Are Class D's Disposable?

[spatters]

What everyone forgets is that for every old Sunn, Peavey, Eden or whatever old amp still out there, there were many more thrown in the trash.

It's called "survivor bias."
The only old amplifiers we still see are the ones that didn't fail and get tossed in the trash.
Thus, we think "Amps were made better back then," but that's not true: what really happened was that all the junk was thrown away decades ago, and the amps that remain are either the good ones, or the ones worth fixing.

Consider:

• Where are all those terrible Ampeg and Marshall transistor heads and combos from the 1970s and 1980s that we've all agreed to forget existed? In the trash.
  
• Crate probably sold more Crates than Peavey sold Peaveys, back in the day. Where are they now? In the trash.
  
• How about Holmes? KMD? How about the dozens of off-brand Japanese solid state amps from the 70s and 80s? In the trash.

I'm willing to bet that modern Class D amps are more reliable, on average, than amps from 30+ years ago. Furthermore, they sound better and are more powerful.

If the timing gets too close (each half of the bridge must turn off before the other half turns on) you get cross-conduction which is a quick way to lose all that expensive magic smoke.

Remember that the higher quality class D amps are switching at between 300kHz and 500kHz, so we are talking about very, very short time intervals between the pulses.

This is amongst the many differences between "it works in theory" and "it works in the field." Class D amplification is a very simple concept! But it took decades for the technology to take over the audio world, despite the large and obvious efficiency advantages.

 

[baddarryl]

Few years back, during my class "D" bass head search, I had a BAD run with "two" of Amp*g's PF500 "D" amps.

Both heads burnt out.

Loud/Amp*g replaced the first one, second one I had to take to a certified Amp*g service center.
Service techs actually told me to toss it!!!!!

I've now (somehow) 4 class D amps.
A Mesa 800+, a G.K. MB Fusion 800, a G.K.MB2-500, and a G.K. MB500.

They've all proven to be real work horses in the gigging band I'm with.

I am the OP and discussion of those amps and their issues is what brought this up.

 

[Marty Forrer]

I dunno what the heck you folks are doing with your amps. I've been gigging since 1963, some of it as a fulltime pro, and I have never once had an amp (or speaker cab) go down, even in the 70s and 80s when things got stupid loud. Maybe it's because I live in a temperate climate?

 

[joelns]

I've got an old Hartke Kickback 120. It cuts out regularly, farts out, and has other issues. I keep meaning to take it to a shop for an estimate, but I'm afraid the estimate/cost is going to be more than it's worth. Maybe good for $200 if I'm lucky. I'm feeling like this one is indeed disposable.

 

[Phlipper]

It's called "survivor bias."
The only old amplifiers we still see are the ones that didn't fail and get tossed in the trash.
Thus, we think "Amps were made better back then," but that's not true: what really happened was that all the junk was thrown away decades ago, and the amps that remain are either the good ones, or the ones worth fixing.
I'm willing to bet that modern Class D amps are more reliable, on average, than amps from 30+ years ago. Furthermore, they sound better and are more powerful.

Dunno the statistics. But it sho don't seem like this is true, re reliability. Also, seems many feel they need 600 watts and a 4x10 just to be heard nowadays. Ignoring better/more efficient combos, small cabs, and micro heads.

But what do I know? I'm effing old.

 

[2strings2frets]

One old gentleman I knew coined what he called the fifty hour rule. “If you get to fifty hours with an electronic device , the next fifty thousand hours are free.” Might be a small exaggeration but not far off just the same. My much criticized Bh3000h is ten years old with no issues. I traded someone my Acoustic mini stack 160 , years ago and it is still working today in spite of long and hard use by several musicians. Class D forever until the next advance in technology.

 

[lug]

One old gentleman I knew coined what he called the fifty hour rule. “If you get to fifty hours with an electronic device , the next fifty thousand hours are free.”

Called the infant mortality rule by the more gruesome of the bunch.

 

[baddarryl]

OP here: Sometimes I go through my old Content on my profile and found this. After reading the whole thing I am looking at my 72 V4B and my Vietnam era SVT and wondering if I should just sell them. I use my Quilter BB800 much more because I don't want to put too much wear on them and honestly it might sound better than the SVT. I did just have the V4b serviced and re tubed so there you go. There is a reliable, reasonable tube amp tech here in town and I fully understand that without him I'd be screwed. Thinking.................

