Announcement

Collapse
No announcement yet.

Tweeter blown. Too much power? Too low/high a Frequency?

Collapse
X
 
  • Filter
  • Time
  • Show
Clear All
new posts

  • Tweeter blown. Too much power? Too low/high a Frequency?

    I just did something dumb and blew up some expensive tweeters (morel 338 104)

    The amp I was using was the Lepai LP1601S (200W)

    The tweeters are rated at 200w. I had the amp turned to 3 o'clock on the volume. Doing test sweeps in REW.

    First attempt at 300hz to 30kHz. (Since the tweeters datasheet went from 300 on up). The instructions I was following asked for a top end of 40kHz for the sweep.

    That blew up. So I put in the other tweeter.

    #2 i set to sweep from 1800hz to 30kHz. (Low end listed as start fs on PE specs)

    In both cases i got one full sweep before thing popped.

    I feel like it may have been the high FS that blew things up. I knew low was too much for tweeters. But maybe too high is, too?

    Can someone help me understand which part likely went wrong - or if that was all a recipe for disaster? I really need to learn as much as I can here 😳


    Thanks πŸ™


    (Incidentally, that amp makes all the high Freq noise on its own!)

  • #2
    3 O'clock? Distortion maybe?

    Comment


    • #3
      Originally posted by AllisterMcRae View Post
      I had the amp turned to 3 o'clock on the volume.
      That tells you absolutely nothing about how much power you were putting into them. You should have been monitoring the amp output with a voltmeter, keeping the voltage well below the driver limit.

      www.billfitzmaurice.com
      www.billfitzmaurice.info/forum

      Comment


      • #4
        Best practise for repeatable measurements is to work to a reference voltage from the amp. Personally, I use the industry standard 2.83volt. Connect a voltmeter to the amp output, open the tone generator in rew, set it to say 60hz, and adjust the amp output until you see the desired voltage. Not only does this mean you can repeat your tests but it also ensures you are only working with 1-2watt for most speakers.

        @2.83 volts you'll be generating around 90db from most tweeters which is perfectly fine for measurements. I usually conduct a sweep from 100 or 200hz at this level and have never had an issue but your ears will quickly tell you if the tweeter is protesting at the lower end. You can/probably should add a large capacitor in series to really protect the tweeter from any unwanted low frequencies (particularly things like leaving the tweeter connected and powering on the amp or connecting and disconnecting sources which can cause a destructive "pop" that could blow a tweeter).

        I would be pretty surprised if it was the upper frequencies that are the issue here. In any case, what instructions are calling for a sweep up to 40khz?? Only the most expensive measurement microphones reach that far (perhaps you have one??). The regular Dayton, omni, Behringer offerings top out at 20k. And most people can't hear anything above 15-16k.
        Constructions: Dayton+SB 2-Way v1 | Dayton+SB 2-Way v2 | Fabios (SB Monitors)
        Refurbs: KLH 2 | Rega Ela Mk1

        Comment


        • #5
          It takes a lot to blow up a Morel tweeter. I had to work to blow one up, think a square wave sweep from Fs to 20k at rated power for about an hour. The distortion from your amp is probably what killed them. The amp specs it's rated power at 10% distortion which isn't really that good.

          You need to get some replacement domes for your tweeters. I'd probably buy 3 or 4 of them and use the two of them that measure the closest.

          ​​​​​Like others have said make your measurements at a known voltage, usually 2.83V. If you're unsure of your measurement skills\technique order a couple of cheap close out drivers when you order your replacement domes and practice on them. Try for repeatable measurements first then try turning up the power and watch the distortion rise.

          Comment


          • #6
            Originally posted by DeZZar View Post
            Best practise for repeatable measurements is to work to a reference voltage from the amp. Personally, I use the industry standard 2.83volt. Connect a voltmeter to the amp output, open the tone generator in rew, set it to say 60hz, and adjust the amp output until you see the desired voltage. Not only does this mean you can repeat your tests but it also ensures you are only working with 1-2watt for most speakers.

            @2.83 volts you'll be generating around 90db from most tweeters which is perfectly fine for measurements. I usually conduct a sweep from 100 or 200hz at this level and have never had an issue but your ears will quickly tell you if the tweeter is protesting at the lower end. You can/probably should add a large capacitor in series to really protect the tweeter from any unwanted low frequencies (particularly things like leaving the tweeter connected and powering on the amp or connecting and disconnecting sources which can cause a destructive "pop" that could blow a tweeter).

