HiFi Hell..please help with Kefs

Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.
Lamepops,
Are you running other equipment (video, tv etc) from the same power outlet ?.
Also, are you routing the speaker cables well away from line level cables ? (this can cause inaudible high frequency feedback/oscillation).
I would also be suspicious of a dry solder joint in your amplifier causing the instability - there is nothing like a car ride to show up/temporarily cure crook solder joints - the solution is to blanket resolder your amplifier.

Just about any amplifier will test stable into a resistive load, so when you test for instability with your oscilloscope, you would be best to do so with the speakers connected.

Regards, Eric.
 
Hey Mrfeedback - thanks for hanging in there with me on this one. It has been quite an ordeal, and frankly my checkbook and patience can't take much more of this.

I've always been pretty careful about routing cables. I'm kind of a neatness freak and don't like excess cable just lying around, so I use cable ties a lot. I never group the speaker wires with the interconnects, but I will check that they are separated.

During the previous two KEF meltdowns, my system was plugged into a dedicated 20-amp circuit in a master bedroom, with only my audio components plugged into a Monster cable power center. Prior to this last meltdown, I had moved everything into a spare room, thinking it might be "environmental". The only other thing plugged in was a small clock radio.

I'm supposed to pick up my replacement speakers tommorow, but I am very nervous about even hooking them up until I get some idea about what's causing this.

Remember - I took my entire system down to their store this past weekend and blasted away for about 2 hours. The KEFs performed flawlessly and my ears were pretty numb for about an hour after. Yet it only took about 15 mins at 1/4 volume to damage them at home. It's just crazy.

I've really got to focus on diagnosing the problem. I have an old pair of Technics speakers I can hook up to look at the power amp under load. Hopefully I can get to this on Saturday morning.
 
No Worries..........

I never cable tie system cables together.
Tieing cables together could cause enough coupling to enable feedback/coupling caused oscillations.
I have seen high power PA system tweeters and crossovers 'cooked' because of wrong cable dressing causing an ultra-sonic oscillation condition.
Also do not 'shorten' cables by looping them and tying the loop.

Did you have exacty the same cable conditions at the store ?.
One thing to note is when a system is oscillating, it can have a 'loaded down' character in the sound, and not as 'free' sounding as it should be.

Eric.
 
Mrfeedback - they actually convinced me to buy the new cables because they modelled them with my system in the store and I was suprised at what a difference they made. Even I could hear that the overall sound was more dynamic or brighter. Also, I felt that dropping $250 on cables was a reasonable price for all of the help they are giving me. Perhaps they did more harm than good though, making my speakers more sensitive.

This morning, they sent the KEFs back to the factory and I will be receiving brand new 2004 models of the same speaker, which they claim has a "beefed up" tweeter. To me, this implies that they knew of a problen with the old design.

One question about cables...how do you deal with the excess cable? Say I have a 10-foot cable and my speakers are only 5 feet away...what do you do with the excess?

Anyway - while they are swapping my speakers, I am going to spend some time looking at my setup at home. Since my system performed like a champ in the store, but breaks almost instantaneously in my home, I am inclined to believe that I have an AC power problem. I have a few things to try....

- look at the AC when my system is both idle and under load
- look at the amp outputs at both idle and under load
 
Nah, Cables Can't make Your System Sound Different can They ?........

"One question about cables...how do you deal with the excess cable? Say I have a 10-foot cable and my speakers are only 5 feet away...what do you do with the excess?"
Run 5' cables, or else try to lay them so that they are not making tight loops, and certainly do not coil and tie them.
You could try running you system with your Technics speakers and take a look with your CRO and see if anything untoward is happening.
You could also add capacitance (say 1 uF) across the speakers to see if you amp is going unstable.
You could also drive an earth stake in to the ground outside your stereo room, connect a wire and measure for difference between this and your wall socket earth.

Eric.
 
mrfeedback - right now I have fairly long speaker cables, and they are coiled (several times). They are wide flat copper conductor cables from Nordost and I don't have the ability to terminate them. I guess I could check with the dealer to see if I can buy terminators for them. I am tempted to go back to the monster cables which I cut to length and terminated myself....at least for the upcoming tests.

I am anxious to look at the AC and the amp outputs with the oscope. Plus it might be fun to destroy the Technics speakers (LOL).

You did bring up another question which is bothering me. While the power station I am using has a grounded plug as does my amp, my preamp and cd player do not. Is that a problem that needs to be addressed as well, or is that simply a safety requirement?

I like the earth ground idea...I suspect it might be a grounding or interference issue. Thanks.

I forgot to mention something I did over the weekend. I took my old large Advent speakers from my daughter's room and hooked them up to my system. I cranked them up to just below half volume, and though I noticed some distortion, I didn't break anything. I let them run for a while...perhaps 10 minutes...and it was real loud. Just another data point for what it's worth. The Advents appear to be able to handle the clipping or whatever it is, while the KEFs cannot.
 
diyAudio Moderator Emeritus
Joined 2001
How about making a CD with some test tones like 5K and 10K, running the CD through the system, and seeing if any other frequencies show up on the oscilloscope?

