Building my first set of speakers this weekend - Nat P's

[zegebe]

So I finally placed an order last week for parts, and everything arrived today! I decided to go with the Nat P's since I was able to use PE's pre-built cabinet, and the crossover wasn't too crazy.

I worked on laying out the crossover this evening - just have to make sure I'll be able to fit it though the bracing to the back wall of the cabinet, and still leave room for the port's hole, before I mount and solder everything.

Read a neat trick about using a battery for a quick polarity test on the drivers, so I'll be doing that before I start wiring things up too.

Quick question: How much foam should I need for these? I went with the unshielded RS180's, which from what I read can be used as drop-in's, but need less volume in a ported enclosure. Does that mean less foam? Also, should I be using foam sheets on the sides of the enclosure, or stuffing?

I'll try to post some progress pictures tomorrow.

[AdelaaR]

"this weekend" sounds like you want them to be all done by monday morning ... good luck with that
It doesn't seem logical to me that less volume means less foam.
Less volume sounds like smaller enclosure i think.

[ThomasW]

Use foam to line the interior surfaces not covered by the crossover.

Add stuffing to fine tune performance to 'taste'

[zegebe]

@AdelaaR: Ha.. I know I probably won't finish this weekend.. can always hope though I thought part of the reasoning for foam was to make the inside of a cabinet appear larger to the speakers.. course I think I heard that many years ago so could be making it up. I had thought about screwing some extra wood panels into the cabinet too.

@ThomasW - thanks, that makes sense

[fbov]

Communication's a wonderful thing. Adding stuffing/foam makes the volume of a box act a little bigger, so if you have less foam/stuffing, it acts smaller.

In a sealed box, you add/remove stuffing from the volume (middle) of the box. In a ported box, you put absorbant foam on the walls, leaving the volume open to resonate. In some cases, you may add volume fill only away from the port air path for fine tuning.

If you're reacting to the reported differences in bass alignment between shielded and unshielded driver, the differences are small, and easily adjusted by changing port length. Clamp the fronts on and listen before you glue everything up. I like both drivers in a box at this volume tuned to 40-45Hz. Port length to achieve this will depend on the diameter of your port.

Have fun,
Frank

[zegebe]

Thanks for the info Frank.

I ended up not having as much time as I wanted to work on the speakers this weekend.. but did make some progress. I cut the port hole out of the back of the cabinets, and cut the baffles as well, including beveling the back of the RS180 openings (just need to cut some notches in the tweeter holes).





I also assembled the first crossover, and decided to build it on 2 boards to make it easier to fit. I didn't really optimize my caps, so I think that took up some extra space (i.e. used 20+12 for 32, instead of 30+1+1, so needed space for 2 big caps instead of 1). Not sure how much it matters, but I also kept my resistors in the air so they have more air circulation.

[fbov]

Lots of things make no audible difference, but we do them a certain way anyway, like 20+12=30+2. I use what's on hand that will work.

Resistors in series will see driver current, but there's rarely one in the woofer path, and that's where the electrical power is. You've got a good low gauge coil on the woofer...

Hang on a second.

NatP is a series-parallel XO. It can't be split into independent circuits. Something's seriously wrong, unless you're not showing the board-board connections. I'm terrible at tracing circuirts from pictures, but this is an easy one. Look at Jon's layout and you'll see the tweeter circuit is nested in the woofer circuit. Both woofer and tweeter "see" C1.

Another way to put it is that you can't bi-amp NatP's. Modula MTM has separate circuits, but not the Natalies...

[zegebe]

They're not independent

I'm not showing board-to-board connections.. I accounted for that. The negative feed from the amp goes to one board, the positive feed goes to the other board, and there are some speaker terminals and board-to-board connections to get it all together. I basically just drew a line through the crossover that cut 2 wires, and those connect it all back up. I think if you snake a line almost horizontally through the pic of Jon's layout, that's about where I split it up.

Didn't want to test just the one since I don't have a 4ohm speaker for the other channel (and was too lazy to try and get a couple 8ohm's in parallel), so will do some testing once I get the other put together next weekend

[fbov]

Glad to hear it. Some builders take things literally, others know how to breadboard components and test circuits before the final build. Henceforth, I'll assume you're one of the latter.

