I-Appliance BBS
The Official Source for Internet Appliance Upgrades and Mods
Amazon Honor System Click Here to Pay Learn More
BBS Main List | Sign In | Sign Up | Search | Help | Linux-Hacker.netReply to Thread | Printer |

Home / I-Opener Areas / I-Opener Technical Stuff
Actuall working TFT screen?
Actuall working TFT Screen?

New MessageActuall working TFT screen? (modified 0 times) turbowsr
Profile
Please don't flame. Yes, I have searched this BBS, and read lots of information. BUT, in the end I have not read a single success story with inserting a TFT screen. I have read threads that claim someone had, but when asked how, the answer never came. Decent TFT screens can be had on Ebay for less than $100. I would like to upgrade the screen to 12" or more (depends on price), put on the touchscreen (I have this part solved), do the rest of the upgrades (HDD, CPU, MEM, etc.)

I have also followed dead link after dead link. Any info on online links with the information would be appreciated. Don't need specs. I want to see that is has worked. (most post end with "as soon as I get it back out ot the closet, I'll try it........." I guess these units aren't out of the closet yet :)


Thank you for any real tangible information.

04-07-2002 09:44:37

New MessageRE:Actuall working TFT screen? (modified 0 times) Wild_Pencil
Profile | Email
People have tried, without any success yet.

I have a Hybrid BIOS that I'm offering to anyone who's determined and willing to get a TFT panel running. The Hybrid uses code from the IOpener and a similar machine, and has a BIOS-Menu option to toggle between DSTN and TFT. So far, no luck on getting it working.

I'm in the process of relocating to Las Vegas (Yay! No more insane Silicon Valley prices), so my IOpeners are in storage for the time being. Should be settled in the new place in a few weeks.

-WP

04-07-2002 17:51:54

New MessageRE:Actuall working TFT screen? (modified 0 times) turbowsr
Profile
Thank you for the information. My I-OPener is on the way. I also have a Fujitsu Sylistic with the DTSN screen. Tech-Info states that they also shipped with TFT so I am also trying to upgrade this one. But it has a digitizer behind the screen, so it will be harder. I'm not an engineer nor a programmer, but I have over 5 years SMT soldering experience.And the "borrowed" $500 soldering iron. So Lets Get These TFT's in There!!!

:)

04-08-2002 15:23:43

New MessageRE:Actuall working TFT screen? (modified 0 times) GWIZAH
Profile | Email
Wow. Good to know we have another Brain in on this little project...

Okay, now that I have three Bios chips...and a working I-O mobo, Im back in the game. Still havent had a chance to flash your TFT bios WildPencil. I seems to have forgotten how to flash the Bios? If I remember correctly, you need another I-opener running to do a "hotflash" correct? I still have this TFT here with the cable all wired up. I'll try getting it running this weekend. If only codeman would respond to the numerous requests Ive made regarding the code he claims to have written to set the Bios. "sigh".

04-08-2002 20:20:14

New MessageRE:Actuall working TFT screen? (modified 0 times) KneuB_2
Profile | Email
It is good to see another joining in on this project. As much as I would like to see a post from Codeman providing the missing link, I'm really skeptical that he ever actually succeeded, in-spite of the e-mail rumors that abound. His last post on the subject states he still had some BIOS code to write, and since the "KIT" that he and Earth Computers were going to sell last summer never materialized I suspect he got to the horrizontal stripes point that GWIZAH reported and considered that as "running", and that is all the farther he got. Now I do not want to blast Codeman because he has been a heck of a contributor (this board for one) and I respect him a great deal. However, the more I read the Cyberblade spec the more I think that Wildfrog is on the right track, the three pull-up resistors (MA3 - MA5) identified on page 7-1 of the cyberblade spec identifies what kind of screen it will interface with. Of course I have read less than 5% of this spec and I do not understand 99% of what I have read so it is very conceivable that I'm not even in the correct city let alone the right ballpark. Anyway to test out this theory I will be stripping a dead MB (courtesy of GWIZAH's misfortune) and tracing out the MA3 - MA5 pins to find out if that is how the i-opener identifies the type of screen being used. If not I will at least find the missing pins (or test points as Codeman called them) for the last two shades of blue. I'll let you know how I progress once the MB arrives (GWIZAH has notified me that the MB has been sent so it should not be long now).
04-11-2002 20:31:42

New MessageRE:Actuall working TFT screen? (modified 0 times) Finatronics
Profile
I read over a bit of the spec, and I was led to believe that those pins were connected to the DRAM somehow. In other words,they might have a different purpose (memory addressing or data bus) after the chip is initialized. Setting them permanently to a certain setting might not be good. I'm not certain about this, though. Actually, it looked like they had jumpers/switches for these in the example circuits. I think it might be partially hardware related (pins tied high or low) and partially software related (setting bits in certain registers).

My question is how on earth do people trace the pins on these things?! First of all, the board is probably multi-layered, so there may be lines running in between the boards that you can't see. Second of all, the pins of these chips don't go through the board and are covered up by the chip, so it's nearly impossible to see what pins go to what traces!
I'm astounded by the things people have been able to figure out (pads for the Floppy drive, pads for this, pads for that). These people must have a board lying around with the chips removed. (If not, PLEASE explain your technique!).