I really don't know oscilloscopes that well, but I believe that it would detect a signal not on the program material. The 5K and 10K tones would be delivered to the tweeter, of course.

Just for your own knowledge, plus the fact that your dealer's patience has limits, you should want to know that before you hook up the new KEF's.

By the way, just for the heck of it, I know somebody who lived on the ocean who bought a speaker system, and whose woofers kept rotting from the salt water. Every few months, the dealer had to replace the rotted woofer. After 3 replacements, the dealer finally had to give the guy his money back.
 
kelticwizard - thanks for the post.

My dealer's patience seems limitless, but KEF (the factory) has suggested that this would be the last time they would do the repair for free. So the pressure is on to figure out what's going on at home with my setup.

I have set aside the entire day on Saturday to look at it. I've done a fair amount of electronics troubleshooting, so I'm confident that I can figure it out.

I already have an audio test cd that I purchased on Ebay for $11.95 and a TEK oscope that I borrowed from a friend.

-
 
right now I have fairly long speaker cables, and they are coiled (several times). They are wide flat copper conductor cables from Nordost and I don't have the ability to terminate them

I'd be willing to bet that this is the reason behind your problems. You are making an inductor which is probably driving your amp into instability.

NEVER coil speaker cables and if at all possible only have them as long as you need them.
 
Testing underway

Here are the results of some of the tests that I have run so far. I decided to do all the tests with a pair of speakers as opposed to a dummy load. I just happened to have a pair of old Technics 3-way loudspeakers. Unfortunately I don't have the ability to post oscope images right now. As a reminder, my setup is:

Pioneer MD-426 CD Player
Adcom 555-II Preamp
Adcom 555-II Amp
Technics SB-K43 Loudspeakers (for dynamic load only, not for sound quality)
Monster speaker wire, AR RCA patch cables
Tektronix 2465A Oscilloscope, 350 MHz bandwidth
Lasertrack CD2000 Professional Audio System Testing Disc

Test 1: AC Power - Load/No Load - Here I was just looking at the AC power line with a scope, with the power amp off, 0 volume, 1/4 volume, and 1/2 volume. At first I looked at the line with a music selection (Zakk Wylde) and then with the test cd (350 Hz, 1 KHz, and 12 KHz, all at -10db). The AC power measures a little high - 125VAC, but it appears to be stable, even as I switch the source, and vary the volume. Note that this was done without a power conditioner. I spent several minutes staring at the waveform, varying the volume/source, and saw no problems.

During the next group of tests, I was using the audio test cd, and looking for nice clean sine wave with no clipping or distortion.

Test 2: Crossover Check: -15db, Frequency Sweep from 40 Hz to 15 KHz. I measured 720mv P-P @ 1/4 volume and 6.0V P-P at 1/2 volume. I saw no anomolies during the entire sweep, and repeated the sweep at least a dozen times at 1/4 and 1/2 volume.

Test 3: 1 KHz, 0 db. I measured 2V P-P @ 1/4 volume and 18V P-P at 1/2 volume.

Test 4: 315 Hz, 0 db. I measured 2V P-P @ 1/4 volume and 18V P-P at 1/2 volume - no problems.

Test 5: 10 Khz, -10db below operating level, 1/4 and 1/2 volume

Test 6: 12 KHz, -10db below operating level, 1/4 and 1/2 volume

Test 7: 14 KHz, -10db below operating level, 1/4 and 1/2 volume, At 1/2 volume I picked up on some unusual sounds (like a pencil tapping on a table), and saw something unusual on the oscope. Then the Adcom Power amp went into thermal protection and I lost the left channel. After allowing some time to cool off, I found a blown fuse, replaced it, and went back into testing, making sure that any further tests were to be done with short bursts (less than a couple seconds) at 1/2 volume. I also verified that the tapping sound was gone. I realized I had been cranking out continuos wattage at high-freq and needed to be more careful. Duh.

Test 8: 16 KHz, -10db below operating level

Test 9: 18 KHz, -10db below operating level

Test 10: Phil Collins at 1/4 and 1/2 volume

Test 11: Type O Negative at 1/4 and 1/2 volume

I saw no abnormal waveforms and no evidence of clipping or distortion, even at 1/2 volume which is something my KEFs were never exposed to. I understand that the KEFs represent a different kind of load than the Technics speakers, but I have more confidence in my setup than I did before. I was really suprised at how good these old Technics speakers sounded with my setup, though certainly not as clear as my KEFs.

My new KEF speakers are coming in this week, and I am confident that I will not be feeding them crap, and that they are just not meant to be used the way I use them. I am considering selling the new speakers before I use them, and use the money to purhase speakers that are more rugged and more suited to my taste.

Any thoughts or suggestions?
 
