• TalkBass has been independent since 1998. Add your voice.
    Create a free account to reply to discussions, view embedded media, and browse with fewer display ads.
    Join freeLog in
    Want zero display ads or expanded classifieds tools? Compare plans.

Acoustic B20 with blown fuse...

I don't really think it's luck. Because you work with a specific manufacturers product line, you don't get to see all the other crap that's out there.
I was recently working on an AER Compact XL amplifier made in Germany. The driver chip that drives the output darlingtons was mounted on a little tiny heatsink that ran so hot, you couldn't hold a finger on it for more than 1 second. The electrolytics in close proximity to it most certainly get cooked to death.
In checking with the warranty depot, I was informed in not so many words that this was operating normally.
 
I see and qualify a lot of parts and assemblies, there's a lot of good stuff out there. I don't know how I have missed ALL this bad stuff???
 
wierd dude i feel the same way about tires, tires never go bad. tires never go flat. in fact every tire i ever inspected at the tire factory was just fine. cant believe ive missed all these flat tires people talk about. ive never ever seen a flat tire before. and i have inspected literally thousands of them lol.

i think its totally wierd too about broken windows on the internet, when i worked at a window factory all the ones i inspected were just fine. never seen these broken windows. there is real quality windows out there trust me ive seen them. shouldnt have a broken window, must be poor design.
all the windows i built and sold were never broken. broken windows really are not a problem. atleast all the windows ive made. none of them were broken. and ive seen thousands of windows. never had this problem.
 
wierd dude i feel the same way about tires, tires never go bad. tires never go flat. in fact every tire i ever inspected at the tire factory was just fine. cant believe ive missed all these flat tires people talk about. ive never ever seen a flat tire before. and i have inspected literally thousands of them lol.

i think its totally wierd too about broken windows on the internet, when i worked at a window factory all the ones i inspected were just fine. never seen these broken windows. there is real quality windows out there trust me ive seen them. shouldnt have a broken window, must be poor design.
all the windows i built and sold were never broken. broken windows really are not a problem. atleast all the windows ive made. none of them were broken. and ive seen thousands of windows. never had this problem.
You have completely missed the point.

Sure I have specified and inspected new parts, but more importantly I have inspected parts that are over 30 years old and they too are just fine. I am not seeing the exaggerated failures that you have suggested, even after 20 years of operation, over tens of thousands of units.

It's not like tires going flat or windows breaking, an absurd analogy at best.
 
You have completely missed the point.

Sure I have specified and inspected new parts, but more importantly I have inspected parts that are over 30 years old and they too are just fine. I am not seeing the exaggerated failures that you have suggested, even after 20 years of operation, over tens of thousands of units.

It's not like tires going flat or windows breaking, an absurd analogy at best.
I dunno, recap seems pretty safe to me.

Screenshot_2017-06-17_23-03-48.png


Only 154.000 problems on audiokarma.
 
There may be a design issue in the amp you mentioned that is independent of the cap, so it may not be the cap's fault but the designer's fault. It's important to understand what the actual cause is before "knee jerk" placing the blame on "defective caps". That's not something that I can comment on since it's not my design.

Since my designs go through reliability profiling of all circuitry, I know what the reliability metrics should be based on the statistics. So far, I have not been surprised by "bad caps" yet, but it's only been 35+ years. I did get surprised by counterfeit IC's, and a mask update on an IC that changed the relative timing of a critical switching part, but for the most part if you understand the boundaries of a part's performance envelope then there's not much drama or finger pointing to be had. I'm not the only designer that's capable of this either ;)

Let me clarify my position a bit. The B20 is an extreme case, because they, (probably naively in 2007, they should have known by then, strike one), used a batch of cheap caps that happened to get caught up in the dielectric problem. Quality engineering was marginal, but good quality engineering is expensive, and most consumers won't pay for it. This is the possible result. I don't think their ripple current is excessive, or that they are running the cap near it's voltage rating, but I have to admit, I haven't measured it. Generally, those errors would dry up a good cap in time, but probably not make it burst. I fully believe in the dielectric problem as described in the Wiki article. When I was running my web retail business, I had 4 HP computers and a few monitors in the offices all crap out prematurely due to power supply failures, even though my collectors item early IBM PC's (20 something years older) were still running just fine. These HP's were well built, about the same time as the B20s. A computer repair shop near here was thinking of hiring my step-son (who has his A+ computer tech certification) primarily to replace caps full time, so many were failing. But I'm getting off subject.

