Up In Smoke?

[wkhanna]

So last night I was recalibrating the speaker levels on my Rotel 1066 HT pre/pro. I had my RS SPL meter in hand, and was using 75 dB as my reference level. I had just completed adjusting the individual speaker volumes when I heard what I can only describe as a V hi-pitch, low level ‘screech’ coming from the right front Natalie P tower. As soon as looked at it I saw a dull blue-ish plasma coming from behind the grill of the RS28A-4 tweeter. Then, not more that a second or two latter, the arcing stopped, I heard soft ‘Pop’, followed by a little cloud of white-ish grey smoke.

Both mid drivers are functioning normally, and I ran the speaker the rest of the night after removing the tweeter with no ill effects. The left side speaker is also performing as rock solidly as ever. As far as my electronics, everything is operating normally, and there are no issues with balance or channel separation. My Nat P’s have been performing flawlessly since the day they were made a year and a half ago. They are run nearly every day for at least 5 hours (12 or more on weekends) and have never given even a hint of any type of problem. IMHO, they made the Energizer Bunny look downright anemic.

Could a failed XO cause this type of catastrophic failure? I am hesitant to re-install the new RS28A I just ordered at cost of $57 when it gets here, for fear I may end up with two blown tweeters and still no idea of what caused it. Is it possible to verify the HF XO circuit with my Fluke V/O meter before I take a chance of needlessly wasting another speaker?


TIA Guys

[Hdale85]

Wow that sucks....hmm I'm not sure. The only way to really see if something is wrong is to put the new tweeter in . I guess you'll have to keep the volumes down low at first though. That's pretty scary though.

[wkhanna]

I was thinking that a possible shorted XO component might let too much current through. If this is the case, the volume setting will do nothing to prevent another case of micro 'Shock & Awe', with the resultant $60 ‘fried’ tweeter.

Would I be able to put a resistor across the tweeter terminals and measure the voltage drop to determine the condition of the XO?

[ThomasW]

There's very little to fail in a passive crossover that's been operating for 18mos. That said it would be beneficial to compare the 2 crossovers for wiring errors.

Realistically it sounds like the tweeter was a marginal build and it finally bit the dust.

PE guarantees it's drivers for 5 years excluding abuse. Problem is what you describe is usually what happens when a tweeter is driven too hard. And in this situation it might be difficult or impossible to differentiate between abuse and a defective unit.

You should call PE's tech support number and talk with them.

The good news is that PE sells replacement domes, so if they don't offer a new tweeter, you'd only need to buy the replacement dome not a whole new tweeter.

[---k---]

Thomas,

It there a way that someone could put a multi-meter on the tweeter leads at the crossover and then run some music and check for voltages WAY out of bounds?

[ThomasW]

Measuring the AC voltage would tell you very little. I suppose you could compare the voltages between the two speakers but again that's a poor diagnostic tool. The real way to test is look at the ouput to the tweeters using an O-scope.

It's very rare that a component in a passive crossover fails, particularly one in a XO that's been operating 18 mos. Not saying it can't happen but that option would be on the bottom of my problem solving list.

If you have a spare smallish cone driver sitting around wire it to the tweeter output and see if treble comes out. If it does suspect a marginal VC winding on the bad tweeter. Remember the wires on these things are the thickness of a human hair.

FWIW, the RS28 tweeter is $50 not $60. You can have a 'new' tweeter by buying a replacement dome/VC assembly for $20 it's P/N 275-132

[Hdale85]

Well when you factor in the cost to ship it to you probably comes out close to 60 bucks.

[Saurav]

Measuring the AC voltage would tell you very little.

And the standard RadioShack multimeter is probably designed to 'accurately' measure AC voltages in the 50-60Hz range. I've used mine to check amp outputs on 1kHz signals, but I've always wondered how accurate it was at those frequencies. Like Thomas said, you could maybe use it to compare the two with a sine wave input. With music it jumps around a lot. You'll probably need a couple of volts of real signal, or the leads might just pick up ambient RF noise (mine tend to do that).

[wkhanna]

Thanks to everyone for their quick responses, advice and concern. ;x(
As noted, I vaguely recalled that measurement of the XO typically required equipment slightly more sophisticated than V/O meter, but thought I would ask, just in case.

