Help with speakers for newbie

Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.
I'm not at all sure where the OP's head is at at the moment, but when I looked at the driver cost for the Tarkus and realized what a bargain they are, especially when you consider all of the positive reviews of the speakers, I couldn't help but wonder what a Tarkus Squared would be like. So a WMTMW Tarkus. Not so much to play louder but to be able to play loud more effortlessly and with less distortion, although it should indeed get about 6dB louder at xmax than the original Tarkus, so I think that's going to be just over 110dB if one cared to play that loud.

I've done the sims and made the necessary xo changes based on a quick sketchup of a simple new WMTMW rectangular cabinet with only the tiniest changes to the original speakers' FR (not including the 6dB increase in sensitivity). If the OP or anyone else is interested, I'll post the results although the final xo values may depend just a tiny bit on exactly how one wishes to design the new cabinet and the placement of drivers. Maybe on the woofer cabinet size too because the sims suggest that you can actually get a little bit more LF out of those 10" Peerless if you go with a slightly bigger box and a lower tuning. And doubling up the woofers should give a little bit more leeway for those extra LF's at higher SPL's.

Driver cost is still only about $360 with the xo cost remaining about the same - I'll be generous again and call it under $200. So that's still actually below the OP's $600 spending goal for a Giant Tarkus. Pretty sweet for what you're going to get I would think. It will be a big speaker though. :D
 
Hey mazvydas1234

First a caveat - I'll just say that a simulated design has an element of risk that a proven design has gotten rid of. So there is a small chance that you might have to tweek my xo if it doesn't quite sound right to you.

Having said that, I think everything should be fine since what I've done is just a relatively simple adjustment to an existing design. Doubling up the mid and woofer drops each set of drivers' impedance from 8ohm to 4ohm and basically the rule of thumb in that case is to double the existing inductors and half the capacitors. However, cabinet dimensions are no longer the same so that also changes the diffraction responses of primarily the mid and tweeter so I've also made changes to accomadate those circumstances.

Attached below:

1) new cabinet dimensions and driver layout
2) summed FR
3) summed impedance
4) xo schematic

Things to note:

1) How you design the cabinet is really up to you. You can make it all 1 box or break it into 2 or 3 smaller sections that will make it easier to work with and move around. One box that size is going to be heavy.

The main compromise to this design is the large center to center spacing of the mids - they're always better off as close together as possible in an MTM design so the first thing I might do is remove the tweeter faceplate and cut it down to the same shape as the truncated mid frames and that should get the mids about 20mm closer I think. If that's something you're not comfortable with or can't do, don't worry about it - compromising the center to center spacing with MTM's is done so frequently with so many speakers that I don't think it's that big a deal.

You could also make the MTM section thinner or appear thinner with some large chamfers on the sides.

I think you could also choose a slightly lower tuning for the woofers if you want but that will mean bigger boxes and given the already large size, that may be something that you don't want to do.

2) You may notice that the simmed FR is not exactly the same as Paul C's measured FR on the Tarkus site but that's ok. Driver FR's are often slightly different in reality from the manufacturers' curves and that's why using real in cabinet measurements is always the best way to design a speaker. But because I'm altering an existing design, what matters here is that the new simulated response of the double Tarkus (black) matches the simulated response of the original Tarkus (brown) which it does exceedingly well. That means that in reality, the measured FR will pretty much be the same as the original but just raised 6dB in sensitivity due to the doubling up of drivers.

3) Impedance minimums mean that a 4ohm amplifier is required. Not a problem for you.

4) Probably most important in the xo is that you'll need an iron core inductor with low resistance on the woofers. My sim was set at .36ohm.

Let me know if all of that makes sense to you.
 

Attachments

  • cabinet dimensions.jpg
    cabinet dimensions.jpg
    109.8 KB · Views: 108
  • summed FR.jpg
    summed FR.jpg
    119.2 KB · Views: 109
  • summed Z.jpg
    summed Z.jpg
    95.3 KB · Views: 109
  • xo schematic.jpg
    xo schematic.jpg
    42.7 KB · Views: 100
Let me know if all of that makes sense to you.

Everything makes perfect sense, and i think this will be the speakers that i build. Definitely in 3 different cabinets, for easier transportation. So pretty much the top and bottom woofer cabinets can be identical, and similar to the original Tarkus bottom cabinet.

The crossover for so many speakers is actually a lot easier than i thought.

Great looking project ahead. :D
 
-3db at 50Hz?

Yes, I noticed the same thing when I first started doing sims. How does F3 at 30Hz in the box modelling get to F3 of 50Hz in the xo modelling?

Two things (well actually 3):

Baffle step loss, which will vary with baffle width, will always drop (in SPL) what's happening at 30'ish Hz compared to what's happening at 100Hz.

And particularly with a 3-way, the bigger inductor on the woofer also tends to raise the woofer level somewhere in the 100Hz zone (depends on the driver's impedance), often allowing you to raise the overall SPL level of the design, albeit at the expense of some LF extension. If the effect is too egregious however or if the loss of LF is too hard to live with, some, like Troels for eg, will choose to kill that peaking with an impedance compensating LCR. Big values though so also a big expense. Here with the Peerless 10", this effect is relatively benign so an extra LCR isn't considered worth the trouble or the expense.

But you may have heard other people talk about also looking at F6 and F10 and there is good reason for it - the sims don't include boundary gain or room reinforcement effects. When you take those into account with this particular woofer, in-room F3 goes back down to the 30Hz neighborhood. This is also the reason it is often recommended that with a big woofer with a nice low F3, an alignment with a more gradual rolloff instead of a sharp knee is usually to be desired.
 
