collapse

Author Topic: Zobel Question  (Read 3927 times)

phatt

  • Legendary
  • ******
  • Posts: 2039
  • Chip Points: 242
    • View Profile
Zobel Question
« on: July 18, 2016, 06:47:38 AM »
Tekky Q,
          This Blown LM3886 circuit snippet is from a BiAmp powered speaker.
Supply rails are ~  +/-40VDC
The output is latched to the rail and too hot to touch within a minute of turn on.
The other two LM3886 are driving the woofer which is still working so half tempted to have a go at fixing this. I figure there is no point in trying to replace the power chip if it's just going to happen again.


Would I be right to think that with no Zobel network on the output it may well have been oscillating causing the chip to fail?  The tweeter is fried. :'(

Thanks, Phil.



(R14, R15 and D9 just go to the limit/overload circuit)
« Last Edit: July 18, 2016, 06:51:55 AM by phatt »

J M Fahey

  • SSGuitar Global Mod
  • Legendary
  • ****
  • Posts: 4142
  • Chip Points: 429
    • View Profile
Re: Zobel Question
« Reply #1 on: July 19, 2016, 09:53:13 AM »
Some datasheets boast "no Zobel needed" ... don´t trust them.  :trouble
Worst case, it does not hurt, so use them always.
Replace your chipamp but test it with no load, until you are certain there is no DC at the output.

teemuk

  • SSGuitar Global Mod
  • Legendary
  • ****
  • Posts: 1074
  • Chip Points: 282
    • View Profile
Re: Zobel Question
« Reply #2 on: September 03, 2016, 01:53:10 PM »
Generally I would agree with Fahey...

But in this case lack of Zobel might have something to do with the amp section's purpose of driving a tweeter only. Generally the basic RC zobel introduces some capacitance to compensate increasing inductance of a generic dynamic loudspeaker at higher frequencies. ...but what happens when the load is something like a piezo tweeter and behaves very much like a capacitive load.

This is somekind of Behringer's active powered speaker, bi- or tri amped, right? I'm sure they did not produce several thousand units that were prone to fail from the start. I'm also suspecting there's a generic Zobel in the "LF amp" section, which probably parallels a couple of LM3886 chips for increased power.

phatt

  • Legendary
  • ******
  • Posts: 2039
  • Chip Points: 242
    • View Profile
Re: Zobel Question
« Reply #3 on: September 04, 2016, 07:46:49 AM »
Thank for the input chaps,, much appreciated.

Teemu the Horn is not a piezo,,, maybe they designed it for a Piezo but swapped it later in planning for a VC horn and no one thought to check. but then the LF section also has no Zobel.

Here is the whole power amp section of that unit.
There is no Zobel network I can see.
Yes Teemu the Low freq drive is 2x LM3886 chips and they still work fine.

The owner said the hi freq no longer worked but low freq was still working.
The Hi freq output had stuck to the rail and had burnt out the VC. Chip running Very hot. Obviously dead.
I was concerned that if I replaced the parts there was no guarantee it would not happen again.
The owner decided to put that money towards a new system.

I got lucky and found the full Schematic for the speakers so I've removed the Burnt LM3886 chip and bypassed the biamp section so the remaining amp now works at full bandwidth.

I happen to own a pair of small 10" PA speakers, one unit is powered while the other is passively driven from the first. So I've re-purposed the remains of the Behringer amp and now both my speakers are powered units. So I've scored well.  <3)
I am a bit concerned about the lack of the Zobel network and now it's running at full bandwidth it may be an issue.

My other concern is the 39 Volt supply rails which seem rather high for these chips but they are now driving an 8 Ohm speaker whereas the Behringer had a 3.8 Ohm woofer so it would be less stress on the amp.
If you can see any major flaws in my grand design feel free to enlighten me. 8)
Time will tell if it stays working,, xP
Oh if anyone needs the full PDF of all these Truth series power speakers just PM me.
Phil.

picture uploading

Enzo

  • Legendary
  • ******
  • Posts: 1868
  • Chip Points: 195
    • View Profile
Re: Zobel Question
« Reply #4 on: September 04, 2016, 03:59:29 PM »
Quote
maybe they designed it for a Piezo but swapped it later in planning for a VC horn and no one thought to check

Oh, come on, Behringer is not two guys in a basement drinking beer while designing products.  I can't improve on what teemu said above:

Quote
I'm sure they did not produce several thousand units that were prone to fail from the start.


This is not the first, nor will it be the last, chip amp that failed in some product.

phatt

  • Legendary
  • ******
  • Posts: 2039
  • Chip Points: 242
    • View Profile
Re: Zobel Question
« Reply #5 on: September 05, 2016, 03:53:30 AM »
I did Notice a pair of Truth series speakers going cheap,, one tweeter not working. humm is this a pattern forming?
I may not have your qualification Enzo but I'm afraid I don't place my faith in a Brandname.

Have you noticed how planes still fall from the sky? Oh and I notice another rocket just blew up. What about a world recall of ALL new Samsung phones?  And the way stuff is pumped out now that list will just get bigger. :-X

I worked in manufacturing most of my working life and I could fill pages with design flaws and complete stuffups by highly qualified people. (one case 13 people died)
Phil.

