なかのにっき

最新版はこちら。 突っ込みは各日付の BBS エントリのほか、 メール (nakano@st.seikei.ac.jp) や フォーム からどうぞ。 なおスパム除けのため、BBS 機能には 緩い認証を入れて います。 検索エンジンから来た方は、エンジンの方のキャッシュを見るか、 下の簡易検索を試してみてください。


hns - 日記自動生成システム - Version 2.19.5

先月 2003年01月 来月
1 2 3 4
5 6 7 8 9 10 11
12 13 14 15 16 17 18
19 20 21 22 23 24 25
26 27 28 29 30 31
Namazu for hns による簡易全文検索
詳しくは 詳細指定/ヘルプを参照して下さい
検索式:

2003年01月11日() [n年日記]

#1 [URL] Mozilla/Galeon Rendering Speed Tweak Mini-Howto

あきらさんのところ から。
しかしうちの gnome2 は相変わらずウィンドウリサイズができない。なんでじゃ。

#2 [labo] テフロンスパッタ

岡田さんに lecture。 10:00 start。

14:15 点火。だいぶ MgO とは RF マッチの条件が違うけど、 ちゃんと点いた。2 Pa-100W で無事放電中。

15:30 頃、シャッターを開けるかという時間に消灯。 その後異常な on/off の繰り返し。 中をのぞくと、プラズマ点火時、ターゲットの10時方向あたりで 緑色の小さなギャップ放電様のチラチラした発光。 こいつが原因っぽい。しょうがないのであきらめて消灯。 チャンバーを開けることにする。

ターゲットを保持しているツメのあたりだろうか。 テフロンに溶けた形跡が。 一通り掃除して閉めて RP で排気。100W はちょっときつかったかな。

#3 [dept] kernel update

身の周りのマシンのカーネルをワラワラと更新。

#4 [labo] OES 結果解析

324nm のピークは 435 ch あたり、 510nm のピークは 976 ch あたり。 いずれも 0 スタートでのカウント。

この両側±3 ずつ取って、7 点に対して gaussfit を呼べば良い。 結果は、
測定位置 A B C W
というデータにして、ディレクトリに 435ch.dat, 976ch.dat という名前で出力すれば良いだろう。 perl から gaussfit に stdin を与え stdout を利用するには、 IPC::Open2 を使えば良い。詳細は Perl クックブックの レシピ16.8 (p.581)。

#5 晩小屋

マンガ引き渡しで西荻窪で小野澤さん・吾妻さんと待ち合わせ、 晩小屋へ。 ブリの刺身、カマ焼き、ブリ大根のブリシリーズが大変うまかった。 ご馳走になってしまいました。ありがとうございまする。
コメント [全部読む/投稿する]

2003年01月12日() [n年日記]

#1 普通の日記

8:00 に起きて朝飯を食べ、風呂に入ったら湯疲れして昼寝。 14:00 に起きて餅で昼飯にしてまた風呂に入って寝る。 温泉モード、というか、どうにも体の芯が疲れている感じ。

夕方から大学へ。吉祥寺 T-ZONE で switching hub と ついでにケーブルを 1, 2, 3, 5 m 二本ずつ。 8ポートの電源内蔵 hub が4000 円しなくなっているのか、すごいなあ。

大学前のローソンに寄ったら藤本くんがいた。

#2 [freshmeat] 1/11 分の新着メールから

#3 [debian] window manager & gnome2

安部さんから sawfish は安定していない というコメントをいただいた。 確かに使っているのは sawfish でした。 こちら に書いてあった
9. How do I switch from sawfish2 to metacity?

