Class D amp long term reliability

[marX]

The only time that sentence makes sense is if it is followed by a few more sentences.

“Factor of safety is for fools. It gives you a false sense of security. You have to consider novel mechanisms of failure that aren’t accounted for by a simple multiplier.”

Sorry I didn't see this post. This is correct. But in our case, he's just after saving some money.

For example, he would often question why the shaft diameter is bigger than the modulus of rigidity calculation. But it's not that simple. There are more factors affecting the design than just the maximum torque to be transmitted versus the modulus of rigidity.

 

[marX]

It would be neat if someone could figure a miniaturized heat-pipe, radiator and/or liquid cooling implementation that becomes the integral part of an equally small Raspberry Pi case. << None of those 4-letter words. (fans)

How dare you speak ill of fans (and blowers). I was the designated centrifugal blower guy of my company. But, our designs was for pneumatic material conveying purposes. Not for optimum noise versus mass flow rate.

Btw, please take a look at this link.

Lab tour! Go inside AirJet's futuristic solid-state laptop cooling tech

Frore Systems' AirJet tech might just change the way we cool all kinds of gadgets. Get a peek behind the curtain in our lab tour video!

www.pcworld.com

 

[GXAlan]

Sorry I didn't see this post. This is correct. But in our case, he's just after saving some money.

For example, he would often question why the shaft diameter is bigger than the modulus of rigidity calculation. But it's not that simple. There are more factors affecting the design than just the maximum torque to be transmitted versus the modulus of rigidity.

Sigh. The one good thing about the OceanGate tragedy is that it provides a good analogy.

This was your former CTO, but if you still have friends there, I would suggest the following

1) What is the impact to our brand if our product fails in customer hands?
2) What’s the shaft made of? To what ASTM standard do you have for purity? does our supplier test every lot? What about heat treatment? What if our supplier has an error In manufacturing?

You can always blame supply chain as a rationale for added safety. This lets your CTO be “right” and not hurt his ego.

3) Depending on how big your product is … what if the customer‘s site is not level, or a field service tech slightly over torques an installation because the torque wrench or screwdriver has lost accuracy.

You can always blame the customer as a rationale for added safety.

I remember reading that Pringles potato chips are manufactured with the help of computational fluid dynamics. There used to be a speed at which the pringles would fly off the conveyor belts so production was not as fast as they wanted. Using CFD, they redesigned the chip geometry and I think the machines as well so that the chips would not go airborne! In the case of potato chips, answer #1 is OK. I think in every package of Pringles there is a broken chip. No worries. You don’t need a high safety margin if it helps costs stay low — just sufficient safety margin. But if there is contamination in the package? That’s a bigger deal. Food recalls have to be infrequent enough to not harm the brand…

We see some that in audio with some people taking the perspective of “it’s cheap, so I don’t mind replacing when it breaks“ to “I would much rather pay more for a more reliable product even it it measures subtly worse, since that engineering for reliability is more important than engineering for SINAD.”

 

[marX]

Sigh. The one good thing about the OceanGate tragedy is that it provides a good analogy.

This was your former CTO, but if you still have friends there, I would suggest the following

1) What is the impact to our brand if our product fails in customer hands?
2) What’s the shaft made of? To what ASTM standard do you have for purity? does our supplier test every lot? What about heat treatment? What if our supplier has an error In manufacturing?

You can always blame supply chain as a rationale for added safety. This lets your CTO be “right” and not hurt his ego.

3) Depending on how big your product is … what if the customer‘s site is not level, or a field service tech slightly over torques an installation because the torque wrench or screwdriver has lost accuracy.

You can always blame the customer as a rationale for added safety.

