Under the Hood

This section shows the internals of the PD70x. The purpose is to show basic construction, and no effort has been made to reverse engineer or otherwise explain PD70x operation. It's for illustration purposes only. While this section might have been better placed towards the end of the review, I already have the shots, and I'm still working on getting more operational stats, so here it is...

Control/Memory Card Door Removal:

Removing the two screws indicated below allows the memory card door to be removed. This reveals the controls and PCB edges as well as some interesting things about the door assembly itself.

In the image below you can see there are two PCBs (printed circuit boards) in the PD70x. They appear to be of high quality and are epoxy verses cheap phenolic boards found in many consumer electronics. I forgot to hold them up to light or check for blind vias so I can't say if there are more that two layers (only PCB geeks care about this).

The two rails the door assembly screws into run the full length of the unit and hold the subassembly together. They are also used to attach the assembly to the aluminum channel via the side screws.

The metal hinge pin is clearly visible in the image below. Interestingly, it's only visible from the one side. The other side is closed. This might have been done to easily prove they are using metal verses plastic hinge pins.

The back of the door assembly shows the hinge pin retainer. Although it's supposed to be heat welded at the two arrows, mine actually fell out when I removed the disk drive bracket. I didn't bother to reseal it since the disk drive bracket holds it captive when installed.

Below you can see the cavity use to install the hinge pins. This answers a few questions I had about the method used to install the hinge pins and also shows there are two individual pins verses a long metal rod. I can't tell if they are splined or knurled but you can tell from the diameter that this is the static end of the hinge and the door hinge is the active end. ie the door hinge rotates on the pin. Yet another reason the door hinge plastic should be a bit longer in my opinion.

Here's a shot of the aluminum case without the guts. My unit had a small spec of dust on the inside of the LCD cover which I took this opportunity to remove.

Here you can see top PCB, LCD assembly, and battery compartment. Notice the two outer battery terminals. They are soldered to the bottom PCB and need to be compressed when removing it. The two center batter terminals are similar to those on the battery door and are considerably stiffer than the outer ones. This may contribute to the excessive tension that causes the battery door to flex.

The LCD is mounted to a bracket that raises it above the other PCB components. There is an LSI chip under the LCD, but I elected not to remove the subassembly due to the bracket thickness. I was afraid the threads might strip too easily.

Here's a closer view of the controls and LCD mounting bracket. You can just see the end of a chip under the LCD. You also get a better view of how the plastic rails are used to hold the assembly together.

The image below shows more control detail. Here you can see the control wheel is just a switch. Its contacts are connected to the circuit through 10k resistors which probably indicates it's directly connected to logic circuitry.

The power switch is your basic slide switch that is despised by many. However, properly used these switches have a very good service life. This one has the case soldered to the PCB which makes for a considerably stronger mechanical connection.

Notice the LCD ribbon cable disappearing under LCD assembly and the top ribbon cable going to the bottom PCB. The top cable seems to be hiding something. Hmm... 'gital SU--0X6-2' I wonder what that is?

Well I'll be. It's our old friends at SANHO Digital. As expected, they make the PD70x as well as the PD7x. I'm not really sure what affiliation they have with EastGear, but I think I should try and find out. I'll post information on this as I get it.

The literature states the PD70x has dual regulators. Well, from what I can see, if the two devices shown on the brochure are DC-DC regulators/converters, then the PD70x has four regulators. OK, perhaps some are interrelated. In any case the device below is a 1 amp low power-loss 5vdc regulator. I didn't bother to see where it is getting its power source, but I would assume since it is so close the external power connector, it's used to bring the power adapter 7vdc down to 5vdc for operation off of ac. I doubt it's used for charging since 5vdc is probably not enough voltage. However, since I have not studied charging circuits for NmHi batteries to any great degree, I'll just leave this up in the air. Most charging circuits I have worked with require a higher charging voltage than the target voltage.

If this device is ever called upon to deliver 1 amp for any length of time, I would expect it should have a heat sink. If it were rotated 180 degrees it could possibly use the case for a sink, but it's inserted so close to the PCB there is not room to allow for leads to bend. I guess you could add a small heat sink to the device itself, but since I don't know what this regulator does, I'm not even sure a heat sink is necessary. Like I said, I didn't follow traces to reverse engineer anything.

You cannot remove the bottom PCB without first removing the battery door assembly. This is easy enough to accomplish, but is not obvious a first glance. After removing the screws and taking a minute or two I found the tabs that held the unit together and was able to remove the door. I forgot to look underneath to see if it had metal door hinge pins. Obviously, the hinge itself is still plastic.

So far I should be within the guidelines of my warranty since EastGear plans to offer replacement door assemblies for the PD7x which would require the same level of disassembly I have performed so far. I have to be honest though. I would not want most users this far inside the PD70x. I feel comfortable with it since I'm inside this kind of stuff every day.

Once the ribbon cable is disconnected and the battery door is removed, there are just four screws that hold the bottom PCB on to the main assembly. As you can see, there's quite a bit going on here. The first thing you might notice is how shallow the compact flash socket really is.

And hey, the chips all look like they've had their product information erased. Yep, that's pretty common nowadays. I've seen this procedure performed since about 1980 or so. Originally they would use a ink eraser (white and coarse) to remove the data, but this escalated to sanding and grinding since the data wasn't always removed effectively.

Reverse engineering is a tried a true method of 'stealing' designs. It's totally legal in the U.S. given a user has no inside information regarding trade secrets etc. Compaq created an entire industry based on this premise when they reverse engineered the IBM BIOS and invented the first 'legal' IBM clone. For this reason is commonplace to obfuscate designs by whatever method available. However, even without correct chip ID one can browse Orcad, Protel, etc, device libraries and find matches pretty easily. Just count the number of pins, look for power pin locations, data busses, etc (yet more PCB geek stuff).

In the image below you can see the ribbon cable connector is the same as the one on the top PCB. Since it allows the ribbon cable to be removed, it means the ribbon cable can be replaced if damaged. Many designs have a ribbon cable that is terminated with a permanent PCB connector. While this is usually cheaper in large quantities, it requires some serious de-soldering if the ribbon cable is damaged.

Notice the empty solder pads marked 'C32 Options' (capacitor 32). I have no idea what this is but I would imagine the word 'options' should have been 'option'. Just a guess. In any case it's obvious they decided that capacitor 32 was not needed for one reason or another. Perhaps they left a door open just in case the power supply needed additional filtering or bypass.

Here are the two heavily touted regulators which could be more accurately described as DC-DC converters. But look at the larger black chip in the lower right hand corner. This is also a regulator. It appears to be a LM117 knock-off made by Bookly Micro with a fixed 3.3vdc output at 1amp max. Now we know where the 3.3vdc supply comes from.