most of the rms limiters are dsp, built in amps,but there are a few like alesis 3630 or symetrix 501.
the damage on the plate looks like its outside the gap>thats no problem.
if it is in the gap,you have to remove that bit.
don't use sandpaper! maybe a hammer with screwdriver to gently tap it .watch out for the magnetic force.
make sure that no iron flakes get in the gap.
the damage on the plate looks like its outside the gap>thats no problem.
if it is in the gap,you have to remove that bit.
don't use sandpaper! maybe a hammer with screwdriver to gently tap it .watch out for the magnetic force.
make sure that no iron flakes get in the gap.
Last edited:
Posts #34/40
Hi Crescendo,
I don't claim to be an expert in this area, but the damage on the top of the machined pole piece/plate does not look like it would result from the voice coil wire rubbing. That's a comparatively hard steel, and according to B&C a copper voice coil, if I had to guess this damage was there when the driver was assembled. It looks like a manufacturing defect to me.
Regards,
Hi Crescendo,
I don't claim to be an expert in this area, but the damage on the top of the machined pole piece/plate does not look like it would result from the voice coil wire rubbing. That's a comparatively hard steel, and according to B&C a copper voice coil, if I had to guess this damage was there when the driver was assembled. It looks like a manufacturing defect to me.
Regards,
I agree with TB46 that the indentations look to be a manufacturing defect, and if they do extend in to the gap, could be responsible for scoring the inner voice coil, which could cause arcing which would make the coil open.
That is doubtful though, as you would have heard some scraping, and the sloughed off coils just indicate too much heat from being presented too much average power.
The bridged PLX3402 will put out 3400 at 4 ohms, 2200 at 8 ohms without the clip/limit lights illuminating.
Since the thermal limit of the 18SW100 is around 750-1000 watts if the music has only 3 dB crest factor, the voice coil could be heated enough to melt the adhesives without any clip/limit lights illuminating.
Until recently I had not thought any LF music would have less than 6 dB crest factor (the dynamic range of the AES test signal), but sine 143 showed an example (and says he has about 2000 similar tracks!) of LF with only 2.7 dB dynamic range, even less than a sine wave!
There are DSP units such as the XTA Electronics DP series (available in a variety of input and outputs) that have separate peak and RMS limiting.
Many DSP (and most analog comp/limiters) have RMS limiting, if the attack time can be set quite long, like 200ms or more, and the ratio is high, like 10/1 or greater, the peaks will still get through and the amp limiting will take care of them, but the RMS level can be set low enough so music like the "Hunter" won't "kill" your voice coils.
RMS limiters can be set easily using a 50-60 Hz sine wave and measuring amp output voltage. This can be done with no load, but should be confirmed with a load similar to the impedance the amp will be loaded with, as many amps will put out less voltage with a load.
Art
Attachments
The artist he referenced as the killer (Rockwell) has done some work with the MC in hunter int he past, and the tracks I have from him are alll prettttty bass heavy (and likely have low crest factor, will test).
I still wonder why the amp wasn't giving any warning signs. At 1/3 power, those amps draw over 30A off the wall socket. Not to mention, putting out the heat of a blow dryer... which should only be allowed to go on for a few minutes before something trips.
The PLX 3402 draws 39 amps full tilt with sine waves, which would pop it's breaker in a few seconds.
Putting out 1000 watts of sine wave like material (enough to cause thermal damage to the speaker), the amp would draw about 11.5 amps.
Art
Attachments
If there is a problem with drivers and more clients with problems (not only one), I am sure the manufacturer will acknowledge it and sort it out, whatever if because of the type of glue, thermal or other. They may ask for the old burned parts/drivers or pictures for their museum... 😀😀 this, if they really care and investigate*, to avoid future problems.
R&D* to the development of new products, or to the improvement of existing ones
R&D* to the development of new products, or to the improvement of existing ones
Attachments
Hi epa,
Yes, soft in comparison to other steels, but hard in comparison to the coil wire. And the damage looks as if the plate got damaged before it was installed. Just my opinion.
Regards,
Yes, soft in comparison to other steels, but hard in comparison to the coil wire. And the damage looks as if the plate got damaged before it was installed. Just my opinion.
Regards,
I spoke to B&C North America's rep, Bennett, the other day. He confirmed their 4" VC recone kit to be out of stock. 4 MONTH wait time. Yikes. Might have to buy a new driver when I put in that order in case I need my 4th sub sometime soon :l
This has me wondering, had I invested the extra bit of money for the extra 200W/0.5" VC the 18SW115 has, would that have been enough to save me from these failures?
The track playing (Rockwell - Fluf), just before the #2 driver let go, was very near the driver's excursion minima (according to the TH18 hornesp sim) - so, a reason for a bit of heat.
For what this information is worth (if any), I measured my PLX3402 amplifier's output voltage a while back. This was due to my curiosity of how the clip lights and limit lights communicate to the user. With the clip limiter off, not matter how much gain with any sine wave and the clip lights cherry red, I could only get the output to just 1950ish watts. With the clip limiter on, I could only get the output to only 1440 watts. With no signal to the amp, I measured a resistance of 6.1 ohms between the ch.1 + and ch.2 - for proper bridged config. This is the figure I used for the voltage calculations. Again, I don't know if this proves much or is at all a proper method for testing an amplifier without a load.
I took some pictures after cutting away the surround & spider. I'll attach them here:
This has me wondering, had I invested the extra bit of money for the extra 200W/0.5" VC the 18SW115 has, would that have been enough to save me from these failures?
The track playing (Rockwell - Fluf), just before the #2 driver let go, was very near the driver's excursion minima (according to the TH18 hornesp sim) - so, a reason for a bit of heat.
