PDA

View Full Version : Dented tweeter (again)



Andyoz
11-04-2008, 06:12 AM
I know it's been covered here before, but I'd like anyones thoughts on how successful I may be at extracting this small dent out of LSR-25P tweeter.

I have heard about vacuum cleaners, etc.

Someone also recommend getting some strong sticky tape and trying to use that to pull it out.

What actually works?

http://i9.ebayimg.com/05/i/001/17/21/2388_1.JPG

coherent_guy
11-04-2008, 09:46 AM
The vacuum cleaner technique is dangerous on tweeters IMO, due to their delicate nature. Any decent amount of suction could pull the entire diaphragm to much and damage the surround/suspension, if not rip the diaphragm out of the gap/pole piece. The vacuum technique lends itself more to woofer dust caps than tweeters, and even there it is dangerous.

I'm wondering if using a drinking straw attached somehow to a vacuum cleaner nozzle and allowing some air leakage to decrease the amount of suction could work. This technique with human-powered suction is much simpler and likely would work better overall.

Imagine the thin, rectangular shaped attachment nozzle provided with some vacuums for doing corners. Insert the straw in one corner of the opening, hold it in place with tape, and using tape or ones fingers to block the rest of the opening to vary the amount of suction, and CAREFULLY put the straw over the affected area. Rather unwieldly of course, but using a long piece of thin plastic tubing like air line for aquarium pumps in place of the straw would solve that. Now that I've channeled McGyver, I'm thinking human-powered is preferable.

The tape technique has worked for me in the past. Masking tape is good since it does not leave much if any residue. I put some on the eraser end of a pencil, the tape folded over to stick to the eraser, and then trimmed the tape down to an appropriate size. Double sided tape would make things easier, like the tape used to attach the clear plastic film over windows to keep out the cold. I would hesitate to use duct tape, although it is very sticky but could leave residue, it might be Ok if used carefully. If you could get some tape to stick to the end of a 1/8" - 3/16" wooden dowel and trim the tape down to a small spot you might have a decent tool for the job.

I've read that misshaped domes make no difference in performance, but I wonder about that. I'm guessing that the dispersion pattern might be affected by dents, but I have no evidence that is true.

I hate to see dented domes, it just is so wrong to me. A curse upon those whose fingers have purposefully ruined the beautiful perfection of the curve of a dome :die:

Andyoz
11-04-2008, 10:07 AM
Thanks, I like the straw idea.

Hoerninger
11-04-2008, 10:21 AM
I'm guessing that the dispersion pattern might be affected by dents, but I have no evidence that is true.

I hate to see dented domes, it just is so wrong to me.
As long as the dimension of the damage "is small against" the wavelength it does not matter (first estimation 1/10 of the wavelenght).
But it is nasty to have such a dome, it is too visable.
___________
Peter

Robh3606
11-04-2008, 10:28 AM
Someone also recommend getting some strong sticky tape and trying to use that to pull it out.

I would be concerned that there may be some "loose" aquaplas that may be pulled free and make it look worse from a cosmetic point of view. You might want to go in there with a small paint brush and cover the exposed titanium with some matte black paint.

Rob:)

greyhound
11-04-2008, 11:41 AM
cant you take the driver out and push it from behind?

coherent_guy
11-04-2008, 12:04 PM
As long as the dimension of the damage "is small against" the wavelength it does not matter (first estimation 1/10 of the wavelenght).
But it is nasty to have such a dome, it is too visable.
___________
Peter

Thanks Peter for the explanation, although I'm not understanding what is meant by "first estimation 1/10 of the wavelength". Do you mean if a dent is greater than 1/10 of the wavelength, then there may be a problem?

The wavelengths of various frequencies appropriate to tweeters:

2kHz: 6.752" or 17.15cm

3kHz: 4.5013" or 11.4333cm

4kHz: 3.376" or 8.575cm

5kHz: 2.7008" or 6.86cm

7.5kHz: 1.8005" or 4.5733cm

10kHz: 1.3504" or 3.43cm

15kHz: 0.9003" or 2.2867cm

20kHz: 0.6752" or 1.715cm

25kHz: 0.5402" or 1.372cm

30kHz: 0.4501" or 1.1433cm

A wavelength of 1" corresponds to a frequency of 13504Hz.

A wavelength of 0.5" corresponds to a frequency of 27008Hz.

A handy wavelength/frequency calculator, with other possibly useful information:

http://www.sengpielaudio.com/calculator-wavelength.htm

robertbartsch
11-04-2008, 12:05 PM
Has anyone tried this:


Use a lollypoop stick and crazy glue to secure the stick to the cratered dome.
Pull the dent out,
Cut off the stick with a sharp ex-acto knife?
Obviously, you should do a few tests before committing to the driver.

Hoerninger
11-04-2008, 12:21 PM
I'm not understanding what is meant by "first estimation 1/10 of the wavelength". Do you mean if a dent is greater than 1/10 of the wavelength, then there may be a problem?

Yes, I meant "greater than".
But at least I would go with whether I can hear it or not.

The problem with calculation in acoustics is accuracy. Often you have to make some simplifying assumtion for calculating. Or when the problem has a mathematical solution than the effort is immense.

So it is often easiest to consider "large against" or "small against". (Calculating the divergence of an ideal dome is not easy, once I saw it in a scientific assay. And than there is a dent ... )
__________
Peter

coherent_guy
11-04-2008, 05:40 PM
cant you take the driver out and push it from behind?

While I'm not familiar with the tweeter in question, I would think that the diaphragm is not directly accessible from behind without major disassembly, in contrast to a compression driver, like a 375, 242X, 244X, etc. If the diaphragm is replaceable, then it might be possible to perform the fix you suggest.


Yes, I meant "greater than".
But at least I would go with whether I can hear it or not.
.
.
.
__________
Peter

Thanks for the clarification. So it would seem a 6mm - 7mm dent (1/10 of 6.86cm, the wavelength at 5kHz) would be audible at 5kHz and above. But as you say, the proof is with the ear, which I imagine could be checked playing sine wave test tones on the damaged driver and an undamaged one, playing them individually and switching back and forth.

In my experience, a metal diaphragm is more difficult to fix in the manner we have discussed compared to a phenolic, plastic, silk, or other cloth dome, which tend to pop back into shape more readily. Anyone have a similar or different experience?

Allanvh5150
11-04-2008, 10:38 PM
If it is titanium I would flag it. Even if you took the diaphragm out and pushed it out from the back you would still risk causing more damage and maybe even cracking the diaphragm.
Don't put the dent there in the first place is the best insurance.:)

Andyoz
11-05-2008, 03:05 AM
Anyone know how easy it is to replace a tweeter in the LSR-25P?