Have you come across design flaws that make you wonder how such bright people could make such poor decisions?
I just did today, when I discovered the Audio Research VT200 amp. Now, people will rightly point out that its main flaw is not unique to this amp. But, that's not really the point. The point is that the flaw is so egregious that it should have never gotten past the suggestion phase (and shouldn't have been suggested by any engineer worth their salt).
I'm not an engineer and I know better, which says something about how egregious the design mistake is.
What is it? A teeny-tiny fan (small fans are typically noisier than large fans for a certain CFM due to them requiring higher RPM) that's so loud that when the Stereophile reviewer moved it far away it could still be heard. How, on planet Earth, does it make sense to spend thousands of dollars pursuing audio quality and simultaneously throw in fan noise to oppose that goal?
It doesn't. It means the design is a mistake and should be scrapped and completely re-imagined.
My guess is that the fan is (in addition to being small — something that looking at the photo makes clear) a single bearing ball bearing design, which is basically the noisiest type of fan there is (although niche designs like blowers might be worse — I am talking about standard computer-style fans for airflow and for radiators). Single bearing ball bearing fans are often used industrially due to their low cost and long life, even though they're loud. Dual bearing fans are a bit quieter but not as quiet as more modern bearings like fluid dynamic (FDB), the floating air bearing, etc.).
The placement of the fan here also looks similar to how a fan is placed on a computer radiator (it's against a resistant barrier). Placing fans in restrictive situations necessitates the use of fans designed for higher pressure versus higher airflow. Higher static pressure models (which often have the large motor hubs for their size, like the one in the photo) are much more effective at combating resistance but they are also noisier per CFM. That's not good when you're designing something for audiophiles. What seems to be ideal is to have a big fan without much restriction in its placement so the air can flow over the tubes. I would think that having two 200mm fans would be better, in order to get flow over all of them but maybe that's overkill. Better flow coverage means lower RPMs are needed, which means less noise. Having the amp split into two mono boxes seems better because the heat is less concentrated and a separate fan on each means the airflow isn't going to conflict between the fans.
There were 200mm fans in existence (and even larger). Most of them were old-style sleeve bearing fans that must be placed vertically. Sleeve bearing fans, when mounted vertically, can be the quietest fan type at lower RPMs, though. One fan that was not a sleeve bearing was Antec's "Bubba" which was a dual ball bearing and relatively noisy. The problem with sleeve bearing fans, even when mounted vertically, is that they tend to have the shortest lifespan. Ball bearing fans simply get louder over time. But, ball bearing fans also have a fatal flaw. If they are dropped/knocked too hard they can be ruined. Sleeve bearing fans can take abuse.
Since tube amps require annoying tube replacement work it seems totally reasonable to have the customer replace their sleeve bearing fan every so many years. The key to that is to make it easy to remove and replace the fan. I have seen sleeve fans last quite a long time, too — so this hassle would be quite a lot less of one versus having to replace tubes, I would think.
This amp, if designed as it was to put so much heat into a single case, should have had a big (like 200mm) thick very slow RPM vertically-mounted sleeve bearing fan to keep the tubes cool — using an enclosed tunnel to channel the airflow over the tubes (as Dell does in its computers, for instance). Big OEMs like Dell realized that using airflow enclosures keeps noise down and improves cooling reliability.
Noctua, in case no one knows, has released FDB (fluid dynamic bearing) 200mm fans, which can be mounted horizontally and are quieter than sleeve bearings at higher RPMs. For an amp, though, very low RPMS with a very large fan should be sufficient. If not, use two. The horizontal ability of the Noctua means you can build a current-gen XBox-style tower to move hot air vertically as it prefers to. (AMD released a prototype machine before Microsoft did that had a vertical wind tunnel design. SilverStone also worked with a hybrid between the standard ATX case and vertical airflow.) If anyone wants to fix this Acoustic Research design by re-doing it, do it that way. Or, the amp should have been two mono blocks to keep the heat spread out more (possibly with a large fan for each at a very very low RPM). No one should be hearing fan noise from a home audio amp, especially a high-end one! Audiophile means the person loves sound quality. Fan noise is the opposite of that. If an amp is designed for live rock concerts and other high-noise high-SPL situations that's, of course, very different.
Whenever I see amps with small fans I shake my head. When I see one that requires so many expensive tubes because it produces so much heat and has one paltry tiny fan I really shake it.