..... So pretty much the top and bottom woofer cabinets can be identical, and similar to the original Tarkus bottom cabinet....

As I said, cabinet design is up to you. Draw things out. Calculate net volumes. Figure out maybe what the maximum size and weights that you can live with are. Understand that depth is the variable that can be changed with the least amount of design consequences.

My thinking was that since the mid section doesn't need to be too large (~24L), it won't be as deep as the woofer sections (maybe only about 8"), so you could reduce the speakers' height by making the top woofer section into an L shape. See below for a rough sketch.

But if height isn't a concern, you could also make the mid section deeper (I'm actually a fan of oversized mid chambers with the extra volume in the back stuffed with high density insulation) with even some spare space in the back for the xo to be mounted externally (see Troel's big speaker designs for examples) and then indeed make the top and bottom woofer sections the same.

With your last post, I'm starting to wonder if some consideration for usage should be taken into account. With a desire for such high SPL's and mention of easy transportation, I've got the feeling these are almost intended as quasi-PA speakers or some serious party speakers perhaps. If that's the case, they just may end up in some larger rooms where boundary and room reinforcement effects aren't likely to be as strong and therefore the bass might end up a little weak. My suggestion if that is the case would be to tune the woofers lower and with a sharper rolloff (yes that means bigger cabinets) so you'll get enough LF extension and SPL out of them in the bigger rooms. Then in small rooms, run them with 1 woofer ported and the other woofer with the port stuffed so you get a more gradual combined rolloff to better match the extra LF boost you get in the smaller room. Just a suggestion anyway.....
 

Attachments

  • sectioned.jpg
    sectioned.jpg
    65.7 KB · Views: 82
Yes, I noticed the same thing when I first started doing sims. How does F3 at 30Hz in the box modelling get to F3 of 50Hz in the xo modelling?

Two things (well actually 3):

Baffle step loss, which will vary with baffle width, will always drop (in SPL) what's happening at 30'ish Hz compared to what's happening at 100Hz.

And particularly with a 3-way, the bigger inductor on the woofer also tends to raise the woofer level somewhere in the 100Hz zone (depends on the driver's impedance), often allowing you to raise the overall SPL level of the design, albeit at the expense of some LF extension. If the effect is too egregious however or if the loss of LF is too hard to live with, some, like Troels for eg, will choose to kill that peaking with an impedance compensating LCR. Big values though so also a big expense. Here with the Peerless 10", this effect is relatively benign so an extra LCR isn't considered worth the trouble or the expense.

But you may have heard other people talk about also looking at F6 and F10 and there is good reason for it - the sims don't include boundary gain or room reinforcement effects. When you take those into account with this particular woofer, in-room F3 goes back down to the 30Hz neighborhood. This is also the reason it is often recommended that with a big woofer with a nice low F3, an alignment with a more gradual rolloff instead of a sharp knee is usually to be desired.

Yes.
I think f3 around 35 Hz is realistic.
My kickbins also play around 15 Hz deeper indoor.
And 35 Hz is good enough for most types of music.
 
With your last post, I'm starting to wonder if some consideration for usage should be taken into account. With a desire for such high SPL's and mention of easy transportation, I've got the feeling these are almost intended as quasi-PA speakers or some serious party speakers perhaps. If that's the case, they just may end up in some larger rooms where boundary and room reinforcement effects aren't likely to be as strong and therefore the bass might end up a little weak.

Actually not, these will be home speakers, but i usually spend a lot of time at my remote garage working with my car, so i want that these speakers could fit in a car so i could move them to my garage. Both my room and my garage are tiny, 2 meters high, and just the size of a car. I just want as much spl as possible because i'm a metal fan.

I did not really understand what you meant with two ports for the bass cabinets, with one of the ports stuffed. Stuffed as in literally? Then why make the port at all?
 
Actually not, these will be home speakers, but i usually spend a lot of time at my remote garage working with my car, so i want that these speakers could fit in a car so i could move them to my garage. Both my room and my garage are tiny, 2 meters high, and just the size of a car. I just want as much spl as possible because i'm a metal fan.

ok, in that case you may want to stick with the size of the original Tarkus. Tmes 2 of course. But again, totally up to you.

I did not really understand what you meant with two ports for the bass cabinets, with one of the ports stuffed. Stuffed as in literally? Then why make the port at all?

With 2 woofers, you've got a couple of options with the boxes and ports.

If you put both woofers in the same box (2 x the original 62L net), 1 larger port will work. But 2 of the original size ports will also do the same job.

If you put each woofer in a separate box of 62L, then you must also use the original size port in each box as well.

The advantage of using the latter approach, besides making it easier to handle big cabinets, is that you now have 1 extra option in terms of tuning the bass response to the room size, the speaker placement (ie. near or far from walls) or to your own personal preferences.

So for example, if you put speakers with either single or multiple woofers but with just a single port in a small room and find the bass too boomy, you might try stuffing the port which indeed turns it into a sealed response. Maybe this will be satisfactory or maybe you'll find that this is now too little bass.

On the other hand, if you are using 2 woofers in 2 separate boxes with separate ports, you now also have the option of only stuffing 1 port if the bass in too boomy and so you will get a response in between the fully ported and the fully sealed ones. In other words, it just gives you a little bit more flexibility.

I'm not sure if it's clear so I'll just add that if you use the 2 woofers in 1 larger box and use 2 smaller ports, stuffing 1 port in this situation won't work the same way. The port tuning will be drastically changed and you'll end up with a very unsatisfactory result.

In your case you are going with 2 woofers, 2 boxes and 2 ports which is what I would have recommended even if they were all going inside 1 big cabinet. Because - flexibility. :)
 
Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.