Looking at the internals of the B20, I really can't fault the engineering or production too much in there. Yeah, it's small, inexpensive, and I still think the fuse is too close to the hairy edge to last, but generally not bad for a cheap practice amp. The build quality is actually pretty decent for a consumer product, and I am critical. I have one complaint in the way the heat sinking is handled. The audio power IC is mounted on a little shiny (black anodized dissipates about twice the heat), non-finned heat sink, about 3 1/2 x 2 inches. This is not a class D device, class AB or AB1 I think, which means it will generate some heat. That heat sink is screwed down to a nice, big, relatively expensive aluminum chassis that could have dissipated a bunch more heat, but they didn't bother putting a nickels worth of heat sink compound between the small heat sink and the big aluminum chassis to conduct the heat through the joint, so a lot of the heat will never get piped past that joint, and the amp parts will run hotter and less reliably than necessary. What a lost opportunity for a cheap, significant boost in reliability. I intend to add some heat sink grease to mine.

I was one of the guys who did the reliability studies at Tellabs, a top notch engineering company that built a lot of gear that lives depended on. When the manufacturers of capacitors come out and tell you that their electrolytic capacitors degrade significantly in around 2000 hours of operation, less at higher temperatures, you need to believe them. That's only a couple hours a day for around 3 years. Granted a good cap will just lose effectiveness, in a conventional amp, that is likely to just cause more hum, and not pop the top on the cap or blow the fuse right away. The failures we commonly saw 10 years ago (like the B20) were far more catastrophic failures. But it's still a good reason for me to routinely replace the electrolytics if an amp is a few years old and I'm in there anyway, or if a failure would be catastrophic. It's a few bucks, and an hour, two maybe? If it is a normal power supply filter cap, I will also increase the capacitance value if it will fit, the amp will be less noisy, and when the cap loses capacitance, it will still be above what is required for a while longer.

See the following links to the cap vendors reliability information. Lifespan of a lytic is not a mystery, the data is readily available.

Invalid Link Removed
Invalid Link Removed

I can tell you two things from first hand experience in working on failed gear since vacuum tubes were the norm:

I've seen a lot of power supply caps fail. They probably aren't much more reliable than vacuum tubes. As a matter of fact, the 1959 vintage Hammond M3 organ in my Avatar over there is full of original issue vacuum tubes that are working fine, but it hums because the lytics are dried up.

If you work for a company, you only hear about the DIY failures, not the successes. If someone fixes their amp themselves, they aren't going to tell YOU, it'll void their warranty, if there is any left. They will have another beer and play on with a great sense of well deserved pride. If they trash it, they may send it in and try to con you into fixing it under warranty, there are those people out there, then you will find out about it. So what do you think you are going to hear about? The successes? Ahh, Nope. I have never told a manufacturer about my successes, except one design error I found in an HP scope when I was fixing test equipment at Motorola, but that is a different subject...

Marty

P.S. Sorry I'm kinda long winded. I probably shouldn't be PUI of high ABV homemade beer :).
 
Last edited:
  • Like
Reactions: JGbassman
Sorry, I disagree (again).

If you were to do the math, you would see that choosing the right caps and applying the proper aging functions based on the ratio of operating temperature to design temperature it would be obvious (to somebody who actually understands how this works) that the base lifespan gets multiplied by the reduced aging loss coefficient. Even an "average" cap can last 10k hours under properly applied condition. A better than average cap will in fact likely last 20k hours with properly applied aging factors. That's (obviously) 1000 hours a year for 20 years, or about 3 hours EVERY DAY.

Replacing caps every few years is simply absurd, a waste of time, and is likely to damage PCBs within the first couple of attempts. There is no evidence whatsoever that this is beneficial in any meaningful way.

Increasing capacitance might be ok but it might not be as well. You are suggesting that folks who have no idea what they are doing second guess the engineer that has both the education and knowledge to make this decision on the product. Many protection systems are designed around the amount of energy stored in the power supply. Increasing this energy can cause the power supply to fail under a fault condition, increasing the risk of damage and possible fire, which would invalidate the safety agency listing. The liability would fall squarely on the person who made the unauthorized modification, as it rightly should.

Regarding DIY repairs, I have unfortunately seen the results of DIY attempts and while some may be ok, the majority I have seen are abysmal. The results are damage that often renders the product unrepairable (or not economically repairable) when the original real problem was a simple, cost effective repair when the tech correctly identifies the fault before hand.

In general, if you don't know what you are doing, leave things alone.
 
Sorry, can't leave this one alone.

Sorry, I disagree (again).