Serious drawbacks:
• The fan!
• Too much heat crammed into too small a physical space.
• Too many expensive tubes to replace.
• Too heavy for safe lifting for many people.
• Ugly (especially the all-black version). If you're going to have all those tubes then show them off and don't use ultra-boring black box exteriors. Tube amps should look retro chic — understated and elegant — not a big boring black box. (Note that retro chic doesn't mean covered with big florid lettering and lots of logo trash which is a contemporary trend with tube amps and even the tubes themselves. I looked through an article that showed affordable tube amp options and, guess what — the only one that looked nice was one designed in the 50s.) Aesthetics are a matter of taste but only to a point. Tubes should be both seen and heard. Otherwise, a person can get everything they can do from solid state much more cheaply.
Yes, my mileage varies. If I spent thousands on an amp and thousands on speakers I don't want to have spent thousands to hear avoidable fan noise. I also don't see how having to find far away placement strategies is a bonus.
I just did today, when I discovered the Audio Research VT200 amp. Now, people will rightly point out that its main flaw is not unique to this amp. But, that's not really the point. The point is that the flaw is so egregious that it should have never gotten past the suggestion phase (and shouldn't have been suggested by any engineer worth their salt).
I'm not an engineer and I know better, which says something about how egregious the design mistake is.
What is it? A teeny-tiny fan (small fans are typically noisier than large fans for a certain CFM due to them requiring higher RPM) that's so loud that when the Stereophile reviewer moved it far away it could still be heard. How, on planet Earth, does it make sense to spend thousands of dollars pursuing audio quality and simultaneously throw in fan noise to oppose that goal?
It doesn't. It means the design is a mistake and should be scrapped and completely re-imagined.
My guess is that the fan is (in addition to being small — something that looking at the photo makes clear) a single bearing ball bearing design, which is basically the noisiest type of fan there is (although niche designs like blowers might be worse — I am talking about standard computer-style fans for airflow and for radiators). Single bearing ball bearing fans are often used industrially due to their low cost and long life, even though they're loud. Dual bearing fans are a bit quieter but not as quiet as more modern bearings like fluid dynamic (FDB), the floating air bearing, etc.).
The placement of the fan here also looks similar to how a fan is placed on a computer radiator (it's against a resistant barrier). Placing fans in restrictive situations necessitates the use of fans designed for higher pressure versus higher airflow. Higher static pressure models (which often have the large motor hubs for their size, like the one in the photo) are much more effective at combating resistance but they are also noisier per CFM. That's not good when you're designing something for audiophiles. What seems to be ideal is to have a big fan without much restriction in its placement so the air can flow over the tubes. I would think that having two 200mm fans would be better, in order to get flow over all of them but maybe that's overkill. Better flow coverage means lower RPMs are needed, which means less noise. Having the amp split into two mono boxes seems better because the heat is less concentrated and a separate fan on each means the airflow isn't going to conflict between the fans.
There were 200mm fans in existence (and even larger). Most of them were old-style sleeve bearing fans that must be placed vertically. Sleeve bearing fans, when mounted vertically, can be the quietest fan type at lower RPMs, though. One fan that was not a sleeve bearing was Antec's "Bubba" which was a dual ball bearing and relatively noisy. The problem with sleeve bearing fans, even when mounted vertically, is that they tend to have the shortest lifespan. Ball bearing fans simply get louder over time. But, ball bearing fans also have a fatal flaw. If they are dropped/knocked too hard they can be ruined. Sleeve bearing fans can take abuse.
Since tube amps require annoying tube replacement work it seems totally reasonable to have the customer replace their sleeve bearing fan every so many years. The key to that is to make it easy to remove and replace the fan. I have seen sleeve fans last quite a long time, too — so this hassle would be quite a lot less of one versus having to replace tubes, I would think.
This amp, if designed as it was to put so much heat into a single case, should have had a big (like 200mm) thick very slow RPM vertically-mounted sleeve bearing fan to keep the tubes cool — using an enclosed tunnel to channel the airflow over the tubes (as Dell does in its computers, for instance). Big OEMs like Dell realized that using airflow enclosures keeps noise down and improves cooling reliability.
