Fosi v3 Mono - User Impressions | Owner's Thread

[burkm]

You probably have never seen the life expectany tables of electrolytic capacitors vs. temperature by these manufacturers
...and You probably seem to have misunderstood what I am "really" saying judging by Your differentate highlighting of my sentence

 

[antcollinet]

As good as the electrolytic caps inside the monos may be, the heat will significantly shorten their lifespan.

Do you have evidence for that? Have you measured the capacitor temperature? Do you know the ripple current? Have you put those two bits of data, plus the other operating parameters (eg voltage) into the lifetime curves in the data-sheet to see what the impact is?

 

[Unbalanced]

Do you have evidence for that? Have you measured the capacitor temperature? Do you know the ripple current? Have you put those two bits of data, plus the other operating parameters (eg voltage) into the lifetime curves in the data-sheet to see what the impact is?

Everything I have read says heat is a major factor. Maybe not the only factor. Are you saying heat is irrelevant?

I was using this as a reference. https://resources.pcb.cadence.com/b...ors-influence-electrolytic-capacitor-lifespan
and this https://issuu.com/wtwhmedia/docs/power_and_energy_efficiency_hb_10-20/s/11276550
and my own personal experience

 

[antcollinet]

Everything I have read says heat is a major factor. Maybe not the only factor. Are you saying heat is irrelevant?

I was using this as a reference. https://resources.pcb.cadence.com/b...ors-influence-electrolytic-capacitor-lifespan
and this https://issuu.com/wtwhmedia/docs/power_and_energy_efficiency_hb_10-20/s/11276550
and my own personal experience

But the question is not if heat will reduce lifetime - that is a known fact, and applies to every single device on the market.

The question is - has the thermal design been done such that the lifetime of the capacitors is sufficient. That can only be known if all aspects of the design (including but not limited to the factors I mentioned) are known, and the lifetime is calculated.

There is a ridiculous amount of FUD being created about the thermal performance of this amp - based on nothing more than touching the case and thinking "hmm that seems a bit warm" Making categoric statements like yours is just adding to the FUD.

Try touching the heatsink of any other device you have - such as your PC, or a Class A amp - or even your phone when you give the processor something meaty to work on. They will all be "a bit warm" also. Are you on the relevant forums worrying about the lifetime of those devices?

 

[burkm]

What You mean by "sufficent" ? "Sufficient" as in functioning as supposed to or what ? Lifetime of an amp is not a consideration therefore although some people keep their amps for years sometimes even for decades ? The inerts (and the amps) should "live" (functioning as supposed to) as long as normally possible, meaning that halfing the lifetime by upping the temperature 10°C makes quite a difference for a neighboring electrolytic capacitors, doesn't it ? The inerts (components) of a V3 mono amp have at least the temps of the case, usually even quite a bit more, don't they ?

 

[Unbalanced]

But the question is not if heat will reduce lifetime - that is a known fact, and applies to every single device on the market.

The question is - has the thermal design been done such that the lifetime of the capacitors is sufficient. That can only be known if all aspects of the design (including but not limited to the factors I mentioned) are known, and the lifetime is calculated.

There is a ridiculous amount of FUD being created about the thermal performance of this amp - based on nothing more than touching the case and thinking "hmm that seems a bit warm" Making categoric statements like yours is just adding to the FUD.

Try touching the heatsink of any other device you have - such as your PC, or a Class A amp - or even your phone when you give the processor something meaty to work on. They will all be "a bit warm" also. Are you on the relevant forums worrying about the lifetime of those devices?

I’m shocked that a simple statement I made about component lifespan, that you said was a known fact, seems to have cast me into the camp of a contributor to ridiculous amounts of FUD. Yes, I do make comments about design concerns in other forums for products that I have firsthand experience with. Isn’t that the purpose of these forums? To get information from other users as opposed to the polished views of manufacturers and professional reviewers?

I am running a lot of 40+ year old HiFi equipment. That stuff was built to last. I expect the same from the equipment that I am buying these days. I want it to operate for decades. I don’t generally buy things and expect to dispose of them in a few years. Your comment suggests some sort of operational lifespan. I would accept that good engineering would take this into account. In that case if I am buying a consumable I would like to know what that operational lifespan is. Is it 1 year, 5 years 10 years?