Currently, the only way to switch window managers is to kill the one that's running, start the window manager of your choice and save you session. Example: 'killall -9 sawfish; sleep 1; metacity &' and then save you session. You can save your session by running the 'gnome-session-save' command or by logging out and checking save session in the logout box.
で metacity にして、ちゃんと window resize ができるようになりました。ありがとうございます。

この辺は debian 本家の方でも いろいろ あったみたいですね。

#4 [dept] 学科サーバ

apdrelay:

電源を入換えて 2.4.20 にアップグレード。

proxy:

2.2.20 にアップグレード。
コメント [全部読む/投稿する]

2003年01月13日(月) [n年日記]

#1 [URL] Galeon: Smart Bookmarks

Galeon handles Smart Bookmarks quite elegantly. In light of that, we offer this archive of bookmarks that others have thought handy. To add one, simply right click on the image or the link and choose "add bookmark".

#2 自戒 (じかい)

「自戎」という間違いを複数箇所で見たのだけど、 どこかの IME の辞書に間違って載ってたりするのだろうか?

#3 [labo] OES 結果解析

gaussfit は stderr も使っていたので、 Perl クックブックのレシピ 16.9 の方、 IPC::Open3 と IO::Select を使うやり方で。 出ていた例を丸写ししただけで動作。ありがたや。

できた。~/data/03oes/2003-01-09/sroes.pl しかし (いつものことだが) 自分にしか読めない perl コードになってしまったな。

むー、やはりターゲットから離れたほうが 510 の相対強度が大きくなってるなあ。 やっぱりターゲットから離れたほうが Cu 密度が低いのかな。 直感的な理解 (=拡散によって薄くなる) とは逆なのだけどなあ。 まあ背景だけでも予稿にはなるだろうから、さっさと書こう。

いや、そういうわけでもないか。 ターゲット近傍は 324 が強くて、一番 510/324 が大きいのは 5mm 上あたりかな。いい感じか? そんなことはなかった。むに。

一応こっちの予稿は書きおわり。

#4 [labo] MgO 予稿

うへ、どうも合わないと思っていたら、 lorentz 関数のところにでっけえバグがあった。 屈折率が波長依存性を示さないので気がついた。藤本くんごめん。

#5 [labo] 応物申込

done。 boss も今日来ていたので、提出前に見てもらえた。 返信メールは +misc/2353〜2356。 パスワードは秘密日記にもメモ。 しかし 10 分も開けずに申し込んだ 2 件の講演番号が 4854 と 4883 だからな。 かなりの殺到ぶりとみた。

pdf 送信は ssl 経由だと dns エラーになってビビったが、 暗号化無しの方で送ったら無事に受理された模様。

17:00 になったらページごとズガっと止められるかと思ったら、 まだアクセスできるみたいだな。

#6 [labo] sroes 測定コマンドマニュアル

書いた。 ほんとは roff で書くべきだったかなあ :-)

#7 [URL] 田中耕一さんの研究内容。

わかりやすい。 工藤研なら guidance とか recruiting に使えそうだ。

#8 [dept] 液体窒素当番

出てきてたので、フォルダを受け取ってしまう。
コメント [全部読む/投稿する]

2003年01月14日(火) [n年日記]

#1 [labo] Vacuum Products 訪問

行きがけ、交差点で左にいた *1 車二台が信号の変わった後に直進し、 交差道路の右側にいた白バイに二台まとめて捕まっていた。 デイツーに寄っていろいろ買い込んで登校。
*1: 自分は右折車線にいた。

#2 [linux] カーネル 2.4.20

うちは都合 8 台 2.4.20 にしましたが、特に問題なく順調に動作しています。 ちなみに tarball を直接とってきたのではなく、 それぞれ woody, sid の kernel-source-2.4.20.deb です。

その debian のカーネル 2.4.20:

なんか最近 debian varsion が続けざまに上がってるなあ。

#3 [labo] PC

最近利用率が上がってきたので h2, h3 を LAN に戻す。 VB6 の入ってた h2 の windows update に disk 不足で失敗、 適当にコントロールパネルから hotfix を消したら立ち上がらなくなった。 しょうがねえ、あとで g4u で書き戻そう。