Noctua, in case no one knows, has released FDB (fluid dynamic bearing) 200mm fans, which can be mounted horizontally and are quieter than sleeve bearings at higher RPMs. For an amp, though, very low RPMS with a very large fan should be sufficient. If not, use two. The horizontal ability of the Noctua means you can build a current-gen XBox-style tower to move hot air vertically as it prefers to. (AMD released a prototype machine before Microsoft did that had a vertical wind tunnel design. SilverStone also worked with a hybrid between the standard ATX case and vertical airflow.) If anyone wants to fix this Acoustic Research design by re-doing it, do it that way. Or, the amp should have been two mono blocks to keep the heat spread out more (possibly with a large fan for each at a very very low RPM). No one should be hearing fan noise from a home audio amp, especially a high-end one! Audiophile means the person loves sound quality. Fan noise is the opposite of that. If an amp is designed for live rock concerts and other high-noise high-SPL situations that's, of course, very different.
Whenever I see amps with small fans I shake my head. When I see one that requires so many expensive tubes because it produces so much heat and has one paltry tiny fan I really shake it.
Serious drawbacks:
• The fan!
• Too much heat crammed into too small a physical space.
• Too many expensive tubes to replace.
• Too heavy for safe lifting for many people.
• Ugly (especially the all-black version). If you're going to have all those tubes then show them off and don't use ultra-boring black box exteriors. Tube amps should look retro chic — understated and elegant — not a big boring black box. (Note that retro chic doesn't mean covered with big florid lettering and lots of logo trash which is a contemporary trend with tube amps and even the tubes themselves. I looked through an article that showed affordable tube amp options and, guess what — the only one that looked nice was one designed in the 50s.) Aesthetics are a matter of taste but only to a point. Tubes should be both seen and heard. Otherwise, a person can get everything they can do from solid state much more cheaply.
Stereophile said:
Yes, my mileage varies. If I spent thousands on an amp and thousands on speakers I don't want to have spent thousands to hear avoidable fan noise. I also don't see how having to find far away placement strategies is a bonus.
Last edited:
A 200 mm FDB fan at low RPM and still enough CFM should do the trick. You are right that the original fan is poorly chosen. Mounting the new fan against the upper cover will not work optimal either (for the fan itself that is as it will take the heat). How nice it may sound, I would not want such an energy wasting heat generating stove in my home certainly when it needs a fan (and knowing I only listen to 1W maximum) 
After many exchanges of sleeve bearing/ball bearing fans I ended up with Sunon brand industrial fans with MagLev bearings as these rarely fail. It is just like with caps: no bearing is the best bearing. Passive cooling is the holy grail however as any fan is an annoyance, be it directly as it is noisy or after a few years when it fails.
After many exchanges of sleeve bearing/ball bearing fans I ended up with Sunon brand industrial fans with MagLev bearings as these rarely fail. It is just like with caps: no bearing is the best bearing. Passive cooling is the holy grail however as any fan is an annoyance, be it directly as it is noisy or after a few years when it fails.
Last edited:
Yes, I agree. The Noctua seems that it would be the one to go with, since Noctua has a good reputation for making more effort to build high quality fans (particularly with an emphasis on cooling per decibel and also on build quality).
When that amp was made, though, Noctua's fan probably hadn't been put onto the market yet. So, the 200mm models to choose from were the old-fashioned sleeve bearing (fine if used vertically, as long as the fan's build quality is good enough) and double ball bearing models (noisy, especially over time).
Given the lack of a modern bearing it seems the designers had the following choices:
1. Separate the amp into two monoblocks, to reduce the heat concentration — probably using a 200mm sleeve fan vertically (one for each).
2. Get an OEM to make a modern bearing for them in a 200mm or so size (pricey).
3. Go for lower wattage or a hybrid design (solid state and tube).
Normally I'd also suggest using a big passive aluminum heatsink but that wouldn't provide airflow on/around the tubes.
Anyway... I'm not an engineer but when I saw that photo my mouth dropped open some.
I have seen sleeve bearing fans last for a really long time in some Apple tower computers. The build quality of most of the 200mm fans wasn't very high. Noctua made a point of saying it took a lot of development time to make a good 200mm fan. The dual ball bearing Antec Bubba, though, had good build quality. So, Noctua was overstating its case for marketing.
And, given a decent level of build quality, a sleeve bearing fan when mounted vertically should outlast a lot of vacuum tubes. Although the build quality of the pre-Noctua 200mm sleeve fans wasn't stellar it should have been good enough — certainly better than the fan the designers chose for this amp.