If you were to do the math, you would see that choosing the right caps and applying the proper aging functions based on the ratio of operating temperature to design temperature it would be obvious (to somebody who actually understands how this works) that the base lifespan gets multiplied by the reduced aging loss coefficient. Even an "average" cap can last 10k hours under properly applied condition. A better than average cap will in fact likely last 20k hours with properly applied aging factors. That's (obviously) 1000 hours a year for 20 years, or about 3 hours EVERY DAY.

Not sure if I've been impressed with intelligence or baffled with B.S. here :). Of course you'd have to know the engineering specs on the caps that are in your amp, and you'd have to know that the caps were actually meeting their specifications, which without a professional parts qualification team, you don't. All you have is faith in the team that designed and built your amp and the reputation of the parts manufacturers. Figures don't lie, but liars figure. Even in the case you state, your assuming 3 hours a day? Well, maybe for a practice amp, but my Linux workstation here is on 24x7, so should I expect the caps to die in 2 1/2 years? That's still about the most unreliable part in there, except maybe the fans. Anything with fans is a failure waiting to happen. They don't last.

In the case of our B20, there is very little information on the caps. Apparently no batch number, no date code, even the name on them seems to lead back to a distributor that buys whatever cheap parts they can get. They have a number, CD110, who knows what that is, and a temperature range, -40 to +85C, but without a spec, or certification from a reputable lab, that is worth the cost of the ink it's printed with and that's about it. Unfortunately, this is probably typical of the quality of the parts that go into gear that is built to compete on price above all.

Replacing caps every few years is simply absurd, a waste of time, and is likely to damage PCBs within the first couple of attempts. There is no evidence whatsoever that this is beneficial in any meaningful way.

Did I say I would replace the caps every few years? Did anybody else? I believe I stated that if an amp was a few years old, I was in there anyway, and the caps that were in there were not top quality parts, I'd change them. I stand by that. After replacing the failure prone crap with good stuff, I wouldn't expect to be back in there for another 10 years if I only used the amp occasionally. Our experience with the B20 makes a pretty good case for this. Generally, I let reality be my guide.

If someone's soldering skills are so bad that they can't replace a part once or twice without trashing the circuit board, either they are a newbie that didn't heed my advice to practice up on junk first, or they are a klutz that should be forever banned from handling tools. Soldering really is an art, a matter of pride for those who master it, any good tech. A competent tech with a quality soldering iron can replace a part a number of times without trashing the PCB. I don't think I have lifted a trace since about, maybe, 1973?

Increasing capacitance might be ok but it might not be as well. You are suggesting that folks who have no idea what they are doing second guess the engineer that has both the education and knowledge to make this decision on the product. Many protection systems are designed around the amount of energy stored in the power supply. Increasing this energy can cause the power supply to fail under a fault condition, increasing the risk of damage and possible fire, which would invalidate the safety agency listing. The liability would fall squarely on the person who made the unauthorized modification, as it rightly should.

Wow. A couple comments.

I find it intensely amusing that all this caution is coming from an employee of Mesa/Boogie. According to a piece in the book "The Tube Amp Book" by Aspen Pittman (good book by the way), Mesa/Boogie came about when Randall Smith, a technician, stripped the guts out of a Fender Princeton, and stuffed in the circuitry of a Fender Bassman, and a JBL D120F speaker. My kinda guy :). Sounds like something i would have done, if I could have afforded a D120F. Somehow, I doubt that the project was blessed by a real honest to god holier-than-though engineer, or that it passed a battery of regulatory compliance tests (which our B20 didn't either). This kind of innovation generally comes from free thinking individuals who are not getting all caught up in regulations and "what-if"s". Sure he might have gotten sued if he sold it, some dolt mis-used it, and hired a lawyer that thought Randall had enough money to be worth trying to screw out of it. There is a certain security in not being rich enough to be worth a lawyers efforts :).

I think the odds of doubling the size of the filter caps becoming a hazard are about the same as the odds of being hit by lightning. The upside is that the amp will exhibit less noise, and last significantly longer before the caps age to the point of being unable to do the job of keeping the hum level down. Odds are the engineer, if he/she were a music person, was forced to chose values lower than he/she would have preferred to begin with, for cost reasons. The engineers don't have final say in these matters, the bean counters do. Such is the reality of consumer electronics. "Build something as good as you can for $X." Thats how it goes. Granted if the power supply is extremely marginal, the additional start up current of the bigger caps could be an issue, but it's pretty unlikely. Personally, I'll take that risk, or improve the power supply if necessary.