Sinewaves unless driven to clipping will typically not show up transient instabilities.
Square wave testing can show up ringing etc instabilities that may be cooking your tweeters.
Perhaps you should try coiling your speaker leads to try to re-create the problem.
Another suggestion is to parallel load resistors across the loudspeakers to get the load down to 4 or 3 ohms and see what happens.

Just some suggestions, Eric.
 
I was wondering about that Eric, and I appreciate your suggestions. I was warned earlier in this thread that the KEF might present as much as a 2-ohm load. If nothing else, I have a baseline now. It seems that the coiled speaker cable test and parallel-loading resistor test are pretty easy to check out. I'm not too sure about square-wave testing. I'll have to think about that.
 
2 Ohms Is Too Low For Most Amplifiers..........

Briefly driving into hard clipping should show instabilities.
Look for 'stiction' wher the waveform is coming back out of clip.
If the Kefs are drooping as low as 2 ohms, this could be triggering output stage current limiting and nasty artifacts, or plain showing up instability with low load impedence.
Be very careful when running into clip at higher frequencies as this can burn mids/tweeters real quick - 400 Hz should be reasonably safe, but only for short duration at hard clipping.

Eric.
 
The last thing I did before returning the oscilloscope was to hook up the Technics speakers with my 8-ohm dummy load in parallel...to simulate a 4-ohm load.

I played some music at about 1/4 volume for about 15 minutes to get everything warmed up. Then I went through the same steps as my last round of testing, with no signs of clipping or distortion at half volume.

Holding my breath, I moved the volume past 1/2 and just shy of 3/4 volume (for just a few seconds). No sign of clipping or distortion at 1 Khz.

I'm worried that my inexperience as an audio person will lead to damage beyond this point so I am done. I have no answer as to why I keep blowing the KEFs. I think I have no choice but to part with the brand new KEFs (when they come in) and try a different set of speakers.

One last thing I will point out, is that I have seen the KEF Q5 which appears to be the logical Q55.2 upgrade. One notable difference is that it seems to have a port of some type above the Uni-Q driver...on the very top of the speakers. I was wondering if this was done for appearance, or if it serves a function (like thermal management).
 
seeing all this, I'm inclined to suggest that the offending signal voltage be measured at the driver terminals at a level somewhat below that which has caused it to 'blow' then buy a heavy duty transzorb rated at that voltage and solder it across the terminals. That way, you'll get occasional distortion but save your tweeter.

Either that or replace speaker and amp with professional audio speakers and a flea power SE amplifier:)
 
KEF Q5 - NIB

Well, I received my KEF Q55 replacements, the Q5. While I am tempted to open these babies up and give them a test drive, I haven't isolated the cause of the blown drivers. Since I've replaced all other components, have run out of money and time, and don't expect to be able to fully understand the issue anytime soon, I'm going to sell them, and move to different speakers. I've been told by the dealer that the Q5 has a more rugged hf driver, but I don't want to take the chance.

I spoke with a gentleman at an Adcom/KEF repair center in New Jersey, and he seemed to understand my problem. He owns the same amp/preamp and guaranteed me that the Adcom could easily handle the < 4-ohm load, and said that the KEFs, while sonically very pleasing, were somewhat "delicate"...more of a finesse speaker than I am looking for, and that I really am just over-driving them. He suggested I move to smaller high-quality amp or more rugged speakers.

Thanks to all the input, and for putting up with my babbling over the past few months. Off to Ebay with the KEFs....
 
Re: Stop the insanity!

MRehorst said:
Something is broken and needs to be fixed.

Special line cords, filters, wood pucks, directional cables, and other pieces of audio-voodoo are not the answer. You may as well pray for the problem to fix itself- it's cheap, but considering the state the world is in after thousands of years of millions of people praying, you're probably better off trying science.......

I am in total agreeance with you on this MRehorst. All the suggestions about the power source and speaker cables have nothing whatsoever to do with the cause of the problem. The only way to determine the cause is by process of elimination. Also, the setting of the tone controls will not be the cause either - after all, the amps being used aren't exactly low quality which one would expect to blow speakers.

Heck, if we had to make sure we didn't crank the tone control settings over a pre-determined maximum (given we are using good equipment and not feeding more power to the speakers than they are rated to handle) then I would venture to say that more people with the same amps would be blowing their speakers.

I would temporarily fit a Polyswitch RXE250 in series with the x-over input on each speaker and then try a different source (either change the CD player or use tape or vinyl) and play at similar settings. Whatever you do the first thing you must do is protect the speakers during your experimentation.
 
rherber1 - I've been without a functioning stereo for over a year and a half now because of these problems, and replaced all components (cdp, preamp, amp, cables, power conditioner, etc). I've spent quite a few hours with an oscilloscope staring at waveforms, and trying to break other speakers with my setup. Though I never saw any problems with the Adcom amplifier outupts or with other speakers, I did learn a few things during the process.

In the end, in my humble opinion, the KEFs cannot hold up to the volume and type of music I listen to.
 
Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.