#4 [JM] minori

定例メンテナンス。昔のアーカイブを消す。 あと一部の perl スクリプトに
ENV{'LANG'}=C
を追加、んでもって Makefile にあった、 これらのスクリプトに対する依存を削除。

#5 [dept] 実験成績集計

最終的な実験担当表をもとに、 このとき の手順で $(teacher)-final.xls を作成。

#6 [linux] mime-construct 上級編

プレインテキストの preamble に xls ファイル 2 つを添付して送りたい場合の処方箋。
mime-construct --header "From: nakano@apm.seikei.ac.jp" \
 --to nakano@apm.seikei.ac.jp --subject 'EXP1 Report Table' \
 --type "text/plain; charset=iso-2022-jp" --encoding 7bit --file 'nkf -j preamble.txt |' \
 --type application/vnd.ms-excel --attachment nakano-final.xls --file nakano/nakano-final.xls \
 --type application/vnd.ms-excel --attachment 2002exp1.xls --file ../2002exp1Report.xls
という感じで実行すればよい。 --file が添付するファイルのオプション。 これは man mime-construct にある "Part Output" のひとつ。

この part に対するオプションを "Per-part Header" で与えることになり、 これが上記の --type とか --encoding とか --attachment とか。 ここで
Each of these applies only to the next part output. After each part is output they are reset to their default values. It doesn't make sense to use them without a following part, so mime-construct will sputter and die if you try to do that.
であることがミソ。

ちなみに --attachment に与えたパラメータは
Content-Disposition: attachment; filename=FUGAHOGE.xls
の FUGAHOGE.xls になる。 --encoding は特に指定しなければ適当に選んでくれるようだ。 --type は debian なら /etc/mime.types を見るとよさげ。

#7 [labo] 今日の卒研

藤本君、尾崎君:

MgO の AFM 測定。 ちと stub がずれててプローブのトライアングルが隠されてたり、 例のチリチリ振動が起こったりで、あまり捗らなかった模様。 だんだん慣れてください。

岡田さん:

テフロンスパッタ。2x10^-2 Torr (表示値) 50W 1h。 今日はちゃんとできたかな?

しかしどうも内壁でチラチラ異常放電が起こっているのが気になる。 こないだはターゲット周りのリングで起きていたが、 今日はアングルバルブのあたりだった。 緑色に光る。絶縁膜だし F が入ってるプラズマだから、 いろいろ厳しいのだろうが。

テフロンのスパッタは金原先生たちもいくつか学会発表していたようなのだけど、 論文にはなってるかしらん。
コメント [全部読む/投稿する]

2003年01月15日(水) [n年日記]

#1 [labo] h2 (PC)

g4u cloning して、サーバで smbpasswd -x h2$; smbpasswd -a -m h2 したのち、ホスト名を変更してドメインに再参加させた。

#2 [Windows] Documents and Settings 以下の個人ファイルを消す

h3 のを消したら 1G 近くも disk が空いた。 しかし administrator で入っても
  1. プロパティ→セキュリティ [詳細]→所有者 で「サブコンテナとオブジェクトの所有者の置き換え」にチェックを入れてから所有者を administrator に変更
  2. 実際に削除
  3. ゴミ箱に入ったのをもう一回削除
という手順が必要。一手一手でいちいち待たされ、非効率このうえない。

#3 生活時間帯

せっかく超朝型になったのに木阿弥か。 まあ予想通りではあるが、試験期間は早起きのほうが都合いいんだけどな。

という訳で(?)研究室に泊まってみた *1 。2:00-8:40 睡眠。
*1: 眠くなって帰るのが面倒くさくなったともいう。寒いし。

#4 [labo] OES 測定

film 設定:

測定をしてもらうユーザに適宜権限を委譲。
% addgroup scsi
% chown root.scsi /dev/scsi
% chmod 660 /dev/sg0 /dev/ttyS0
して、scsi グループと dialout グループに kodyssey, hideaki を追加。

aic7xxx.o がロードされてなかったので /etc/modules に追加。

位置合わせ:

5mm 単位で動かし、ch435 のピークを見ながら左右に微妙に移動。 pmameas に --width オプション (測定範囲を限る) がいるな。

測定:

1 Pa (表示値) 50W で。bottom は -7mm、そこから 5mm おきに +18mm まで。 基板ホルダの有無の効果を見てもらう。

#5 [linux] rwiki と mod_encoding

mod_encoding を入れるとやっぱり rwiki の文字化け が起こる。 どうやら 検索語によって namazu が not found になる のも同根の問題らしく、 あきらさん に教えていただいたところ、解決策は こちら とのこと。

自分は httpd.conf の LoadModule encoding_module の行をコメントアウトして解決にしたが、 一応 BTS もしておいた。

この件に関する あきらさんのコメント 。つうか日記はじめたんすか :-)

#6 [labo] 今日の卒研

上記 OES 測定のほか。

岡田さん:

試料取出しの実演。 テフロンはやや茶色がかった膜になっていた。一応向こうは透けてみえる。

佐藤くん:

実験結果の解釈について+明日の実験の打合わせ。

rwiki:

研究室の方のに、各種連絡用のページを作ってアナウンス。

#7 [debian] gnome2 @ tp600x

自宅のノートは長いこと gnome 1.4 で hold していたが、学校の gnome2 環境が ほぼ問題なくなった ので、こちらも gnome2 にしてみた。 gnome-control-center (2.0.3.2-1) と gnome-panel-data (2.0.11-2) で /usr/share/pixmaps/advanced-directory.png が conflict していたので、前者を --force-overwrite でインストールしてしまった。 学校では衝突してなかったんだけどなあ。なんでだろ。 nautilus は起動しないことにした。というか remove した。
コメント [全部読む/投稿する]

2003年01月16日(木) [n年日記]

#1 [Windows] Winbiff 2.4.2 リリース

V2.41からV2.42への変更点(2002/12/10)
変更点 修正した不具合 

#2 [labo] 今日の卒研

佐藤君:

XPS 実験用 Cu, In 蒸発源のチャージ + In 蒸発源の焼き入れ。
取出してミニチャンバーに設置するときに In 蒸発源の頭が折れた(泣)。

工藤さん:

出てきたレポートにコメント付け。

#3 [freshmeat] 1/15 分の新着メールから

#4 [JM] mirrordir.1

空き時間にちょっとずつやってたのでだいぶかかったが、 やっと終った。 んー、man -l -Thtml mirrordir.1 で変換したのだけど、 .nf のあたりとか、ちょっとおかしいところがまだあるかなあ。 roff 版は こちら

うへ、いきなりだいぶチョンボってた。なおしました。

#5 [CD] "Angel Beetle" The High-Lows

最高。
コメント [全部読む/投稿する]

2003年01月17日(金) [n年日記]

#1 [LDP] 1/8〜1/17 の updates

updated FAQ:

updated HOWTOs:

#2 [JM] mirrordir.1

さらに修正(^^; して cvs commit。

#3 モンティ・ホール・ジレンマ

小松さんのとこ から。なるほどおもしろい。
あなたがゲーム番組へ出ていて、3つのドアのうち1つを選ぶとします。1つのドアの後ろには車があって、あとの2つのドアの後ろにはヤギがいます。
あなたはドアを1つ、たとえば1番のドアを選んだとします。番組の司会者は残ったドアの1つ、例えば3番のドアを開けます。司会者は、それぞれのドアの後ろに何があるのかを知っています。3番のドアにはヤギがいました。ここで司会者はあなたに、「2番のドアに変えますか」と聞きます。
さて、2番のドアに変えた方がいいでしょうか。