Frankly, there's a ton of things this chipset is capable of that we aren't using. We could easily install a PCI and/or ISA slot and many other useful and fun hacks from the information in that spec, assuming there are pads (a big assumption, I know).

Eric

04-12-2002 11:51:49

New MessageRE:Actuall working TFT screen? (modified 0 times) turbowsr
Profile
Yes, they are Multilayered. Hold them up to a light. You notice how you can see through the board in spots, and not in others. This is because you have only one internal layer (remember, these are cheap PC'S) This is a ground plane within the Mobo. In spots that you can see through, you'll notice the through-holes goto traces on the other side. Mode inexpensive boards are built this way, or like older electronic (or even cheaper) devices have only one layer. The cost of layers is very high. each layer almost doubles the manufacturing cost.
So this is basicly a double sided board. Thats how you trace a line. :)
04-12-2002 16:37:43

New MessageRE:Actuall working TFT screen? (modified 0 times) turbowsr
Profile
Ok, I saw that someone found something about a Compaq that has a similar spec. and tried to flash the bios. This led to blank screen.
Anyhow, I searched around and came across two other Compaq Laptops with the same specs as the IO. The Presario 1200 series and the Armada 100S.
The largest diff's are: PCMCIA, TFT, K6-450 (refering to factory IO)
So I read some post's, and did some thinking (took some asprin :)
Are these flashed bios's actually bad? Is the bios looking to boot to a VGA monitor (using MVP4 bios)and hangs or halts (for error) during post? or looking for the TFT Panel (using Laptop equivlent bios)?
Has anyone hooked up a VGA monitor, and then flashed with MVP4 bios?
Anyone with TFT hooked up tried a laptop bios?
I don't have a TFT to try out. As soon as I find the CN1 VGA pinout, I'll rig up to CRT and try that. I have a V3B IO. so gots ta de-"goop".
Another thought also. Since most laptops are OEM, the BIOS will be OEM and stripped. But since there are various mobo manufactures that sell to system builders, the BIOS should be full and not a stripped version. Just a thought.

Looking for some input on this.

04-12-2002 19:03:47

New MessageRE:Actuall working TFT screen? (modified 0 times) *SF*
Profile
I-Opener Technical Info Reference (MODERATED) / CN1 Pinout - VGA Out Connector

Take a deep breath & reread Wild Pencil's first post in this thread.
He is your Best Hope in my opinion.
You want to Request this BIOS very Soon as he is moving.

Good Luck * StarFish *

04-12-2002 20:16:12

New MessageRE:Actuall working TFT screen? (modified 0 times) GWIZAH
Profile | Email
KneuB_2 Just for clarification purposes I'd like to post the message I received from codeman in November of 2001:


ok the hookup looks ok but you still need some software to set the chipset
to tft mode
i will find the cyberblade pdf and put it on my site it's large and will
also find the code
i wrote to set tft mode..

codeman


It seems that he read over my info posted on my site and noted that he had written code to set the "chipset" (presumably Bios) to TFT mode. Also, I currently have two I-o's and one is hooked up with the VGA out cable. So If any one wants to send me Bios images or anything else, feel free. Also, please let me know when you get that mobo.

04-12-2002 22:06:19

New MessageRE:Actuall working TFT screen? (modified 0 times) turbowsr
Profile
StarFish and GWIZAH(responce below Starfish's),

>People have tried, without any success yet.
self explanitory, hence my post.

>I have a Hybrid BIOS that I'm offering to anyone who's determined and willing to get a >TFT panel running. The Hybrid uses code from the IOpener and a similar machine, and has a >BIOS-Menu option to toggle between DSTN and TFT. So far, no luck on getting it working.
Quote "no luck". and TFT, which I remind you, I don't have, yet :)

>I'm in the process of relocating to Las Vegas (Yay! No more insane Silicon Valley >prices), so my IOpeners are in storage for the time being. Should be settled in the new >place in a few weeks.

>-WP

So I don't see how this post relates to what my last post stated. BUT, thank you for the link, THAT will help. I am working on that right now.

GWIZAH,
Quoting the spec's on Trident's page:

"Interface to external LVDS drivers capable of supporting TFT panels with 1024 x 768 resolution (XGA)."

Not sure how to take this. Could mean drivers for 1024x768 resolution(for beyond VGA)? or drivers for TFT? Sounds like it needs software beyond the Sys. Bios (Video Bios?) to run TFT.
What does LVDS stand for (I'm sure I know it, but a cranial search returns nothing :)

04-12-2002 22:41:21

New MessageRE:Actuall working TFT screen? (modified 0 times) *SF*
Profile
I read, You were looking for a BIOS answer.
Wild Pencil, has alot of time into the BIOS &
a good idea what laptops are close to the I-opener.
If you could get his assistance, You could save alot of time.
was my thoughts. Just trying to Help in my own Lame way

Not sure If this is your LVDS answer, I will let you decide.
http://www.pcwebopedia.com/TERM/L/LVDS.html

04-12-2002 23:09:51

New MessageRE:Actuall working TFT screen? (modified 0 times) turbowsr
Profile
K, I see now.

