PC110 "No.9: Kongo" FSTNized success !
The "compensation film" for FSTN was available and I tested it , but in order to confirm its practicality, as the foundation using " No.9:Kongo " I tried to complete as it is.
Except for the fact that the screen is a little dark (feels), this FSTN color screen seems to be practical enough ! (Fig_1)
The "STN color LCD screen" (Example : the screen on the right of in Fig_2 [click]), which had disappeared due to the vinegar syndrome, has been revived normally ♪
(Although this "LCD panel" + "control & drive board" is a diversion/modification product from another discarded model, I'm thinking of replacing it later for further improvement.)
" No.10:Katsuragi " will eventually be converted to FSTN, but "improvement of backlight" to brighten the screen and "film" will be attached neatly. I want to do it after devising something for that.
−−−−−−−−−−−−−−−−−−−−−−−−
+++ Procedures, placement, etc. +++
Originally, in order to increase the front brightness, two "lens sheets" (vertical and horizontal dome type) and one "diffusion sheet" (luminous flux dispersion type) are placed in the "backlight part"。
But there were few sheets that could be used, so I decided to omit them.
For " CCFL/Cold Cathode Fluorescent Light tube ", I used newly bought .
It's a bit bluish, but it probably has a longer lifespan.
At the beginning, the "glue surface" is pasted on the "back side of the liquid crystal glass" of " X.Compensation Film ", but this quadrangle was cut out with right angle from the big one.
This Film is "100mmx75~77mm quadrangle", and then the "glue surface protection sheet" on the surface was removed, and pasted (I was careful not to let dust enter).
Before assembling the "backlight part", the "protective sheet" behind the "Compensation Film" has also removed.
Finally, the front and back and tilt angle of the " Front Polarizing Film " were examined while actually looking at the displayed image.
The " Front olarizing Film " must be cut out as an "angled quadrangle" with the front and rear "protective sheets" attached, but after that, when the "glue surface protective sheet" is removed, the tilt angle changes greatly, and the color tone also changes significantly.
Therefore, after examining the sheet with the "sheet" peeled off, a quadrangle tilted "about 25 degrees" to the left from the vertical line was cut out.
(However, it feels a little bluish, so it might have been better to increase the tilt angle a little more.)
On the contrary, if the "glue surface protection sheet" is not removed and used as it is, it would have been better if the tilt angle was small, but this time, "glue surface & non-glue surface" should be used so as not to reduce the "screen brightness" as much as possible. All "protective sheets" were removed.
+++ Defective part +++
The "Compensation Film" side was not very noticeable, but on the "Front Polarizing Film" side, a "bubble pattern" remained and it failed at first !
The second time, it still remains. (Fig_5)
I tried to do it the third time, but I stopped because there was something wrong with the "LCD panel assy".
The image disappeared in the lower half of the screen. (Fig_6 [click])
Normally, the full screen is displayed, but on rare occasions the lower half of the display disappears.
Probably, it is a terrible defect of "terminal peeling" of "LCD driver line", so it will be difficult to repair.
Since the phenomenon is rare, I'm thinking of keeping it as it is.
If it doesn't work, I'll use another FSTN-LCD for " No.9:Kongo ".
Besides, I'm thinking of retiring this machine from the role of a " test bench ".
(This is a perfectly healthy body now, but deterioration due to natural fatigue and mistakes of mine will be inevitable.).
Therefore, I will try the next "backlight improvement" in " No.10:Katsuragi ".
PC110 No.9:Kongo memory error found
Until now, this " No.9:Kongo " was only operated on " Pasonaware ", but from this time on, I prepared to run it on the " Win95 ".
... However, I couldn't set up except with "16 color display".
When I try to force it to "256 colors", it will malfunction and eventually it will get a "fatal error". (Fig_7)
Initially, I re-installed the "display driver of Chips " as if it was broken, and after that, " Win95 I tried to reinstall it.
However, the situation did not improve at all.
Finally, I thought there might be a problem with "16MB extended memory". (Fig_8 [click])
This 16MB memory is handmade, and the one I gave to "Mr. M " (I don't remember if it's paid or free) is back this time. For a moment, "Anxiety : may be bad this ?" passed through my head.
−−−−−−−−−−−−−−−−−−−−−−−−
+++ 16MB confirmed +++
In order to confirm the correctness of this 16MB memory, I installed it on two other machines such as " No.10:Katsuragi (FSTN modified)".
As a result, no abnormality was found on other machines. (Fig_9)
When I install it on " No.9:Kongo " and start " Win95 ", 16 colors will be displayed. and each time it gets a warning that the driver is incompatible.
... This abnormal phenomenon might occur because the execution of " XPATCH " to change 16 colors to 256 colors is not done well. I had such question.