I am a mostly happy new owner of a pair of V3 monos. One unit was defective out of the box and needed to be replaced. They don’t run a little warm. They run hot. Most devices that run hot like this are designed to leverage thermal convection to aid in heat dissipation. The audio devices and computers have air flow vents above and below to facilitate passive cooling. Many include active cooling as well. The goal being to get the heat out of there.

Now the V3 monos use their case as a heat sink. Lots of devices do that. They run hot. Lots of devices do that too. However, for devices that run hot I would expect venting on the top to release the heat. I would be concerned about promoting stacked operation as any way you slice it the top unit is going to get warmed by the bottom unit. The case unfortunately traps heat and the air vents along the side are not as effective as vents on the top would be. The air won’t move sideways without some help.

As an owner having spent a lot of time watching a lot of talking head reviewers talk about how fantastic these units are I thought I would share my experience. I’m pretty sure I said they sound fantastic. But products have other things that they have to live up to. I was amazed at how effective a 120mm case fan was to cool the units. They went from too hot to touch to cold. I think that’s fantastic too and I wanted to share.

 

[SMen]

I’m shocked that a simple statement I made about component lifespan, that you said was a known fact, seems to have cast me into the camp of a contributor to ridiculous amounts of FUD. Yes, I do make comments about design concerns in other forums for products that I have firsthand experience with. Isn’t that the purpose of these forums? To get information from other users as opposed to the polished views of manufacturers and professional reviewers?

I am running a lot of 40+ year old HiFi equipment. That stuff was built to last. I expect the same from the equipment that I am buying these days. I want it to operate for decades. I don’t generally buy things and expect to dispose of them in a few years. Your comment suggests some sort of operational lifespan. I would accept that good engineering would take this into account. In that case if I am buying a consumable I would like to know what that operational lifespan is. Is it 1 year, 5 years 10 years?

I am a mostly happy new owner of a pair of V3 monos. One unit was defective out of the box and needed to be replaced. They don’t run a little warm. They run hot. Most devices that run hot like this are designed to leverage thermal convection to aid in heat dissipation. The audio devices and computers have air flow vents above and below to facilitate passive cooling. Many include active cooling as well. The goal being to get the heat out of there.

Now the V3 monos use their case as a heat sink. Lots of devices do that. They run hot. Lots of devices do that too. However, for devices that run hot I would expect venting on the top to release the heat. I would be concerned about promoting stacked operation as any way you slice it the top unit is going to get warmed by the bottom unit. The case unfortunately traps heat and the air vents along the side are not as effective as vents on the top would be. The air won’t move sideways without some help.

As an owner having spent a lot of time watching a lot of talking head reviewers talk about how fantastic these units are I thought I would share my experience. I’m pretty sure I said they sound fantastic. But products have other things that they have to live up to. I was amazed at how effective a 120mm case fan was to cool the units. They went from too hot to touch to cold. I think that’s fantastic too and I wanted to share.

I confess I don't understand why Fosi did not combine the side vents with top holes as per the V3 stereo (top holes).
I am sure the convection generated would be the equivalent of the fan you mention (or comparable to the use of a fan).

If cooler is better (and I am sure everyone agrees with this statement) then why not do cooler? I am sure Fosi could have come up with 'cool' looking top vents also.
But maybe the temperature is not an issue. Most of my kit runs pretty hot in summer because it can get pretty hot down here in the South of France . My old A07 ran hot. As does an A08 (with side and top vents). FUD perhaps ... it is what it is ... LCB better ;-)

 

[antcollinet]

I’m shocked that a simple statement I made about component lifespan,

The other part of your comment was:

it is poor from a cooling point of view

Other than the fact that the case temperature gets warm to the touch, which as it is being used as a heatsink is not a huge surprise, what evidence do you have that the cooling design is poor? Especially compared to other similar components?

 

[Zek]

what evidence do you have that the cooling design is poor?

Because the entire case is used as a cooler, so all the electronics inside it are exposed to high temperatures, instead of the heat being dissipated independently of the rest of the device.

 

[Sokel]

Because the entire case is used as a cooler, so all the electronics inside it are exposed to high temperatures, instead of the heat being dissipated independently of the rest of the device.

The obvious bad choice probably because of the limited space is the caps lying on their side directly above the main source heat.
PCB is not an insulator,heat goes by just fine,specially when is the only barrier between the chip and the caps and they are both in touch with it or very close.