But now I look for another bios answer. Its DOA. I degooped, didn't tear up the socket, but must have killed the bios chip. Guess I'll be getting a from Bad Flash.
Thanks *SF*. I don't know if that link answered it. I assume that is what Trident ment. But now I don't understand LVDS drivers? If LVDS is -/+ Volts. I don't understand the connection there. Well I won't dwell on it, I don't think this will solve the TFT situation.
Some thing else I found. Did a Yahoo/Google search for "Cyberblade DTSN" and FreeBSD lists drivers for Cyberblade i7 and drivers for Cyberblade i7/DTSN.
So now I am back to what others kinda hint at, there is a hardware and software issue here.


keep trying,
Turbowsr

04-13-2002 01:32:52

New MessageRE:Actuall working TFT screen? (modified 0 times) Finatronics
Profile
AFAIK, when the specs say the chip includes LVDS (or any other) drivers, they mean the driver circuitry, not driver software. Looking through some of the schematics they gave, the LVDS and other appear to be interfaces for LCDs, such as the much more common DVI port.
If it makes you feel any better, I've tried to revive interest in getting the floppy drive interface supported by the BIOS, but it doesn't look like anyone's read my posts, or it was decided long ago that it was nearly impossible. At least you're getting responses! I thought I'd try the floppy connection just to see if they got it working in the newer BIOS that I have. I soldered everything up and made it look pretty decent and spent pretty much the whole afternoon/evening working on it. But the end result is that apparently it never was figured out, and they just dropped it :(
I'm beginning to wonder if the I-Opener is now considered boring to the people who used to find it exciting... I can't remember seeing anything new and useful since 2001. I guess they've hacked as much as it can be hacked...

Eric

04-13-2002 01:44:09

New MessageRE:Actuall working TFT screen? (modified 0 times) turbowsr
Profile
Personally, I believe Floppy is Dead. For the price of 50 floppys, you can find, looking hard enough on the web, that you can buy 100 CD-RW's. I have paid less than 20 cents a piece for CD-RW's.
Other routes could be USB Floppy, LS120, or a new one I am looking into getting is the
LS-240. It has the capability to write 32MB to a regular 1.44MB Floppy, and can write on the 120 and now 240MB disks. It is USB only.
I know the problem with these is no boot floppy. And Bios doesn't seem to support Boot LS 120 or boot USB. Does it even support Boot from CD-ROM? I didn't even check. Can't now, bios died.
04-13-2002 08:15:01

New MessageRE:Actuall working TFT screen? (modified 0 times) GWIZAH
Profile | Email
okay guys...Boo-hoo the I-opener is dead... Cmon how can you say that when we still have innovators like CKbone on here? Okay, Im willing to bet there is a way to add TFT functionality to the I-o. And I bet its closer than we think. I have a Toshiba laptop here that Im going to tear apart this weekend...Ive been reading up on bios tweaks and I think I have a pretty good idea of what I need to look at to get the bios off of there. Its pretty much the same specs as the I-o, LCD TFT, VGA out, CDrom, etc. Maybe Im just beating a dead horse :(
04-13-2002 09:30:08

New MessageRE:Actuall working TFT screen? (modified 0 times) Finatronics
Profile
LOL Floppy is dead... HAHAHAHA! They've been saying that for years (since the iMac)! I work in IT, and I know for a fact that CD's will never _replace_ floppies in all situations. Only about half of our computers can even boot from CDs, and alot of them don't even HAVE CD's! Making a bootable floppy disk is quick and simple, making a bootable CD takes special software, a CD-Burner, a BLANK CD (floppies can be found almost anywhere just laying around in drawers and stuff, but CD's are only where the CD-Burner is). What if I'm trying to fix someone's computer which doesn't have NIC drivers, and I'm three stories away from my CD-Burner and blank CDs? I'll go to another computer in the room, dig out a floppy from the drawer and put the drivers on it. Floppies may be slow and outdated, but they're still quite useful. Calling Floppies dead is like saying VHS is dead! Or like saying there's no reason to understand DOS just because Windows is everywhere. If you work in IT you will HAVE to know and understand AT LEAST the basics of DOS. Have you ever seen a boot CD which boots directly into Windows? Have you ever made one? I've tried, and discovered it's not possible because Windows HAS to be able to write to the disk it's installed on.
You're right, as a home user who has no other computers to be dealt with, there is no reason for floppies. But in the real wide world of computers, getting rid of floppies completely would be a sin. I never did understand why LS-120 didn't become more common. It's certainly a wonderful idea.

Sorry to go off topic, but I can't stand to listen to ignorance like that!
Eric

04-13-2002 11:09:03

New MessageRE:Actuall working TFT screen? (modified 0 times) turbowsr
Profile
Finatronics,

Your not worth it.

I've better things to do.