I remember reading that Pringles potato chips are manufactured with the help of computational fluid dynamics. There used to be a speed at which the pringles would fly off the conveyor belts so production was not as fast as they wanted. Using CFD, they redesigned the chip geometry and I think the machines as well so that the chips would not go airborne! In the case of potato chips, answer #1 is OK. I think in every package of Pringles there is a broken chip. No worries. You don’t need a high safety margin if it helps costs stay low — just sufficient safety margin. But if there is contamination in the package? That’s a bigger deal. Food recalls have to be infrequent enough to not harm the brand…

We see some that in audio with some people taking the perspective of “it’s cheap, so I don’t mind replacing when it breaks“ to “I would much rather pay more for a more reliable product even it it measures subtly worse, since that engineering for reliability is more important than engineering for SINAD.”

This has already happened (re. broken shaft) and he blamed the customer (it is mentioned in the operation manual that the machine should start and be stable first before feeding the raw materials to it but the customer did not do that. He started the machine after filling it to the brim). The design was done according to his logic - the machine would start without any load.

BTW, I'm more in the camp of “I would much rather pay more for a more reliable product even it it measures subtly worse, since that engineering for reliability is more important than engineering for SINAD.”

 

[EJ3]

heat pipe/water cooling both unnecessary. Both are a solution for tranferring heat from CPU to heatsink. But you still need a heatsink.

With the PI - the whole thing is small enough to have that heatsink in direct contact with the CPU - like this:

Aluminium Armour - Heatsink Case for Raspberry Pi 4

Your Raspberry Pi 4, as small and powerful as it is, still requires passive cooling and case protection to function efficiently and maintain its computing power.

thepihut.com

But in any case - for audio applications if not doing DSP - additional cooling is not necessary.

WOW!: In many different colors & inexpensive.

 

[EJ3]

Its not like they are expensive to replace anyway: 180 euros for a SMPS1200 power supply for example.
https://www.diyclassd.com/product/smps1200a400/10

So, I get paid (don't know Euros, I haven't been to my born home of Salzburg, Austria since they changed from Shillings) $36.91 an hour. To replace the amp: I have to order, find or go to the store to replace it, no matter how you cut it it will take at least 3 hours of my time to obtain a replacement & to actually replace it.
That is it cost whatever the price of the replacement is + $110. 73 ($36.92 X 3) for time involved away from whatever else that I would choose to be doing for the 3 hours that I cannot get back.
Not to mention the time that I did without being able to enjoy my music. That is if everything goes smoothly.
What if when it fails, it takes out another piece of my gear. Say, perhaps my APC that it is pluged into.
Now we are back to whatever the replacement costs of both units (the amp & the APC) [Lets say that it is easier to deal with the APC & it only takes 2 hours of my time].
Now we have the cost of the components + the cost of my time (3 Hours +2 Hours)= 5 Hours x $36,91 = $184.55 + the fact of time that I cannot get back during which I could have been doing something that I wanted to do (as long as it wasn't listening to music, so there is now substantial time that I could not listen to music) + the time it took (that I cannot get back) involved in getting the replacements & replacing them.
I do not see a scenario that can justify the equipment going down.
It is also why I have at least 1 duplicate (or substitute) of everything that that is in my system. So that on the occasion something goes out (or several things go out at the same time), I can simply swap something in in fairly short order & deal with getting a replacement at my leisure.
To me, longevity is very important, certainly worth paying more for.
EJ3

 

[EJ3]

Hey, if it's beyond your skillz....but wtf cares after you're dead?

Because it should last at least until I am dead, not before.

 

[egellings]

That is a 4-letter word, and if you continue using it; I will be forced to press the "Unlike" button...

3-letter word--fan. If the fan comes on only during high power operation, then it would have negligible impact on S.Q. If the fan is always on, then it could be annoying during low volume listening.

 

[EJ3]

3-letter word--fan. If the fan comes on only during high power operation, then it would have negligible impact on S.Q. If the fan is always on, then it could be annoying during low volume listening.

