The box, as best I could calculate it is supposed to be 170 liters (6 cubic feet).
The two ports are flared one end, 3.95" in internal diameter, and 6" long from the lip of the flare to the inside end (total length).
The box has a impedance resonance at 28 Hz when measured with a sine wave swept through a 1K resistor so I can measure current across the resistor.
Interior stuffing is 1" polyfill stapled to all walls.
Curiously, when I plug that data into BassBox Pro my resonance point does not line up with the observed resonance of 28 Hz. The box acts as if the internal volume is 235 liters (8.3 cubic feet).
Calculating the internal volume of the cabinet is tricky. It is a truncated pyramid and only two walls in the whole cabinet are parallel and they are the smallest walls! Great for reducing standing waves, but a nightmare to calculate.
I'll try recalculating the internal volume again.
is this better than using other methods?
fs and Q is quick and easy but Vas using a small sealed box or added mass is more of a problem
I did a fast scan of Dickason's book and a simple sweep with an inline resistor can give you free air resonance.
The other parameters are possible, too, with more work.
The question is, what are the really important or critical parameters I need to focus on for this evaluation?
I can also use Fuzzmeasure (I have a Mac) to get T/S parameters, but I think the calculations produced by that program are a little suspect.
Total volume - Less the volume displaced by driver ~6L (0.2ft3) and the volume of the (x2) ducts.
Calculate Enclosure Volume
You can also use some material to fill the inside and measure it outside as someone did in diyAudio. (Protect the woofers in this case)
It really isn't that easy. Look at this cutaway of my cabinet. There are corner braces (mostly 3/4" by 1-1/2" pine) that are not shown that are glued into each corner where two panels meet.
No two panels meet at a right angle.
An externally hosted image should be here but it was not working when we last tested it.
Great for combating internal standing waves, but harder to calculate the true internal volume. I also need to verify that the internal braces shown are the right dimension, too.
Front view:
An externally hosted image should be here but it was not working when we last tested it.
Last edited:
Okay, I did the best estimate in CAD, based on everything I know and it comes out to 6.68 cubic feet.
This is subtracting the driver and ports and all internal bracing.
It does not include the crossover or any volume the filling takes up.
The crossover is probably less than 1/10 of a cubic foot, so 6.60 cubic feet (187 liters) is probably realistic.
Tested 2235H Woofer Fs = 22 Hz.
Rechecked the cabinet and the cabinet is tuned to 28 Hz.
I can't account for the mismatch for the cabinet tuning, port size and length, and internal volume. It doesn't match the predicted Fb for the cabinet with a volume of 6.6 cubic feet and the port diameters and lengths that I have using BassBox Pro.
BassBox Pro predicts an Fb of 31 Hz with F3 at 33 Hz for the box volume and port sizes. Maybe that is as close as the program gets, I don't know.
Rechecked the cabinet and the cabinet is tuned to 28 Hz.
I can't account for the mismatch for the cabinet tuning, port size and length, and internal volume. It doesn't match the predicted Fb for the cabinet with a volume of 6.6 cubic feet and the port diameters and lengths that I have using BassBox Pro.
BassBox Pro predicts an Fb of 31 Hz with F3 at 33 Hz for the box volume and port sizes. Maybe that is as close as the program gets, I don't know.
Last edited:
Loren, I imagine that you discounted Mid enclosure volume (vol?). I would not like it (Audax) working as a passive radiator?!
Yes, the mid is totally in its own enclosure and that enclosure volume is not part of the 6.6 cubic foot volume calculation I cited, which is only for the woofer.
The dark gray rendering is the interior space of the mid enclosure:
An externally hosted image should be here but it was not working when we last tested it.
I plan to, but my setup needs more work. I need a quiet place to perform the measurements and a better setup.
Getting the Fs at least tells me that the driver isn't grossly wrong.
I also plan to move the cabinet into another room; our music room where we have all our acoustic instruments and try running another sweep there. The room is a little smaller, but it is carpeted and has more objects that create a less live room. If there is a significant change in the knee where the bass rolls off in that room I can probably credit the lack of bass to the room acoustics.
I would also try to sweep it outside, but in Florida there is always a breeze and therefore not very quiet.
That's a bit high for them and closer to a 2234 where the FS is 23Hz. Try modeling a 2234 and see it what you are getting makes more sense. Could be stiff surrounds as well if they have been refoamed. My new recones were sub 20Hz right out of the box.
http://www.lansingheritage.org/images/jbl/reference/notes/tech1-3a/page09.jpg
Rob
That's a bit high for them and closer to a 2234 where the FS is 23Hz. Try modeling a 2234 and see it what you are getting makes more sense. Could be stiff surrounds as well if they have been refoamed. My new recones were sub 20Hz right out of the box.
http://www.lansingheritage.org/images/jbl/reference/notes/tech1-3a/page09.jpg
Rob
Hmm, taking your lead, doesn't this look interesting! Red trace is the 2235H and the cyan is a 2234H...
An externally hosted image should be here but it was not working when we last tested it.
I bought these and I have no idea who did the recone and with what, but my guess is that they used those generic eBay recone kits and they do not have the mass ring that makes it a true 2235H.
Your sim has an error in data entry.
What's a mass ring anyway? Can it be seen in a photo at the JBL site.
I think if they reconed with a non JBL part or a cheap VC then sensitivity would suffer, maybe that's what happening.
Measurement of Q would be revealing and my be more accurate than Fs. uses exact same setup as fs
find BandWidth 0,707 down from peak at fs. Q = fs/BW
Measure Vas in your box with all air holes plugged.
What's a mass ring anyway? Can it be seen in a photo at the JBL site.
I think if they reconed with a non JBL part or a cheap VC then sensitivity would suffer, maybe that's what happening.
Measurement of Q would be revealing and my be more accurate than Fs. uses exact same setup as fs
find BandWidth 0,707 down from peak at fs. Q = fs/BW
Measure Vas in your box with all air holes plugged.
What's the error?
Mass ring is just what it sounds like. It is an extra mass added to the end of the voice coil under the dust cap that drops the Fs.
It is not something that you can buy with a generic recone, so when someone recones a 2235H with a non-standard JBL kit you get a JBL 2234H, not a 2235H. This is very common thing and theer are many eBay 2235Hs out there that are portrayed as a reconed 2235H, but they are not. I think I got stung when I bought them a long time ago, but I had no idea back then about the difference.
Correct recones cost about $250 per driver to do with JBL parts. You can get it reconed for $150 using a generic kit, but they don't have the mass ring. Live and learn.
I'll do a complete work up of the driver at some point and compare it with a 2234H, but I would bet that is what happened here.
It's a 50 gram ring mounted inside the top of the Voice Coil. Can't see it without talking off the dome. The JBL kits comes with it loose. If you want to make a 2235 you put it in. To make a 2234 you leave it out. You use the same kit for both drivers.
You may be able to get your hands on a pair of them. All you need to do is remove the dustcap and put them in. Try asking over on Lansing Heritage.
Rob
Attachments
Last edited:
Vas
I would assume if an Ebay seller doesnt have documentation about the re-cone regarding JBL parts, then its not up to snuff. The value of JBL parts would be a proper VC with the rectangular wire. I dont think that the generic coils could offer that. The mass cone ring being there is not as important IMO. FWIW There are many other good ways to add 50 grams to a cone. Which may be useful to test Vas.
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.