Last edited:
That was then and this is now. One can use whatever 200 mm fan (with enough CFM and right bearings) of good quality that is quiet and may be confident it will give better results than the manufacturer designed. Mounting the tubes upright sticking through the upper cover would indeed have been the most optimal choice.
Air flow with a 200 mm fan will be much wider instead of cooling only the tubes/valves in the middle. The original fan is a 3 blade type which are not the most silent ones as well. Simply not well though-out. Whatever we think of this is irrelevant, people buy stuff when the reviewer tells the device is OK.
Air flow with a 200 mm fan will be much wider instead of cooling only the tubes/valves in the middle. The original fan is a 3 blade type which are not the most silent ones as well. Simply not well though-out. Whatever we think of this is irrelevant, people buy stuff when the reviewer tells the device is OK.
Last edited:
I don't know of any 200mm fans that use modern bearings other than the FDB Noctua. I haven't followed the PC gear news very closely for some time, though.
PC gear is best for PC's as these are written off in a few years. Industrial stuff is really a better choice in general. If not available I would use the 200 mm Noctua. Even without complex calculations and predictions one can be assured that any larger fan even with the same CFM will work out better. Best is of course to avoid this amplifier.
BTW Imagine audio devices having those fans with RGB LEDs ....
BTW Imagine audio devices having those fans with RGB LEDs ....
Attachments
Last edited:
Noctua makes industrial fans so there is overlap between PC equipment and industry. Do you know of any 200mm fans that use modern bearings other than the Noctua?
I haven't heard of any. They definitely weren't available when that amp was designed. Not even the Noctua was.
There may have been some large fans with high build quality that were designed for things like RVs but low noise was probably not a major factor in the design choices. Ball bearings would have been used, which likely means too much noise for audiophile equipment. For audiophiles, the only affordable option at the time was sleeve bearing fans used in PC equipment, as far as I know.
Big audio companies might have had the budget to get an OEM to make a custom design for them with a modern bearing. I don't know how much that would have cost but given how long it took Noctua to introduce a 200mm fan with a modern bearing to the PC market it seems the inertia was very high.
Many industrial fans are designed for low expense, high airflow, and high reliability. They're not good fits for audiophile gear. Similarly, enterprise-grade motherboards are designed to be very reliable but to also use cooling systems so noisy that they would be intolerable in a home environment. The low cost incentive is why single bearing ball bearing fans have been the most popular in industrial usage, despite their high noise.
For that, less reliability less of a problem versus low noise. If a fan fails it should be easy to replace (if the design of the unit is good), which includes being easy to recognize in terms of it needing to be replaced. The main thing is that it should be as quiet as possible during its useful lifetime while also doing enough to cool the components (particularly when it comes to extending the life of a lot of expensive tubes).
As far as I know it's a more recent bearing type than FDB. It has become popular even more recently. It does have a good reputation, at least in high airflow designs. I don't know how well it holds up to higher restriction scenarios (e.g. against radiators) versus FDB and similar modern competitors. Dual ball bearing designs with big motors tend to still find use in static pressure designs (high resistance) despite their noisiness.
Interesting decision. In the eighties these Sunons had a quite bad reputation failing within 500 hours of activity. Some weeks ago I had to replace a rattling sunon MagLev 40x40x20mm. I choosed an EBM-Papst instead.
"In the eighties"......
In the seventies japanese cars were laughed at by german car producers. Things can change in time. The fact that A380 has many of them says at least something. The one you had that failed could be the 1 in a million, who knows. It could be the dusty/aggressive environment where it was used etc.
I did service large industrial devices for a living and these used EBM-Papst AC and Dc fans exclusively and in large numbers. I am quite fond of the brand and the quality but ... of course I experienced a premature failure now and then even with stuff like EBM-Papst. It is about the numbers. We replaced them at 50.000 hours but in clean environments they could live quite some years longer. In an aluminium factory they needed replacement way earlier. I once experienced an abandoned device in a factory in Germany that had been powered on for 23 years without any service...and still worked (in a way).
Something underestimated: transport of ball bearing fans has influence on their longevity. Simple. A courier throwing the box on the floor of the van can make the difference really.