But no one can tell how long they will last.At the end it's the cost that dictates any compromise.
And of course cost is the main (and maybe the sole) strength of this amp.

 

[antcollinet]

Because the entire case is used as a cooler, so all the electronics inside it are exposed to high temperatures, instead of the heat being dissipated independently of the rest of the device.

And again - without details of what temperature those devices are getting to, what they are rated for, and what that means in terms of lifetime calculations - no-one knows if that means "poor" or "excellent" compared with the competition. I thought we were about science and engineering here. Not evidence free assumptions and judgements - which has already got people scurrying to waste their time and money on almost certainly un-needed cooling "enhancements"

The obvious bad choice probably because of the limited space is the caps lying on their side directly above the main source heat.
PCB is not an insulator,heat goes by just fine,specially when is the only barrier between the chip and the caps and they are both in touch with it or very close.

Which is probably as good a reason as any for conducting the heat from the chip downwards and away from the caps into the housing. It may well be that the approach taken is better than other chip amp designs which don't do this.

It is also almost certainly better to conduct the heat into the housing, from which it can be dissipated to the external ambient, than to dissipate it into the air inside the case - which is another alternative.

As always - the devil is in the detail.

 

[Bleib]

money on almost certainly un-needed cooling "enhancements"

I had a couple of bigger heatsinks lying around, bought thermal tape made for computer components for almost nothing and put them underneath. Cost almost nothing, and guarantees more heat dissipation early on from the main source. It can also be removed if I sell the amps. Not exactly a high cost for this. No idea if this affects longevity but it certainly won't make it worse.

 

[SMen]

I thought we were about science and engineering here. Not evidence free assumptions and judgements

Which is probably as good a reason as any for conducting the heat from the chip downwards and away from the caps into the housing. It may well be that the approach taken is better than other chip amp designs which don't do this.

It is also almost certainly better to conduct the heat into the housing, from which it can be dissipated to the external ambient, than to dissipate it into the air inside the case - which is another alternative.

These are both very basic assumptions regarding heat dissipation (conducting heat downwards or using the case as a cooler) ...or possibly deductions ... is this not really basic stuff? Even I understand it .

If it feels hot it is probably hot. I remember the Fosi marketing of these at the outset of the Kickstarter campaign. Mine are hotter than expected. I would have discovered and thought this regardless of anything read on this forum thread, because heat dissipation was a marketing selling point.

Could I cool them easily ... yes. With the most basic evidence free, schoolchild science assumptions. I think it cuts both ways here.

Could Fosi have done better at close to zero increase in cost (or less because they were already stamping the v3 stereo cases) ?

Yup.

Post edit: But I appreciate I have a schoolboy level assumption about the science of convection going on here - heat rises, let it escape etc.

 

[Negatywny]

And of course cost is the main (and maybe the sole) strength of this amp.

Then other amps have none xD

 

[Unbalanced]

The other part of your comment was:

Other than the fact that the case temperature gets warm to the touch, which as it is being used as a heatsink is not a huge surprise, what evidence do you have that the cooling design is poor?

I thought we were about science and engineering here. Not evidence free assumptions and judgements - which has already got people scurrying to waste their time and money on almost certainly un-needed cooling "enhancements"

Out of an abundance of curiosity and because we are about science and engineering I decided to do some testing to provide evidence around the issue of cooling design.

Method
I measured the scenarios listed below. All measurements are of temperature in ºC over time in minutes. Temperature measurements were made using a non-contact infrared thermometer. For consistency I measured the midpoint of the front of the two units. I could not get reliable readings of the internal temperature, but it was usually ~2ºC higher than the external point I was measuring.

1. Typical stacked configuration starting from cold. Units were off overnight.
2. Active cooling, fan blowing right to left across stacked units.
3. Units mounted sideways in a rig I made out of Lego, starting point was warm units following scenario 2. The rig holds the units 4cm above the surface they are sitting on. The rig is mainly open on the bottom to allow airflow. It's just Lego though, so some of the vents are blocked on the bottom.
4. Active cooling fan resting on top pulling air through units
5. Active cooling fan 2cm above units blowing down.
6. (not tested ) Active cooling fan blowing up from bottom.

