Ouch. I'm not even working as an engineer any longer and I see maybe half a dozen misconceptions or conflations promulgated on this thread that could use some clarification. I will share what info I have, and apologies in advance if I too am promulgating any misconceptions or conflations. I never designed AV equipment, just for full disclosure.
Originally Posted by djp2k7
Pioneer elite avr's are using class D amps now which run quite cool
The efficiency of a switcher approaches double that of some commonly used linear power amps and is theoretically the highest efficiency available that I am aware of. I am glad this market shift is finally possible now that audiophoolery regarding tubes and linear amps is dying out under the pressure of Atmos etc. object-oriented surround where consumers want more and more amplifier channels crammed into a single device at the lowest possible cost.
The first few attempts at affordable full spectrum switching amps introduced stability, reliability, and audibility issues that made the entire industry gun-shy for at least a decade. Pioneer sold stereo receivers with switchers in budget equipment years ago that worked fine and I've used one of them that was still working fine even after someone's coffee wiped out the tuner. All plate amps in powered speakers that I know of use switchers.
A dog might not appreciate the switcher whining away in the background but if you scritch behind his ears that will mask the whine.
I doubt any human will hear any issues under any conditions with modern switchers, particularly with the prevalence of hearing damage today where the US navy is using frequency shifting so that recruits who have already damaged their ears as kids with earpods can still do sonar.
My supposition is that advanced computer simulation has made predictable and reliable switcher designs far easier to accomplish. I don't know if switchers apply to higher powered amps in general but I do know there is at least one high powered design out there that people have raved about for years. As with all higher powered audio addicts (let's raise our hands everyone) I suspect these enthusiasts may have damaged their hearing to the point they wouldn't know the difference anyway even if switching noise was a factor for someone else.
Originally Posted by noears
What's the big deal if the amp runs hot?
No one called attention specifically to the actual concern! SMH and
It's the electrolytic capacitors. The electrolyte chemically degrades under prolonged elevated temperatures. Electrolyte was also behind the 'capacitor plague' where a stolen, incomplete formula was used by multiple manufacturers. The part of the formula that was missing was the part that stabilizes the electrolyte.
Commercial spec capacitors are rated for 1000 hours of operation within rated tolerance at 85C (translates to... um... roughly six weeks) whereas commercial spec integrated circuits will run basically indefinitely at that junction (not case) temperature. Capacitors don't dissipate appreciable power internally, and they don't usually experience anything like that temperature, but they still get baked to death eventually inside a hot chassis, particularly if the PCB they are mounted to is doing double duty as the primary heat sink for the integrated circuits on it.
This is the problem with Onkyo receivers IMO. Their backplane is so hot that the commercial spec local regulator filter capacitors degrade drastically with normal use. The HDMI/DSP board becomes plagued with bit errors from degraded power supply filters and eventually fails entirely.
My TX-NR929 that includes better power management in the HDMI/DSP is doing slightly better than my TX-SR706 did, so far. Adding DIY heat sinking to the chips, or putting a fan on top of the case that blows directly at the HDMI board, helps keep the capacitors cooler by keeping the PCB cooler.
The internal fans of my 929 that could help with this problem have not once turned on because I am in an apartment and I cannot run the amps loud enough here to dissipate that much heat in the amplifiers and activate the fans. This receiver might paradoxically last longer if I can turn it up loud enough to activate the internal fans because the improved air flow might actually cool those vulnerable backplane capacitors down a little.
You will not notice that your capacitors are degraded until it gets bad enough to cause issues, but rest assured, they are fading slowly all the time and much more quickly if operating at elevated temperatures.
Mil spec are rated for 2000 hours. (disclaimer: I have not designed for a while and may be in error). Big whoop. Yes, they are better, probably much more than twice as reliable under elevated temperature than one would conclude from the spec, but they will still fail eventually if abused with a de facto oven environment. I try to buy PC components like motherboards and power supplies that advertise mil spec caps. Why invest in high performance equipment that will fail in 5 years? I am not a power user on the upgrade treadmill.
Originally Posted by noah katz
Life of semiconductor materials decreases exponentially with increase in temperature.
This statement may be technically true (or not, depending on the context) but is totally irrelevant to the average user who is not putting an inertial sensor into a drill bit at the bottom of an oil well. Semiconductors are very stable in consumer equipment. They do experience elevated temperatures because they are a primary (likely the
primary) heat source in equipment, but with normal operation and heat sinking, their dissipation contributes far more to the degradation of capacitors nearby than to their own degradation IMO.
The next is a delightful mix of truth and misconception. I will focus on the misconception that I perceive.
Originally Posted by noears
If you are worried about the components in your AVR being temperature stressed, then you are basically accusing the engineers who designed it of being incompetent.
Corporations are legally driven under articles of incorporation to maximize profit as their sole responsibility to investors. It's not about incompetence or planned obsolescence either, it's about shaving margins.
The reliability/product engineers who do the dirty work make a product just as reliable as it needs to be to last 5 years without excessive design or maintenance cost (manufacturers in the USA have to stock parts for repair of products under 5 years old or buy it back, unless I am completely wrong or something changed recently that I missed). Leave us design engineers out of it please. We only get our hands dirty when the company is too small to afford marketing, product, and/or reliability engineers to collaborate in shaving those razor-thin safety margins to the limit.
