Often, people get worried too quickly when they feel a device becoming warmer than around 45 °C. However, electronic devices, like CPUs, can handle higher temperatures, up to 90 °C. It's crucial to stay within the specified operating temperatures provided by the manufacturer. RME designs their devices well, ensuring they perform according to their specs within the documented temperature range.

Refer to the RME manual for technical details, regarding operating conditions, taking UFX III as an example here:

    Temperature range: +5° to +50° Celsius (41°F to 122°F)
    Relative humidity: < 75%, non-condensing

If you're placing devices in a poorly ventilated rack (which is common), it's strongly recommended to leave some space between them to prevent excessive heat buildup.

For example, I use only three devices in a 4 RU rack, providing approximately one finger's width of space between each device.

In short, don't overthink it. Just ensure there's some space between devices for a little airflow.

For reference:

Same for the two Lexicons below, 2 devices in a 3 RU rack unit.

I would also take care of, that the racks are not too deep and just fit for devices and cabling.
So for the typical RME devices I use Thomann Ecorack, which is 23 cm deep.
The Lexicons are taller and required a rack which is at least 35 cm deep.

Careful if you put one rack onto the other, for this they are not being made.
The feets of the upper rack puts too much pressure on the rack below.
So I had to use a wooden plate between them to stabilize it.

One of those typical threads where people kind of overreact.
The devices simply have good heat dissipation via the casing and that's it.
All you need to provide is to allow for a little airflow in the rack.

Heat and wear, however, are hardly avoidable side effects of electronics, especially when devices are designed to be high-performing. This requires a powerful CPU that generates heat, which is naturally dissipated through the casing – a sensible design choice.

That's why I say yes, you should make sure that if you have a passively cooled rack, that the devices have some distance between them so that the heat doesn't build up.

It's interesting how people don't react as strongly to tube or Class-A amplifiers.

One should perhaps not have wrong expectations and compare a highly capable ADI-2 DAC/Pro with an "ordinary" converter that only performs basic D/A conversion. Naturally, the latter runs cooler when it doesn't have much else to handle.

Always remember that this series of devices has been meticulously designed to operate this way, and they have proven to be successful and reliable in the market for many years.

I would simply trust RME's many years of design and studio experience here.

I don't find it helpful when links are shared without indicating what you want to express with them. What are you actually trying to convey to cupakm?

Guys, thanks to both of you for your answers.

I conclude answer to my own question "Am I correct?" - yes, "apart from using rubber feet / generous rack spacing, or active ventilation like fans, there isn't much to be done about it." - they will get warm, some more, others less, but most probably adequate to their Typical power consumption, stated in manual(s).

Maybe one could also slap a heatsink with thermal pads onto the unit, in order to dissipate the heat more effectively, but it probably isn't really needed.

Heat management in homerecording is often underrated while room acoustics get overforced.
Walls coated with foam, window stays closed (neighbors shall not hear the bad playing) and 20 rack gear is running all day long.
Sometimes one finds himself in a Sauna !
You have to deal with it.

Is this your experience, or where you get the numbers from?

Got one Multiface II lying around, actually wanted to use but all the analog inputs hiss quite a lot - pretty much like described here https://forum.rme-audio.de/viewtopic.php?id=27915
Asked RME support about how can let it service (because I never before in 10 years needed to service RME stuff...), but didn't get an answer, sadly.

Anyway I wonder why I measure UCX II draws constant 11 W (and is pretty warm) even when I mute ins, outs, turn off P48, DSP processing... the device appears not to distinguish any heavy load, light use or idling. No matter what it takes its 11 or 12 watt and that is it.

cyrano I'm not quite sure what are you responding to with the X32, Presonuses and MOTU history of yours... do you want to demonstrate FF400 lasts 20 years if recapped? I don't get it...