speedbump Posted February 1, 2002 Share Posted February 1, 2002 Hello all, I am in the process of designing my own subwoofer from the ground up, with the exception of building my own driver. I am building a Class D switching amplifier based on the National Semiconductor LM4651 & LM4652 Overture design. This should allow me to push approximately 125W RMS continuous into a 4 ohm load. I have looked at several drivers, including the Shiva Mark II, The Dayton Titanic, The Dayton Titanic II, and the Peerless SWR308. After plugging in the Thiele Small parameters (most of them anyway), It looks as though I am going to use the Dayton Titanic 1200. It seems to perform the best given my amplifier size, but requires a larger enclosure than the other drivers do. Here's the rub. I have very little experience with the software (WinISD) I am using to get enclosure size, port width and length, MaxSPL, Etcetera....and I would love it if one of you more experienced SW builders could verify my data. For the Dayton Titanic 1200 Subwoofer my data is as follows. QTS = .41 QMS = 8.23 QES = .43 VAS = 280.2 liters XMAX = .014 meters FS = 16.26 Hz RE = 3.66 Ohms LE = 1.96 Z = 4 Ohms SPL = 90dB PE = 350 Watts BL = 9.47 DIA = .24 meters SD = .046 square meters I would be using an enclosure with a volume of approximately 8 cubic feet, with 2 4" ports with a length of 22" each. This gives me a theoretical -3db point of approximately 16 Hz and a max SPL of 111db at 125 W/1 m. Does this sound right? I would use a larger amp, but I am building this one as an engineering project for school, so I may as well use it for now. Enclosure would be made of 3/4" MDF, ports would be PVC. Any help you could provide is greatly appreciated. Alternate driver selections for higher SPL and smaller enclosure size are also welcome. Thanks, Speed ------------------ Klipsch Chorus I Mains Klipsch Academy Center Klipsch Forte II surrounds Marantz SR19EX Receiver Yamaha 5 CD Changer Mitsubishi WS-55807 55" HDTV Playstation 2 for games/DVD This message has been edited by speedbump on 02-01-2002 at 04:03 PM Quote Link to comment Share on other sites More sharing options...
Audio Flynn Posted February 2, 2002 Share Posted February 2, 2002 I wish I had your time to look at that kind of detail. I do not have the answers but here are a few thoughts. 1. Parts Express is supposed to have 800 number support for design question if you buy from them, I heard. 2. Consider sealed design like a copy of an M&K design if you like music more than low frequeny capability. Quote Link to comment Share on other sites More sharing options...
djk Posted February 2, 2002 Share Posted February 2, 2002 "Any help you could provide is greatly appreciated. Alternate driver selections for higher SPL and smaller enclosure size are also welcome." Your grade for comprehension is 0.The relationship beween box size/cut-off frequency and SPL is fixed, ie if you decrease box size by 50% the SPL will decrease by 3dB.The relationship beween cut-off frequency and size/SPL is proportional to the cube of the cut-off frequency, ie going from 40hz to 20hz is a 8:1 ratio or 9dB.An easily visualized commercial example of these laws would be the Sunfire vs your idea, for roughly the same cut-off frequency we are looking at 0.4 cu ft vs 8 cu ft or 13dB difference.The Sunfire comes with 2.7KW amplifier power, 13dB less would be 135W.So you are in the ballpark anyway.As a side point, if this is an engineering project for a grade why should you be given a grade? Lifting a design from the National Semiconductor data book doesn't qualify as 'an engineering project for school' anymore than copying from Cliff Notes constitutes a 'book report'. Quote Link to comment Share on other sites More sharing options...
speedbump Posted February 2, 2002 Author Share Posted February 2, 2002 djk, First and foremost, let me apologize for being much less experienced in speaker design than you are. This in itself warrants self loathing on a grand scale. Add to this great transgression the audacity to use a reference design from a semiconductor manufacturer, for a class whose main purpose is to teach circuit board layout and construction techniques using the Cadence software package, and you really have me pegged. The entire class must build this circuit, and I will pass your recommendations for the Purdue syllabus on to those in charge. The calculations for enclosure size were based on what little I had learned about the software I was using, the size of the amplifier I was required to build, and what I thought looked like a flat gain curve. Thanks for your help..... Speed ------------------ Klipsch Chorus I Mains Klipsch Academy Center Klipsch Forte II surrounds Marantz SR19EX Receiver Yamaha 5 CD Changer Mitsubishi WS-55807 55" HDTV Playstation 2 for games/DVD This message has been edited by speedbump on 02-02-2002 at 02:24 PM Quote Link to comment Share on other sites More sharing options...
