【冷凍庫PC】ヤバそうな結露と、コメ反映版今後の予定など【#3】FreezerPC

【冷凍庫PC】ヤバそうな結露と、コメ反映版今後の予定など【#3】FreezerPC


This time, I would like to introduce the known dew condensation problem of the freezer PC and its countermeasures, the future schedule updated based on the comments received, and talk about the comments I would like to reply to. At the time of posting this video, the freezer PC is being remodeled. I wanted to explain the condensation problem early. In addition, I received comments and reviewed the schedule and changed it, and there were comments I wanted to reply. So this video only talks about my thoughts. It’s been about 2 weeks since I posted the last video. I was very happy that many people watched and received comments. So I was able to increase what I wanted to do and options. For future planss, it will take more time than before, but it seems that I can challenge many things. I am honestly surprised by the fact that I am planning to implement a method that was not planned two weeks ago. Let’s get started! First is condensation. Here I will talk about how condensation became a problem, what caused it, and what to do about it. First, why condensation has become a problem. As a matter of fact, at # 2, it was known that dew had formed even if the lid was closed when returning the freezer PC from freezing to room temperature. There was no particular change in PC parts at the start of freezing or during freezing. However, after confirming the operation in a frozen environment, when the PC was viewed from the observation window when the temperature was returned to room temperature, it was found that metal components such as heat pipes and heat sinks had condensed. However, when returning from freezing to room temperature, the PC was shut down and the power of the PC power supply itself was turned off, so even if there was condensation, I did not mind that much. However, while preparing for the water-cooling retrofit, it was discovered that a white powder had formed between the M / B and the M / B fixing screw. The lack of an insulating washer seems to have affected this, but I was worried that it would have corroded due to condensation. At this point, I finally realized that dew condensation was at a dangerous level … and decided to think about dew condensation in earnest. First, I thought about the cause of condensation. If you move the freezer with the lid first, it naturally starts to cool the inside of the freezer. Then the air, which was originally at the same temperature and humidity as the room, cools down and reduces its volume. This is the opposite of a hot air balloon warming the air and gaining buoyancy. If the food on the plate is wrapped and heated in the microwave and left as it is, it is compressed by the wrap. This is the same phenomenon. In other words, the air inside the freezer cools down, reducing its volume and lowering the pressure. Then the air in the freezer draws in the air in the room trying to balance the pressure. The moisture of the air originally in the freezer is condensed on the inner wall surface of the freezer and is further cooled and stored as frost, but the moisture of the air that has additionally entered is further stored as frost. There will be a lot of water in the freezer. Moreover, when I was checking the operation, the temperature was changed to room temperature, frozen, room temperature, frozen without removing the lid. So the moisture in the freezer increases and the humidity increases. If this idea is correct, warming the air in the freezer by benchmarking, etc., seems quite risky. The warm air that has been formed will cool down if you stop benchmarking, so it seems that more moisture will be added. In other words, if air can flow in and out, the risk of condensation increases as the temperature changes even in a frozen state. I also think that the difference in temperature change for each material such as air, resin, and metal will have an effect. Air changes easily in temperature, whereas metal does not. In addition, metal parts are quite large and heavy, such as heat sinks, so the temperature change is even slower. When returning from freezing to room temperature, the air warms faster than the PC parts and contains frosty water as water vapor, giving heat to the cold PC parts by condensation. I’m wondering what an unpleasant situation is happening. I wonder if condensation will happen this way. On the other hand, the current freezer PC did not take any particular measures, and ignored the condensation. I was thinking about a dehumidifier using a Peltier device as a countermeasure for dew condensation, but I didn’t make it because it was troublesome. When it is structural, I made projections on self-made lid to make it difficult for air to enter, but it was not as effective as it could be sealed. Even after putting the lid, I simply pasted the four corners with the curing tape. Actually, I was thinking about sticking around the outer circumference with aluminum tape, but I didn’t do it because it seemed troublesome to remove. And the gap tape used for the cable penetration part such as ATX24pin may have influenced. This tape comes with a sponge and will deform to some extent depending on the shape of the cable. But a small gap was inevitably created. Since there was a wire net for fixing the power supply, it was difficult to close the hole, so the gap was roughly closed. If you look closely at the gap tape itself, the holes would not be independent, so air would have passed through from here. Well it will be easily deformed because the sponge hole is connected, but it is not suitable for airtightness. The effect is minimal compared to the others, but the acrylic panel on the observation window actually allows only a small amount of air and water to pass through. Plastic materials, though large or small, allow gas and water to pass through. To be precise, it is transparent. It seems that the current observation window is not good for heat insulation or dew condensation. I think it is not suitable for regular use. At present there seem to be many problems. From here, I will talk about measures to be taken for condensation. In terms of structure, I would like to change the freezer lid of my freezer to a metal that can be sealed if possible, but it is difficult to make. So, basically, it seems that aluminum tape will be used a lot to eliminate the gap between the freezer and the lid. If stick it directly, it will be difficult to peel it off, so I will put a curing tape first and then put an aluminum tape on it. As a method of sealing the cable penetration part, I would like to use putty used for air conditioners. Putty is used to fill holes in the wall when plumbing the air conditioner. After putting the curing tape, put the putty and fill the hole. Or rather, I think that I can put together cables such as HDMI and USB with putty and make it like a single cable. The viewing window does not take any action at this time. Over time, I will reconsider the structure itself and remodel it. In addition to the structural countermeasures, I will also take measures to reduce or prevent condensation, such as moisture absorption by silica gel, dehumidification by Peltier devices, and making PC parts warmer than air. I would like to make a dehumidifier using silica gel with reference to the blog of the person who made it. Perhaps a dehumidifier for home appliances would be better in an environment above 5℃, but use silica gel with a low risk of failure even at 0℃ or less. This will be used to dehumidify the original air and the air that is added when cooling, and to remove the water vapor from the frost melting when returning from freezing to room temperature. We plan to install a silica gel dehumidifier outside the freezer. Just because the cable sealing method is going to be easy, I hope that the preparation, operation and maintenance will be better. The Peltier device will be installed in the freezer as a dehumidifier, and will be condensed instead of PC parts until the