No fans.
That is like using nitrous on a turbo car. If it can't be done with one power adder put 2 on. It increases complexity & if something happens, such as the air/fuel mixture leaning out the failure is catastrophic.
If you can't design a heat sink that dissipates the heat naturally, then you add a crutch that increases complexity & makes a catastrophic failure happen when something happens to it. Such as the fan doesn't com on when it should.
No crutches!

 

[Deleted member 21219]

3-letter word--fan. If the fan comes on only during high power operation, then it would have negligible impact on S.Q. If the fan is always on, then it could be annoying during low volume listening

The phrase "four letter word" is a euphemism for crude, vulgar or otherwise undesirable language. The object of the comment might be four letters, three letters, or sixteen letters, it makes no difference.

Jim

 

[egellings]

No fans.
That is like using nitrous on a turbo car. If it can't be done with one power adder put 2 on. It increases complexity & if something happens, such as the air/fuel mixture leaning out the failure is catastrophic.
If you can't design a heat sink that dissipates the heat naturally, then you add a crutch that increases complexity & makes a catastrophic failure happen when something happens to it. Such as the fan doesn't com on when it should.
No crutches!

Heatsinks don't fail and fans do, but fans allow for a lighter more compact product build, and oftentimes, the product may be later resold long before fan failure occurs. Big thing with fans is that air filters, or the fan itself need to be cleaned on a regular basis.

 

[EJ3]

Heatsinks don't fail and fans do, but fans allow for a lighter more compact product build, and oftentimes, the product may be later resold long before fan failure occurs. Big thing with fans is that air filters, or the fan itself need to be cleaned on a regular basis.

I still have my NAD 2200 Class G amps that the one of mine tested pulled a 95 SINAD through the LAB inputs, see no reason to get something else, I do not sell working gear. It either functions or can be made to function or is destroyed.
You made my point precisely: Heat sinks don't fail.
Obviously I don't give a rat's a$$ about a lighter, more compact build.

 

[pseudoid]

How dare you speak ill of fans (and blowers).
Btw, please take a look at this link.

Lab tour! Go inside AirJet's futuristic solid-state laptop cooling tech

Frore Systems' AirJet tech might just change the way we cool all kinds of gadgets. Get a peek behind the curtain in our lab tour video!

www.pcworld.com

Good link, ty
Confession: For decades now, I had held high hopes for them wuchoomicallemz ThermoElectricModules (TEMs, TECs, etc.) ... for adding/removing BTUs from electronics.
I think fusion will become a reality before practical TEMs.

 

[rdenney]

This has already happened (re. broken shaft) and he blamed the customer (it is mentioned in the operation manual that the machine should start and be stable first before feeding the raw materials to it but the customer did not do that. He started the machine after filling it to the brim). The design was done according to his logic - the machine would start without any load.

BTW, I'm more in the camp of “I would much rather pay more for a more reliable product even it it measures subtly worse, since that engineering for reliability is more important than engineering for SINAD.”

Hmmm. A business model that requires blaming the customer.

Not sure how that will work out.

Rick "bulletproof is desirable even when the bullets violate the user manual" Denney

 

[rdenney]

Heatsinks don't fail and fans do, but fans allow for a lighter more compact product build, and oftentimes, the product may be later resold long before fan failure occurs. Big thing with fans is that air filters, or the fan itself need to be cleaned on a regular basis.

That's true with heatsinks, too. Even unforced convection currents are enough to deposit dust onto the heatsinks over time, reducing their cooling efficiency substantially. And if the owner has cats, it will happen in about 12 minutes by my personal observation. Maybe 10 if it's a Persian. And let's hope nobody smokes cigarettes.

And given that most heatsinks are sawcut from aluminum extrusions with burrs and sharp edges all over them, cleaning them turns out not to be all that easy.

My old B&K amps get warm, but by all reports they have adequate heatsinks. Nevertheless, I installed an additional thermostatically controlled fan that only comes on when the amp is making excessive heat. Why? Because there is no "critical" temperature for component life--the relationship between heat and component failure is both stochastic and nonlinear. Cooler is better even if it is not formally or theoretically necessary. But the output transistors in those B&K amps are now unobtanium.