04-13-2002 12:54:20

New MessageRE:Actuall working TFT screen? (modified 0 times) oldman
Profile | Email
GWIZAH et all:

the TFT screen from a Toshiba laptop may work if you can get the pin-outs for the connector and you can build a bios.

the pin outs: I have been saving pin-outs i have found for different lcd's including Sharp, Toshiba, Hitachi, NEC, Optrex. the pin-outs are difficult to obtain and there is no standard, even for panels made by the same company. i did obtain a Sharp 10" panel that i have the pin-outs for. i also have the pin-out for the Sharp i-o display.

the bios: while the cpu runs the operating system, the chipset is the computer. the bios runs the interface between the i/o devices, the chipset and the cpu. the i-opener has a bios written by Award so we need a bios from Award written for a computer that has a tft panel and the same VIA chip set. it does not matter if the i-o does not have all the hardware as the computer we get the bios for. the extra code will not function unless it has some hardware to work with. back about a year ago, people were trying to add to screen memory through the bios, someone found that the Compaq Presario 7200 had a VIA chipset like the i-o, with a Award bios. if you want, you can search for the thread. i downloaded the Presario bios and saved it.

i have all this stuff to try a tft hack, but i never took the time to work it up.

the Presario bios should also have useable code for a floppy drive.

if you want copies of thi stuff, e-mail me.

have fun,


oldman
04-13-2002 18:11:20

New MessageRE:Actuall working TFT screen? (modified 0 times) turbowsr
Profile
Thank you Oldman. Glad to get some real information on the topic I needed.
As I said before, and we are in agreeance, we need a bios for a similar board. But I do have to disagree about using a Bios from any PC Manufacturer like Compaq, Dell, HP, etc. Because they tend to strip the bios down. (esp. Gateway). Of course, just to get TFT working, these should be fine. But I also want options for the other hacks :)
I think that a bios from a mobo manufacturer would do better. They seem to allow more options. I found and d/l one from EPOX for the MPV4, I'm going to give it a try.
I would like to get those pinouts from you. TFT's from 8' to 15' are selling on Ebay for less than $100, some less than $30.

Thnak you.

04-13-2002 19:07:48

New MessageRE:Actuall working TFT screen? (modified 0 times) turbowsr
Profile
Thats 8" to 15" (inches not feet :)
04-13-2002 22:29:34

New MessageRE:Actuall working TFT screen? (modified 0 times) oldman
Profile | Email
turbowsr:

i actually have a MSI motherboard for Pentium III, socket 370, with the same VIA chipset. the driver cd has a copy of the bios. since this mb was designed for a desktop installation, it probably does NOT have code in the bios to run a tft screen.

the presario is a laptop. usually, a laptop has capabilities a desktop does not have, such as pcmcia and tft. the presario also has usb. this bios may also fix the i-o usb timing problem.

it seems to me that what is needed is a Award bios for the same VIA chipset as the i-o from a laptop that has a tft display.

my email address is listed in my profile.

keep having fun


oldman
04-13-2002 23:42:46

New MessageRE:Actuall working TFT screen? (modified 0 times) turbowsr
Profile
The MVP4 chipset was Socket 7. I didn't know VIA also had a S370 with the blade. You sure this isnt the S3 Virge integrated chipset. VIA bought out S3 and integrated it into thier SAVAGE line on chips.
After my posts I got to thinking about the TFT. And it would be more viable to use laptop.
From what I've been reading, I think what needs to happen is:
VGA Hack (Known good UI)
Or build an adaptor(for TFT to switch between displays easily) and use DTSN, which ever works easier.
TFT Hack
Bios update (I will be using BIOS SAVIOR for Dual bios :) like dual boot)

Is there actually anyone doing any Bios swapping and having success. I think the failures before were caused by the bios defaulting to VGA or TFT displays, leaving a blank DTSN screen.

04-14-2002 00:00:15

New MessageNarrowing down the options (modified 0 times) KneuB_2
Profile | Email
GWIZAH et.al.

The MB arrived last night; I shall start on it on Tuesday’s lunch break. First problem; I remember there was a great deal of discussion on removing the cyberblade chip without damaging the mother board, but I can't find the resolution of how it was done, does anybody remember? How are these things installed anyway, they do not look soldered, are they just pressed on?

On the subject of TFT pinouts; Oldman, I would love a copy of whatever TFT pinouts you have acquired. I've stopped bidding on TFT screens on e-bay until I can find the spec sheet with the pinouts for one of them, and so far I have not found a single one, even the link to the pdf file that GWIZAH posted here for his TFT screen is now a bad link. I suspect that others are in that boat as well. Is it possible that you could start a public repository for TFT pinout spec sheets that others could add to; I think that would help a lot of people.

I've noticed an interesting phenomenon during my searches for a TFT screen. There are several big sellers that from time to time will post an auction that says something to the effect;

"We do not know what laptop model it came from but we plugged it into ours and it is in perfect working order with no bad pixels."

With all the prior statements made about no common pinout standard even within the product lines of one manufacturer, it seems to me that there must be some sort of pinout commonality by the time you get to the connector otherwise the ebay sellers could not be making that type of claim without getting a lot of bad feedback. This sure sounds like the beginning point for a TFT Rositastone. (OK so I can’t spell, but you get the idea. What is the correct spelling for rosetastone anyway? I can't seem to get close enough for my spell checker to find it.)

