DIY Surround speaker : please help

Status
Not open for further replies.
I am having 3 set of speakers left after av setup upgrade. Following are the specs (from onkyo website), both are having almost the same specs.

SKF-320F: 2 Way Bass Reflex Front Speakers (2 No.)
Impedance: 8 ohm
Maximum Input power: 100 W
Output sound pressure level: 76 dB/W/m
Frequency response: 70 Hz-50 kHz
Crossover frequency: 4.5 kHz
Woofer: 3-1/8 inch Cone
Tweeter: 1 inch Balanced Dome
Magnetically Shielded

SKF-320C: 2 Way bass reflex Center Speaker (1 No.)
Impedance: 8 ohm
Maximum Input power: 100 W
Output sound pressure level: 78 dB/W/m
Frequency response: 70 Hz-50 kHz
Crossover frequency: 4.5 kHz
Woofer: 3-1/8 inch Cone x 2
Tweeter: 1 inch Balanced Dome
Magnetically Shielded

My question is:
- Is it possible to use the drivers from these speakers and make a good bipolar / dipolar surrounds. Is direct firing or bipolar / dipolar the best for surrounds? I am a novice in electronics. Since I dont want to waste those speakers, i am going for diy path. Is it recommented to use those drivers for surrounds in a new box (bipolar / dipolar)? Else which driver is better for that and their availability, i am from Bangalore. Please help..
 
sure you can make some surrounds from those. If I read it right you have 4 of those 3" midbass drivers and 3 tweeters. Lots of different configurations could be made from these, so I will just suggest a few. One arrangement would be to have the midbass drivers firing to the sides and run full range (no crossover), with the tweeter on the front. This arrangement would have the tweeter facing the listeners and the midbass drivers facing toward the front and rear walls (with side mounted sides).

Alternatively, I would bet that these small speakers have a fairly full range to them, and you could argue that no tweeter is needed. In fact, Dr. Geddes has mentioned that his view on surrounds is that they should not be full range and suggests using 8" drivers run full range. The idea being that intentionally not having the higher frequencies helps to make the surrounds seem more diffuse, to disappear more, and create the impression of more space. It's inconsistent with my past world view on surround speaker design, but its worth a shot.

I recomend against using the tweeters in a dipole or bipole configuration, its not been my experience that this is very effective. You don't have enough tweeters, but in case this concerned you, I think its actually a good thing.

Yes I would make new boxes for this purpose, I'm not sure how you would keep and use the old ones.
 
Hi PJPoes

Hi PJ Poes,

Thanks for your reply. You are right, I am having four 3" mid/bass speakers (not sure if they are full range speakers or not) and three tweeters. They are from the Onkyo HTIB HT-S 570. I dont have any problem in building a new box for these. My purpose is to have a good upgrade to the existing surrounds (currently using the front satellites as surrounds, upgraded the fronts with towers). Could you please tell me what are the possible ways to make a good surrounds with these speakers (as you suggested, if required, i will not use the tweeters). Also whether it will a noticeable upgrade to the existing setup?

Since I am a novice in electronics and speaker building, i am confused about how to find the optimal volume or the shape for the boxes and also confused about the crossovers. The amp I am using is Onkyo HTR 320 (part of HTIB HT-s 570). Specs : http://www.onkyousa.com/model.cfm?m=HT-S570&class=Systems&p=s

Please help me .. 🙂


Thanks n Regs
Sajith Mathew
 
pjpoes said:
In fact, Dr. Geddes has mentioned that his view on surrounds is that they should not be full range and suggests using 8" drivers run full range. The idea being that intentionally not having the higher frequencies helps to make the surrounds seem more diffuse, to disappear more, and create the impression of more space. It's inconsistent with my past world view on surround speaker design, but its worth a shot.


The idea here is for small rooms where the surounds are pretty close to you. One of the "ques" for distance is HF content so a slow roll off of HF will give the impression of the source being farther away. The worst thing that you can do in a surround is to have a wide directivity tweeter (as virtually all of them are at the bottom of their passband) in a speaker only a few feet from you - this ruins the effect for me as the surround seems "in your face" and that is certainly not what the mix intended.

In a big room where the surrounds are ten feet or more away a direct radiating tweeter is fine, but not in a small room.
 
Just a comment on a solution I have used that worked for me.
We have a very small room at the moment and my surrounds are at our shoulders, by using the speaker stands at full height and turning the speakers to face the ceiling that directivity problem was moderated enough to use the existing speakers "AS IS" and is an excellent cost effective solution and if the speakers are rear ported as mine were just stuff the port
 
I assumed that the reason for the full range driver was to give a rolled off highend as a result of the psychoacoustic effect it has on distance.

My current surrounds (which I'm not overly happy with) use the M&K tweeter which is basically a cheapened version of the waveguide loaded tweeter made by peerless. I replaced the diaphragms and reinstalled them, but wasn't happy. I then made a waveguide adapter/insert that continued the guide out to be even with the edge of the enclosure (The tweeter was recessed), remeasured, redesigned the crossover, and things were much better. This created controlled dispersion of the highs above around 2800hz, not ideal, but better than it was. I've since added a small value cap, something like .47uf in parallel with the tweeter which creates a shallow roll-off and it has considerably increased their ability to disappear. The perception of depth has increased as well. This was my weekend project, and would consider it a partial proof of concept for Dr. Geddes idea. The speakers don't fit his design world view I don't think, but having improved the high frequency dispersion control and reducing high frequency output has pretty drastically improved the sense of depth.

