm kinda wondering if the dayton RS style packing boxes will be enough to keep these from getting damaged in shipping...

[cotdt]

wow those drivers look gorgeous!

[Dylan Matlow]

wow those drivers look gorgeous!

8x)
Thanks!

[Hdale85]

12mm of xmax on a 6.5" mid....very interesting indeed :B

[wildfire99]

12mm of xmax on a 6.5" mid....very interesting indeed

Like those Extremis things with 13mm of xmax!

I guess that begs the question of is it really a mid anymore?

[Hdale85]

At 6.5" it surely doesn't have much SD so I'm sure it can't go to low. But I guess it could be a woofer. But I would still call it a mid

[Dylan Matlow]

I'm still trying to get the leads woven perfectly to avoid any possibility of leads breaking or interfering with the excursion which is taking a few extra days... It is a shame this is taking longer than I hoped...

I ran across a rather curious aluminum cone but I'm still not 100% sure what to make of it...




It is still a mid, and hopefully there isn't too much of a bump in the harmonic distortion associated with that peak in the fiber cone but we will see...

[Dylan Matlow]

Everything seems to take longer than it should...

I have zaph's measurements posted now.

Not as awesome as I expected, but not bad... The frequency response deviated between the sample and the production quite a bit, for this reason I'm going to avoid pulp mixes as well as textile domes in the future to help out batch to batch consistency... At least until I can afford to watch over the mixes and coatings in person. The distortion below 80hz could stand to be a bit better but in all it looks to do fairly well at the price point.



Comparing with the seas W15CY001 the shortcomings are obvious:



It is more fair to compare against the Dayton Aluminum given the price:



The peak at 4khz can be dealt with, without too much issue. I also think the dip in the frequency response helps mask the 3rd harmonic distortion associated with it since my ears aren't registering much in the way of anomalies there. I'm playing with several designs at the moment.

I find it curious that the scanspeak has 2nd harmonic distortion associated with the beginning of its edge resonance and mine has 3rd... Granted my pulp mix has less damping and quite a bit of rigidity... That is a mystery worth looking into...









I've been working on this budget project and pulled out more BL than I was quite expecting so I have a curious 7" with t/s suited for ~.3cuft tuned to 60hz and 88.7dB/w if anyone has any input. ^^ Given the cone's breakup pattern I would expect fairly low HD.



Latest news on the 6.5":







A few steps forward, a few steps back...

The spider now is capable of more throw and is of the Flat type :T. So now the CMS is a bit off (I guess they just used the same treatment concentration not accounting for the profile change).

The spider is woven, but it should have weaves 180 degrees apart for increased symmetry. The wrong material is present at the glue joint, note the change in the frequency response with the addition of felt there. I'm just going to go with a clear joint until I can tool a foamed polypropylene ring for production after I find some people interested in helping out financially or I sell enough of these 5.5s to move forward :W

I'm currently working on getting a webpage together. With Fall Quarter at OSU is quickly approaching I'm in a bit of a time crunch. All I can do with the 6.5” though is to hurry up and wait… Wish me luck.

[Dylan Matlow]

Due to the stiffer than expected suspension I took a more in depth look at how the compliance of a midwoofer impacts the transfer function, max spl and group delay and thought I would share the results...

Yellow has the loosest suspension at 2mm/N
Grey has a moderate suspension at .77mm/N
Orange has a stiff suspension at .44mm/N

Red is the Dayton RS180-8
Green is the AA Poly 6.5 t/s on zaph's site

The two enclosure sizes are .8 cuft tuned to 38hz and .3 cuft tuned to 45hz






It is apparent that a looser suspension has a clear advantage at least for this motor, mass and surface area combination, however given the very long throw nature and the woven leads there is a limit to how soft the spider can be for quality issues without moving to a thermoplastic elastomer for the spider. I have a feeling the old CMS in the range of .75-.9 is as loose as will be attainable at this particular manufacturer but I will gladly accept 1.5-2mm/N if they can pull it off with a combination of weave type and treatment concentration. I'm sure their "engineer" is getting frustrated with my quest for perfection in the sample stage and isn’t used to dealing with such specific demands...

[noah katz]

"It is apparent that a looser suspension has a clear advantage at least for this motor, mass and surface area combination,..."

I'd venture to say that it's a gemeral truth.

The system stiffness is the sum of the suspension's and the box air spring's.

Lowering the suspension stiffness thus has exactly the same affect as increasing box size, as evidenced by your curves.

So people are misleading themselves when they see a high Vas (which a soft suspension will contribute to) and think it means they need a bigger box.

Also it should reduce distortion because the more linear air spring will provide more of the stiffness.

Very impressed with your knowledge and drive, best of luck in this endeavor.

[Dylan Matlow]

I'm still working on this little project and I figured I would run the changes by everyone here.

I would love any feedback on the cone color and motor/frame finish if anyone cares one way or the other.

Somehow the surround got switched out however it should still handle the 12mm of xsus required of it, if it wont in practice it will just require another sample cone/surround...

I'm moving closer to production step by step, although rather slowly since I'm being a bit of a perfectionist about this and acting on my own

After the motor and suspension are handled I'll be looking at a few different higher end cone options though if anyone has suggestions I'll keep it in mind.

[dlneubec]

I think it's a very attractive looking driver, FWIW.

[Dylan Matlow]

As always the design process involves a series of steps forward followed by steps backwards. Hopefully this is going to end now with one last final step forward.

After this I can quickly move to handle a shallower car audio (or IB) adaptation of this unit followed by an 8" woofer.

As much as I would like this unit to have a smaller diameter motor, I have been avoiding neodymium magnets as much as possible due to the current volatility in the market price prices. If it appears that the market price is more stable due to changes in the environment, I will begin to explore the neo option in future productions of this as well as other units.

Here is where I stand:



Here is where I need to be:



Hopefully the frequency response holds up just fine. If it won't there are plenty of options but that will require another revision which I would like to avoid if possible.

The lead wires in the current weave might need a little more slack but that is a fairly simple task to handle. Quality control might be a little more complicated, however.



Should the next fiberglass cone not be adequate in terms of the desired frequency response, I am exploring the option of mineral and hollow glass microbead (3m) loaded polypropylene as well as Akulon (dsm) Ryton, and a few higher end engineering thermoplastics that will remain un-named for now and possibly :edit: -- I'll let you know later...

If this fiberglass cone is acceptable, (which I believe it will be due to the high grade of the fiber used) I will simply have to wait for the 8” to experiment with other cones

There is nothing quite as much fun as tinkering!