Unless we have reports of a high number of AVRs failing of shutting down due to overheating... or higher power consumption than the competition.
See 'Onkyo' above.
If the Yamaha feels cooler to the touch, its because the designers implemented heat sinking which doesn't couple to the case as much as the Denon, not because its generating less heat.
Maybe true, maybe not, but omits vital information about processors.
Newer products that incorporate smaller transistors in the processing path can dissipate much less power while doing the same work, or do much more work while dissipating the same amount of power. Modern receivers do a lot of DSP.
Unless we have detailed knowledge and a compelling reason to use it when making a purchase, we probably will not benefit from this factoid in the slightest, but we might detect the change when our latest and greatest new toy runs delightfully cool compared to the obsolete boat anchor that is holding the door open while we take out the trash.
Originally Posted by noears
I would love to see the data behind this claim: "Most studies show that every 10 degree increase over 85 degrees F leads to a whopping 40% reduction in your equipment’s life span.".
See 'capacitors' above. It's not the same thing as implied in this running conversation about the reliability of semiconductors, but IMO it is the principle factor regarding lifespan vs. operating temperature of the AVR.
I'm not trying to be a pain in the ass.
Me neither. I'm just a pedantic PITA though. It's in my nature.
Originally Posted by torii
when I crank my avr to 0 master volume for a couple hrs it gets up to 150 degree F. the marantz tech support laughed at me impossible til I sent them a vid clip of it real time...then they said my infra red thermometer must be defective...this was 2 years ago +...still working but I want it to die for upgrad itus
150F is ~66C. That is way too hot for the case/chassis from my perspective, but I would not be surprised, particularly if measured in an enclosure during warm summer months with no AC. Maybe you cherry-picked a hot spot on the case or took a reading through perforations/louvers and got a higher reading than the case actually experienced.
Tech support often gives the wrong information. Sometimes even the closeted engineers behind that tech support public firewall don't know what they are talking about. I've been one of them at various times and stumbled into them even more frequently. It happens. The ones who are really good at the art won't generally talk to customers on the help line and often showed disdain for someone of my caliber too who occasionally needed their advice, particularly when I'm being an ignorant, overly ambitious, pedantic pain in the ass.
Originally Posted by noears
And unless you are measuring those components temperatures directly, you have no idea how close to their reliability limit they are.
Originally Posted by noears
So yes I could cool my ASICs, capacitors, inductors, MOSFETS, etc... from 80% of the rated temperature to 50%, but the increase in reliability would be so small its not worth investing the fans or material to do so.
Just trying to discuss some data behind this before people run off and spend money they don't necessarily need to on cooling devices.
Keeping the electrolytic capacitors cooler would bring the biggest benefit IMO and is worthwhile any time the case feels uncomfortably warm to the touch when sensed far from any obvious major heat sources like external/case heatsinked power transistors, or when the power transistors aren't really dissipating much power at lower volume. Miniaturization of capacitors has put more demands on the electrolyte and moved it closer to heat sources, while surface mount technology couples more heat into capacitors from the PCB too.
That is, worth it if you are planning on keeping the equipment five years or more, or you intentionally bought something with a known thermal issue that causes premature failure. Otherwise I would not bother.
I am wondering if failing aspects of Onkyo HDMI design are being incorporated into the newer Pioneer products, or if reliable aspects of Pioneer HDMI design are being incorporated into the newer Onkyo products, or if the two lines are still operating independently. That is insider information I could use when considering my next purchase (psst hint hint Russia I'd love it if you put it on Wikileaks
Originally Posted by M Code
....today's AVRs lack the over-design that previous generation AVRs included.
I don't think it has anything to do with 'over-design'. That is a feature reserved for very high-end niche equipment for industrial use where reliability trumps cost, or for the idle rich who don't want to waste effort replacing failed components instead of golfing. The professional PA at a Metallica concert is likely somewhat over-designed to avoid blowing a live show where they have to issue gobs of refunds and lose everything they invested in the gig. A consumer AVR is almost certainly not overdesigned, at least not intentionally, regardless of when it was built. That would violate the corporate mandate to maximize profit and the CEO would be floating to earth on a golden parachute.
For example, once an output device's case temperature heats up to like 60 degrees C, its efficiency drops rapidly as its SOA (safe operating area) decreases while the user responds by increasing the volume level that accelerates the subject process.
The safe operating area is a fixed specification on a component, not a moving target that changes with the temperature of the case. I think you stumbled on your language but I get the idea you were trying to convey.
I find your statement about the user increasing the volume in response to decreasing device gain baffling, unless you are assuming no feedback around the device to stabilize the gain of the stage.
Even then it is still baffling because the gain of a typical output voltage follower stage that is commonly used to boost the output current in a low/no feedback amplifier (here I am wandering out of my depth maybe) is near unity due exactly to internal feedback within the semiconductor, where the input node and output node of the feedback path are physically identical in that topology of circuit. The gain of the stage doesn't change.
The output power does not drop unless the thermal protection activates and then it is probably going to just click off until the temperature drops to a safe level again.
I hope that helps. Apologies for my pedanticisms.