BTW, the only time you really need the spec'ed drivers for XO testing is if you're evaluating the design - verifying you got the right impendence curve, for example. Any driver will put on a load, and no modern amp will blow if you short across the output. The trick in final testing ported boxes is to keep the XO external and run the leads through the port until you're sure the circuit's right.
Frank

[jmik26]

Did you get a chance to listen to them yet?

[zegebe]

I ended up on a business trip this week so haven't been home since Sunday =/

I should have the other crossover wired up friday night or saturday, and will test then before starting to put them together.

It'll be interesting to hear how the foam affects the sound.. I've never experimented with that before and I guess it can be a bit of trial and error.

[zegebe]

Finished building these last night! (well, aside from the foam)


So I finally got everything assembled and tested, picked up some nicer quality speaker wire for the internals (had a spool 2-conductor speaker wire on order from monoprice, but it hadn't arrived yet, so got some 2-conductor audioquest wire from bestbuy - the Magnolia section had it by the foot)

I realized I didn't have any good foam for use inside the cabinet though - I glued some that I had lying around to the walls for now, but I think it's too open to really do much, and didn't seem to have an audible difference, even when stuffing it right behind the drivers. So I've got some proper acoustic foam on order from foambymail.

So far, I'm really liking. They have much more of a sound stage than the little JBL bookshelf's they were replacing (which I bought probably 15 years ago or so). They are a bit more lacking in bass than I was hoping for, but I knew that was a compromise with bookshelf style speakers going into this - my long term plan is to move these to surrounds and build some towers for the front.

I think my next project will probably be a new CC though, as that's now severely lacking!

Here's some pics of the rest of the progress, as well as a comparison to what they replaced:

[jmik26]

Nice, thanks for the update. I ordered all my stuff and its coming in this week, I can't wait to get started!

[fbov]

...I realized I didn't have any good foam for use inside the cabinet though ...I've got some proper acoustic foam on order from foambymail....They are a bit more lacking in bass than I was hoping for, but I knew that was a compromise ...I think my next project will probably be a new CC though, as that's now severely lacking!...

Bass is lacking? Run them hard for a few days if you haven't manually exercised the cones past Xmax. A ported 1 cu ft box should tune nicely to an F3 of 45Hz, and room gain will give you all the extension you'd need for musical instruments. Using the shielded drivers, my 65l towers are tuned to 32Hz and the port impedence peak is at ~23Hz, but that's an extended bass shelf (EBS) alignment. Regardless, these aren't going to give you the subsonics of a subwoofer....

I use carpet underlayment foam. Two layers of 1/2" to start, and it's cheap when you buy damaged rolls. That said, even ported boxes like a little acoustic fill; I use a bag of pillow stuffing from a fabric store. I use it around the base of the bottom port in my 40" towers, to minimize internal standing waves without interfering with box resonance.

If you want a matching CC, the Modula MTM CC is the only quality choice. I spent 2.5 years with a third NatP on its side, and was very pleased until I replaced it with a purpose-designed CC. Centers are all about dialog, and even the wife mentioned the improvement in dialog calrity.

http://htguide.com/forum/showthread.php4?t=11475

Have fun (and ENJOY!),
Frank

[DancesWithBeers]

I agree with fbov. Two 7" woofers per side shouldn't be lacking in bass coming from small bookshelf speakers. If you briefly connect an AA battery to the binding posts, to both woofers go in or out in unison? What about on the other side? Are they wired in the correct phase at the amp? How far are the backs of the enclosures from the wall? Have you run an impedance sweep to see where they're tuned? Maybe the previous speakers had a bump in the bass region that your used to?

[zegebe]

I did do a batter test to all speakers before wiring everything together, to make sure the polarity was marked correctly. Just did a sanity check too and they all move out. Backs are about 10" from the wall right now.

Have not done an impedance sweep..

I'm using the PE pre-built cabinets and haven't sealed the front yet (just hand tightened) and also haven't sealed around the mounting hole for the port yet (the plastic had a very snug fit), so I suppose there's a possibility of air leakage.. how drastic of an effect would that have? I plan on sealing everything up once I get some real acoustic foam inside the cabinets (I have a roll of speaker gasket foam that I think might work on the front baffle if needed).

