Major College Project

[garybeard333]

Hi there,

My name is Gary Beard and i have just discovered this website and the excellent help everyone on the forums seems to be willing to give. I am currently studying my last year of a BTEC National Diploma in music technology and am currently working on my major project.

I have decided to base my project on speaker cabinet designs and to test how different shapes of cabinets have an effect of the sound produced. In order to do this effectively i need to try and keep the frequency response as flat as possible.

I have already purchased HiVi Research M6a Mid-Bass Drivers (http://www.swanspeaker.com/product/htm/view.asp?id=78) and HiVi Research TM1A Mid & Tweeter combination unit (http://www.swanspeaker.com/product/htm/view.asp?id=21).
I did not realise at the time of purchase that the TM1A's are 5 ohms whereas the M6a's are 8 ohms, will this be a problem? will the resulting resistance not be 8 ohms?, will there be an issue with balancing the bass, mid and high of the speaker? and is there any way to fix any issues that may result?

I was considering purchasing Visaton HW3/120NG-8 (http://www.visaton.com/en/chassis_zu...ichen/286.html) 3-way cross overs, as i am not particularly knowledge about electronic components. Do you think this would a good choice?
I would be willing to make my own if i was told how and what components to use, as i have good soldering skills, perhaps you guys might have some suggestions for me?

Each of my cabinet designs that I have decided to make for my experiment, which are rectangular, triangular, and trapezoid, will have the same volume, height and baffle width. I have chosen these shapes as they are the easiest I can build with the tools I have available to me. Having the same baffle width will look ugly on the rectangular and trapezoid cabinets but this illuminates another variable, how it looks is not that important for this experiment.

Eliminating as many variables as possible is essential, and as such I am in the process of creating an Excel document that will calculate the dimensions of each shape of cabinet, keeping the height and volume constant in each design.

This is where my problem lies with the cross-over, since I will need a "generic" cross-over design that will work relatively well in each design. This is necessary in order to illuminate more variables since it is my intention to measure the effect that the shape of each cabinet has on the sound produced.
I intend to use the same crossover in each cabinet, as using a different cross-over in each cabinet will of course throw my results as another variable will have been added.

I understand that you guys have no reason to spend the time to help me, but if you could find some time to do so i would be immensely grateful, thank you very much in advance, i look forward to a reply.

[ThomasW]

SpeakerBuilder/AudioXpress magazine has a comprehensive article covering the effects of different shaped enclosures...IIRC it was Lubos R. Palounek, “Enclosure Shapes And Volumes,” Speaker Builder 3/88.

Might be something for you to reference...

If you're not interested in sound quality only measured effects, using a generic crossover will work

Software programs like 'The edge' and or 'Baffle diffraction simulator' would also be something to check out

[evilskillit]

Well I don't know anywhere near as much about this as most of the other people here, and I might be a bit late in saying this as you have already bought drivers. However the best thing for you to do might be to get a single full range driver and eliminate the crossovers all together. I don't know the scope of your project, and maybe going with a design that simple might affect your grade. However a generic crossover will probably not compliment the speakers at all tho at least they'll sound bad in the same way from one enclosure to the next if you use the same crossover I suppose.

[garybeard333]

Thanks for the replys

Thank you for your replies.
For the purpose of this project sound quality is not paramount but i would like the frequency response to be as flat as possible.
A full range speaker would probably have been a good choice, however i want to make use of these drivers once the project is complete and i can't see that i would use full range speakers for my own uses. I have a unique cabinet design that i would like to make once i have complete this project. I will post some images and details in a few weeks for you guys to look at.
what i really need is a cross-over design designed for the drivers that i have already purchased, and that would give a relatively flat response in all three cabinets. Each cabinet will be the exact same height, the same volume, and the baffle will be the same size on each. This means that the only variable would be the shape of the cabinet.

[---k---]

Vance Dickenson's Loudspeaker Cookbook also has results of several experiments with different baffle shapes.

I agree with the others that a fullrange driver would have been better. A TB W4-1347 (the midrage in the Statements & LineUp) would have made an excellent fullrange choice. You could use the notch filter from Jed's D4 and have very listenable speakers.