In the seventies japanese cars were laughed at by german car producers. Things can change in time. The fact that A380 has many of them says at least something. The one you had that failed could be the 1 in a million, who knows. It could be the dusty/aggressive environment where it was used etc.
I did service large industrial devices for a living and these used EBM-Papst AC and Dc fans exclusively and in large numbers. I am quite fond of the brand and the quality but ... of course I experienced a premature failure now and then even with stuff like EBM-Papst. It is about the numbers. We replaced them at 50.000 hours but in clean environments they could live quite some years longer. In an aluminium factory they needed replacement way earlier. I once experienced an abandoned device in a factory in Germany that had been powered on for 23 years without any service...and still worked (in a way
).Something underestimated: transport of ball bearing fans has influence on their longevity. Simple. A courier throwing the box on the floor of the van can make the difference really.
Last edited:
Heatsinks dissipate tube heat effectively enough? Airflow over the glass and around the bases isn't really needed? I have seen a lot of tube amps that don't use fans but the wattage is lower and/or they have the tubes spread out in an open layout with no cover (or just a thin wire cage).
I would want to do everything possible to avoid a fan in audiophile gear, too. But, if the design must have one because of the company's goals of high wattage in a small space (especially one enclosed by box-like covers) I would do everything I could to choose the quietest fan possible (big, low RPM, good build quality, quiet bearing type) and design the unit to utilize it as quietly as possible (lowest required RPM).
I definitely agree about hard disks. I find their noise (at least the 7200 RPM and higher versions) intolerable. I paid a lot for 240 GB drives when that was a huge size for an SSD and ended up with OCZ units that were defective because the controller couldn't cope with the 64-bit NAND the company did a stealth bait and switch with. But, fortunately they have become more reliable. (Not the OCZ units. Even the replacements are bricked.)
Then redefine what YOU want. The company's goal of high wattage in a small space might not be your cup of tea as the device is ugly, technically not well thought out etc. If you need only 10W max why bother with very heavy space stealing energy wasting heat generating crates filled with smelly stuff filling up your living room (adding background noise as a bonus)?
It might be that a small 2 x 15W tube amp with glowing lamps sticking out of the cover makes you way more happy.
It might be that a small 2 x 15W tube amp with glowing lamps sticking out of the cover makes you way more happy.
Last edited:
I was looking at the amp from the point of view of an engineer hired by a company and required to design according to the goals of the managers.
If I were in such a position I would make it clear that a unit that needs a fan due to design choices would have to have one that is inaudible from a practical standpoint.
It is true that sometimes the engineer (or someone else) needs to just say no altogether (like Steve Jobs' insistence on designing the case of the Apple III before the innards and not giving it vents for aesthetic reasons). But, even with a fan this amp could have had a much quieter implementation of active cooling.
Even the best engineers and the best brands make mistakes, often because of stubborness or ignorance of decision makers. Can't reveil details but I once was nearly electrocuted by an industrial device I was trained on but that apparently had one series that had a section that was live even though it was switched off. Of course I found out when I was replacing parts and saw my screw driver making blue light. It happens...
Some things should not happen but still they happen, often because of everyone focussing on what shouldn't happen
Some things should not happen but still they happen, often because of everyone focussing on what shouldn't happen
Last edited:
Airflow engineering is/was probably not the strongpoint of the electrical engineers who made this amp. (Fan noise to performance ratio is also an issue of ergonomics, another field of engineering.) There is also the issue of perspective. Steve Jobs felt very strongly that aesthetic appeal was crucial for succeeding in a crowded computer market, especially when facing the faceless design language of IBM.Even the best engineers and the best brands make mistakes, often because of stubborness or ignorance of decision makers. Can't reveil details but I once was nearly electrocuted by an industrial device I was trained on but that apparently had one series that had a section that was live even though it was switched off. Of course I found out when I was replacing parts and saw my screw driver making blue light. It happens...
Some things should not happen but still they happen, often because of everyone focussing on what shouldn't happen
He was right about that. Form and function should both be harmonious. But, he also wasn't an engineer so his stupid ideas about putting aesthetic appeal above more practical matters (like adequate cooling) led to several embarrassing failures (cracks in the G4 Cube, the Apple III overheating and the company telling buyers to drop the machine from several inches off of their desks to try to reseat the memory, iMacs that ate hard drives in computer labs, etc.).
Engineers often neglect the aesthetics and managers often neglect the engineering.
Last edited:
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.