First: I want to say thanks for all the help everyone here has given.
Second: I know this has probably been covered amillion times here, but here goes.
I changed out my MTX sw2 12" driver with a Dayton 12" DVC. For the time being I'm using the SW2 incloser as a test with WINisd. Using WINisd, I put in all the parameters of the driver, the measurements, airspace of the box and the tuning I would like to tune the sub to. WINisd gave me a nice little graph of what the fr curve should look like and specs of the port. The SW2 has two 3" ports about 3" long and I'm driving it with my Hafler DH-200. I plugged one the ports and added the correct length to the other, according to WINisd's specs.
Here is the problem. According to WINisd 1 watt\1m I should get certain db readings at certain frequencies. I calibrated my sub via the pink noise on my Onkyo AVR. I sat about 3ft from the sub and played my test tone cd (10hz-120hz). The readings I received from my Radio Shack SPL meter were way off from the readings of WINisd and I know it wasnt the spl meter because I could hear it with my own ears. So I moved locations and sat in my normal seating position, same problem. The spike I get @ 50hz and 80hz from the spl meter is much higher than the graph shows on WINisd. My xover setting is @ 80hz. The roll off is quite abrupt above 80hz, so thats normal. The graph I plotted was nothing like the WINisd graph. For example: 90 dbs is my ref point.
80hz 104dbs (this is about where my Polk 7's started coming in)
If you put that on a graph , the line is all over the place. This cant be right. I did this in 3 different positions of the living room. All positions came out with a similar graph.
Did I do something wrong?
Just some things to consider:
A simulation will give a perfect smooth or flat response while this will never be the case in practice. The ripple is depending on many factors. For instance a flat line in WinISD Pro could well mean a ripple of +/- 3 dB or more.
Did you measure in your room? If so, the room will have a large to very large influence on your measurements (resonances/nodes), measuring outside is much more realistic towards the simulation in that way (looks like you have one of your room resonances around 50 Hz). Did you simulate the use of your filter on the woofer? Is the Shack meter calibrated? How do you know your applied power?
With kind regards Johan
Johan, simulate the filter? I'm not quite sure what you mean.
I know there is alot more to it than what I did. I calibrated my whole HT with the spl meter and it came out excellent. I guess my question is:
What can I do, besides changing positions of the sub or myself (WAF), to the box to get a better responce? I also tried plugging both ports, but I got a very low db reading and the spikes were still there. I guess I could stuff the sh$t out of the box itself. I guess all those little things could help out.
Any pointers would be appreciated.
I thought upgrading to a Dayton would give me a much better responce, I guess not. I also thought that 2.5cuft would be plenty of airspace for the time being. I will be building a new larger sub box for the Dayton, but I wont be listening to it outside...LOL.
Let me ask this. At 90dbs, since the sub is 89dbs efficient, I should be at 1watt @1m (3ft) right? Thats how I judged the wattage.
I gues I was hoping for the more relaxed curve, a curve more like what WINisd gave me.
I do appreciate the help.
Here are the specs of the Dayton 12". 2.5cuft airspace and one 3" port 11" long.
Should I shorten the port or lengthen it to reduce the spikes? I tuned the sub to 21.7 hz, thats how I came up with approx 11" for the port.
Specifications: (Note: All specifications are with voice coils connected in parallel.) Power handling: 350 watts RMS/per coil, 600 watts total * Voice coil diameter: 2" * Voice coil inductance: 1.81 mH * Nominal impedance: 8 ohm per coil / 4 ohm total * DC resistance: 2.69 ohms * Frequency range: 20-450 Hz * Magnet weight: 112 oz. * Fs: 21.7 Hz * SPL: 87.4 dB 1W/1m, 90.4 dB @ 2.83V/1m * Vas: 4.25 cu. ft. * Qms: 12.53 * Qes: .38 * Qts: .37 * Xmax: 15.1mm * Net weight: 18 lbs. * Dimensions: Overall Diameter: 12-1/4", Cutout Diameter: 11", Mounting Depth: 6-3/8", Magnet Diameter: 6-1/2".
A simulation, like WinISD Pro, is made in half space. Outside, on flat ground, open space, grass/soil is relatively close to this half space condition. Getting your graph like your simulation would require the conditions (matrix), to be the same as far as possible. Your room is not(hing like that), so just don't expect similair results.
If you measure outside and your graph is still giving spikes at 50 and 80 Hz it's speaker/enclosure related. If not, it's room related and changing your port and/or enclosure won't change your room.
You can adjust electronically (filter, EQ of some sort), use a basstrap (no personal experience) or use the as-long-as-it-sounds-good-approach. There might be other ways I'm not aware of.
With simulating the filter I mean simulate the effect of the filter in WinISD Pro a7.
All together I would concentrate more on getting the graph to be relatively alike (same shape) rather than absolutely.
With kind regards Johan
Might be of interest
with kind regards Johan
I will do some test with stuffing and port sizes this weekend. I think if I stuff the (you know what ) out of it, I will get rid of most of that "chamber" sound. It might even lower the spikes abit. Instead of my "custom" made pvc port, I'm buying a 3" and 4" dia port from parts express and see wich one best suites this box.
Your right though, close is what I'm aiming for, not excact.
Thanks alot for your help,
|All times are GMT. The time now is 07:39 AM.|
vBulletin Optimisation provided by vB Optimise (Pro) - vBulletin Mods & Addons Copyright © 2014 DragonByte Technologies Ltd.
Copyright ©1999-2014 diyAudio