With the drivers you have, and for experiment purposes, I would think a 2-way would be sufficient. Baffle diffraction is primary an upper midrange, tweeter issue. I would just use your TM1A and call it a day.

You're not likely to get a custom 3-way crossover designed for those drivers here. No one here has measured them and has the data to do it right.

[garybeard333]

Confusion

I think i have perhaps explained this wrong. It isnt the baffle that i am changing the shape of, the baffle will be the same in height, width, depth and shape on all three designs. What will be changing is the cabinet shape. Looking down at the top of the speaker cabinets they would be rectangular, triangular and trapezoid, from the front they would look identical.
I have already completed the designs for the cabinets, im just stuck on the cross overs.
A problem i could have with a full range speaker is the fact that the frequency range would be limited, most notably the lowest and highest frequencies, and im trying to measure what effect the internal cabinet shape has on frequency response and resonance. I know that i can find information on this an probably the results, but thats not the point of doing the project. I have to set my self an investigative task and find the results, this is what i will be graded on, the fact that i have found this information out by doing my own experiment.
Sorry for the confusion, i may have rushed my messages slightly as i have a tonne of stuff on today.
Thanks again for your interest and suggestions guys .

[Operandi]

Cabinet shape won't affect high and low frequencies (somebody will correct me if I'm wrong) so a full range driver is a good suggestion. Jed's F4 TB W4-1347 makes for an very detailed, excellent sounding speaker.

Regardless If you are looking for a flat response I think it would be best to start with a proven speaker that is known to be flat then you can modify the shape of the cabinet from there.

[garybeard333]

What i could really use help with is the cross-over.
if some one could give me a good design for a cross-over for use with the drivers i have (i cant buy more, i'm on a pretty tight budget), i would be really grateful.
I'm good with a soldering iron so i someone has the time and ability to tell me what components to buy and perhaps make me a schematic i could buy the parts and then build it myself. It doesn't need to be absolutely perfect but as close as possible would be great. Its either that or i buy a pre-built cross-over.
It may be important to note that TM1A is a unit comprised of a soft dome tweeter and a soft dome mid-range
The data for the drivers i have can be found here:

http://www.swanspeaker.com/product/htm/view.asp?id=21

and here:

http://www.swanspeaker.com/product/htm/view.asp?id=78

The data for the separate drivers in the TM1A assembly can be found here:

http://www.swanspeaker.com/product/htm/view.asp?id=37

and here:

http://www.swanspeaker.com/product/htm/view.asp?id=23

Again thanks for your hep and suggestions guys.

[Teran]

Are you going to be measuring the finished speakers? If so, can you mount the drivers in a box, measure them, and post the results? Then, someone may be able to help with a crossover.

[mx2]

How about an active crossover such as one of the cheap Behringer stereo 3-ways? A woofer which plays flat higher up and without the breakup would make it feasible and simple. Remember to make the front baffles the same...to eliminate that variable too

Btw, I just joined this forum too...about to start building my first pair of speakers. The guys here do seem very friendly and reasonable. :B

Thomas

[Undefinition]

What i could really use help with is the cross-over.
if some one could give me a good design for a cross-over for use with the drivers i have (i cant buy more, i'm on a pretty tight budget), i would be really grateful.
I'm good with a soldering iron so i someone has the time and ability to tell me what components to buy and perhaps make me a schematic i could buy the parts and then build it myself. It doesn't need to be absolutely perfect but as close as possible would be great. Its either that or i buy a pre-built cross-over.

If you are absolutely married to that TM1A thing, then I suggest you post where you live, and maybe a DIYer in your area can measure the speaker and design a good crossover for it.

The other alternative would be to sell that TM1A and buy just a good, solid tweeter with a low Fs: Seas 27TDFC, Dayton RS28a, Vifa DX25 come to mind. Then, maybe someone out there can do a simmed crossover that should get you in the ballpark. But since the TM1A is pretty much discontinued, and no one's really used it, it's hard for any of us to even guesstimate how it would behave.

[ThomasW]

His IP addy is in the UK

[garybeard333]

Yes i am in the UK - specifically Chippenham, in Wiltshire.