Horizontal testing

[Manoet]

    My RMAF guy must've gotten busy today as I haven't heard a peep out of him requesting my bringing it to him. Maybe he's just going to be too busy and I'm ok with that as I enjoy continuing to learn about it.

Tomorrow if the 200 remains with me I want to start desktop testing. See if and how much temps vary from vertical mounting to horizontal mounting. First with no air assist then later with a laptop cooling pad like above. The pictured one is a cheap-o. Only cost $15, has a single 140mm fan and infinite speed control between 700 & 1,400 RPM and is USB powered. Advertised at 15-21dB but I doubt it makes that much noise. I can't hear it at all from a foot away but that may make a greater statement about my hearing than the pad's quietness. Plan is to sit the Devialet on top of it, with & without feet removed and try it both ways, ie; pushing/pulling air. I will also make a sealed gasket-to-bellypan interface from 12.4mm thick closed cell foam to insure all air is getting into or pulling from the Dev. No idea how effective this will be over passive convection cooling but will be interesting to find out. If results prove a decent increase in cooling I may go back and purchase the next higher CFM cooling pad. It moves twice the air but at 35% more noise and costs $10 more than this one.

[hk6230]

It looks so cool. I want one like yours, too.


从我的 iPhone 发送，使用 Tapatalk

[Axel]

It looks so cool. I want one like yours, too.


从我的 iPhone 发送，使用 Tapatalk

Same here! If I give you the wool, can you knit me one too, please? 

[Manoet]

    Moved the 200 from wall to desktop with no interruption in power, volume or streaming. *NOTE* Unit has been powered up and streaming continuously at volumes of -16 to -34dB, 24/7 since last Monday at 1:50pm. Test lasted 5 hours (from 6am-11am) and temps climbed exactly 2*C across the board in horizontal position, ie; A/D/S. Measurements taken at 25.5C ambient, DPM off, SAM off, volume -26dB. Unit is mounted on cooling pad that is 28mm thickness to more approximate isolation/decoupling device heights, not related to SQ but no power applied to internal fan via USB cable in pic.

Unit went back on test at12pm with power supplied (fan on) and will complete at 5pm with results to follow.

[Manoet]

Cancelling test early. I knew soon as I turned the fan on that results would be negligible even if measurable as I could "feel" it and that's been borne out in the subsequent 4 hours. Not one iota difference in temperatures but I know why. There's a 6mm gap between top of cooling unit and bottom of belly pan. Additionally the cooling fan is centered between the two viewing ports (convection inlets) in the Dev belly pan that I'm using for convection air intakes. When fan is 'pushing' (blowing), the air is hitting the belly pan between the two intake ports and scattering air laterally/omni-directionally outwards from between the Dev and cooling pad unidirectionally. If I switch the fan direction to 'pulling' or sucking it simply pulls air in from all directions thru that gap instead of only from the Devialet ports. Liquid or gas (air) will always take the path of least resistance... its the law! There's a simple work-around just by making an interface sealing gasket from foam. It will feature an open area between fan inlet/outlet and Devialet inlets/outlets but it will seal or block any/all air from getting in or out of the gap between the Devialet and cooling pad. Sadly that will have to wait till Monday.

My friend at RMAF called and wants me to bring the D200 over tomorrow mid-morning and have lunch with him. He admits it was silly trying to fit this in during RMAF. But wants to see & hear it privately before returning home. I feel like I've dodged a bullet so I'm happy to do this for him. I'll bring it back home with me tomorrow and it'll be "life as usual" again for me tomorrow afternoon.

Remember that steak & beer I talked about last weekend? I'm thinkin tonight's the night!

[Manoet]

I'm less than 10 minutes into round 2 of the horizontal forced cooling temp test and suspect this may provide the coolest temps yet! Reason being, when wall mounted I've never been able to 'feel' warm convection air escaping from the cooling ports in the front of the D200. Nor has it ever made the flame from a candle flicker in any significant way. I noticed immediately in horizontal mode just now with fan on and interface in place I could feel air escaping from the cooling ports immediately... even without intentionally feeling for it. A candle flame placed in front of those bores now flickers vigorously! Not enough to threaten extinguishing the flame but serious flickering! Heheh now I wonder about about that next size up cooling fan!?! Same diameter fan (so interface will work) but TWICE the air CFM! No way I can avoid giving it a whirl (pun intended)!

If this is born out in testing looks like I'm going to have to make a new wall-mount that will accommodate a custom CNC'd housing with a reversible fan that'll blow or suck the icing off a birthday cake from across the room!!