04-14-2002 13:27:04

New MessageRE:Actuall working TFT screen? (modified 0 times) turbowsr
Profile
Rosetta Stone, A rock near Mesopotamia(not sure on either spelling).
Any how, most all the parts go through what is know as a Wave Solder machine. There is a template that a machine (or person) uses to put on a solder paste. Next comes gluing on the SMT (Surface Mount Technology) then Thru Hole(i.e. pins, or any part that goes thru the entire board). The board then goes thru the wave solder machine which heats up the solder paste and and the parts are soldered on. There are machines that most mobo manufacturers use that will put out about 100,000 boards a month. They are batch tested, packed and shipped. Of course for simplicity I left out many steps, but thats about what happens. Now lets go to the NorthBridge (Trident chip), it uses BGA (Ball Grid Array) technology. All the contacts are on the bottom of the chip, and is soldered on via the above mentioned process. To remove, you'd need a High Temp heat gun, and would have to be careful not to allow other parts to fall off. The glue that originally holds the parts on before wave solder is just a tacky substance and is usually washed off (another part of the process).
Grainger carrys these heat guns for about $50+/-.
If the chips works, don't mess with it. The traces match that chip. It is not upgradable. I would assume that this would match the pinout of the VIA chip (can't remember the number right now). But then you just put the same chip back on that you removed, it would just say VIA bla bla bla, instead of Trident bla bla bla.
I built these things for years, both on wave solder and by hand. We didn't have a high output, maybe 60 units (600 +/- boards per unit) a month, so alot were hand soldered.
04-14-2002 14:24:35

New MessageRE:Actuall working TFT screen? (modified 0 times) oldman
Profile | Email
KneuB_2:
i got out my lcd data. i have pinouts for Toshiba LTM10C042 & LTM09C015-1, Sharp LQ10D011/LQ10DH11, LQ9D011, LQ14D311; NEC NL644AC30-06 these are all 640X480. the panel i found is a Sharp LQ10D363, but i need to check it again with a meter to determine which pin-out will work. as far as there being a standard, some panels have two connectors that are both used and some have only one, even from the same mfg.

another problem is finding the correct connector to fit the lcd panel. my panel had the old ribbom cable attached, so i was lucky.

GWIZAH still has a lot of stuff on his web site, and i added the pin-outs from his site to my collection: NEC NL8060AC26-11.

i intended to use the i-o display cable (since it has the right connector for the mobo), the connector off the dead panel from one of my i-o's and solder wire to the connector on the ribbon that was with my tft panel. now that i have all this stuff out, i may go ahead and do that and flash a bios chip also.

turbosr:
the chips on my MSI mobo have the same numbers as the chips in the i-o and the same 4-in-one, sound and display drivers work.

keep having fun,


oldman
04-14-2002 15:18:41

New MessageRE:Actuall working TFT screen? (modified 1 times) Finatronics
Profile
KneuB_2,
Since this board is already dead, you don't have any reason to keep any of the chips on, right? The reason I ask is that a heat gun like Turbowsr mentioned is what I'd suggest. It would probably burn out the chip and any nearby, since you'd have to heat the backside of the board (or the top of the chip) for quite some time until the heat melts the solder between the board and chip. I've used this method quite a few times for desoldering standard SMT ICs, and it works well. The chips have so far ended up working perfectly afterwards, but I never had to heat them as much as you would to get the solder on the backside of the board to melt. It's quite likely that any surrounding chips, resistors, etc, would all come off, too. Frankly, if we could have a board with nothing on it to follow traces with, that would be wonderful! But, no sense in wasting so much time desoldering everything if you just want to trace a few pins. Maybe you could scan (highish resolution, ~300dpi...) both sides of the board after you desolder the chip? If you don't have a scanner, or whatnot, I'd be willing to spend a few days this summer desoldering all the chips and scanning the entire board high quality (I'd probably do up to 1200dpi, but save it in several different dpi settings for people who don't need the HUGE filesize). On the other hand, I can't guarentee that the heatgun method won't cause some damage to the board itself. I have caused the layers of some boards to seperate in a few places. As Turbowsr mentioned, they probably have some sort of adhesive holding the chip on. This will probably melt while heated, but might not. So, you might have to use a flathead screwdriver to pry the chip up. The stickiness will remain after it cools again, so don't put it directly on your scanner ;)

Eric

04-14-2002 18:39:04

New MessageRE:Actuall working TFT screen? (modified 0 times) GWIZAH
Profile | Email
Hey I fixed that link on my site. Silly tripod.

Whoa, Man what you guys are talking about is nuts :) Still, just a few years ago I thought the idea of cracking a machine open and adding resistors, Extra USB ports, etc. was crazy too. Good luck.

04-14-2002 20:53:42

New MessageRE:Actuall working TFT screen? (modified 0 times) Finatronics
Profile
I forgot to mention that heat guns aren't anything special. The one I use is actually a paint stripper we used to use for our boat. It works great and is quite handy to have around not only for desoldering nearly everything, but also for shrinking heatshrink tubing and removing sticky labels and tape from things.
For desoldering, I've used the heatgun to remove things with as many pins as zif (CPU) sockets and ISA slots from motherboards, which would be nearly impossible with a solder sucker and soldering iron (I've never tried solder wicks, does anyone have any experience with them? Can they be used for things with so many pins?). And I can usually unsolder anything this way in a matter of less than 30 sec. If you do it quick enough, you can even remove ICs (through-hole or SMT, doesn't matter) to be used for other projects. I have yet to burn one out, but I'm sure it's possible at those temperatures.

Good luck, KneuB_2! I can't wait to hear (and hopefully see...) what you discover!
Eric

04-14-2002 23:22:01

New MessageRE:Actuall working TFT screen? (modified 0 times) Programmer
Profile
There are no floppies in my house and there are 15+ computers, and 3 cd burners.. floppy (for me) is dead


Turbo3 has an IO mobo that I completely stripped (I took EVERYTHING off that puppy).. I used a heat gun that was intended for embossing from michaels craft supply.