I wish I had known that a replacement diaphragm was available before I ordered the entire speaker, I just called PE and the order I placed this morning had already been processed. They said I could return it after it arrived, and just order the diaphragm now, which was kind of them, but I still would be eating the shipping (both ways) and in the end would save less than $15 when it was all said and done. So I will chalk this up as a $57.16 (with shipping) ’learning experience’ ops:, and wait to hear from you guys next time before making panicked and uniformed decisions again. :W

I find ThomasW’s words encouraging, and think I will go over to Rat Shack and pick up an inexpensive speaker (if not, I have an old cheapo clock-radio in the garage I can ‘sacrifice’ a speaker from just for testing purposes) to verify that all is well with my XO before installing the replacement RS28A-4.

On the bright side, I am part-way with the BOM on a new set of speakers should I decide to get the replacement diaphragm down the road. The wife said just the other week that she wanted a set of NatP’s in the kitchen. Not V practical size-wise :roll: , but at least she appreciates good sound. :T

[Hdale85]

Why not Modula MT's? even they are large but not as large as the Nat's. Did you see if PE would replace the tweeter? They may also if you wanted you could reject the shipment and they would return it to sender so you wouldn't have to pay return shipping. Thats what PE told Bob to do.

[pedroskova]

I wouldn't rule out the amplifier as a possible culprit. That "screech" could be the amplifier oscillating in that channel. Were you still feeding test signals at the time, or was the system idling?

[cjd]

I'm also suspecting something went haywire in the amplifier - a shorted input or a fried component, or even a short to rail on the output (that would suggest a bad or bridged component) - some possibility that it was shown a dead short, which could toast something and flip around to toast the speaker. If you were seeing an arc that's likely a LOT of juice to the tweeter - probably high frequency, which may have saved your mids.

C

[augerpro]

On a side note, I don't think you are supposed to put power to a crossover with a driver missing, particularly the tweeter. Don't remember why though...I think it looks like a dead short or something to the amp.

[cjd]

On a side note, I don't think you are supposed to put power to a crossover with a driver missing, particularly the tweeter. Don't remember why though...I think it looks like a dead short or something to the amp.

Should be super high impedance, not low, outside the pass-band. Series crossovers might be more of a problem, or other oddities like just using parallel caps or something...

[Homebrew]

Put your meter on ohms, with power OFF measure between the speaker leads from the amp. If there is a problem with the amp you will notice a large difference from left chanel to right.
Then disconnect the tweet from both cabinets and measure the resistance between the tweeter leads, you will notice a short or an open on one tweet if you blew a tweeter, large differnce in reading from the good one to the bad. If it is a short or open in the tweeter than just replace it. You can also measure the crossovers the same way. Compare one with the other, WITH THE TWEET DISCONECTED.

[Saurav]

I knew I'd read this somewhere too.

http://sound.westhost.com/lr-passive.htm

A word of warning is worthwhile here. Never operate an amplifier into a crossover network with the drivers disconnected. It may be tempting to look at the response, but at a frequency equal to the series resonant frequency of the inductor and capacitor, the network may present almost a dead short circuit to the amplifier (depending on the filter type - second order filters are the greatest risk).

Current is limited only by series resistance, and dangerous voltages can be developed across the capacitor and inductor. These can be sufficient to damage the capacitor (due to over-voltage), and can give you a very nasty electric shock. The amplifier may not survive this abuse either, so it could be a very expensive temptation indeed.

With only 10V RMS applied at the resonant frequency of the 3kHz filter shown below (and assuming a total series resistance of 1 ohm), the amplifier will be supplying 8.3A RMS, and there will be 98V RMS across the inductor and capacitor. Provided the Zobel network is left in place, the resonance is damped so heavily that any risk is eliminated.

I don't know if that really applies if you only remove one driver and your crossover has both HP and LP sections in it. It probably does, if one part of a parallel circuit looks like a short. Now that I think about it, I've done this several times, but probably never with a circuit that just had one L and one C in it.

[cjd]

Ohhh, you meant measure across the output of crossover without a driver? Yeah, that's bad. If it's open. hmm. no, that should be bad too. I guess I was tired and not thinking straight yesterday.