この問題は、
A君とB君がいます。A君は素直に1番のドアを選びました。しかし、B君は残りの2つのドアを選ぶと言い張ります。揚句の果てに、司会者に1つに決めるから、ヤギの入っているドアを先に教えてくれと頼む始末でした。司会者はいい人で、3番のドアを開けてヤギが入っていることを教えてくれました。B君は2番のドアを選んで、こう言いました。「これでA君と同じように3つの中から1つのドアを選んだんだ。公平だろ」と言いました。
あなたは、A君の立場を選びますか、B君の立場を選びますか。
という話と等価である。
B君の立場の方が有利であることは言うまでもない。
うーん、いまいち納得できん。ほんとに等価か? というわけでちと google ってみた。

法医学実習によるモンティ・ホール・ジレンマの理解:

「選択を変えない場合に当たる確率は1/3。選択を変えてハズレるのは最初に当たりを引いていた場合で,それ以外は当たるわけだから選択を当たる確率は1-1/3=2/3となる。したがって選択を変えた方が有利」j

モンティ・ホール ジレンマの誤解:

ruby プログラムやリンクがある。

或いは科学的な日々:

2002-01-22〜24 にかけて。
結局,答えを知っている相手(司会者)による選別を経たほうが確率が高いということだ.自分が最初に選んだ一つだけは相手の選別の対象となっていないので,確率が上昇しない.

The pleasures of probability:

八田(ま)さんのとこ で紹介されていた本。 むー、ペーパーバックはないかいな。

#4 [URL] サンバック

真空装置メーカ。DM から。

#5 [labo] PC 修復

k2 が立ち上がらなくなったとのことで、 件のマシンの hda2 に入れといた norton ghost イメージを書き戻して *1 さっくり復旧。 ついでに h1 にも g4u イメージを書き戻して諸々設定。 これはバックアップマシンだな。
*1: フロッピー起動して local→partition→from image と選び、 destination に new ではなく 1st drive の 1st partition を選ぶ。

#6 [linux] ssh -N

-N      Do not execute a remote command.  This is useful for just
        forwarding ports (protocol version 2 only).

#7 [labo] VP 電話

来週火曜日に電話してくれるとのこと。水曜日は試験監督が無いから、 水曜日にいけるとありがたいのだが、と伝えておく。

#8 [paper] PRL 90(1)

Reversing the Brazil-Nut Effect: Competition between Percolation and Condensation:

We report on experiments on vertically shaken binary granular mixtures, which separate into their components due to the external excitation. This well-known phenomenon, where large particles rise to the top of the mixture, is called the Brazil-nut effect. Recent theoretical findings predict also a reverse Brazil-nut effect, where large particles sink to the bottom of the container. We choose spherical beads of various diameters and materials in order to observe the transition from Brazil-nut effect to its reverse form. The direction of demixing depends sensitively on the external excitation, so that it is possible to switch between both effects for a given mass density ratio.
実験報告。おお。

Rapid Roughening in Thin Film Growth of an Organic Semiconductor (Diindenoperylene):

The scaling exponents α, β, and 1/z in thin films of the organic molecule diindenoperylene deposited on SiO2 under UHV conditions are determined. Atomic-force microscopy, x-ray reflectivity, and diffuse x-ray scattering were employed. The surface width displays power law scaling over more than 2 orders of magnitude in film thickness. We obtained α= 0.684±0.06, β= 0.748±0.05, and 1/z = 0.92±0.20. The derived exponents point to an unusually rapid growth of vertical roughness and lateral correlations. We suggest that they could be related to lateral inhomogeneities arising from the formation of grain boundaries between tilt domains in the early stages of growth.

