This is a document I have written for a customer, that explains, how plain text files can be shared easily between the HP Palmtops and a Windows system, without losing special characters in the text:
Here is a document written by me, which describes a self-made car charger for the HP Palmtops:
See also: Tutorial: Disassembly of a 200LX
This article has been written by Carlos Izzo Videla, Buenos Aires, Argentinia. He asked me to host this valuable piece of information here on www.hermocom.com.
I have never met or heard of an owner of a HP200LX who does not have a broken latch problem. I have only seen an original working latch once and it was in a practically unused HP200LX.
The 'latch problem' seems to be, along with the omnipresent and hideous hinge crack and broken battery covers, the most common problem encountered (sooner or later) by all HPLXers. I am, of course, no exception to these (only the last two, fortunately) problems which I personally find difficult to forgive HP's engineers for. Much more so if we take into account the amount of money one of these little marvels went for when they first came into the handheld computer market.
There are a quite a few fixes and/or solutions to the broken latch problem out there and this is just one more of them, just another idea. It takes a bit more time, requires some knowledge/experience in taking apart the HP200LX, manual dexterity and a certain knack for fixing things in general. It involves more risks but, in my opinion, the end result justifies all the trouble.
If you are reading this, chances are that you have already read the previous page. If not, then please do read it, particularly this same section.
I won't go into the details of how to dismantle a HP200LX to fix the latch as it has been thoroughly documented by two very experienced LXers: Daniel Hertrich has done it here and David Sargeant has done it here.
Please do take the time to read what they have to say about taking apart, hacking and repairing a HP200LX, generously sharing their experience with the rest of us.
When I got my HP200LX, one of the the first things I did (I'm a 'it's not yours until you take it apart person') was take it apart and this put me in contact (among other interesting things) with the latch problem. At first I did not see it as a big issue, but in time and with everyday use I eventually came to the conclusion that it was not a good thing to have the latch in that condition with the screen 'open', so to speak.
Or maybe it was just the itch to do something more with my HP200LX ....
So I started looking for a way to get the latch in working condition again. Unfortunately, none of the solutions I found (silicone blob, rubber band, foam, etc.) did the job very well. The latch never closed properly and thus the screen was not secured in it's closed position.
I really hate to say this, but after taking it apart and seeing how the latch was 'supposed' to work, I concluded that HP's design for this latch is really not a good one, to say the least. Certainly not what one is (was?) used to expect from a company such as HP and in something as expensive as a HP200LX.
The 'spring back' force the latch relies upon to stay closed is actually provided by the same thin plastic rods that form the latch's hinge (a horizontal pivot) through a couple of 90º bends at their ends which are held in place and kept from turning along with the pivot when the latch is pushed on.
So every time you push on the latch to open the HP200LX, the thin plastic rod that serves as a hinge is subjected to torsion (certainly not for this type of material) with the end result being that when the plastic reaches it's fatigue limit and breaks, you not only lose the 'spring back' force needed to keep the latch closed but you also lose the hinge the latch pivots on. If I didn't know better, I'd say that it was designed to break after some use.
Which is exactly what it does. ;-/
The problem is then twofold. In order to get the latch working properly again, two things have to be recovered: the latch's 'spring back' force and the hinge the latch pivots on. I think it is the lack of this hinge that keeps the other solutions available from working properly.
Close inspection of a broken latch and it's lodging will show all that I've described above plus one thing I had not noticed before: a small cone shaped protuberance on the surface of the underside of the latch.
Being most probably the injection point for the mold used to make the latches, it is (by design) right in the centre of the latch and thus in a perfect place to secure a coiled spring in order to give the latch the force needed to remain locked. Curiously enough, when the latch is in place, this little cone shaped protuberance is situated right opposite the centre of a square flat piece of plastic which is part of the upper side of the screen bezel and serves as a perfect point of support for the spring to work against.
Couldn't be better.
Here's a shot showing what it would look like once in place:
How it ends up will obviously depend on the type of spring you use. I used one of the battery connector springs from the mainboard of a dead Palm III and cut it progressively shorter till I found the right combination of force and radius. You need for it to exert just enough pressure but you don't want it to be so large that it shows outside the bezel or keeps the lid from closing.
I'd say that most any old electronic appliance using 1.5v AAA batteries can provide you with an adequate spring that, being properly designed for the job, won't break after just 250 cycles. 8-/
Depending on the spring used, you can glue it on to this small 'cone' or try to have it stay in place by means of it's own pressure. I put a small drop of a cyanoacrylate based adhesive (Super Glue, etc.) to help it stay put, especially while I was assembling the lid.