I still haven't gotten around to finishing my cable..

The lines that are used for determining the lcd type should not have their resistors removed.. the resistor is there so that the chipset can latch whatever voltage is on them at power on, then reuse the lines as IO lines, though I highly doubt that they're actually using the lines for anything else..

04-15-2002 07:38:33

New MessageRE:Actuall working TFT screen? (modified 0 times) turbowsr
Profile
Can we get this thread back to getting a TFT to function?

Thank you

04-15-2002 15:15:56

New MessageRE:Actuall working TFT screen? (modified 0 times) Wild_Pencil
Profile | Email
When searching for compatible BIOSes.. keep in mind that the IOpener doesn't use the whole MVP4 chipset -- it only has the southbridge in common. The Cyberblade i7 is the northbridge, and also controls the display.. so if you manage to flash an MVP4 BIOS, with "Halt on No Errors" set, it *might* boot but it won't display anything useful 'cause the display isn't initialized.

In my own searches, most of systems using the i7 are using Phoenix BIOS, which is completely different from Award BIOS.

The Advantech BIOS is the closest cousin I've found, and is fully functional. It *CAN* boot the IOpener, but only after the CMOS is in a known-good state. It is also the other half of the Hybrid BIOS I've cobbled together to work around other compatibility issues. If you're looking for compatible BIOSes to experiment with and pull code from, this is the best bet.

-WP

04-15-2002 18:18:36

New MessageRE:Actuall working TFT screen? (modified 0 times) vwbug19
Profile
perhap try check out the bios from netier thin client? the netier has same chipsets of main and the video too
netier use via82c686a with trident cyberblade 7i
if you want to eamine the bios ,go to ftp://netier:netier@rebellands.net it's under bios
note that this bios has lan booting feature(pxe)
04-15-2002 18:59:14

New MessageRE:Actuall working TFT screen? (modified 0 times) oldman
Profile | Email
Wild_Pencil & vwbug19:

the Trident Cyberblade i7 northbridge chip is identical to the VIA VT86C8601, except that it may be a low power version intended for laptops and other power critical applications. the VIA 8601 actually is the northbridge/display controller on the MSI mobo in my desktop.

i think the VIA driver checks for some i.d. in the chip and the Trident driver does the same. the Trident driver for the i-o would not instal on my desktop, the but the VIA driver has all the same files as the Trident, with Trident written all over them. i believe the chips are interchangable. i think VIA makes the chip and puts the Trident markings on and both companies market them.

i still do not think the MSI bios has code for tft, but this expands what we can look for in a possible donor bios.

still having fun,


oldman
04-15-2002 20:06:47

New MessageRE:Actuall working TFT screen? (modified 0 times) turbowsr
Profile
Oldman,
The MVP4 north bridge is VT8501 for Socket 7, not VT8601 for Socket 370. Hears VIA's website with the facts. BUT I have found several laptops with PIII and Cel with the CyberBlade I7.

http://www.via.com.tw/en/apollo/MVP4.jsp

I found a very detailed spec of the CyberBlade i7 here:
http://help.nec-computers.com/uk/pri/item_spec_TridentCB_i7.asp

VT8501 DataSheet
http://www.via.com.tw/pdf/productinfo/501brief.pdf
VT82C601
http://www.via.com.tw/pdf/productinfo/dspm601brief.pdf

WP,
According to BOTH VIA and TRIDENT the MVP4 is desktop version and the CyberBlade I7 is the mobile version of the same chipset.

Doing a search for MVP4 mobo's:
Epox EP-MVP4 A/F/M/(M/J)===>2Mbit Award Bios
Gigabyte GA 5VMM===>2Mbit Award Bios
ECS and all its affiliates (PC Chips, Alton, Amptron, etc) only used Award Bios for Socket 7
Tekram P5M4-M+===>2Mb Award Bios
FIC PAG-2130===>2Mb Award Bios
Shuttle only uses Award Bios on Socket 7
AdvanTech PCA-6752F 2Mb Award Bios

CyberBlade Laptops
NEC Versa Ax (K6-2) Pheonix Bios (owns Award)
HP Pavilion N33xx, N32xx Unknown Bios
Compaq Presario 1246, 1247, 1400, 1200 Armada 100S Unknown Bios
Sony PCG-F610/630 Unknown Bios

04-15-2002 21:54:09

New MessageRE:Actuall working TFT screen? (modified 0 times) turbowsr
Profile
It looks like this will be a lot harder than thought. By reading posts, I was led to believe that hooking up a TFT to the pin Header for DTSN plus a bios able to recognise TFT would solve this problem. But reading the spec, there appears to be separate pins for DTSN and TFT. Since the Trident chip DOES need to be removed to figure out where these pins connect to the board, and at which point these pins are visible and can be soldered to in order to make a small daughter board (needing the required components) adaptor. Soo now we need high resolution pics of this. Programmer or Turbo3 has a stripped board :) hint hint. If we could get both sides. He is also on to something about those resistors. They appear only to be used on power up, then ignored. I don't believe that a Bios update will be needed. Would Netpliance (or whomever built these boards for them) actually spend the extra $$$ to pay programmers to remove something that can be ignored? I bet that the bios is fully functional, and may only need enhancments to recognize new products (ie large harddrives >8 MB). AND sometimes engineers leave things that marketing does not believe the consumer wants/needs (ie the CF slot).