speedbump Posted February 2, 2002 Author Share Posted February 2, 2002 I have always read DJK's posts in the past and found them informative. I am disappointed that he felt the need to call my scholastic integrity into question. If this were a design project, rather than an arbitrary circuit used to teach schematic capture and layout software, I would agree whole-heartedly that his "Cliff's notes" analogy would apply. In addition, I am the first one to admit that I have no idea what I am doing when it comes to enclosure design, that's why I asked for help in the first place. **rant over** So let's see if I am catching on.... If enclosure size is halved, SPL is decreased by 3dB. This 3dB drop could then be compensated for by doubling amplifier power? I'm not sure if I am getting this next part. Is the 8:1 ratio you mentioned the increase in size for an enclosure tuned to 20Hz as opposed to an enclosure tuned to 40Hz? 20Hz^3 = 8000 40Hz^3 = 64000 64000:8000 = 8:1 If this is what was meant, does that mean I could reduce the size of my original enclosure to 4 cubic feet from 8 cubic feet if I change my cut off frequency to 20 Hz? 16.1Hz^3 = 4096 20Hz^3 = 8000 8000:4096 = 1.96:1 8 ft^3/1.96 = 4.08 ft^3 If I type the new frequency and enclosure size into WinISD, my gain graph grows a little hump from about 28Hz up to over 100Hz. It was my impression that it was desirable to keep this graph as flat as possible, which is how I got my first numbers. I have attached the new graph as well as the old one to show the difference. Port length is a bit fuzzy as well. I am going by the "Vent mach" number in winISD. This seems like the wrong way to go about this, but I have not found any good info on this subject yet. Thanks again to all that have helped. Speed ------------------ Klipsch Chorus I Mains Klipsch Academy Center Klipsch Forte II surrounds Marantz SR19EX Receiver Yamaha 5 CD Changer Mitsubishi WS-55807 55" HDTV Playstation 2 for games/DVD Quote Link to comment Share on other sites More sharing options...
speedbump Posted February 2, 2002 Author Share Posted February 2, 2002 mdeneen, Thanks for your clarifications. There are many things I am sure I am not taking into account, but I guess that's part of the learning process. I thought I read somewhere that a good "middle of the road" tuning for a sub enclosure falls somewhere between the two graphs I posted (around .65 to .70). Your remark seems to verify this. Something else that sticks in my head is that "room gain" tends to add quite a bit of gain to the frequencies around FS. Wouldn't that make a flatter gain curve more desirable? If there is one bit of information I have gained in reading this post, it's that I need to do a bit more reading. Any and all comments on the subject are appreciated. ------------------ Klipsch Chorus I Mains Klipsch Academy Center Klipsch Forte II surrounds Marantz SR19EX Receiver Yamaha 5 CD Changer Mitsubishi WS-55807 55" HDTV Playstation 2 for games/DVD This message has been edited by speedbump on 02-02-2002 at 11:59 PM Quote Link to comment Share on other sites More sharing options...
djk Posted February 3, 2002 Share Posted February 3, 2002 Ahh, the truth comes out.It's a CAD (glorified drafting) class.So you are doing the mechanical engineering on this project rather than the electrical engineering.Are you planning on measuring EMI and RFI on your finished project? AudioSource had to stop building their switching amp for these reasons. "Alternate driver selections for higher SPL and smaller enclosure size are also welcome" This statement indicated a lack of understanding on how things work.Cut-off frequency, efficency, and box size are locked together.You did pick up on this quickly, and while true that raising the cut-off frequency 1/3 octave will halve the box size, it will need a different driver.In general the equation for vented box size may be expressed as (Qts^2*Vas)*4.5 = Vb and system cut-off is roughly Fc = Fb = Fs with (Fs/Qts)*.312 = Fc also being a good rule of thumb.One a driver is chosen there are really only a couple of choices, sealed or vented.Just changing the box size will mainly cause a change in the flatness of the frequency response.Looking at the 295-400 (Titanic) it looks like 8 cuft F3=17hz, 4 cuft F3=22.3hz with 1.2dB of peaking around 40hz.It should be mentioned that the 90dB SPL figure on the 295-400 is for 2.83V, not 1W.So if it was -3dB at 16hz that would be 84dB + 21dB (125W) or 105dB, not 111 (no room gain). Quote Link to comment Share on other sites More sharing options...
speedbump Posted February 3, 2002 Author Share Posted February 3, 2002 djk, Yes, I am doing the board layout, enclosure design, power supply circuitry, some active filtering, and other additional tidbits for the amplifier portion of this project. I wasn't trying to confuse anyone. As far as EMI and RFI go...not on this project. It's only a two credit hour class, and I have 4 other classes to be concerned with. I'm just trying to make the best of a situation..... That was a good catch on the 2.83 V/1m rather than 1W/1m. There is a way to change that number on WinISD and I must have done it. quote: and while true that raising the cut-off frequency 1/3 octave will halve the box size, it will need a different driver Why is a different driver necessary in this instance? Is this because of the Fs of the driver? Would it be accurate to say that the enclosure should always be tuned to Fs for whatever driver is to be used in said enclosure, or does this just provide peak efficiency? Is there any wiggle room here? Speed ------------------ Klipsch Chorus I Mains Klipsch Academy Center Klipsch Forte II surrounds Marantz SR19EX Receiver Yamaha 5 CD Changer Mitsubishi WS-55807 55" HDTV Playstation 2 for games/DVD Quote Link to comment Share on other sites More sharing options...
speedbump Posted February 3, 2002 Author Share Posted February 3, 2002 Anybody know how to calculate size and weighting of passive radiators? ------------------ Klipsch Chorus I Mains Klipsch Academy Center Klipsch Forte II surrounds Marantz SR19EX Receiver Yamaha 5 CD Changer Mitsubishi WS-55807 55" HDTV Playstation 2 for games/DVD Quote Link to comment Share on other sites More sharing options...
Recommended Posts
Join the conversation
You can post now and register later. If you have an account, sign in now to post with your account.
Note: Your post will require moderator approval before it will be visible.