            I would be pretty surprised if it was the upper frequencies that are the issue here. In any case, what instructions are calling for a sweep up to 40khz?? Only the most expensive measurement microphones reach that far (perhaps you have one??). The regular Dayton, omni, Behringer offerings top out at 20k. And most people can't hear anything above 15-16k.
            Thanks for the detailed response. I'll be hooking up an old Denon and doing these measurements. I'd been sticking with the lepai since I'd taken some other measurements with it already. Fantastic thought on the capacitor.

            I'm just running an EMM-6 Mic for the moment. I'll scale back to 20K since the distorition is an experience noone wants to look at above that.

            The instructions are from VituixCAD how to with REW docs; the image on page 4. Though its later on that it says to export 5-40kHz.

            Cheers
            Last edited by AllisterMcRae; 06-26-2022, 11:47 AM.

            Comment


            • #7
              Originally posted by devnull View Post
              It takes a lot to blow up a Morel tweeter. I had to work to blow one up, think a square wave sweep from Fs to 20k at rated power for about an hour. The distortion from your amp is probably what killed them. The amp specs it's rated power at 10% distortion which isn't really that good.

              You need to get some replacement domes for your tweeters. I'd probably buy 3 or 4 of them and use the two of them that measure the closest.

              ​​​​​Like others have said make your measurements at a known voltage, usually 2.83V. If you're unsure of your measurement skills\technique order a couple of cheap close out drivers when you order your replacement domes and practice on them. Try for repeatable measurements first then try turning up the power and watch the distortion rise.
              Great info. I saw replacements domes available - but didn't know if thats all I needed to replace because I couldn't smell the burning before I pulled one off. Though couldn't see any damage to the coil other than some brown which rubbed off - so I assumed some part of the internal structure smoked in there.

              To your comment on how you blew up some Morels, I was playing dumdum yesterday with a slightly lower volume (again, unknown: "volume at high noon"), runing Fs from 5Hz-25kHz. I was definitely watching closesly and ready to change my shorts then - but no blowups.

              Defintely a funky amp for distortion. Even running the UW1258 and the 686ND it was having moments where it was in/out on the sweep. Sounded like impedance issues. But only twice out of 18 sets of sweeps.

              Do I really only need to replace the domes? That would be super great! [Edit: I see something about Ferro Fluid in another google result. Though didn't see anything come out when I pulled the dome. What else am I looking for?]

              Thanks for this information. Super helpful!

              Comment


              • #8
                The brown that rubbed off is likely ferrofluid. When you remove the domes, you should be able to see the fluid in the gap.
                Wolf
                "Wolf, you shall now be known as "King of the Zip ties." -Pete00t
                "Wolf and speakers equivalent to Picasso and 'Blue'" -dantheman
                "He is a true ambassador for this forum and speaker DIY in general." -Ed Froste
                "We're all in this together, so keep your stick on the ice!" - Red Green aka Steve Smith

                *InDIYana event website*

                Photobucket pages:
                https://app.photobucket.com/u/wolf_teeth_speaker

                My blog/writeups/thoughts here at PE:
                http://techtalk.parts-express.com/blog.php?u=4102

                Comment


                • #9
                  Originally posted by AllisterMcRae View Post
                  I just did something dumb and blew up some expensive tweeters (morel 338 104)

                  The amp I was using was the Lepai LP1601S (200W)

                  The tweeters are rated at 200w. I had the amp turned to 3 o'clock on the volume. Doing test sweeps in REW.
                  At what SPL? If it was hard to listen to or you had to put earplugs in or leave the room then you were pushing them WAY to hard. Regardless what the spec sheet says a dome tweeter is a fragile device not designed for sustained sine wave reproduction at anything beyond a watt or two.

                  As for the amp you were using, it's rated at 2x100 but an online bench test I saw suggested it may only produce 75w/ch depending on the speaker load. In any case that is more than enough to blow a tweeter as you have discovered. And it's class D which uses a steep lowpass filter at a little over 20khz to filter out switching noise, so there is no getting 30khz to 40khz from this thing... at least not in any form that would be useful for musical purposes. If you want to test the HF extension of a driver out into this range you will need an old school analog amplifier, many of those were flat to 50khz or even higher.

                  So bottom line on blown drivers is.... the cause is ALWAYS overpowering, it could be too much low freq power that leads to mechanical damage, too much burst power which leads to a single open circuit in the voice coil, or too much sustained average power from clipping or compression that results in a burnt coice coil. And there are all kinds of ways to achive one of these failure modes, the amp couuld be way to powerful for the driver overall, or underpowered and driven into clipping, or it could be perfectly suitable but the source signal doesn't have enough dynamic range or the duty cycle is too high and the driver doesn't have enough time to cool off between power cycles. I think if you follow the advice given above and use a measured 2.83v output and closely mic the driver you won't have a repeat of this experience.
                  Paul O

                  Comment


                  • #10
                    Thanks Wolf and Paul O .