I can't help but take issue with the engineer worship going on here. Yeah I became a Senior Member of Technical Staff at Tellabs, an engineering title not awarded to many engineers. But for a number of years of my career, I was jumping on planes and going into critical outage situations that were more often than not the result of an error on the part of engineering (granted the usual hardware failures and application errors were typically solved before I got involved). Engineers are people, and people make mistakes. People well trained in theory without a whole lot of real world experience don't always get it right. Our product got a whole lot better after I created a test group, staffed by a couple degreed engineers, and a couple techs with significant real world field experience, to beat the heck out of the product before we sent it out. That is what it took to achieve the 99.999% reliability that we eventually achieved.

Regarding DIY repairs, I have unfortunately seen the results of DIY attempts and while some may be ok, the majority I have seen are abysmal. The results are damage that often renders the product unrepairable (or not economically repairable) when the original real problem was a simple, cost effective repair when the tech correctly identifies the fault before hand.

In general, if you don't know what you are doing, leave things alone.

When you say leave things alone, you are suggesting what, tossing the amp in the trash without even trying, or paying someone more than the unit is worth to repair it? Unfortunately, that is often the reality.

Again, the majority of DIY repairs that you have seen. Most likely, you would not have seen the successes. Your sample group is biased.

What we are talking about here is an amplifier that on the whole was built pretty well, but got caught by sourcing cheap caps that were part of the plague of Asian caps made with bad electrolyte. Doo-doo can happen when the price of the product will not support top notch components and quality control. I really feel for the guys that had to take that gamble, and lost. I might have taken the same gamble, given the choice they had. It was a gamble trying to deliver a price competitive product, they lost this time.

Now when someone comes across one of these amps, they have a choice:

1. They can take it to a shop, which probably needs to charge $50 or $60 an hour to pay the staff, the rent, the heat, and keep the lights on. That is the reality of the situation. Keeping a repair shop operating costs money.

2. They can scrap it, the speaker isn't bad, I intend to test mine and model the Thiele-Small characteristics to see what else it may be good for, before I put this thing back together.

3. They can take a look inside and see if their amp looks like so many other B20's, the cheap caps failed, are bulging on the top or have blown open, started drawing excessive current, and blew the fuse. Ironically, the fact that so many of them fail in exactly the same manor is actually an indicator of good quality. They all only failed because of the same failed part. If the cap supplier hadn't failed them, most likely they'd all still be up and running. The goal in quality is consistency, even in failure.

For the price of a decent six pack they can replace the caps with a better brand, and the fuse, if they know how to solder, or are willing to learn.

I came up through the school of hard knocks. I learned to troubleshoot when my stuff didn't work. I got good when I wanted stuff that didn't exist or I couldn't afford to buy it. So I built it. It served me well as a profession, and as a bass player.

I'm near the end. My time is nearly up. I want to encourage another generation to blow off the naysayers and take on the challenge of first fixing what's in front of them, and then going a step beyond that, creating something that is new, that advances the art of communicating emotion through music. Somebody needs to do it :).

Marty
 
You want to encourage the next generation to blow off the naysayers, I'm encouraging the next generation to get an education, learn about how/why things work. This is the responsible thing to do IMO. You might not agree, that's fine, you don't have to.

Reminds me of guys that blow off naysayers... usually starts with "hey, watch this".
 
Was having coffee this morning within a couple of engineering buddies and this topic came up.

Increasing the capacitance on a large power supply or a switching power supply is probably more of a risk because impedance are much lower and currents are much higher. Why might this be so important you might ask? Because in order to meet line transient requirements for agency certification, large power supplies have some form of inrush current management that is calculated based on known parameter. Doubling the capacitance as you suggest would double the inrush current envelope and quadruple the power that the inrush elements must manage., putting them at risk of catastrophic failure. This alone would be a direct violation of the product's safety rating, and jeopardize the consumer.

You keep mentioning 24/7 operation... that's not how MI amps are aged. Apply reasonable, real world aging factors (you keep saying real world versus "us engineers) and properly specified caps should last 20-40 years in today's quality MI products.
 
  • Like
Reactions: staccatogrowl
@Marty Hewes Are you a trained, knowledgable engineer? You’re arguing with one of the best in our industry. I’ll take his word over yours any day and twice on Sunday! As a somewhat ancient engineer my self I understand where Andy is coming from. I will admit that he leaves me with a headache at times but in all cases he has been proven right.

It distresses me when I see someone so disrespected who is giving freely of a lifetime of learning and experience. I too have seen units with DIY repairs. They ranged from sad to destroyed.

I remember an amp with MOSFET outputs where the owner had decided that he could get more power by adjusting the bias. He switched both banks of devices fully on which destroyed them all. That was an expensive repair to what was a fully functional amplifier.

No one is bashing a knowledgable amateur from repairing his/her amp. It’s knowing when you are out of your depth that’s the trick.