Now, the spring by itself will not solve the latch problem. The lack of a hinge on the latch's horizontal axis for it to pivot on does not allow the catch to hold in place as the whole latch is pushed upwards and not just the part with the catch.
To fix this (actually the trickiest part of this screed) you have to practise some surgery on the latch itself. It is the trickiest part because it is not possible to use just one single (and very easy to align between just two points) piece of plastic or wire to rebuild the hinge: it just won't fit in place due to the existence of two mounting tubes which are used to fix the screen bezel in place.
All these descriptions I am induging in may seem a bit confusing if you haven't taken apart a HP200LX's lid and screen bezel, but with the parts in front of you it all makes sense right away. A look at David and Daniel's thoroughly documented instructions will quickly dissipate any doubts you may have.
The thing is that you have to manually align two separate pieces of plastic or wire along the same axis and fix them in place by melting the plastic using heat or with resin, epoxy or whatever else you can come up with.
I reconstructed the pivot on the latch using a piece of steel wire I had left over from when I made 'Z' type springs to hold down coolers on top of Pentiums with broken ZIF sockets and a combination of heat and resin to glue them in place.
If you are lucky, the end result will look much better than the frankenhatch I was able to come up with. It's damn ugly, but it goes unnoticed once in place and most important, it works quite well.
This is a close up of one of the ends:
Here's a view of the finished frankenhatch. Notice that the wires are not exactly aligned along the same axis (the ideal situation). In this case, it does not seem to affect the way the latch works, probably because they're off by just a bit. As always, ymmv.
Once in place, this is what the fixed latch looks like from the outside. Notice that it sits flush with the rest of the cover/case whether it is open or closed:
Here you can see the spring once the latch is in place:
That's all ... ;-D
Copyright © 2008-2010 Carlos Izzo Videla. All rights reserved.
Page now hosted by www.hermocom.com
See also: Tutorial: Disassembly of a 200LX
This article has been written by Carlos Izzo Videla, Buenos Aires, Argentinia. He asked me to host this valuable piece of information here on www.hermocom.com.
After going round in circles for a while, I finally found a viable and inexpensive alternative to the different commercial memory upgrades for the HP200LX that are available on the web. In a relatively short time and with a limited investment, I was able to add 4Mb to my stock (SG-June '86) 2Mb palmtop by "piggybacking" two 2Mb memory chips atop the on-board memory chip.
I do not advise anyone doing this nor do I endorse this procedure in any way. If you are not naturally curious and don't have a strong knack for and experience in opening up things even if only to see what's inside and try to learn how they work, please heed my advice: do yourself a favour and stop reading this right now. In carrying out this procedure, you'd be opening up and practising major surgery on your precious HP200LX palmtop in a way that would absolutely void the warranty on any PDA, HPC, electronic or electrical appliance and running a very high risk of causing irreparable damage to it.
All the information I have put forth in this screed is freely available on the web: by NO means and under NO circumstances will I accept ANY responsibility for ANY damage to anything or anyone that carrying out this modification to your HP200LX could eventually cause.
I've been naturally prone to folly and risk from a very early age as well as overly confident in my manual dexterity. This is probably the reason why I saw no impediment in exploring the possibilities that this 'do it yourself' memory upgrade could open for me and thus decided to go ahead with it. It's also quite possible that the Gods were smiling upon me that particular day or that it has been a fluke of sorts that I didn't toast my HP200LX.
It is at this point that I'll insist once again: there is more than one purveyor of memory upgrades for the HP200LX on the web and I'm quite sure sure any one of them can do a good job in a cost effective manner and without any risk to your HP200LX palmtop.
In short: if not for the Gods' fickle humour or a fluke, you are on your very own.
When I purchased my 2Mb HP200LX, the first thing I did (after taking it apart, of course) was to search the web for the different hardware hacks that could eventually be applied to it. A memory upgrade and a DS modification seemed to be the most important ones, but my unit's configuration had only one 2Mb chip, no empty places to fill on the motherboard, no expansion board, not even the memory expansion connector. So after installing a 32Mhz crystal, I went on to look for a viable memory upgrade solution.
This post (broken link, will be updated once the HPLX-L archives are available again) to the HPLX list by Michel Bel (04/03/2002) and subsequent follow-ups by others gave me an idea for expanding the memory of my HP200LX. The critical information I needed to make this memory expansion idea a reality arrived in the form of a very interesting and well written page posted here (broken link!) by a fellow by the name of Rudy Moore to whom I would like to express my gratitude.
Thank you for sharing your work, Rudy!
Please note that I consider Moore's article to be mandatory reading for anyone who is considering running the risk of going ahead and attempting this memory upgrade.
