Argon one V2 Vs ice tower cooler Test. Part 2 

I'm glad my hunch about those cables was confirmed. I have both V1 and V2 versions of that case with the m.2 add-on on some of them and they work properly without issues, (I'm using an Argon USB-C 5.25V 3.5A PSU from their site) plus they look really nice and tidy on my desk.

Since the top of the Argon One V2 is basically a big heat sink; I'm not surprised it keeps the RPi cooler. I've been very happy with the performance of mine (The M.2 version) over about a year and a half. It's a very nice and tidy piece of kit, too. 👍

Thank you for the thoroughness of the benchmarking/testing. I do agree with the comments that say "let's see how it does after the case becomes saturated with heat". My house is warm: thermostat is set to 74f at night and 76f during the day. My Pi4 idles ~50c. I do not agree with having 3 options for fan speed and not utilizing them all. If the ICE cooler has only one setting, that is on them. I do see 55c often; seldom more than that and I use the suggested values. I'm too old to hear the fan kick in at 10% rpms and I have never seen 60c in the case. But: this is a toy for me and I do not use it as my daily driver. Your temps may vary.

Glad to see you worked out the mystery. Always enjoy your videos.

I love my Argon cases. However, I think it may have had an unfair advantage in your test. The Argon case has a lot more thermal mass to work with than the competition. Your case starts out at room temperature that's well below the PI operating temperature. It would be interesting to see whether the Argon still wins a much longer, say overnight, heat/stress test.

thank you for excellent part 2 ive been too chicken to try highest overclock as so dam hard to get a new pi and afraid to kill my kodi player,.also thanks to LilRedDog for making possible.. im going to move my 8gb pi to my argon case and use my ms2 with lees os as it just works so well after changing the kde connect + wallet settings. 10/10 both for info

You had too many variables from the start. If you do similar again use the same pi, cables, PSU etc. Just swap whatever you're testing. Those PSUs were designed for older pi's. With all those adapters you will have a fair voltage drop.

I can respect the effort to get the comparison right but it is funny to have you talking about a possible conspiracy theory then have you tease us and make us wait a day for part 2. From this video I swapped my 8 gig pi into the Argon case and put the cooler on the 4 gig pi I use as a bitcoin node so many thanks great videos

Yeah, Argon cases are still my favorite.

Okay so I just deleted my previous two comments about doing fre-- sh installs on identical SD card (again... like I was saying on the first Part) now that I'm about 11:30 into this vid and we see that it was your cables the whole time.... lol I think that if you were to have shown us your cables in the first video most of us would have said to make them just the power brick wire and only that. lol I too would have guess that the usb-A would have caused the issue over the usb-C for sure... but maybe usb-C is super sensitive to cheap manufacturing and so HAS TO be made to the highest quality where any ol usb-A wire just needs to have shielding and 4 wires and that's about it LOL - Finally finished the vide oand WOW this was REALLY FUN!!! I really wouldn't have thought it was a cable causing problems like this... but it did!!! :D

I have this argon case and ice tower cooler. My experience with cooling performance is the ice tower cooler performs better without a doubt. If I run for retro gaming or for a desktop pc, the argon case gets hotter.

Excellent vids love the argon not long had one and its amazing. Need to get the add on for m.2 drives would make great media server

My Argon one case has a lot of thermal mass, but the fan isn't too good at bringing the temperature down. So it has a very long averaging time compared to other coolers.

I am mildly unhappy with how Raspberry Foundation designed the power input for the Pi 4. 5V with absolutely no PD protocol to negotiate higher voltage / power output. For the Pi3, this was just bearable. However for Pi 4 it is quite capable of getting spikes of 3A which is completely outside the USB spec. You can even see that they designed their own power supplies to 5.1v, not 5v, because RPi is just hitting the limits of Vdroop and at its current draw and losing regulation, and for their official supplies they wanted the slight extra headroom so everything would work. For any normal 5V USB supply, acceptable voltage is 4.8-5.2V. 2A is max draw without further negotiation. In fact in many situations a strictly designed USB supply (they hardly ever are) would cut power to an RPi at 2.4A. If you get unlucky and have a perfectly capable (power-wise) USB charger that is sitting at 4.9V, within spec, just random luck on the variation resistors in the regulator, it's likely you'll have issues powering a Pi 4. So basically Pi 3 sits just inside the spec for vanilla USB power, and Pi 4 sits just outside, and hopes for the best. Ofc whenever anyone complains you'll get a bunch of people ask "are you using the official RPi PSU??" If it was necessary to use a proprietary single-vendor PSU for the RPi to get normal results, this should have been advertised with the product. And as you have found out, when you start drawing serious amps at only 5V, the quality of your usb cable starts mattering. The more resistance it has, the exponentially more likely you'll get power issues. A cable which works 100% fine for 99% of tasks might fail when the Pi 4 is relying on getting high enough 5V draw to regulate. Hence I also stay away from all the ones with charging leds and meters, that only makes the problem worse. A moderately cheap way to measure USB cable resistance is to get an UM meter from Ruideng, UM24/25/35 and one of their loads. They have a little wire impedance test built-in which is a bit finicky but very useful, to actually see the quality of your USB cables. Lower impedance = lower voltage drop over the cable when drawing high current = less power issues. When you add extra power draw from SSDs and fans it becomes extremely dicey indeed, as you have just found out. I know it would add to the BOM cost, but if the foundation had just designed in USB PD all this would be a non-issue. Buy any PD capable supply and any RPi could run 100% fine and dandy. As it is now you have to tip-toe around power requirements, what you're adding to the supply, buy only official supplies (which sucks because they only provide one output...), get short and/or highest quality cables (and hope cable manufacturer isn't lying...) and/or hope for the best. Even an absolutely premium USB charger rated to charge laptops at 80W is luck of the draw because Pi will only draw at the stock vanilla USB 5V2A limit and you have to hope you get lucky that Vdroop of the 5V rail won't kill it or underpower it. While overall argon one is a great product, the fact it's transferring power over yet another board and connectors (adding resistance) and powering a fan at boot-up adds that bit of extra % of power draw that can cause problems like these. And unfortunately due to the design of the pi there's not much to do about it. /rant