Enzo

  • Legendary
  • ******
  • Posts: 1868
  • Chip Points: 195
    • View Profile
Re: Zobel Question
« Reply #6 on: September 05, 2016, 10:34:45 PM »
Yes, we can all point out failures of products.  Most however are not systematic failures.  Airplanes fall from the sky now and then, but MILLIONS of them don;t.  Not only that, we have to eliminate the ones that went down not of their own fault, like those with bombs in them or that collided with something. 

Behringer makes a TON of product, many with VERY similar designs, and as far as I know, none of their line has gathered a reputation for just blowing up a lot.

In fairness, the rocket is not a commercial mass produced design.  It is still experimental cutting edge technology.  But I get your point.  Nothing is absolutely 100% all the time reliable.  On the other hand, many people look on the internet and find the most popular products are also the ones with the most complaints.  They incorrectly assume it means the product is not reliable.  But in fact it just means there are so many of them that a normal failure rate means more of them will appear than a less popular item.

What is the most common repair I have had to make to PA speakers over the last 30 years?   Blown tweeters.   Guys, especially DJs keep turning them up without regard for limits and eventually the tweet gives up.   In powered speakers, this can also result in failed electronics.  I don;t know that your particular units were over driven of course, but it would be the common cause.

Whenever someone points out that model XYZ had several failures on the internet boards, I feel obligated to point out that 10,000 of the same unit did not.  And that is the pattern I usually see.

phatt

  • Legendary
  • ******
  • Posts: 2039
  • Chip Points: 242
    • View Profile
Re: Zobel Question
« Reply #7 on: September 07, 2016, 04:44:29 AM »
Thanks all for the advice;

Ok one more Q,, There is no Zobel on the woofer section and I've left it as built because ideally these need to be close to output pins but this may prove to be impractical on such a tight space PCB. xP

But if I was to install a zobel network on this paralleled output then How?
Do I need 2 separate networks? (Before R1 and R2)
I'm mainly asking for future reference.
, Phil.

J M Fahey

  • SSGuitar Global Mod
  • Legendary
  • ****
  • Posts: 4142
  • Chip Points: 429
    • View Profile
Re: Zobel Question
« Reply #8 on: October 13, 2016, 07:45:33 AM »
Those woofer 3886 amps are in parallel  :loco , outputs mixed through R1 R2 .
I´d add a Zobel from mixing point to ground or even across woofer terminals if no space available.

As of Behringer´s design team, that Company is famous for ripping other designs, so they might very well think "if XXXX brand product works and is successfully sold, why waste time developing/improving it? just fire the 3D copier and churn out a ton of them at 1/3 price"
Problem is that they are not *exact* copies, for one most *circuits*  can be copied but related *art*  can not, specially PCB design.

And we all know that tracing a track "here"  instead of "there" or grounding "here"  instead of "there", one inch away, can bring instability.

If it´s that bad that 100% or at least 10% fail, or even 2%, that will be detected at once and corrected, no doubt about that and I don´t think them as fools, quite the contrary, but if product is just somewhat less stable, it may take longer.

FWIW both "no Zobel" , parallel outputs and parallel bridged ones are suggested in the data or applications sheet itself, by the chip maker, no less.

Yet we all know the reliability problems Marshall MF350 has, so much so that nobody wants to repair them (including Enzo ;) )... yet I bet thousands do work and even more, the parallel bridged configuration IS suggested in the official applications sheet.

I just suggest playing safe , following established rules does not hurt any day of the week.

Overkill?
Maybe, just as much overkill as carrying a fire extinguisher in a car ... although Millions of cars run happily and never catch fire.


phatt

  • Legendary
  • ******
  • Posts: 2039
  • Chip Points: 242
    • View Profile
Re: Zobel Question
« Reply #9 on: October 14, 2016, 07:12:15 AM »
Thanks heaps,
                    Yes it's all built and works fine but I will open it up again and check out the possibilities of adding the zobel network. There might be just enough room to tap a wire from the junction of R1&R2.
But not enough room for the Fire Extinguisher.  :lmao:
I'll get some pictures next time I have it open.

I actually carry a fire extinguisher in our car,, I was in once in a friends car that caught fire and lucky the chap behind us had one on hand which saved the day. Needles to say we replaced his extinguisher and an extra couple for us. ;)
Phil.

R.G.

  • Master
  • ***
  • Posts: 87
  • Chip Points: 241
    • View Profile
Re: Zobel Question
« Reply #10 on: December 04, 2016, 08:39:49 AM »
The Zobel network is a protection against oscillation caused by an overly inductive load. The damped inductor LR network in series with the output of many amps is protection against overly capacitive loads. Pretty much 100% of hifi amps use both.

It's entirely possible that this is simply a thermal problem, too. If heat sinking is inadequate, even the internal thermal shutdown may not save the chip after zillions of overheat cycles. That can be as simple as incorrectly torqued mounting bolts holding down the semiconductor to the heat sink, or too much force that tilts the metal tab on the heat sink.

Just speculating.

 

* User Controls

 
 
Welcome, Guest. Please login or register.
Did you miss your activation email?

* Recent Posts

Laney VC30 Output Transformer by phatt
[Today at 12:03:26 AM]


signal tracing by g1
[December 10, 2019, 02:26:30 PM]


Eight Audio Kit TDA7294 and TDA2050 from India by laikairawk
[December 10, 2019, 01:57:45 AM]


The Dangers of Speaker Impedance Mismatching by Enzo
[December 10, 2019, 01:15:39 AM]


Fender Mustang II v.2 by Enzo
[December 08, 2019, 04:46:37 PM]

* Sponsors