My design is a sort of clone of the M&K tripole concept, but using mostly better drivers (they were still all cheap vifa and peerless buyouts for the most part). One issue I have found is that the side firing drivers are too high in level, don't have a smooth response, and the polar response is terrible. Given that they fire out of phase, I expected some of this, but I think the problem is that they are recessed into the cabinet quite considerably so the grills can fit flush. My first fix will be an insert made to fit into that cavity which will offer a flared shape, and I hope reduce diffraction problems. I suppose it will act a little bit like a waveguide, but would be too small to have any real effect, so, its for diffraction.

Ok but to help this poor chap out. Make the enclosures small, and keep them sealed. These are 3" drivers designed to work in relatively small enclosures. Without even knowing the T/S parameters I can guess pretty accurately that a box of between .05 and .1 cubic feet would be fine for a single driver, and probably still work for a pair, just increasing the system Q a little. Since we are talking about running them full range, there will be no crossover in this design, so its really a very simple setup. Simply connect the speakers terminals to the terminals you put in the enclosure, and you are done. If you decide to do a pair and run them in bipole or dipole, then wire the drivers in parallel (dipole would have the + connected to the - of the other driver, so that one is out of phase with the other). If you wanted to add more roll off (most 3" drivers, regardless of design as a midbass or fullrange, have very extended top end response and its often not pretty), then add a parallel cap of fairly small value. Like I did, use something in the .22 to .5uf range. If I had measurements of your drivers or at least could see the impedance plot I could give you more exacting values for the caps for the best roll off, but those will work. Small amount of series resistance can work too.

If your receiver can not handle a 4ohm load or the drivers are 4 or 6 ohm impedance to begin with, then don't wire them in parallel as suggested above. Wire them in series. If you want them in series and out of phase, then instead of having the positive of one go to the negative of the other, just have the positive of one go to the positive of the other.

This is way more than you probably wanted to know, but since I did the measurement, I thought I would share. If you mount to drivers back to back on opposite sides of a box, and then take measurements of the response and acoustic phase from a given distance roughly center between the two drivers, the drivers will not be in phase with each other. If the test I did they were already roughly 80 degrees out of phase with each other, and wiring them electrically out of phase did not put them 180 degrees out of phase with each other, but what turned out to be more like 240. The way to fix this is somewhat complicated and would require a ladder delay circuit, so don't worry about that. Just maybe nice to know I guess.
 
pjpoes said:
The perception of depth has increased as well. This was my weekend project, and would consider it a partial proof of concept for Dr. Geddes idea. The speakers don't fit his design world view I don't think, but having improved the high frequency dispersion control and reducing high frequency output has pretty drastically improved the sense of depth.


I think that your approach is fine. What I object to is the use of a small tweeter which is basically omni-directional at its lower end which then ends up right in your ear even if the speakers don't face you - which, of course, they never should in a small room. Your control of my objection is a good solution. It is a little more than I would bother with for surrounds, however, but if you have the time and money, go for it.
 
Hi all, Thanks for your help..

Thank you all for your help.

@PJPoes, Let me put what i understood from your message. I can use the two 3" mid/bass drivers for a surround speaker (without the tweeter) in a bipole or dipole box (shape : \_/, each speaker facing 45 degrees left and right) with box volume 1 cubic feet (28.32 liters) connected in series without any crossovers, right? Could you please send me the dimensions of the bipolar or dipole box required for this volume? You have mentioned Q will be high if 2 speakers are used with volume 1, what does it mean or how it will affect the performance and how to fix that (whether it will be fixed if I increase the volume, say 1.25 cubic feet)? If I use a normal 6"/8" woofer(s) and connected in series, whether the performance will be better than the suggested setup?
 
Its a typo error, .1 cu.ft (2.8 litres)

@All, I am not building the speaker box from the scratch. Its already in a company fitted satellite box with 2-way setup. Since I am not satisfied with the performance, I am looking for building a new box with those 3" mid/bass woofers and tweeter(if required).

@PJPoes, Can you reply for my prevoius post, regarding the Q issue and the dimensions of the box (\_/)
 
I'm sorry I didn't reply, I got busy with some life stuff and forgot to check back.

You could use 1 cubic foot, but thats really way too big. I would suggest using a speaker box that is .15 cubic feet and no bigger than .25 cubic feet for these two drivers. Even smaller is really fine, I mean, if I was doing it, I think I would probably go with roughly .05 cubic feet, and no, thats no typo, that means an internal dimension of 3x6x4 inches, which is plenty for one 3" driver, double that volume for two. That means .1 cubic feet for one, or .15 if you want a little wiggle room for parts and such.

Q refers to the QTS or total sysem quality and takes into account the mechanical and electrical "quality" or Q of the driver in a box. For your purposes, don't worry too much about this, but as an example, a 1 cubic foot box with two 3" drivers will likely have a system Q equal to that of the drivers qts, similar to being in an infinite baffle. This means that the box creates no "air cushion" to help control cone movement at lower frequencies. With such small drivers (probably limited to 100-150hz), I would consider a higher q box which will bloat the bottom end a little (not always a bad thing), but also add some excursion protection in the form of an acoustic suspension.
 
Status
Not open for further replies.