I have not added the fan to the Buckeye amp, but it doesn't get nearly as warm in use. I wouldn't be opposed to it, however.

Rick "cleaned a lot of dust-impregnated cigarette tar from radio equipment heatsinks over the years" Denney

 

[mhardy6647]

The nefarious influence of cats on audio hardware.
... and woe betide he (or she or they) with a chain-smoking Persian kitty.

 

[egellings]

That's true with heatsinks, too. Even unforced convection currents are enough to deposit dust onto the heatsinks over time, reducing their cooling efficiency substantially. And if the owner has cats, it will happen in about 12 minutes by my personal observation. Maybe 10 if it's a Persian. And let's hope nobody smokes cigarettes.

And given that most heatsinks are sawcut from aluminum extrusions with burrs and sharp edges all over them, cleaning them turns out not to be all that easy.

My old B&K amps get warm, but by all reports they have adequate heatsinks. Nevertheless, I installed an additional thermostatically controlled fan that only comes on when the amp is making excessive heat. Why? Because there is no "critical" temperature for component life--the relationship between heat and component failure is both stochastic and nonlinear. Cooler is better even if it is not formally or theoretically necessary. But the output transistors in those B&K amps are now unobtanium.

I have not added the fan to the Buckeye amp, but it doesn't get nearly as warm in use. I wouldn't be opposed to it, however.

Rick "cleaned a lot of dust-impregnated cigarette tar from radio equipment heatsinks over the years" Denney

I remember cleaning amps used in bars long ago when indoor smoking was allowed. What a mess they were, and it was just not possible to remove the smoky stink from them.

 

[pseudoid]

...smoky stink from them...

Sniffing power amps? Odd!
Was that once a fad amongst audiophile... before the invention of ClassD amps?

 

[egellings]

Sniffing power amps? Odd!
Was that once a fad amongst audiophile... before the invention of ClassD amps?

Didn't have to sniff the bar room amps. They stunk up the room.

 

[EJ3]

That's true with heatsinks, too. Even unforced convection currents are enough to deposit dust onto the heatsinks over time, reducing their cooling efficiency substantially. And if the owner has cats, it will happen in about 12 minutes by my personal observation. Maybe 10 if it's a Persian. And let's hope nobody smokes cigarettes.

And given that most heatsinks are sawcut from aluminum extrusions with burrs and sharp edges all over them, cleaning them turns out not to be all that easy.

My old B&K amps get warm, but by all reports they have adequate heatsinks. Nevertheless, I installed an additional thermostatically controlled fan that only comes on when the amp is making excessive heat. Why? Because there is no "critical" temperature for component life--the relationship between heat and component failure is both stochastic and nonlinear. Cooler is better even if it is not formally or theoretically necessary. But the output transistors in those B&K amps are now unobtanium.

I have not added the fan to the Buckeye amp, but it doesn't get nearly as warm in use. I wouldn't be opposed to it, however.

Rick "cleaned a lot of dust-impregnated cigarette tar from radio equipment heatsinks over the years" Denney

When I built my Pontiac 400 (built to 406, for you metric folks 6.7 Liters) that made 473 HP (and got 21.7 MPG (HWY, at 85 MPH [18 MPG @ 55 MPH, 18 MPG @ 110 MPH] in the early 80's, when I ported my heads, I also deburred ALL of the block, heads & chamfered ALL of the oil holes. Deburring a heat sink (for me at least), is just a rainy afternoon project.
It's boring but you just have to keep thinking that it will make life easier in the long run.
As to that engine, the car was stolen without the engine or transmission in it. So I still have them: Doug Nash 5 spd with 4.03 first & 1 to 1 5th.
And now I have a 455 (metric 7.4 Liters) engine to work with.
In a year or so, I'll start looking for a couple chassis/bodys to put them in.