 

[rutrho]

Class D stuff in general is something that can be serviced but the skills need to do it are different than what is required to work on older tube amps or lead sleds. In general the newer stuff is better designed and more components are working within their design envelopes, which helps to lengthen life.

Some of the issues I've seen with class D stuff so far have all be with switches, pots getting dirty, output jacks having cold solder joints or seeing fatigue/damage from handling or PCB failures from handing. When I opened up my Rumble 500H I saw a lot less potting/staking of parts, leads, wires/etc than I'm used to (I deal with electronics and mechanisms in ultra harsh environments) but nothing that'll limit the amp head from having a long service life as long as I don't throw it down a flight of stairs, out a window or spill beer on it.

Called the infant mortality rule by the more gruesome of the bunch.

Yep.
We do all kinds of burn in testing and functional testing to find those issues before mission start. Consumer devices... don't get it to the same degree.

 

[msb]

Since I posted in this thread I’ve had my little GK MB200 exhibit the pink ring of doom. It was sent to a qualified tech , and is back and working wonderfully. This is a practice head , and is usually run at low volume for several hours every day. (the joy of retirement)
The fan had become gunked up , and the power chip had to be replaced (probably due to the fan not working properly)
When the amp died I researched other similar amps and decided I liked the MB200 enough to replace it with another if it could not be fixed.
It was fixed , for a fraction of the replacement cost.

 

[rutrho]

Since I posted in this thread I’ve had my little GK MB200 exhibit the pink ring of doom. It was sent to a qualified tech , and is back and working wonderfully. This is a practice head , and is usually run at low volume for several hours every day. (the joy of retirement)
The fan had become gunked up , and the power chip had to be replaced (probably due to the fan not working properly)
When the amp died I researched other similar amps and decided I liked the MB200 enough to replace it with another if it could not be fixed.
It was fixed , for a fraction of the replacement cost.

This actually hits on one of the trends with the class D stuff that I dislike. The cooling fans. There's a lot that can go wrong with fans as they age, and they're often one of the more moisture/temperature/environmentally(dust) sensitive parts of the the whole cooling system in them.

I really really wish that some of them had passive heatsinks vs having fans. I'd much rather a couple extra pounds of CU and AL to make a better heatsink vs having to rely on a fan, but, thats a trade off a lot of people won't make for weight and the fans are a LOT cheaper than additional heatsink material.
That doesn't get into the whole thermal conductivity aspect and what efficiency differences would be there in active vs passive cooling, but that is a totally different kettle o fish.

 

[agedhorse]

This actually hits on one of the trends with the class D stuff that I dislike. The cooling fans. There's a lot that can go wrong with fans as they age, and they're often one of the more moisture/temperature/environmentally(dust) sensitive parts of the the whole cooling system in them.

I really really wish that some of them had passive heatsinks vs having fans. I'd much rather a couple extra pounds of CU and AL to make a better heatsink vs having to rely on a fan, but, thats a trade off a lot of people won't make for weight and the fans are a LOT cheaper than additional heatsink material.
That doesn't get into the whole thermal conductivity aspect and what efficiency differences would be there in active vs passive cooling, but that is a totally different kettle o fish.

This is not completely true.

A heatsink still has to dissipate the heat into the environment, and as amps get smaller with higher rated power, the surface temperatures climb no matter how you try to dissipate it passively. I designed a very small class D amp a while back and there was no way that I could meet the UL/IEC chassis surface temperature without using a fan. No safety certification makes such a product a "no go".

Good quality fans are not necessarily cheaper than a heatsink either, in fact it's generally more expensive.

 

[rutrho]

This is not completely true.

Which part? I'm genuinely curious... though I think we could spare the thread a dissertation on factors that go into thermal management and design thereof. My background on thermal stuff deals with black body radiation of heat which is a different beast then dealing with thermal management in a gas. I'd have to run numbers on how much surface area is required to dissipate heat generated by some amp running at say, a watt load, and what the thermal conductivity coefficient is through that material... But that's a project for another time.

The regulatory body certs was something I totally didn't think about given I have zero background in consumer hardware engineering.

Do you have any insight on fan life with those environmental factors mentioned above?