As you have surely grasped, Moore's careful and well documented approach does not use the "piggyback" method I am using here. IMHO, this one is more convenient (albeit more technically challenging) as it allows for the use of two chips without the resulting wire clutter that would otherwise be generated in the very limited space available under the HP200LX's motherboard.
Nevertheless, it does have a couple of drawbacks:
1. You have to get your hands on a specific chip type (1Mx16 1K refresh 3.3v EDO DRAM) and it's not too easy to find. You either pull it from unused older equipment or purchase it from whoever has it in stock. I have seen them with asking prices ranging from US$15.00 to US$25.00 apiece (much too dear, I think). In this case, I used a pair of Toshiba TC51V18160AFTS-70 chips that I unsoldered from a T1910's 8Mb memory card. This chip is, AFIK, identical to / compatible with this Micron chip which in turn is identical to / compatible with the on-board chip my HP200LX has.
2. The chips have to be soldered directly on top of the existing one (hence this technique's name), which is not an easy task to carry out. A stable pulse and sharp eyesight as well as very good soldering skills are a must.
3. One memory chip on top of another and so on could, in theory, impede proper dissipation and acumulate heat. I don't think that a DS HP200LX could cause the three piggybacked chips to overheat, but I really don't know for sure.
I won't make any of the obvious recommendations (ESD protection, back-ups, etc.) for I think that anyone who is up to doing this you should know what to do. I won't delve into the gory details of how to dismantle a HP200LX either: it has already been thoroughly documented by two very experienced LXers: Daniel Hertrich has done it here and David Sargeant has done it here.
With Moore's general procedure and instructions as a guide plus the Micron data sheet (thank you for the tip, Mack Baggette!!!) for the chips I had, I set out to map out and verify the on-board chip-to-memory connector pinout. Once I checked and double checked that the pinout was correctly mapped out, I drew up the wiring diagram shown below:
I have used electronic soldering equipment a myriad of times but never acquired the sort of soldering skills this job requires. The use of the right soldering wire, paste, wick, proper size tips and temperature are of prime importance so if you lack the specific knowledge involved or your pulse and eyesight are as challenged as mine, please think it over and have someone fit and knowledgeable perform this task for you. I may be prone to folly and risk but not dumb: off I went to see my good friend César Viglianco at his downtown electronics lab. He once performed an AMD5x86 SMD chip transplant into an ageing 486DX laptop for me and this job was a walk in the park for him.
Once you've dismantled the palmtop and carefully taken out the motherboard, turn it over and secure it adequately on a stable working surface. Here's a photo of the motherboard showing the on-board memory chip and the memory expansion connector pads:
The memory chips used must be carefully examined with a loupe to make sure that there's no extra plastic along the borders (very carefully shave them off with a sharp x-acto type blade) that could keep them from fitting perfectly in place and in full contact with the one below. After that, all the chip's pins (with the sole exception of the RAS# pin) must be bent downwards so that once it is in place they come into contact (even if very lightly) with the pins of the chip below it. Work slowly and be very careful when you do this !!!
If all this is done properly, once the soldering is finished the electrical continuity between the chips will have been assured, the limited space available underneath the motherboard won't be compromised and you'll have no problems when it's time to assemble the palmtop again. Like in any similar situation, be sure to check and double check everything before deciding that it's OK to go ahead and use the soldering pencil on your palmtop's motherboard. Patience and Prudence will be your very best girlfriends when doing this. If necessary, stop, go have a beer with them and return to the worktable afterwards.
The critical part of this upgrade job comprises a two step process which basically involves stacking two additional memory chips on top of the existing on-board memory chip, wiring up their RAS# pins to the RAS1_ and RAS2_ pads on the motherboard, closing up and then checking to see that everything works properly.
In the first step, one of the memory chips is soldered on top of the existing on-board chip and it's RAS# pin is wired to the RAS1_ pad on the motherboard.
Here's a photo showing the result:
Closing up the HP200LX, checking to see that everything works properly and that the system reports an additional 2Mb clears the way to the next step.
In the second step, the other memory chip is soldered on top of the first one and it's RAS# pin is wired to the RAS2_ pad on the motherboard.
Here's a photo showing the result:
Here's another photo (not too good, I'm afraid) showing the finished job from another angle: two 2Mb chips soldered on top of the original on-board memory and with their RAS# pins wired to the corresponding RAS1_ and RAS2_ pads on the motherboard.
The memory upgrade is finished when you close up the HP200LX, check to see that everything works properly and make sure that the system reports a total of 6Mb available memory, as shown in this screen capture below:
That's all there is to it ... ;-D
Rudy Moore's site.
David Sargeant's page.
Micron data sheet.
Copyright © 2008-2010 Carlos Izzo Videla. All rights reserved.
Page now hosted by www.hermocom.com