[ThomasW]

Removing the tweeter won't look like a short, it will instead be an open circuit.

If the other drivers are in the circuit there's no problem since they create the load on the amp.

[wkhanna]

There seems to be nothing wrong with my electronics. I am running a Rotel RB-990BX amp for the front main channels, and a Carver C-19 Pre for 2-ch sources, along with a Rotel RSP-1066 for HT duties. I pulled a cheap tweeter from a crappy set of ‘White Van’ speakers I’ve had hidden in the basement for ages, and dropped it in the vacant opening created by the offending RS28A-4. While there is definitely a noticeable difference the sound quality as a result, the temporary replacement has been acting normal, with no indication of unusual distortion or hints that it is being overdriven.

My new RS28A-4 is scheduled to arrive on Tuesday, and I will drop it in as soon as I get home from work that evening.

I did take the diaphragm off of the ‘fried’ unit, and the VC coil windings must have shorted against the cooper ring, as there were burn marks in one area. ThomasW wasn’t kidding when he said the VC wire is small diameter, and there is V little clearance between the coil wires and the copper ring, where the short took place. Any slight misalignment or stray bit of VC wire that may have worked its way loose could easily cause the unit to short out.

I wouldn't rule out the amplifier as a possible culprit. That "screech" could be the amplifier oscillating in that channel. Were you still feeding test signals at the time, or was the system idling?

Yes, at ~ 45 dB at the time of the incident. I had just finished calibration with my RS meter @ 75 dB, a process that lasted maybe about 5 minutes. I had just turned the volume down and was getting ready to switch over to the tuner source. The white noise, or whatever the test signal that the RSP-1066 produces as its test tone, may have been the proverbial ‘last straw that broke the camel’s back’, if one of the VC wires had worked loose over the hours of use, or aggravated some type of misalignment.

While my speakers have seen a lot of hours over the past 18 months, it is rather rare that they spend any significant time over the 80 or 85 dB level. Not that I baby them, but my room is not V large, so I don’t have a need to ‘crank’ them much. Also, 80% of my listening is acoustic jazz, and at anything over 85 dB, you tend to loose some of the authentic character of the music. Though I must admit, my Dire Straits LP, Brothers in Arms, at ~ 95 dB is a fun thing to do once in a while.

[wkhanna]

Just a quick update……….

I installed the replacement driver last night.
All is well and right again in the galaxy for this padawan. :T

Must have been a nanno blip in the Force, that or maybe a Gamma spike from the flux capacitors I put in the XO. :huh:

[Hdale85]

Or maybe just a faulty tweeter

[JJones]

I would have said to just try some cheap less than $10 small speaker that fits the tweeter cut out, but good to know it didn't blow the replacement...

About the RS28 Zaph said: " I tested 8 of these, and there was a consistency problem. " - maybe they are a little overpriced then? Would expect high consistancy and QC at that price, for a tweeter...

[ThomasW]

I would have said to just try some cheap less than $10 small speaker that fits the tweeter cut out, but good to know it didn't blow the replacement...

That recommendation was made in the 6th sixth post in this thread

About the RS28 Zaph said: " I tested 8 of these, and there was a consistency problem. " - maybe they are a little overpriced then? Would expect high consistancy and QC at that price, for a tweeter...

That was news in 2005..

Long ago PE implemented a factory based QC system where each tweeter is individually measured and there's a corresponding QC sticker on each tweeter.

[JJones]

That recommendation was made in the 6th sixth post in this thread

It's helpful to see that more than one person agrees with a certain idea... Something wrong with getting a "2nd opinion"? :T

That was news in 2005..

Well, that's what it still says on Zaph's page; and, his problem with the tweeter does seem like a QC issue.

[ThomasW]

and, his problem with the tweeter does seem like a QC issue.

The QC issues Zaph talks about were apparent with new in box tweeters. And that problem was remedied with the inspection process.

This is the first report I've seen posted where the tweeter has been in operation for 18 months then failed. Given the hundreds of these tweeters currently in use, his single tweeter failing probably isn't indicative of a QC issue worth noting.