 

[beans-on-toast]

The Sony C800G microphone uses a Peltier cooling system that helps cool and lowers the noise floor. It would be interesting to use that technology in a musical instrument amp. Not too cost effective though.

 

[Sparky Mark]

My GK MB110 combo has a 100 watt class D amp without a fan. As class D output stages are much more efficient than class AB I reckon you could get to 200 watts with passive cooling but the heatsink would probably still need to be a considerable size.

The heatsinks on my passive cooled Trace Elliot 200 and 250 watt class AB heads weigh as much as a 500 watt class D head on their own and they get pretty warm when driven hard.

Modern fans in combination with smaller internal heatsinks are extremely reliable and checking them for accumulated dust/gunk (even spiders!) periodically is a fair preventative maintenance measure.

 

[Hot Rats]

Nothing is "perfectly" reliable. I repair medical equipment. Nowhere on Earth are greater precautions taken than in an OR. If there's a way to make something more reliable, and have backup after backup after failsafe after backup, and OR is where you'll find it. And, even there, things occasionally break down. The same holds true for space programs and nuclear facilities. No matter how hard we try, we haven't "perfected" ANYTHING, yet.

You have just perfectly described the old " ghost in the machine" phenomenon.

 

[agedhorse]

Which part? I'm genuinely curious... though I think we could spare the thread a dissertation on factors that go into thermal management and design thereof. My background on thermal stuff deals with black body radiation of heat which is a different beast then dealing with thermal management in a gas. I'd have to run numbers on how much surface area is required to dissipate heat generated by some amp running at say, a watt load, and what the thermal conductivity coefficient is through that material... But that's a project for another time.

The regulatory body certs was something I totally didn't think about given I have zero background in consumer hardware engineering.

Do you have any insight on fan life with those environmental factors mentioned above?

When you take the gas away, the thermal equations change a bit...

Part of the issue is contending with high ambient conditions, say when the amp must operate in 105 deg F ambient things get more challenging than say 70 degrees F. At some point you reach the point of diminishing returns.

Quality fans (ball bearing) have a rated lifespan of 50k hours. Low quality sleeve bearing fans have a rated lifespan of 5k hours. Guess which ones are more expensive? Companies designing only to a price point may have to use the less expensive option, which reduces reliability. I have used ball bearing fans exclusively for over 10 years, in MY opinion it's worth the extra cost.

 

[fdeck]

There are very tight timing requirements within the gate drive circuits on both the SMPS and the power amp. With the power supply, this affects efficiency and radiated noise. With the power amp, this affects distortion and radiated noise. If the timing gets too close (each half of the bridge must turn off before the other half turns on) you get cross-conduction which is a quick way to lose all that expensive magic smoke.

Remember that the higher quality class D amps are switching at between 300kHz and 500kHz, so we are talking about very, very short time intervals between the pulses.

I wonder what was more important in the history of reliable class-D amps: High performance MOSFETs, or high performance oscilloscopes.

 

[S-Bigbottom]

I really really wish that some of them had passive heatsinks vs having fans. I'd much rather a couple extra pounds of CU and AL to make a better heatsink vs having to rely on a fan, but, thats a trade off a lot of people won't make for weight and the fans are a LOT cheaper than additional heatsink material.

I love that my head can be on for three hours, turn it off go right into its soft carry case and I am off to home.
It never gets warm to the touch.

Unlike the old SWR's with their fins.
Them things you can place a cheese sandwich on them and play the first set.
At break, flip it.
Play second set to get the other side.

After the gig you have a grilled cheese sandwich to eat while you wait for the amp to cool down enough to pack it up.
Even the knobs got hot on them things.

 

[baddarryl]

When you take the gas away, the thermal equations change a bit...

Part of the issue is contending with high ambient conditions, say when the amp must operate in 105 deg F ambient things get more challenging than say 70 degrees F. At some point you reach the point of diminishing returns.

Quality fans (ball bearing) have a rated lifespan of 50k hours. Low quality sleeve bearing fans have a rated lifespan of 5k hours. Guess which ones are more expensive? Companies designing only to a price point may have to use the less expensive option, which reduces reliability. I have used ball bearing fans exclusively for over 10 years, in MY opinion it's worth the extra cost.

On a chance Andy do you happen to know what the Quilter BB 800 uses for a fan?