I tried running " XPATCH " five or six times (with two saved FDs), but every time I said "_4" Only come out. (What ?)
I've never failed, so in this situation I don't know any other way than trying the same thing over and over again.
As a test, I put "4MB extended memory" to it. (Fig_10 [click])
In that case, no abnormality will occur !
In other words, if I reduce the "extended memory" to "4MB" (total: 8MB), " Win95 " will start up, and even set it to 256 colors, it does not occur abnormally !
(When I put "8MB" on it, it will start up for the time being, but it will freeze after a while, so it seems that "4MB" is the limit.)
May be a congraturation! after that, but there are still some problems.
I saw the "flow of horizontal waves" on the screen.
The cause may be "capacity loss" of the "electrolytic capacitor" on the "liquid crystal control board".
If I replace the "LCD module", it may be easy to get rid of it, so I'll leave it as it is for a while.
PC110 No.9: Kongo screen display, improvement
Each " PC110 " at hand aims to "keep the seeing the internet" as possible as.
(The other day, for that reason, https://garakutaen.sakura.ne.jp/ was slightly improved so that a part of the "soliloquy collection" could be viewed.)
I put a wireless LAN card (BUFFALO WLI-PCM-L11GP) in this " No.9:Kongo " so that I could do it, and started maintenance. (Fig_12 [click])
However, the "screen brightness" is not enough (because of the CCFL backlight without the lens sheet), and the characters are very difficult to read. (Fig_11)
I used to use a magnifying glass without stopping, but as if to prevent it, the difference in brightness between the upper side and lower side of the screen became noticeable, and the condition worsened.
I made another " FSTN-LCD module " with a bright LED backlight and tried to replace it, but it didn't work and I couldn't achieve it.
Also, I stopped directly modifying the "control circuit board" of " No.9:Kongo " because there is a risk of breaking it.
Then, I just changed the "backlight" from "CCFL light source" to " LED light source ".
With this, the screen became "bright", but the "brightness difference between the top and bottom of the screen" also became large.
It seems that we still have to continue repairing and improving.
−−−−−−−−−−−−−−−−−−−−−−−−
+++ Brightness improvement +++
The "LED backlight" of " No.10:Katsuragi " has "series resistance: about 11Ω" to suppress "blueness", but this " No.9:Kongo " has prioritized screen brightness over that, and set it to " resistance : about 7Ω ". (Fig_14 [click])
(The extremely small connector for connection (pin pitch 1.72 mm, 4 pins) was bought at Nipponbashi, Osaka the other day. \ 40/piece)
Connect this " LED backlight " and " CCFL backlight " to the same DC power supply as a single unit, and compare the "backlight surface brightness". I tried it. (Fig_13)
Applied voltage
LED
CCFL
Total current
7.5V
0.8μW
50μW
290mA
9V
65μW
50μW
295mA
10V
130μW
50μW
300mA
All values are about (approx.) values.
Moreover, 10V or more was not measured.
However, when connected to the actual "backlight power supply" of " No.9:Kongo", the actual current flowing through the "LED backlight" is about 250mA . It turned out to be .
Estimating from this, the backlight surface brightness will be quite high.
I think the reason why the current is higher than expected is probably because the voltage is a little over 10.5V.
I accidentally shorted +,- and was about to blow the fuse, so I didn't measure it properly.
This terminal voltage once became 0V due to my ciucuit short, so I was prepared for a "fuse blow", but it recovered naturally later. Lukey !
+++ The elimination of the vertical difference in brightness is incomplete +++
I think that the difference in brightness between the top and bottom of the current screen is obviously a defect of the "control circuit", but (I) bear the " jinx that will always break if you touch it ".
I want to leave this " ( semi-defective ) FSTN-LCD module " as it is.
I tried to prepare a working "FSTN-LCD module" separately with the intention of replacing it, but this did not work.
(It seems that the "control circuit board" that was confirmed to work in the experiment before is not working properly.)
I'm afraid that this " ( semi-defective ) FSTN-LCD module " has only two "small electrolytic capacitors" on a trial basis.
I tried to add it, but there is no effect ! (Fig_16 [click]) Replace all" electrolytic capacitors " like Mr. Tenmani-chan ", I may have to be prepared.
However, in order to attach or detach this "control circuit board" for experiments, I think that it is necessary to devise a jig for it.
This is because there are places where the actual product will become unusable due to "land peeling" if repeated as it is.
(What kind of good method is there ?)
PC110 No.9:Kongo display part replacement
The "display brightness" of " No.9:Kongo " was different between the top and bottom (Fig_18 [click]).
So I changed the display (LCD module) to more better FSTN display of " No.7:Bamboo " (which the main body is currently out of order).
And then replaced it with " LED backlight ".
" No.9:Kongo " has finally reached the "target level".♪
However, the method of attaching the "surface polarizing plate" of the "LCD module" was not good.