In most cases I kept measuring until the temperature stabilized. It might have gone a wee bit higher given another couple of hours, but I wanted to get all of these done today.

Obervations

Summary

• Side mounting to leverage convection reduced peak temperature by about 5ºC. In other words, convection has an impact. I thought it would have a greater impact, but the overall temperature gradient was not high enough to generate a strong airflow. I didn't measure the bottom of the top unit consistently. At the end of the test I was able to get a reading of 56ºC from the bottom of the top unit.
• Active cooling had a very big impact and very quickly. That's why I shifted to measuring every 5 minutes rather than every 10.
• Having the fan blow air towards the units seemed to work better than having it pull air away from the units. Airflow and fan design are tricky things. The difference was small ~ 1ºC. There was no enclosure around the fans and the V3 Monos to help constrain the airflow to pulling air specifically from the V3 Monos.
• In summary active cooling reduced the temperature of the units by ~25ºC. Passive convection reduced the temperature by ~5ºC.

Conclusion

I will be running active cooling with a fan or two as I am confident that running the units at an operating temperature that is 25ºC lower than the default stacked configuration will lead to a longer operating life for the units.

Images

1. side mount
2. side mount with fan resting on top blowing up
3. side mount with fan 2cm above blowing down

 

[Negatywny]

this is pretty cool, but still proofs nothing

Active cooling lowers temperatures, who could have expected that?

My marantz av recieiver is equaly (or more) as hot as the V3 monos, and he is 1000 times more complicated, and works almost 8 years now and it developed small problem with the potentiometer, rether not temperature releted


So two things, is there anything indicating real temperature problems, besides usual - temperature bad, look at my 70'C class A amp from the 80s, its a beast

And second: what speeds was those coolers from the test and why there's no cooler on the bottom - the most logical for me - option tested?

 

[nitpicker1]

Cooling the bottom surface will yield best result. The single test you avoided to do. It's practically the equivalent of a CPU lid of of the unit, cool that, and the rest will largely follow. Your own ~2c internal/external observation supports this.

If I wanted actual airflow, I'd use 2x NF-A4x10 FLX 40mm fans in a push pull config on the mesh sides - 100% silent and proven extremely effective in my other uses. But I see no need for it. Maybe I'll 120mm the bottom if they start getting uncomfortably warm.

I worry the power bricks will die long before the units.

 

[antcollinet]

I will be running active cooling with a fan or two as I am confident that running the units at an operating temperature that is 25ºC lower than the default stacked configuration will lead to a longer operating life for the units.

Well that was above and beyond the call of duty. But I certainly appreciate the effort.

Your convection result was not a surprise for me. I know from a 35 year career in electronic design and project management that the trickle of convection in a typical electronics housing does very little. It can be more useful when enclosure vents are in the vicinity of particularly hot components. But overall, this is why I'm not particularly concerned about holes on the top of the housing which would do even less due to the space filling PCB blocking pretty much all airflow in the conventional orientation.

You are absolutely correct and have clearly demonstrated that forced air cooling makes a huge improvement to cooling, but again this is not a huge surprise. And it will of course improve the thermal life of the amps. One thing though that you will have to worry about is the dust that will accumulate in the devices as a consequence of the airflow, which is also a reliability issue. In my former life we had design quality standards forbidding forced blown air over electronics. You might need to plan to open up and vacuum out a couple of times a year.

However:
1 - We still don't have data to tell us that the lifetime of the device without the forced air cooling is not good enough. I'm prepared to give the benefit of the doubt to the FOSI engineering team here until evidence shows that I shouldn't
2 - Your conclusion will equally apply to ALL heat generating electronics. There is no reason to assume the FOSI needs forced air when (for example) your TV doesn't. Or my mini DSP flex which also gets noticeably warm to the touch, and only has a couple of square inches of ventilation holes on each side.

 

[Negatywny]

One thing though that you will have to worry about is the dust that will accumulate in the devices as a consequence of the airflow, which is also a reliability issue.

We need coolers AND dust filters

 

[Negatywny]

They went from too hot to touch to cold.

Too hot to the touch? Well, i'd say mine ar very warm, but no problem touching. And reading these temperature posts makes my touch them everytime I turn them off. So is that an overstatement, or are yours carrying a grater load?

I'd say wee need temperature/load graph now