Optical Properties of Ultrathin Films: Evidence for a Dielectric Anomaly at the Insulator-to-Metal Transition:

Optical properties of ultrathin layers (<50 Å) of Au and Pb quench condensed on amorphous germanium (α-Ge) have been measured in situ at 10 K. The development of these films from an insulating state to a metallic state is traced as a function of the film thickness as well as the sheet resistance, R□. Of particular interest is the regime of R□ near 3000 Ω where there is an anomaly in the optical transmission. This anomaly is due to a singularity in the dielectric function when the system undergoes an electronic percolation or insulator-to-metal (I/M) transition.

#9 [paper] APL 82(1-3)

実際に拾ったのは全部 No.1 から。

Nanoindentation of silicon nitride: A multimillion-atom molecular dynamics study:

Nanoindentation of crystalline and amorphous silicon nitride films is studied using 10-million-atom molecular dynamics simulations. A rigid pyramid-shaped indenter tip is used. Load-displacement curves are computed and are used to derive hardness and elastic moduli of the simulated crystalline and amorphous films. Computer images of local pressure distributions and configuration snapshots show that plastic deformation in the film extends to regions far from the actual indent.
力技。

Time of flight secondary ion mass spectrometry study of silicon nanoclusters embedded in thin silicon oxide layers:

Si nanoclusters have been formed by 5 keV Si+ implantation at a fluence of 1×10^16 atoms/cm2 into a 200 Å thin thermally grown SiO2 film on Si (100), followed by thermal treatment at 1000 ℃ with different annealing times. All the annealed samples show a broad photoluminescence spectrum with increasing intensity as function of annealing time. The use of a dual beam time of flight secondary ion mass spectrometry in negative mode with Cs+ ions at low energy for sputtering allows us to observe variations in Si_n- signal due to excess of silicon atoms introduced by implantation. With the high sensitivity achieved using this instrumental configuration it is possible to follow Si signals which give information about the chemical enviroment of the Si atoms. The possibility of studying the time evolution of the nucleation and growth of nanoclusters has been investigated.
Sin- クラスターのシグナルがミソ。

昨日か一昨日の日刊工業新聞に載ってたネタだが、 群馬大のグループが Si と SiO2 の co-sputter で Si ナノクラスター分散 SiO2 を作っていたな。

Direct evidence of a buried homojunction in Cu(In,Ga)Se2 solar cells:

The built-in electrical potential of Cu(In,Ga)Se2 (CIGS) solar cells was measured quantitatively and resolved spatially using scanning Kelvin probe microscopy. Profiles of the electrical potential along cross sections of the device demonstrate that the p?n junction is a buried homojunction, and the p/n boundary is located 30?80 nm from the CIGS/CdS interface in the CIGS film. The built-in electric field terminates at the CIGS/CdS interface, indicating that the CdS and ZnO layers of the device structure are inactive for the collection of photoexcited carriers.

#10 [paper] JVST 20(2)

去年の March 号。うひ。

Gas heating and throw distance for the sputter deposition of aluminum and tungsten:

The throw distance of Al and W atoms was determined (A) by investigating the pressure-dependent deposition rate and applying the Keller-Simmons one-dimensional model. The power density at the target as a parameter was varied over more than one order of magnitude. Whereas the values determined for Al are not influenced by power, the values for tungsten strongly increase with power. This effect is related to strong gas density reduction related to gas heating for W sputtering, which reduces the probability of collisions between sputtered atoms and the gas. The gas heating was experimentally determined by simultaneous measurements of the pressure in the sputtering chamber and at the end of a tubular probe inserted into the hot plasma region. For tungsten, a strong heating and related gas density reduction was found. An analytical expression describing the spatially resolved temperature profile in a sputtering discharge is developed by means of the Keller-Simmons formulas. The influence of various process parameters on gas heating can be easily studied by the model. There is good agreement between the model calculations with the experiments performed as well as with Monte Carlo simulations and experimental findings from the literature. Forward simulations by the model result in a power-independent throw distance (B) for both elements, which are in agreement with the values (A) resulting from the pressure-dependent deposition rate. Existing differences are explained by the rough assumptions of the model, which is critically discussed.
Al だとほとんど効果がないというのがちょっと意外。

Cross section for removing chemisorbed oxygen from an aluminum target by sputtering:

By chemisorption of oxygen on an aluminum target, an absolute target voltage increase is noticed which can be attributed to the reduction of the secondary electron emission from the target. By following the target voltage behavior during magnetron sputtering of the aluminum target precovered with a chemisorbed oxygen layer, the cross section for removing the chemisorbed layer could be calculated based on a proposed model that relates the measured target voltage to the target coverage.