I tried again pealed the front polarizing film to remove the "bubble pattern".
Then, it seems to have improved considerably.
The "final goal" is to make the "readable of internet site" using a "wireless LAN card".
But this " PC110 " has some faults in various places, so at first, I felt it would be difficult.
However, I was able to read the site of my another site (https://garakutaen.sakura.ne.jp/).
At this time, I've reached my goal ! Yei !
One of the faults is that I can't increase the memory capacity beyond 8MB .
Also, it seems that " wireless LAN card " cannot be used in the upper slot for PC cards (although the lower slot can be used).
These defects, I can't expect any further improvement (with the limitation of my technical ability), keep it as it is, and I will maintain and make use of it.
−−−−−−−−−−−−−−−−−−−−−−−−
+++ LCD module replacement +++
During I'm trying to check the cause of immobility of " No.7:" Bamboo "", I noticed suddenly the display part (LCD module) of this was (refleshed) FSTN and might be able to display.
Immediately, I decided to replace it with that of " No.9:Kongo ".
(Fig_19 Left:"No.9:Kongo", Right:"No.7:Bamboo")
When I replaced it, there were a lot of "bubble patterns" ! And it was a "CCFL backlight".
It was easy to replace it with an "LED backlight", but it was difficult to erase the "bubble pattern" !
So, I was warmed the "film surface" with a "hot blower", peeled off once, and then reapplied while slowly leveling from the edges.
The result seems to have improved considerably.
However, the warming was insufficient before peeling it off, so it was difficult to peel it off.
So small "light leakage" occurred from the part strongly sandwiched by the "hand nippers". (Fig_20 [click])
... It should had to warm it up enough and then peel it off, and reattach it.
The "LED backlight" has a little more current, so it seems that the "blueness" has become stronger, but I would stand this.
If I get more preferred "LED array", I'll repalce it because it's easy to replace it.
Also, since the surface of the "front polarizing plate / film" is smooth, I think it will be easier to see if it is replaced with "matt treated film", but this is also an issue for the future.
+++ Internet connection +++
The " Access Point " is made by the AP of Cisco and the power supply unit of Buffalo . (Those were gifted by "Mr.M ", Thanks again !)
For the " wireless LAN card ", " Buffalo WLI-PCM-L11GP " was used. (Fig_22 [click])
(Actually, short head type Sony PCWA- C150S I wanted to use it if there was an extra , but there is no extra and it is not available now, on the contrary, I have a lot of Buffalo products, so I will use this)
Until now, "wireless LAN cards" have been used mainly on the "upper slot" of "PC110", but in this " No.9:Kongo ", the card is not recognized on that (upper) slot. (The cause is unknown)
I tried the "lower slot" and it worked, so I decided to do "Internet browsing" (http: // site only) for the time being.
BTW: I could use the Internet explorer 3.0 for browsing.
[2021/01/28)]
PC110 No.9: Kongo lens sheet inserted
I tried to increase the "display brightness" by inserting two "lens sheets" (for a large liquid crystal) between the "backlight" of the "FSTN-LCD" and the back of the liquid crystal.
As a result, the brightness seems to be slightly increased, and white characters appear to be glittering.
However, if the sheet is placed incorrectly, "interference fringes" will appear strongly, and " hard to see " will increase. (Fig.24 [click])
The original "lens sheet" was dirty with black tape residue and dust and could not be used, so I disassembled the large LCD module and used the "lens sheet" collected from it.
However, apparently, " the pitch of mountain range lens " (for 10.4 inch XGA) is " the pitch of LCD cell dot " (for 4.8 inch VGA). It seems that it was different, and it seems that it caused interference fringes.
In other words, not all "lens sheets" were acceptable. (Dumm !)
+++ Situation +++
The "lens sheet" has back (smooth surface) and front (mountain range type convex surface), and the front side has "vertical mountain range type convex surface" and "horizontal".
It seems that there are two types of "direction mountain range type convex surface". (The name as "mountain range" is my creation and not standard)
In any case, it has the effect of increasing the forward brightness by moving the luminous flux that is going to be scattered to randam direction to the center, and it is a kind of "Fresnel lens" and uses two lenses in the X and Y directions.
However, if the "step pitch" of the "mountain range type convex surface" is different from the "pitch of the liquid crystal element", "interference fringes" will inevitably occur.
To prevent that, I put a "diffusion sheet" in between, but it seems that it is still insufficient. (Fig.25)
When they ("lens sheet" and "liquid crystal surface") are brought into close contact with each other, the "interference fringes" are severe and the screen becomes extremely hard to see. (Fig.26 [click])
After all, it seems that the "lens sheet" must use the original one (if it is beautiful).
I'm going to keep this No.9 as it is for a while.