                    Today, I did measure the outputs on a Denon 1801 to 2.7V partly because one decimal is the limit on the digimeter on hand today, as well as the volume on the amp clicking in above (I think it was 2.9...but I'm not in the mood to test theories on voltages today ;))

                    I should have taken the hint when the mic was clipping during the sub tests. I was definitely running red - probably above -10dBFs in REW yesterday, if memory serves. Today, things are a much more civilized -30. And seeming to handle the noise quite well.

                    That experiment was run since I couldn't get the sensitivity on the mic to be adequate and so had everything loud for SNR.

                    Which leads to this side note: The Motu M4, when I was adjusting gain for calibration, was staying below -12 [ref] until I cranked the input gain to full and then brought it back down. Then it was coming in at just over half gain to get to -12. In case anyone ever needs to know that.

                    Thanks for your input everyone. I think this thread has brought me some great knowledge. Super appreciated!

                    Comment


                    • #11
                      When you say "blown" what do you think happened? A short in the voice coil?

                      Also, 3 o-clock on the dial on a little Class D amp is probably clipping like crazy. You probably were in square wave territory.
                      Isn't it about time we started answering rhetorical questions?

                      Paul Carmody's DIY Audio Projects
                      Twitter: @undefinition1

                      Comment


                      • #12
                        You've already got your solution, but I'll share how I do this in case another drops in here. I've done some scary tweeter sweeps myself and haven't blown anything yet (knock on wood). Rather than monitoring the voltage at the amp output, I watch for SPL. No real reason for that, my measurement system isn't level calibrated so the dB reading is way wrong and doesn't matter. Also, I think my multimeter was dead at the time I was setting up measurements for a project. So... this was my alternative method:
                        1. Start with volume at 0!
                        2. Turn on the generator in the software (REW, ARTA, whatever) They all have a generator mode to turn on.
                          1. I used pink noise with a high pass at 200Hz
                          2. A mid band sine wave would work as well, then you could do a multimeter to 2.8V at the same time
                        3. Bring up your volume slowly and monitor with your ears and a SPL meter
                          1. I brought the noise up to between 80 and 85dB on the meter
                          2. For most tweeters that should only be in the couple watt RMS or less range
                        4. Kill the generator
                        5. Run the sweep
                          1. I didn't limit the lower end on my sweep at all, since trial mode ARTA didn't have that option. I puckered my cheeks and it worked without damage
                        Electronics engineer, woofer enthusiast, and musician.
                        Wogg Music
                        Published projects: PPA100 Bass Guitar Amp, ISO El-Cheapo Sub, Indy 8 2.1 powered sub, MicroSat, SuperNova Minimus

                        Comment


                        • #13
                          Originally posted by wogg View Post
                          You've already got your solution, but I'll share how I do this in case another drops in here. I've done some scary tweeter sweeps myself and haven't blown anything yet (knock on wood). Rather than monitoring the voltage at the amp output, I watch for SPL. No real reason for that, my measurement system isn't level calibrated so the dB reading is way wrong and doesn't matter. Also, I think my multimeter was dead at the time I was setting up measurements for a project. So... this was my alternative method:
                          1. Start with volume at 0!
                          2. Turn on the generator in the software (REW, ARTA, whatever) They all have a generator mode to turn on.
                            1. I used pink noise with a high pass at 200Hz
                            2. A mid band sine wave would work as well, then you could do a multimeter to 2.8V at the same time
                          3. Bring up your volume slowly and monitor with your ears and a SPL meter
                            1. I brought the noise up to between 80 and 85dB on the meter
                            2. For most tweeters that should only be in the couple watt RMS or less range
                          4. Kill the generator
                          5. Run the sweep
                            1. I didn't limit the lower end on my sweep at all, since trial mode ARTA didn't have that option. I puckered my cheeks and it worked without damage
                          Thanks wogg . This almost makes more sense than starting with the multimeter - between two of them, there was more than a .4v difference. 😳

                          Cheers

                          Comment


                          • #14
                            Originally posted by Paul Carmody View Post
                            When you say "blown" what do you think happened? A short in the voice coil?

                            Also, 3 o-clock on the dial on a little Class D amp is probably clipping like crazy. You probably were in square wave territory.
                            Hi Paul,
                            I replaced the coils in the original Morel housings. everything seems back to normal.

                            No visible issues with the coils ... Assuming the blue was an insulator more than fries metal. So too much power seems the only reasonable choice - maybe distorted coils as the new ones I'd in easier than when I tried to put the old ones back in.

                            Comment

                            Working...
                            X