temperature returns from freezing to room temperature, and the water vapor in the freezer will be collected as frost and condensation. The Peltier device is an electronic device that absorbs heat on one side and supplies heat on the other side by supplying electricity, so the PC parts are heated by the warm air generated by the heat. I expect that dew condensation will be reduced if it is exposed to warm air. Keeping PC parts warmer than air is a measure to consider if the previous two plans were not enough. Specifically: Benchmark and stress test PC parts until the temperature inside the freezer rises to about -5℃ when returning to normal temperature from freezing, that is, before the temperature range where frost starts to return to water. In the temperature range where frost returns to water, the PC shuts down, and the PC power itself is turned off, reducing the risk of short circuit. Although I came up with this idea, I am afraid of the breakdown, so I will consider the situation and consider whether to implement it. Honestly, I don’t want to do it. Heating the PC parts with a heater is also an effective means, but when it comes to heating the whole PC parts, it is necessary to attach a lot of heaters to the PC parts, so I will suspend it now. By the way, at present, I consider waterproofing PC parts and immersing them in oil as the last resort. I will not discuss the oil immersion here, and will explain it later in the “future planss” section. I don’t want to do waterproofing if possible because it seems to be very difficult to restore. If possible, I want to use an alternative to reduce condensation to an acceptable level. That’s all for the dew condensation measures I are currently thinking about. In the next video, I’ll make a freezer PC with dew condensation measures and check it. If dew condensation can be effectively prevented, there is a possibility that natural low heat sources can be used without using a freezer. On the other hand, there is a risk of failure if condensation measures cannot be taken. If there is a risk of failure, it is not possible to achieve the usual use of the freezer PC’s target of “can cool well and be used regularly”, so there is concern. Well, even if the measures introduced this time are not enough, I plan to carry out dew condensation again, so I will continue to check the operation. By the way, as suggested in the comments in the previous videos, I think that the risk of condensation is lower in refrigerators than in freezers. Perhaps the dehumidifier for home appliances can be operated at all times, so it is safer in terms of condensation. Although the amount of heat generated by the operation of the dehumidifier increases, the cooling capacity of the refrigerator will be severe. However, if a freezer is used together, a system called “ice thermal storage” can be built, so I believe that there is no problem. It was suddenly introduced as “ice storage,” but details will be explained in a future schedule. If you use two, a refrigerator and a freezer, a normal two-door refrigerator is good. That sounds interesting too …. That’s it for condensation. From now on, I will talk about the updated future planss. First of all, thank you for watching, rating, subscribing and commenting on the previous videos! The motivation was raised as many people watched it, and various advices were given as comments, so there are changes in “future planss” introduced in the previous video. In the previous plan, the next challenge was to use antifreeze liquid cooling, but first I will take measures against dew condensation and then check the operation with Intel retail cooler. I completely forgot the existence of the retail cooler until I got a comment. If I attach a retail cooler, I can check the operation, so I will check before doing water cooling. I am looking forward to the results. Next is antifreeze water cooling. In the previous video, I wrote the antifreeze liquid cooling item with one arrow, but I will actually do two items. The first is to simply apply water cooling using antifreeze to the PC in the freezer. And the second is to introduce the ice thermal storage system I mentioned earlier. Although detailed explanation is omitted this time, ice thermal storage is a cooling system used in building air conditioning. In the freezer PC, the idea is to store the cooling capacity of the freezer in the form of ice, and to process the heat (that cannot be handled by the freezer) into ice. By the way, it will be frozen not with water but with antifreeze water whose melting point can be adjusted. Initially, this ice storage was the trump card for the freezer PC cooling system, but it has now changed. After the water cooling is finished, try the three cooling methods. I will talk about the details after the water cooling is over, but I will try a cooling method I have not heard. There are still DIY devices, but it is likely to increase. By the way,Iare thinking of using oil, water and ethanol. At first, I was going to choose the cooling method I had tried so far. However, it seems that gas cooling and Peltier device cooling, which were suggested in the comments, can be challenged. Well, this schedule is a long way off, so the schedule may have changed before that. If it becomes possible, it will likely be incorporated in some way to cool the PC. I was reluctant when I was offered a suggestion in the comments, but it seems to be available now, so I am thinking positively. I also had a suggestion about Peltier device cooling in the comments, so I will try to challenge it. That’s it for the cooling method to try at the moment. Conversely, some are passive. Oil immersion cooling and liquid immersion cooling. Oil immersion cooling is perfect as a dew countermeasure. However, I want to cool the latest high-spec PC at that time with the cooling system completed as the last schedule, so I immediately hesitate to submerge immediately after purchasing. So I will do my best to reduce the condensation. About submerged cooling. Fluorocarbon-based liquids used for immersion cooling refrigerants have characteristics such as toxicity and greenhouse gas. Neither use nor management seems to be able to respond alone, so I have not considered introducing it. In other words, it’s about the memory malfunction that I mentioned in the previous video. In the comments, I was told that Micron’s memory has low temperature resistance, and I was able to see an article that OC was performed by cooling the memory and CPU with liquid nitrogen cooling. The following is a summary of the low-temperature operation for each PC part that is currently known. The PC parts that have been directly cooled with liquid nitrogen are the CPU and memory. Although not a direct cooling record, M / B has a setting for liquid nitrogen cooling. Information about low-temperature operation for each PC part was increased. Well, the malfunction of the memory is only difficult to start at -30℃, and it may not be a problem if it cools down while it is operating, but in addition to measures to heat and keep the memory warm, I was able to increase the options . I just want to give priority to trying different cooling methods, so memory measures are expected to be postponed. Information was obtained from comments on SSDs that were unknown about low-temperature operation until the previous time. Upon examination, I found that the typical operating temperature range was 0℃ to 70℃. It was usually written on the manufacturer’s specification table ….. It’s working now, but if the SSD malfunctions, I’ll use a cable as a last resort to install it outside the cabinet. But when it comes to cable extensions, is it just storage with SATA connections? I’m using SATA connection M.2 SSD, but I want to use NVMe connection …. That’s it for the updated schedule. I got advice from many comments. Thanks again for watching and commenting. The video content from here is a local topic in Japan. I’m sorry, but I didn’t make English subtitles from here to the end.