1) We have the spec sheet.
2) There's enough talent here to figure out the pins
3) We NEED mobo pic's
4) design the adaptor board
5) We'll get TFT :)

Is anybody else really trying? Lets get this going.

04-16-2002 15:29:47

New MessageRE:Actuall working TFT screen? (modified 0 times) Finatronics
Profile
Are you suggesting that Netpliance had a pre-made BIOS to use and modified it according to their situation? I never thought about it that way. I don't think that's entirely true. My guess is that whoever made the BIOS (Netpliance, Award, whatever) had some sort of BIOS-Creation utility made by and for Award BIOS's. The way I imagine it, they have a program that asks them to select what hardware there is on the board, lets them set IO ports and addresses, and lets them choose settings that are appropriate for the board. I envision it much like compiling a Linux Kernel using MenuConfig. If they choose not to add support for a certain thing because it was never intended to be part of the computer/MoBo (such as the Floppy port, since that's what I'm learning the most about recently, or the TFT drivers), that wouldn't be "compiled" into the BIOS image, and therefore not in the BIOS at all. This seems to be the issue they've run into with the Floppy port... Not only was the option (to enable the drive and FDD controller) disabled permanently (originally), but apparently the actual floppy control routines aren't actually in the BIOS at all!

I've read a tiny bit of the CyberBlade Spec relating to the LCD, and I agree with you. The impression I got was that all you had to do was tell the chip that you have a TFT display installed and it would take care of the rest. The way this was done was with the address lines (I think it called them DRAM pins, but that didn't make any sense to me...). Logically, the computer should have no reason to NEED to know/control what type of display to use. The display is the display no matter what, or one would think. But, looking at the specs, it had different features available to different types of displays (CRTs, LCDs...). The OS/BIOS would NEED control over this in the case of a laptop, especially, which would have the possibility of using the built-in LCD, an external CRT, or both... switchable via a key-combination on the keyboard. (On my laptop you press Fn-F1 to switch display modes). However, it does seem silly to think that the OS/BIOS would need to have special control over what type of LCD or type of CRT is being used.

I don't know what to suggest about enabling a feature which isn't already available. Like I said before, I imagine they compile (assemble?) the BIOS specifically for the MoBo. If someone got hold of this utility, I'd bet we could add nearly anything we choose, assuming one could find the proper "drivers" or modify them appropriately. If this somewhat "modular" approach to the BIOS is true, there might be little blocks of code for different features which could be replaced and moved around as we pleased. I don't know much about that. I imagine this is what WildPencil means by his "Hybrid" BIOSs.

As WildPencil said before, legally speaking, we probably can't share these hybrid BIOSs among ourselves, since they contain code which was probably specifically licensed to that one particular motherboard that it was made for. On the other hand, software piracy is hardly even considered wrong by most people these days, it seems. And, this is really rather harmless (not like selling MS Office on the black market like some OEMs do, no matter how much MS deserves it). Just be warned, and be prepared to accept consequences should they happen.

Eric

04-16-2002 19:21:04

New MessageRE:Actuall working TFT screen? (modified 0 times) turbowsr
Profile
Wish OpenBios was further along. I really can't believe that an Open Source Bios isn't readily available. A bios with the powers of Open Source. WOW. Small, stable, User upgradable.
I don't know how the companies lisense their Code.
Thats one part of a PC I never really worried about, I booted up, it worked, it found my stuff, Windows started, then crash, then Windows restarted, then Fdisk, format, install Windows, crash, damnit, try linux. Just not Desktop ready, yet.

Looking at the spec. pg 7-20 JLVDS Pins 1-2=TFT 2-3=DTSN. These pins look as tho they come from the Sodimm. ???
[shrug]
Won't let that slow me down. Until I find out that there is no way to get this done, I won't quit until we all have TFT availablity.

04-16-2002 20:18:54

New MessageProgress - or lack there off (modified 0 times) KneuB_2
Profile | Email
Finatronics

Yes the board is dead. Before attempting to remove any of the chips I made a cursory review of the board. There were several broken traces by the keyboard connector as if it was used to catch the i-opener after a fall. As far as a reason to keep any of the chips on the board .... Ultimately no, but I do want to keep most of the little stuff (resistors and the like) on the board while I follow traces, they make good reference point.

Turbowsr

In wave soldering, is the "wave" a physical wave of molten solder or is the wave some sort of transmission wave that will heat a solder paste to the melting point without really heating the surrounding board or chips, sort of the way a microwave heats food without really heating the plate?

Also, is Ball Grid Array (BGA) a variant of the pin in a hole technology or a variant of the surface mount technology. What I really want to know is; Will I need to pry the chip up when the solder melts or will it simply fall off?

oldman

Can you e-mail me copies of those specs? please


Progress - or lack there of

I found a co-worker that owns a paint stripper heat gun he will let me use, but as yet I have not made the attempt.

At lunch today I tried to photo sections of the board (around the cyberblade chip) through a microscope. The microscope was a little to strong; on the lowest power it required four shots per each edge of the chip to circle the whole chip. But it was fascinating to be able to see every miniscule dust fragment down in the bottoms of those itty bitty via's, and reading the fine print on the smallest of resistors was not even a challenge. It's such a shame that when I dumped the CF card to the HD at the end of lunch, every last shot was black.