;-)

[Manoet]

After 5 hours of 'active' (forced air) horizontal cooling I've recorded the 2nd coolest temps yet but not by any significant margins. All specifications for cooling remain in effect, ie; 5 hours, FW 8.1.3, factory config, DPM/SAM off, 25.5C ambient, yada-yadda. The results below:

Wall Mount (passive cooling): A/D/S 29/29/31C
Desktop (passive cooling): A/D/S 32/32/33C
Desktop (active cooling): A/D/S 30/30/32C

While I'm surprised the fan didn't manage to eek out slightly lower temps I have to realize its a very lightweight fan designed for 'noiseless operation' first and foremost. Tomorrow I'll pick up the upgrade fan that's a mere 2 dB louder with a hefty 50% increase in CFM.

What I do remain entirely happy with is how close all these numbers are but even more importantly how consistent they are, how little the numbers change between vertical/horizontal, passive/active, and how perfectly repeatable they are when reverting back to a previous test. I have to believe the Devialet today is a meticulously well balanced enclosure temp-wise to achieve these temp consistencies under varying conditions.

What I'd like to do just once is get all three A/D/S temps in the high 20's. Not that I'll use it that way but simply braggin' rights... to say "I've done it." I still suspect cooler contributes to better SQ... I'm pretty sure of it in fact. But after the next 'active/horizontal' test with the higher CFM fan I'm going to throw in the towel on continued testing as I think I've pretty much exhausted my curiosities as they relate to these things... let others rule the roost at Tweaker's Corner for a change. Tho don't count me out on ever becoming satisfied, perfectly content/complacent or too comfy within my existing temp & SQ achievements. That'll NEVER happen!

[rwjr44]

Cooler may not be optimal. Too cool is a possibility. Given the range of temperatures we are discussing here, you may not reach that lower threshold but staying within a specified temperature range is important.

[Manoet]

Totally agree but until and if that lowest number becomes better or absolutely defined the only way I know to find it is to go lower until SQ worsens appreciably then back up a degree or two. Going south with temp research is certainly a more benign/innocuous pursuit than going north ;-)

[Manoet]

Today's testing finished on a high note which bodes well since this is the last regularly scheduled update. I found the secret handshake between the D200 and laptop cooling fan. In retrospect the additional CFM of the 20mm larger fan didn't hurt but the biggest cooling increase absolutely came about as reversing the fan direction and pulling air into the Devialet convection air outlets rather than pushing air into the Devialet via the belly pan inspection ports. My guess is cooling efficiency is increased via pulling cool air in over the hottest components first before that air path passes over cooler components near the center-mass of the Devialet on out thru the belly pan rather than the opposite dynamic. Additionally by reversing fan direction allowed the unit to get to its temperature dramatically faster. From 45C to 30C takes all of 8-9 minutes and once there it never waivers. From a 12 hour power down its up from ambient room temp in just at 12 minutes. Of course it doesn't have far to go since todays temps were glued at 26C all day A,D & S! In fact my GUI temps thru-out testing have been showing the A&D boards lagging behind PSU temps by a degree or two. Once I hit 28/29C all three become locked together & identical. Still no decrease in SQ associated by lower temps and if I had to say it I think they're the best yet. I'm not happy about that as it puts me in the perceived shadow of crazy men but there it is! I wish I'd never heard about lower temps and SQ! Of course even when I had heard of the phenomenon I ignored it till I was knocked upside the head BY it! Now I can't unring the bell. Long as I keep this active cooling system going on my desktop I don't even need the Devialet's GUI. If I wanna know Devialet's A/D/S internal temps today all I have to do is look at my Nest wall thermostat which I can read from across the room without budging an inch! A quality smart thermostat should give room temp as well as Devialet A/D/S temps... and it does!!!

But in a few days I'll get back to passive cooling via wall-mount. I just hope I can live with associated 30C temp SQ and haven't spoiled myself cuz I don't want to tear into walls building a combination Devialet vertical cooling/wall-mount setup to hang this bad boy from. That's the down-side to experimenting & testing... tempt not/want not!

**UPDATE** Spent another day trying to get temps below 26C across the A/D/C sensor with no luck. Finally dawned on me. Of course I can't! The room is 25.55C! How could I possibly get Devialet internal temps lower than 26C by pushing/pulling 25.55C ambient air!?! Sometimes I can be so dense! Sheesh I shoulda been all over this yesterday! Especially after going on about how my Nest thermostat now matches Dev A/D/S. Well of course it does... it HAS to! And small wonder the A/D temps started paralleling/matching the Supply temp instead of the A/D temps lagging behind a degree or two like they do at slightly higher temps. I could easily lower temps in the Devialet at this point to cooler than existing ambient temps. I would just have to lower the thermostat (slaps forehead) DOH!

A little humiliated/embarrassed I put the Dev back on the wall this afternoon and thought about not telling. But we've come too far to hide my humanness at this point... flaws & all!