Ion energy distribution functions and Langmuir probe measurements in low pressure argon discharges:

Using an energy selective mass spectrometer, ion energy distribution functions (IEDFs) in a low pressure radio frequency (rf) argon discharge have been measured. This experiment has been done in the diffusion chamber of a helicon source. Two kinds of IEDF have been observed depending on the gas pressure: classical "one-peak" spectra at high pressure (>1 μbar) and "two-peak" spectra at low pressure (<1 μbar). These results have been correlated with the plasma potential and density profiles along the axis of the reactor obtained with a Langmuir probe and the IEDF have been fitted using a simple analytical model. In this way, noncollisional diffusion of energetic ions from the source chamber to the diffusion chamber has been shown.

#11 オーストラリアドル

証券会社のおねーさんから、 10 年もののオーストラリアワラント (年利 4.5% 相当) を勧められる。 ただちょいと 交換レート を調べてみたら、豪ドルって円に対しても 結構下がってる んだな。 今回は見送りかなあ。豪ドルはあまり使いがいもなさそうだし。

#12 [labo] アートビジネス

昨年末 のメールの件をプッシュ。あと牧君が ANSYS 使ってる PC の メモリが足りなさそうなので、256M の追加分を購入依頼。

#13 [labo] 今日の卒研

工藤さん:

レポートにコメントして、卒論にまとめていく方向を話す。 来週いつか CT 計算の実演。

佐藤君:

昨日折れた In 蒸発源の代わりに、新しい奴にチャージ。 しかし取出してみたら充填率が足りなかったという罠で、 現在再チャージ中。22:00 までに終るかな。 終らなかった。23:00 終了。

#14 [JM] 今日の post

蒸着装置の終了待ち時間を利用して、mirrodir の forward.1 。おもしろいな。
コメント [全部読む/投稿する]

2003年01月18日() [n年日記]

#1 [dept] センター試験

8:50 集合、17:30 終了予定。

終った。腰いてえ。
ちなみに担当の部屋はキャパ 90 で、 受験者は 英語 86、地歴 44、数1 45、数2 37 であった。

#2 [labo] Cu-In XPS

試験監督の合間にサブチャンバーのベーキング、 終って蒸発源のガス出し。 詳細はノートに書いてあるが、 という感じ。

#3 誕生日

でした。

どもども :-)
コメント [全部読む/投稿する]

2003年01月19日() [n年日記]

#1 holiday

1 日フルの休みは 1/5 以来。 明日返却の実験レポート (最終回分) の採点をしつつ、 採点に飽きたら Let's ブラボーミュージックスペースチャンネル 5 など。 ゲームしてる時間のほうが長いがな。

不在時に配達されていた奈良方面からの荷物を郵便局まで受けとりに。 ありがとうございます。

そんなこんなしていたら床屋に行きそこねた。

晩飯はなんとなく土鍋で粥を作ってみた。 米:水 = 1:5 で 60 分水に漬け、 強火で沸騰させて弱火で 1h。

スペースチャンネル 5:

「チュー!」ってのは "shoot!" のことだったのか?