3 thoughts on “【冷凍庫PC】ヤバそうな結露と、コメ反映版今後の予定など【#3】FreezerPC

  1. ご視聴いただきありがとうございます!
    また皆様から頂いたコメント、いつも喜びながら読まさせて頂いています!

    動画内で言い忘れてしまったので、この場を借りて補足。
    結露対策として庫内を厳密に密閉するのは、金属製圧力容器とか真空チャンバー並みにしなければいけ無さそうなので、ある程度の空気の出入りは諦めて、庫内空気の除湿をメインの手段にする、といった方針です。

    【下記は皆様から頂いたコメントを参考に、「そのアイディア使わせてもらいます!」的な感じで、次回以降に導入しよう・導入できるかもと思ったものを、動画投稿者がメモ的に書いたものです。文章はチョイチョイ更新されると思います。】
    2020/01/21記入

    1.冷凍庫を囲うケースを作って、その中の空気を除湿器等で除湿。
    冷凍庫に侵入しても結露するリスクが小さい空気を冷凍庫周りに準備する。
    感想:結露対策として庫内の密閉や庫内空気の除湿しか検討していなかったが、ケースを囲って除湿するだけなら、簡単に作れるし効果も期待できそう。(素晴らしい
    冷凍庫をケースで囲うと幅奥行高さが全て1m以上になりそうだが、元々ガス冷却を行うことになったらそのくらいの大きさになる想定なのでガス冷却実施時に適用したいと思う。除湿器の排熱もガス冷却で補おうと思う。

    2. シリカゲル除湿器に低温でも吸湿性が期待される塩化カルシウムなど塩化物も使ってみる。
    感想:塩化物も低温環境で除湿効果が見込めるので使えるか考えてみようと思う。塩化物が吸湿すると融け出し、導電性や腐食性がありそうな液体になるが、シリカゲル除湿器は冷凍庫の外で使えるようにしようと思うのでちょうど良い感じ。塩化物を通った空気に腐食性があるとPCパーツが錆びてしまうので、問題ないか調べて判断しようと思う。

    アイディアを下さった皆様、本当にありがとうございます😂

  2. 冬、結露防止に窓に貼る吸湿シートを冷凍庫の壁面に貼ってみるとかどうでしょう?

    冷凍庫PC、これからも楽しみにしてます。

  3. 安定運用したいなら-30℃DPくらいは欲しいけど
    消耗品の乾燥剤使わないと無理な領域(ペルチェクーラーじゃ厳しい)
    凍結による物理的応力の破壊も怖い。

    冷凍庫を水冷チラー化してCPUに水枕当てた方が現実的かも。

Leave a Reply

Your email address will not be published. Required fields are marked *