I do have a small sub, so that could be distorting my idea of bass response. It's just a little 5 or 6" 60watt sub I purchased from Cambridge Soundworks *years* ago. I do feel like even just listening to Pops music, that some of the lower instruments aren't quite as pronounced as I would expect without the sub. Right now I've got the sub turned off while I do some more break-in listening.

I also didn't exercise the cones before I put everything together, so that might be playing a factor too - running some Erich Kunzel cincinnati pops through them right now, since some of his tracks have some very low frequencies.

[Hdale85]

It can take some time for the RS drivers to fully break in. The suspensions are pretty stiff from the factory.

[zegebe]

It can take some time for the RS drivers to fully break in. The suspensions are pretty stiff from the factory.

I'm beginning to think that may be the case. I really haven't listened to them much at all yet, but I was really pushing them today (as much as I could within reason.. definitely the highest I've ever cranked the volume on my receiver), and I think there's starting to be a noticeable difference in bass response.

As far as breaking them in goes, does low-volume/excursion movement help with that? (as in, would there be any benefit to just letting them play relatively softly during the day while I'm at work)

[fbov]

For driver break-in, you really want excursion. That's why manual movement works; you're reducing the stiffness of the spider and surround (depending on material, of course) near the rest point by making sure it's free to the limits of movement. Mine just took a few minutes to change, without manipulation. Your boxes are still open, so give it a try if you have any residual concerns. Xmax is 6mm, but that's typically the distance where the coil's fully in the gap (or vis versa). Xlim/Xmech - the hard stops - are farther and not an issue when done manually (and gently).

Also, don't discount the affect of past listening on current expectations; you may not be accustomed to flat, low bass extension. Try listening to something with interesting bass; I've come to like Tracy Chapamn's Fast Car after Sean Olive mentioned it's one of Haman's trained listener songs. You should hear a kick drum "whomp" in with a bass line that's clear on every note up and down the scale. I find myself humming along with the bassist a lot, so this wasn't hard for me to hear. If you hear all the notes easily, at moderate listening levels, that's good flat bass.

Finally, leaks do contribute to port output, but they're only a problem if they whistle. The port is a big leak that'll prevent others from whistling, and PE's cabinets are rabbeted in front so you have no worries about leaks if the baffle fits snugly. I've never used gasket tape, but I'm blessed with the tools to get snug fits and flat mating surfaces.

Have fun,
Frank

[zegebe]

So over the past week the bottom end has definitely opened up - I've kept my subwoofer off and it's not nearly as noticeable.

I also got my foam in the mail yesterday - I was afraid to line all the walls behind the woofers and around the port as I didn't want to restrict airflow too much. I did line most of the back and sides behind the cabinet brace though.

[fjhuerta]

Zach, I also built the Nat-P's as my first project. Break-in and other issues aside, I always thought of them as too "forward" and bass lean (and I built a huge tower cabinet). I stuffed them, unstuffed them, padded the tweeter... personally, I think they would have worked great on a room with a bit of absorption, but the one I was using them in was simply too resonant.

I ended up taking the pieces and building an ZDT3.5 instead, which I liked quite a bit more.

[jmik26]

I am running into the same problem with my Nat P's. I lined the entire cabinet with acoustic foam and left them sealed. I anxiously brought them into the house and started playing them with less then desirable results. So I left them playing at high volumes for at least 6 hours while I went shopping to pick up some PVC because I was sure they sounded bad because they were not ported.

I got home ripped everything apart, installed a 3", 11" long port to the rear (with elbow). Cranked them up only to find out they kinda sounded worse? So, I listened for a day going through a collection of 500 cds and nothing left me totally impressed. After a day or two I decided to rip out all the acoustic foam on the sides of the walls and they sound a lot better but I am still not blow away by them.

It makes me appreciate how great the Lou C topaz's are but baffles me that I spent way more building these? It just seems like the Nat P's are lacking bass and when I listen to music I can tell where the speaker is located, like someone is pointing a megaphone at me? I am going to keep trying, If I come up with anything I will post.... Jeff

[ThomasW]

It just seems like the Nat P's are lacking bass and

The ported version offers solid output to ~32Hz in the PE box. As with any 'bookshelf speaker, it goes without saying they should be used with a sub for HT applications.

when I listen to music I can tell where the speaker is located, like someone is pointing a megaphone at me?

Sounds like a room placement/room acoustics issue.