The case directs the airflow, where as the caseless one doesn't have the same heat flow causing air flow in a non-chaotic fashion.

Brilliant idea! Cliffhanger YT.

If you happen to notice the Argon One case getting warm while taxing it, I've found that removing the gpio cover, it cools right off.

Thanks for that. I worried about my ssd temperature this (hot) summer also. I think i'm going to try one of those small heatsinks you can put on your m.2 with some elastic bands and the thermal tape. It worked for my PC. Maybe the temperature is also better because the Argon connects the GPU also to the case?

Some manufacturers definitely fake stuff. I have an NVMe SSD from Kingspec that drops its reported temperature by 20C when it hits a certain value. So under load it seems fine but it is really running very very hot.

For the ArgonOne case, how about doing temperature comparisons between fan configuration. Do one test with the fan blowing down on the Pi and the other with the fan blowing upward?

I wonder if anyone makes a copper case. Since it's a better conductor than aluminum I would think that it would wick the heat more. For that matter, it'd be neat to see heat pipes going through an otherwise solid copper block and have a full case designed around it to direct the airflow. Probably a better idea for cases like that to have an always on fan but set at a lower speed. Eventually the heat will build up too much, and if you keep it on 24/7 that could be an issue in a couple of days.

The power cable is built with too thin of wires inside. It has too much internal resistance, thus unable to deliver the power your pi wanted

Even on my screen I can tell that portable monitor has a good panel.

Argon One case is a Passive Cooler without Fan

I've had short profile USB drives get almost burning to the touch hot, & where the usb drives were plugged into a usb hub, and at the most get warm to the touch, the usb drives, & cable connect point to the Pi only worm? Any idea why? (I cook, so what I call hot to the touch is like 140F, 60C)

I feel weird cuz I've got the LP RGB Ice Tower in a metal Geek Pi case built for the tower & the lowest mine runs is around 32°C due to the heat of my room but the highest mine gets is around 65°C unless I'm doing a heavy video edit then i can push almost 80°C but it doesn't stay that long once my fan kicks in as it drops it back to around 45°C. Thing too is my 4B 8GB Pi has my cooler on the 5V pin plus I've got my CPU at 2147, OV +8 cuz anything lower or higher crashes my Pi & my GPU has a 750mhz OC & I've got the memory set to send 1024mb to the GPU for video.

i now that argon need a argon power supply for better performance or a good one

Argon M2 from manual: "Aluminium alloy top case acts as a giant heatsink connected to the CPU of the Raspberry Pi 4." - why are you surprised? it is much bigger than the tower (surface wise)

Request: Can you make an EndeavourOS (Arch) KDE OS for RPi next?

Can you get two run 100% the same with the same everything? case/fan, cords, drives, etc. So you can test diff roms, OSes, cords, drives, power supplies, etc. going forward?

Hi. Are you going to cover the official Raspberry Pi version of Fedora 37 once it gets out of beta? So far, nobody has covered it running from a SATA SSD via the USB port of the Pi, and I think that will require a shitton of tweaks in the configuration file in the boot folder.

The result isn't that surprising. Finned coolers are that good a passive cooling, look at the changes noctua made to their passive cooler, and you have a fan blocking most of the potential passive air flow. So not only does the case have a larger thermal mass it also has a larger effective area able to passively cool

Can you post a link to your config.txt file. I'm trying to overclock my pi and your file looks more straightforward than the stock one

Wasn't there also a revised CPU in the Pi4 which could be part of the differences?

Why would the Pi throttle itself below 80C?

aren't the CPU's ran at a super low heat levels so kids in schools don't burn their finger tips? ... on the temps the CPU's can handle?

your c to c link where the problem

disk taking too much power probably

say again why you did not use stress-ng to make it 100% of the CPU and to see the temperature changes? when you test 1 thing you should eliminate all the possible variables, no? ;-)