[help] Ported vs Sealed design (Peerless Fabrikkerne, 4" driver)

Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.
Hello everyone, new user here!:)

Please excuse my beginner knowledge in this realm.

What began as upgrading a 23 year-old set of JVC MX-S300 stereo into a Bluetooth system turned into a desire to build my own set of speakers. I am still surprised that this old system sounds just as amazing as it did years ago. Soon I began reading up on drivers types, driver parameters, crossovers, cabinet design, etc.., and bloody hell!!! the more I read, the more confused I got :scratch1:. (as a fresh engineering graduate, half of it was understandable but rest of the acoustical engineering is more or less alien to me)

Introduction aside, I obtained a pair of 'woofers' and tweeters manufactured by Peerless Fabrikkerne (India).

Primary aim of this project is to build a budget speaker system that sounds decent at medium hearing levels (very occasionally rev'up the volume). I plan to use it with my desktop, and at times in the living room in case of small/quiet parties. Secondary aim is to make the speaker have a small footprint/unobtrusive if possible. (like the mini-tower, tower designs). I don't mind a bit of challenge in building my first speaker.

Basically, I want to design a 2 way stereo system using a TDA7492P 25W+25W board with Bluetooth receiver, using the combination below -

Peerless Fabrikkerne 4" KO100 WFR (8 ohms)
Peerless Fabrikkerne 1/2" Polycarbonate Tweeter DW 69 (8 ohms)
with a crossover at 4000 Hz.

These are relatively unknown drivers compared to big names like Dayton Audio, Mark audio, Tang Band, but budget and availability were the main driving factors for this selection.

I have attached a bunch of images below : driver parameters, winisd graph.
From the specs-sheet, it's evident that the 'woofer' behaves more like a full range driver. Also, all 3 curves in winisd are for the same driver, 1 for sealed, 2 for ported. Data was entered by me.

Regarding woofer parameters, the specifications detail a 4 ohm driver, I assumed a similar ratio to be carried over to the 8 ohm, thus the new dc resistance is put as 6.4 ohm. I am ball-parking with voice coil inductance value. Calculated Xmax comes around 0.75mm, although I must say that it seems suspicious to me as I tested the speakers bare for an hour or so at medium-high listening levels in a small room, and the cones seemed to bounce a lot more than that visually.

Found two ways of choosing the 'appropriate' enclosure : looking at EBP or Qts. In my case, EBP of 150 says ported. But a Qts of 0.531 says sealed. Ofcourse, these rules are non-binding apparently, but given how notorious speaker building is known for settling for the right compromises, I don't know how to proceed. I referred diyaudioandvideo's online calculators as well.

I considered transmission line as well, but given the bonuses vs complexity/size/costs, I am turned off by that. I am still jealous of seeing people using 2"-3" FR with ML-TL and getting some ballsy sounds from it. Maybe another time, unless someone says so otherwise.

Can I simply EQ/boost the low end of audio track to about 40-50hz or so based on winisd if I either go ported/sealed? I might get more freedom to do so if I don't listen at really loud SPL, but that's just me imagining a scenario I haven't gone through.
edit 1 : I generally listen to instrumental, progressive metal, djent (sometimes), 80's synthwave/synthpop.

Nasty stories and warnings of unloading, cones rupturing, or flailing wildly but silently give me sleepless dreams :whazzat::eek:

Regards
BN
 

Attachments

  • tw spex.PNG
    tw spex.PNG
    333.8 KB · Views: 230
  • wr spex.PNG
    wr spex.PNG
    337.8 KB · Views: 223
  • winisd1.PNG
    winisd1.PNG
    90.5 KB · Views: 215
Last edited:
Does it work like that though, just adding the Re to the coil?

AFAIK the Qts will be very different if you change the impedance. All the speakers I know with both 4 and 8 ohm versions with very similar Qts have different Mms and Qms, etc. which I assume they did to compensate. Might be wrong. I'm still a bit hazy with the details of T/S parameters and just went along with it when I design.

With that said, how did you plug those values into WinISD without getting an error? It's been a while since I used WinISD though. Might be a it rusty in getting the right sequence when inputting the T/S parameters.

P/S: Xmax usually just means how much the voice coil can travel without escaping the magnetic gap too much. The farther away it is from the magnetic flux, the less in control the motor is. You can push it as much as you'd like over 0.75mm. You'd increase distortion, but as long as it is within the mechanical limits of the suspensions it'll be fine. The actual mechanical limit is called Xmech, Xdamage, Xlim or whatever label that makes it clear.
 
Last edited:
AFAIK the Qts will be very different if you change the impedance. All the speakers I know with both 4 and 8 ohm versions with very similar Qts have different Mms and Qms, etc. which I assume they did to compensate. Might be wrong. I'm still a bit hazy with the details of T/S parameters and just went along with it when I design.

You are right! I completely forgot about that. Check this website
Peerless Fabrikkerne (I) Ltd - KO100 - 4" Coated Paper Cone Woofer

I chose not to follow what was mentioned on the website because the details were filled haphazardly with some parameters missing. Actually, I didn't understand how Vas for a small 4" driver could be 11.306 litres for the 8 ohm version when the 4 ohm version required 4.1 litres.

New EBP turns out to be 140.4 while Qts is 0.45
I wish I had the tools to measure Thiele Small parameters myself :D .

With that said, how did you plug those values into WinISD without getting an error? It's been a while since I used WinISD though. Might be a it rusty in getting the right sequence when inputting the T/S parameters.
I went straight to database maintenance (Ctrl+D), and sequentially filled in the values row wise. (version is WinISD 0.44 if that matters)
 
Calculated Xmax comes around 0.75mm, although I must say that it seems suspicious to me as I tested the speakers bare for an hour or so at medium-high listening levels in a small room, and the cones seemed to bounce a lot more than that visually.
Thou calculated Xmagmaxlin, not Xmagmaxmax, which is voice coil height + (magnet gap height / 2). With a listener's fair tolerance to amplitude modulation, X may exceed Xmagmaxlin for up to Xmagmaxmax / 3.

Since thou already have a pair of stereo loudspeakers, i suggest to build a mono box from thy new parts. Another matter is outfit: Do not just think about size but also about weight. Allowing size for a lo-reaching Xrelieving quarter-wave-line but keeping weight down by caring for efficient strength and sufficient damping is a possible way to follo.
 
Out of curiosity I did a quick modelling of this driver in a ported mini tower using Hornresp. 10cm x 13cm, 80cm tall. Slot port of 10cm x 1.5cm (and 5cm long). Gets an F3 of about 70Hz, but should be usable around mid 50Hz in nearfield. I used stuffing of 80g. Of course this is assuming that the driver parameters are correct.

You can use cardboard to build a prototype.
Screenshot%2081_zpsxg8fknfu.png
 
I have a small doubt regarding crossovers too (don't wish to create a new thread for that).

Is there any rule behind selecting inductor and capacitor values for the cutoff frequency Fc. I mean as long as Fc remains the same theoretically, can one mix and match different values of coils and caps?

For eg: L = 0.3mH and C = 4.7uF || L = 0.45mH and C = 3.3uF || L = 0.25mh and C = 5.7uF (parallel connection with 4.7uF and 1uF caps respectively)
 
Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.