Let's ブラボーミュージック:

曲クリアの達成感が第一作に比べて薄い分、 ややルーチンワーク的かも。

#2 [dept] 第一実験レポート採点

というわけで終了。 ずいぶん進歩した人がいる一方で、 全然ダメな人もちらほら。
コメント [全部読む/投稿する]

2003年01月20日(月) [n年日記]

#1 [JF] Debian Jigdo mini-HOWTO

リリースされた。

#2 [dept] 試験監督

1 限 物理情報演習第 II:

ねむい。

#3 [labo] Cu-In XPS

今日は 65%。9:30 から。

Ar ボンベのバリアブルリークバルブがフルクローズになっていなかったようで、 いつもの調子で 1 回転〜としていたら MC の圧力を 10^-4 Torr まで上げてしまい、インターロックを落としてビビる。 閉めてポンプその他再起動、10^-9 Torr まで通常のペースで下がったので、 特に問題は無さそうだ。

佐藤君に基板エッチングをまかせて試験監督の続き。

Cu に続いて In depo。最初 cv で上げたらまた成膜速度の振動が出たので、 cc でやってみたら安定した。なんでだろうか。

13:40 から加熱前 depth profile。あとはおまかせ。

17:00 まで測定。 In/Mo がいつもに比べてだいぶ高い。 Mo のシグナルが小さいのかな。

うーん、加熱後の depth profile を見ても、 要するにやたら In が多いような。

10:30 に測定終了。最後 X 線源の循環水の都合で 30 分待機する必要があるのを忘れてた...

#4 [dept] レポート返却

12:10〜12:50。 予想通り時間を間違えていた学生がおり、 明日も部屋をあけることに。やれやれ。

#5 肩コリ

首から左肩にかけて。やばいな、いつもの風邪の予兆のような気がする。
ちょっと寝たら少し良くなった。しかし暫く試験監督があるし、気をつけないと。

#6 [labo] 丸地君来室

所用で東京に出てきた、とのことで寄ってくれた。 いろいろ興味深い話 (特に内容を秘す) を聞く。

#7 [labo] 高城さん来室

i1124 を渡し、 加熱制御電源の仕様 (0〜500℃、0〜500W、2 段分のオープンラック) について話す。見積りを出してくれるということに。 メモリも今日明日入荷とのこと。

#8 [labo] AFM トラブル

ステージ昇降が変だ、という報告を受けて見に行く。 ステージはよくある、ステッピングモーター+ボールねじによる昇降。 通常はソフトウェアの up/down ボタンを押すと動き、離すと止まる。 ところが down ボタンを押すと、 down 方向に本来とは違う遅い回転をし、 ボタンを離しても止まらない、という症状。 ソフトウェア的には停止できない。EIU の電源 off/on と、 Windows ごとのリセットが必要。

ちとばらしてみたが、特におかしなところは見当たらない。 どうもこれは上側のリミットセンサが入ったときの挙動らしい。 上げすぎてボールねじが固く嵌まってしまい、 モーターの回転がギヤの空回りで伝わっていなかったので、 あらゆる動かなくなったのだろう。 手動でネジを下げ、リミットセンサを解除したら無事に復旧した。 (その辺の写真)

明らかにソフトウェアのバグなんだけど、 いまうちにあるハードウェアは古いので ソフトウェアも古いバージョンのしか使えないし、 以前別件で Quesant に聞いたときソースコード残ってないって言ってたから *1 運用でカバーするしかないだろうなあ。 「あまり上まで上げるな」ということで。
*1: ちょっと信じがたいが...

#9 [JM] secure-mcserv.1

XPS 終了手順の待ち時間を利用して訳出。 これで mirrordir は一通り揃った。

#10 貴乃花引退

日記的記録として。
コメント [全部読む/投稿する]

以上、10 日分です。
タイトル一覧
カテゴリ分類
book
dept
issp
labo
paper
snap
stock
vsj
Powered by hns-2.19.5, HyperNikkiSystem Project

中野武雄 (NAKANO, Takeo) <nakano@st.seikei.ac.jp> Since 1999-10-07
RSS feed, 更新時刻, LIRS エントリ, アクセス制御 (解説)

中野のホームページへ