Like all of Jon's designs the NatP's are quite 'revealing' of the electronics in the signal chain..

[---k---]

I haven't read the whole thread, only the last couple of posts. So, sorry if I'm covering old ground.

I wonder if there is a problem in the crossover? If it is that bad, there is probably something wired wrong. First thing that came to my mind is are your RS180s wired out-of-phase? Are you sure? I've make at least one mistake with every pair I build - we all have.

The NatP and ModulaMT designs have been heavily demoed at public DIY events alongside other DIY speakers. Every time it was very well received - with some critical ears in the audience. I've personally heard a couple of ModulaMTs. JonW's set in my house and a couple at DIY Iowa. All sounded good to me. JonW had to tweak his to raise the top end (opposite of Javier), but they were good. If bass is weak and they are beaming, I think you have a problem with your build.


I think the detail (and lack of) given in our build threads falsely leads people to believe that our designs are kits and the the only I in DIY is in the box. I know from first hand experience that using different foam or lining can sometimes make a significant difference. I also know that there are tolerances on the drivers that can make a significant difference. I also know that there are tolerances on every single crossover part, that if the stars align can make a difference.

Unless you use the exact same box, foam, crossover parts and drivers, your results MAY sound different than the designer's. There may be tweaking required to get it to sound right. Just a fact of DIY speakers. It is a hobby that takes a lot of time investment to get to nirvana. Many people aren't that critical about their speakers or have a good reference so you don't read a lot about this in posts.

[fbov]

My NatP's have been to the PE show, so it's not just my ears that think they're good. I've also put a mic on them and they're flat to the low 30's without including port output.

Having seen pics of Javier's room, I think he's right about it being too resonant, but I also suspect he's got a finer ear than most and so heard small problems.

Comparing specs, I can see no a priori reason for the Topaz to have better bass capability than RS180's, although there's a lot you can do in the crossover.

The two thoughts that come to mind are:
- these are all built with the new, unshielded drivers. These like smaller volumes and higher tunes to get similar LF curves in simulators, and several builders have blessed them as drop-ins to the XO design. But...
- these are also built with recent-build RS28's. It's possible the tonal balance is off due to a hot tweeter. In a recent project, I used RS28's purchased 1.5 years apart. The newer one needed a 10 ohm resistor in parallel to pad it down, but it was also marked 1dB higher in sensitivity.

Maybe the stars have aligned?

Have fun,
Frank

[fbov]

... It just seems like the Nat P's are lacking bass and when I listen to music I can tell where the speaker is located, like someone is pointing a megaphone at me? ...

Here's what I got, close mic'ed (8" from tweeter) and at listening position. Above 200Hz, the biggest difference is a slight tilt downward at the LP, consistent with an increase in the direct/reflected sound ratio. Below that, things are highly variable... but you can see the port's contribution at the LP.

If you don't have good tonal balance, there's got to be something that can be fixed....

Now, as to sound localization, what you describe is a feature/shortfall of the MTM design, in theory and to my ears. I just built an MT design that also uses the RS28A tweeter, because it tends to disappear. These things have much "better" imaging and great airiness compared with the NatP's, and you jsut hear music, not where it came from. The NatP's are great for multichannel HT, where the ambiance is on the program and you want the added transient response from a somewhat focused loudspeaker.

Think in terms of the continuum from a point source to line array. MTM is the first step away from point source, and it's advantage is reducing energy toward floor and ceiling. The little MT with it's tight driver spacing and small baffle engages the whole room, including ceiling and floor, and the result is a greater sense of space, something that's highly desirable, IMHO in stereo listening.

HAve fun,
Frank

[jmik26]

Thanks for the help. I started a build thread to give my info, sorry OP for hi-jacking the thread. http://www.htguide.com/forum/showthread.php4?t=38382

On the bright side I seem to have solved my problem without trying. I think I had some polarities crossed?

[jmik26]

Think in terms of the continuum from a point source to line array. MTM is the first step away from point source, and it's advantage is reducing energy toward floor and ceiling. The little MT with it's tight driver spacing and small baffle engages the whole room, including ceiling and floor, and the result is a greater sense of space, something that's highly desirable, IMHO in stereo listening.

HAve fun,
Frank

Makes sense to me. I guess I know what my next build will be, ha ha ha ha...

[jmik26]

It ended up being a crossed polarity on mine.