Lunch tomorrow I'll try to get the "before" photos one more time and then I'm going to apply the heat.

Free Icecream April 22nd

Free Icecream May 1st

Free Jamba Juice days

04-16-2002 23:10:43

New MessageRE:Actuall working TFT screen? (modified 0 times) BubbaDog56
Profile
KneuB_2,

Turbowsr is a little mistaken in his description of the assembly processes. The BGA is a surface mount part, it has small spheres/balls of solder on the bottom of the part pads and solder paste is placed on the board pads using a stencil and a screen printer. The part is then placed into the paste and ran through a reflow oven, a long conveyorized oven that can preheat the solder paste and flux and then rapidly heat it to the melting point. Wave solder machines are used for through-hole parts like connectors and such, and is actually a large heated pot of solder with a pump that generates a 'wave' of molten solder that the bottom of the board contacts while moving on a conveyor.

Glue is not normally used on surface mount parts on the top side of a circuit board, usually only small surface mount parts (resistors/capacitors) on the bottom side that can then be soldered with the wave solder instead of solder paste. This allows simultaneous soldering of the bottom side surface mount parts along with the connectors and such. If you have large surface mount parts on both sides (like ICs), you can actually use solder paste for both sides. You do the first side and run it through the reflow oven, then print the solder paste/place the parts on the second side and run it through the reflow oven again. Even though the solder melts again on the parts hanging upside down, the surface tension of the melted solder is strong enough to keep them from falling off.

You should be able to use the stripper gun to remove the BGA without too much trouble, it shouldn't be glued down. Keep the gun moving in a slow circular motion over the part so you don't burn the board, after about 30-45 seconds try to gently lift one corner of the part with a pair of tweezers. If the solder is melted it will lift up easily, don't force it and damage the pads underneath. If it won't lift, continue heating for another 30-45 seconds or as long as necessary. Lift the part up cleanly, if you push on it you'll smear solder over all the pads and make a real mess to clean up to see the actual layout.

B'Dog

04-17-2002 05:21:51

New MessageRE:Actuall working TFT screen? (modified 0 times) vwbug19
Profile
im adding new information about the bios compatiablity,the netier XL1000 thin client has the same bios chip as iopener yep it's sst 39sf020 and the neier bios image has cdrom booting support that iopener bios doesnt have ,i will let you know as i swap the bios to try booting it on iopener(iopener and netier is pretty the same hardware chipsets execpt for built in rtl8139b ethernet
turbowsr: as for openbios ,did you check out linuxbios project at www.linuxbios.org?
ive read and it looks promising but via chipset not supported yet
04-17-2002 05:59:30

New MessageRE:Actuall working TFT screen? (modified 0 times) turbowsr
Profile
BubbaDog 56,
I was just trying to give a quick rundown of how these are built. I built these often enough that I could have spent all day detailing the process, but thought that alot of info was not important. Now that I reread what I put, yes I incorrectly said they are put thru the wave solder machine. I put most of my time in manually soldering on parts.
If you've ever flown in a Boeing, McDonnel Douglas, Fokker, AirBus, Air Force One, Southwest Airlines, etc, my handy work is inside.
Anyhow,
KnueB_2,
BubbaDog is right BGA is surface mount. Looking at the I Opener, I see very few Thru Hole. Pins, some caps (like those near power plug), etc. Finatronics is correct. The same heat gun we used to heat up boards and remove parts is the same gun used to heat up a lot of things. These were bought out of Grainger and are not electronics specific. Just be careful and don't get in a hurry. I would prop up the board so that it sets along the edge. Heat the area, if you see the board browning, its getting too hot in that spot. Parts will start falling off. The middle ground plain should draw off a lot of the heat, this will actually help. If you plan on stripping all the parts off the board, use the same process. Let me know how it goes. I scanned a pic of the back side, its about 17 gig JPG, low compresion.
I looked over the specs last night. It seems that the mobo manufacture followed the specs closly. All the connectors labeled in the spec sheet match connectors on the mobo, even the switches match. I looks like the DTSN IS a LVDS connector and is also the TFT connector. From the Sodimm, pin 133 SCLK, determines DTSN or TFT. Page 7-20 Lower left corner, labeled JLVDS. This corrisponds with what KnueB_2 and WildFrog says about MA(5:3)((JLVDS?)) But this is for Power UP. Suggesting a 4.7Kohm resistor placed here tells the system what display is default.
Any suggestions here?
Hurry KneuB we need that pic :P
04-17-2002 14:51:02

New MessageGo To The Follow-On Thread (modified 0 times) KneuB_2
Profile | Email
THIS THREAD GOT TO LONG
SO TURBOWSR CONTINUED IT WITH THE THREAD
TFT SCREEN

PLEASE GO THERE

04-17-2002 22:12:30

Reply to Thread | Printer |
All times are PSTPowered by UltraBoard v1.62



Copyright © 2000, Netmake Inc. All Rights Reserved.
See Terms and Conditions for more information.




i-opener opener laptop notebook computer help drivers dll free windows dos repair fix linux mac macintosh 2000 95 98 nt pc configure hardware software sound video netscape explorer network networking lan wan software cmos fat bios printer card mouse modem ide scsi cd rom controllers scanner tape hard drive cgi scripts source code mp3