For what this information is worth (if any), I measured my PLX3402 amplifier's output voltage a while back. This was due to my curiosity of how the clip lights and limit lights communicate to the user. With the clip limiter off, not matter how much gain with any sine wave and the clip lights cherry red, I could only get the output to just 1950ish watts. With the clip limiter on, I could only get the output to only 1440 watts. With no signal to the amp, I measured a resistance of 6.1 ohms between the ch.1 + and ch.2 - for proper bridged config. This is the figure I used for the voltage calculations. Again, I don't know if this proves much or is at all a proper method for testing an amplifier without a load.
I took some pictures after cutting away the surround & spider. I'll attach them here:
An externally hosted image should be here but it was not working when we last tested it.
An externally hosted image should be here but it was not working when we last tested it.
An externally hosted image should be here but it was not working when we last tested it.
An externally hosted image should be here but it was not working when we last tested it.
i sugest to immediately put some tape over the gap.
4 months waitinglist.....
same problem with our eighteensound suplier ,some drivers un availeble for a few months .
i'd prefer monkey proof limiter setings.
4 months waitinglist.....
same problem with our eighteensound suplier ,some drivers un availeble for a few months .
maybe ,or maybe in the end you play louder with the same result.
i'd prefer monkey proof limiter setings.
https://www.parts-express.com/bc-r18sw100-recone-kit-for-18sw100--294-6449
how many volts did you measure off the back of your plx3402 (I dont care how many watts it put out).
how many volts did you measure off the back of your plx3402 (I dont care how many watts it put out).
"Only" 1440 watts?
Again, I'll repeat: since the thermal limit of the 18SW100 is around 750-1000 watts if the source has only 3 dB crest factor, you don't need 1500 watts to burn it.
With sine waves, or music with the crest factor of sine waves, the difference between 1500 and 1700 drops to 750 to 850, your amp probably has enough voltage to cook the 18SW115 too, there is less than one dB difference in the power ratings.
4 month wait for parts is a drag, would not have expected that from B&C.
he can try, but i suspect pe wil have to wait for 4 moths also.
unless they have it in stock,wich i doubt.
unless they have it in stock,wich i doubt.
your likely right. sucks, they have the 18sw115 recone kits in stock....
https://www.parts-express.com/bc-r18sw115-recone-kit-for-18sw115--294-6450
https://www.parts-express.com/bc-r18sw115-recone-kit-for-18sw115--294-6450
Hi!
Firstly, sorry for your loss.
But looking at the pictures of the bobin we could see clearly that the cooper color of the bobin has changed completely and it must be due to the heat in excess for long term.
The glue that holds the coil has a maximum heat tolerance of about 225º C.
When there is over excursion (enemy of the THs not tweaked, ie only at the Pe limit) for long periods, cooling by convection is impaired, so heating the coil more and more, causing the dynamic compression that can reach up to about 8 dB, and when that remain for a period, it will cause the glue that holds the wires come loose partially, been the result usually worse the next time you use the driver, when the coil starts to scrape the gap, even cause short circuit.
That's one of the greatest problems at automotive warranty claims when arguments are: -I have played it yesterday and all was OK; today I just started the sound and it stops...
My tests with the TH18 show it clearly, the over excursion at the Pe limit, when the low cut should be set carefully.
With the Jbell TH15 happened the same, and after some tweaks, like throat restriction (Jbell advice), cone correction, I notice some upgrade, but what gives the better result beside this was bracing the mouth till it stops to rattle.
Now the bass guitar notes from 40 to 100Hz play as they should, the cone travels less, and the mouth panels are quiet.
Maybe that's the way to go.
Regards,
Firstly, sorry for your loss.
But looking at the pictures of the bobin we could see clearly that the cooper color of the bobin has changed completely and it must be due to the heat in excess for long term.
The glue that holds the coil has a maximum heat tolerance of about 225º C.
When there is over excursion (enemy of the THs not tweaked, ie only at the Pe limit) for long periods, cooling by convection is impaired, so heating the coil more and more, causing the dynamic compression that can reach up to about 8 dB, and when that remain for a period, it will cause the glue that holds the wires come loose partially, been the result usually worse the next time you use the driver, when the coil starts to scrape the gap, even cause short circuit.
That's one of the greatest problems at automotive warranty claims when arguments are: -I have played it yesterday and all was OK; today I just started the sound and it stops...
My tests with the TH18 show it clearly, the over excursion at the Pe limit, when the low cut should be set carefully.
With the Jbell TH15 happened the same, and after some tweaks, like throat restriction (Jbell advice), cone correction, I notice some upgrade, but what gives the better result beside this was bracing the mouth till it stops to rattle.
Now the bass guitar notes from 40 to 100Hz play as they should, the cone travels less, and the mouth panels are quiet.
Maybe that's the way to go.
Regards,
Notes:
Resistance at spkr terms: 5.2-7 - will use 6.1ohms
Clip lim disengaged:
Volt meter: 109.7
109.7 * 109.7 = 12034.09
12034.09 / 6.1 = 1972.8 watts
Same setting w/clip lim:
Volt meter: 94.6
94.6 * 94.6 = 8949.16
8949.16 / 6.1 = 1467.0 watts
Got about 2800w w/a 100Hz warble tone
And, correct, PE would have to order the 18SW100 recone kit :/
Yes, only. I was expecting more with that test. Limit lights were as bright as I could get em. I never practice this in the field. And with NO limit lights illuminated nor flickering, the output voltage would have been at least 3dB less. At least.
And I definitely didn't expect that from B&C either. Thought they would gladly stock them. Their rep says it's due to no one ever really blowing them. Well, surprise. And I don't think that's a very good excuse either.
- Status
- Not open for further replies.