 |  |
| |
 Notching the Newt for the switch I used a hobby file to file out a notch that is the same size as the switch I was using. You can see the switch in another page. Fitment was basically trial and error. I just filed away until the switch was flush with both edges. Unfortunatly, this panel seats into slots on the other half and you have to make allowances for this. Like I said...trial and error! |
 Switch notch close-up Nere you can see a close-up of the notch I created. Note that this is a very, very small notch. You'll see why in the next picture why! the switch I am using is extremly small! |
 Switch Here we see an old Centenial SRAM card. This card was unique in that it had a rechargeable battery in it instead of a replaceble battery. For some reason, this card would no longer work, so I canabalized it for the switch (see where my stylus is pointing) The toughest part was destroying the case of the card to get to it's guts! I didn't want to ruin the switch in the process. As you can see, it is quite small and very delicate. |
 Zoom-In on the switch Here we can see the switch on a close-up. For those of you with a PDA Panache pen, you can tell the size by comparing the size of the switch to the penpoint! Removal of this switch was quite easy. As you can see, the contacts are out there in the opne and all it took was a hot soldering iron. I used my tried and true method of removing the solder...heat it up and bang board on the desk to jar the solder off. Clean and easy! |
 Switch and wire. Here we see the switch in it's place. You need to be careful with this switch. I super glued it to the case. This switch isn't sealed...this means that if you get super glue inside the switch, it will glue the switch in that position! That's why I later on decided to use hot-melt glue...this is handy in that it comes off cleanly if needed. I also used wire-wrap wire for the connection. The advantage is that it is pre-tinned and very small. I connected the wire to the 'common' part of the switch. The other end of this will go to one of the crystal pads (see the next page) |
 Switch wire and crystal pad. Here we see the other end of the wire. This is connected to the center or 'common' connection of the switch. This will allow the crystal connection to switch between either the left or the right connection on the switch. |
 Crystal rig There has been quite a bit of talk on the NewtonTalk list about keeping the leads short on crystal speed upgrades...With this in mind, I developed this. (see picture) The 5Mhz crystal is connected directly to the original 3.686Mhz crystal I pulled out of the Newton. The other end of that crystal is connected to one contact of the switch. The other lead of the 5Mhz crystal is connected to the other connection of the switch. What may be hard to see here is that I created a small connection that joins the two crystals to the open solder pad. This will become more obvious (hopefully) when you see it installed. What's important here is accurate measurement! I basically laid the crystals on the Newt and took sections of wire and ran them to where they would end up. I then cut the just a tad bit longer than needed (just in-case) and created a 'rig'. The advantage of this method is that it keeps the leads short and creates a clean, easy to install (and troubleshoot) assembly. |
 Crystal and Newton Hopefully this view will help. You can see the switch and the open pad on the Newton. If you mentally move the crystal assembly over the newton, you will see how it lines up. Once you place the crystal assembly, all you do is solder the connections! Be careful though on the wire connection between the crystals...since that's a short piece of wire, it can become unsoldered because of the heat. |
 Crystal assembly installed Here we can see the crystal assembly installed. It fits well because my measurements were accurate. You'll notice though that the wires had to be 'bowed' out somewhat. The reason for this is to allow for clearance on the back plastic. Seems that one of the door-clippy-thingy's sits right where the original crystal is placed (oops! didn't allow for that, did I?) Oh well, some judicious bending and placement puts the crystal out of the way of the clip. This seems to work OK. See the next image for a close-up. |
 Crystal image zoomed This is nothing more than the previous image zoomed. Hopefully it isn't too out of focus to help you see how this was placed. One of the things you will notice here (and probably one of the questions you are asking yourself) is that this design is able to use the existing wires to elevate the 3.686Mhz crystal off of the mother board to prevent any shorting. I tested this and pressed down on the crystal to check and there is no problem here! Of course, your placement may require some electrical tape, but it just happened in my case that was not necessary. The only critical piece here was the wire connecting to the other solderpad. It was almost too short. But as you can see, it seems to fit. |
 Back of Newton Here you can see that the door clip is clearly going to be a problem. Take a look at the next image... |
 Complete I've laid both halves side-by-side to show that now that I've moved the crystal, it will just miss the door clip. This was troublesome in that I couldn't figure out why when I tightened the screws down, the Newton would fail! Well, now it works flawlessly. All that is needed now is to reassemble the Newt and switch back and forth. I've switched many, many times and I have yet to have an occurance where the switching has caused any problems. I don't know, perhaps I manage to flip the switch precisely between electrons and it works, or perhaps I'm just lucky. If you decide to use my method, I would recommend that you spend plenty of time laying out the parts before you commit them to solder. Working in the tight confines of the newton is not to be done lightly! |