Just to let you know what lies ahead....
If you want to go from making a hand-built breadboard or prototype to actual PCB's, you have a lot of hours and anywhere from several hundred to a few thousand dollars cost in front of you, depending on how much you are willing to do yourself.
Schematic capture and PCB layout
First of all you need to capture your design using some sort of schematic capture program, and then design a PCB. One of the more popular programs is EAGLE, which I use. They have a EAGLE Light version ($49), but it can only be used for schematics with one sheet (any size), two signal layers, and 100x80mm (approx 4"x3") routing area. For any serious work, you need at least the EAGLE Standard version, which costs $747. There are probably other less costly (even free) alternatives. There are lots of others that cost thousands or tens of thousands of dollars. In any case you will have to spend considerable time learning how to use the program.
Or you can pay someone like me to do it for you ($$/hour).
PCB Fabrication
Getting boards made is the next step by a PCB fabricator. The problem here is the NRE (non-recurring engineering) costs. Some board houses treat this as a separate figure, and others built it into their per-board quote. In any case, it is almost never economical to have just a few boards made. You might spend $100 for two boards, and $500 for 25. You need to have really large quantities to get down to just a few dollars per board.
The gotcha is, if you make 25 boards, populate just a couple of them for testing and find they don't work (and there is not an easy fix -- e.g. because you laid out a connector backwards), you might end up throwing away the other 23 blank boards away and you would have been better off just getting two. I have stacks of blank PCB's as evidence of this phenomena.
PCB Assembly
Unless you are willing to build the boards by hand, you will need to have them assembled. Surface mount packages are difficult to deal with. If the board has BGA or QFN packages, you probably won't be able to build them yourself unless you have your own reflow oven.
Getting your first two boards built by an assembly house might cost $500. Whereas getting 25 built might cost $1200. (Once again, the problem here is the NRE costs.) Getting down to just a few dollars per board requires (again) large quantities.
And someone else has already discussed the problem of getting parts.
Make sure you use parts that are readily available -- if both DigiKey and Mouser have hundreds of the part available you should be okay. If instead, they have it in their catalog, but it is currently out-of-stock, try to find something else. If you need some special parts that aren't carried by DigiKey or Mouser, make sure you have a reliable source before incorporating it in your product. (Note: the more unusual parts you use, the more likely you will have to add the part manually to your PCB parts library.)
Custom Cases
Do you want to put your board into a case? If you need to have a custom case designed, that will be a couple thou for the designer using a program like SolidWorks (I don't do that, but can recommend someone who can). If you are going to make just a few cases to begin with, you will probably need to go with rapid prototyping, such as Selective Laser Sintering (SLS). Figure at least $100 per case in small quantities. To get down to a few dollars per case cost, you need to have a custom mold made. NRE time again! Plan on spending $10,000 or more for the mold.
And I won't even start on EMC or EMI testing, since I don't know if it applies to your product.
As you can see from all of this, until you get into production, the cost of the electronic parts is usually not the biggest item on a per board basis. Doing your own assembly for small volumes will save you a lot of money. So it is important to design with that in mind -- no impossible to solder-by-hand parts.
To get really low prices for high-volume, generally you need to go offshore -- China etc. But I would avoid doing so in the beginning.
The product has a maximum update rate of 512 Hz, which is really slow. I understand that it uses a timestamp, takes three readings, buffers for FAT filesystem translation, and writes it all in ASCII/CSV format, which sounds like a lot. However, an SD card configured for sustained writing should be able to do at least 2 megabytes per second, and class 10 cards can handle 10 megabytes per second. At 512 Hz, the problem is not the SD card.
In your comment, you mentioned:
The manufacturer knows all about the
problem, and is unable to fix it with
the available hardware resources (it's
only an 8051 that's breathing really
hard). All pauses seem to be due to a)
FAT overhead, and/or b) flash erase
cycles.
Adding a buffer at the SD card is not going to solve this problem. The bottleneck is up-stream of the SD card. The manufacturer's FAQ reads:
I have an old accelerometer, can I
upgrade the firmware?
Yes. Please contact our sales
department for instructions on how to
return the unit to GCDC. We will
upgrade the firmware and return the
device promptly.
Unless you've worked out with them that they'll write you a firmware package which doesn't do any buffering/FAT filesystem translation and mutually agreed upon a protocol for a faster, gap-free data stream, adding a RAM device that plugs into an SD slot will not solve your problem.
However, just because the task is bringing their 8051 to its knees doesn't mean that this is an un-solveable problem. A faster, more powerful, more modern microcontroller could handle the data stream. The on-board RAM on a decent micro should be enough to buffer the data. Most small micros will run at 20 MHz or more, which is probably plenty fast, especially since they'll have hardware peripherals that can handle the DMA, SPI, and ADC tasks for you. Of course, there are other people who have tried to solve the same problem - Take a look at this SO question about logging data at 96kB/s.
You have a few options:
- Dead-bug a higher-performance microcontroller onto the existing PCB, and write some firmware which will handle the task. If the 8051 isn't up to it, an ATmega probably is and a Cortex-M3 certainly is. You might consider ignoring the FAT filesystem and write a simple flat file, and just emulate a USB drive with the microcontroller.
- Have the original manufacturer do this for you. I don't see any mention of this problem on the website, so they could probably be motivated to fix the problem. They have all the schematics and drivers already written, so they could probably do this without too much trouble.
- Build your own. You can have as fast an uninterrupted logging rate as you want, with as much precision as you want, with as many DOF as you want - You just have to pay for what you want. The sky is the limit on this sort of project (and the budgets reflect this). You won't get this nice a package for this much money, but I'm not sure what your requirements are.
- Buy a different one. There are plenty of dataloggers that can handle faster serial streams without gaps, and a lot of accelerometers have as fast or faster update rates. The +/-250 g and +/-28 g range on this one is unusual, however.
Best Answer
I am not very sure what are your manufacturing powers, but here are some considerations:
You can use flexible PCBs. Produced on polyimide base they can be 0.20mm thick for two layers board and 0.13mm for one layer board.
You can implement the connection contacts as traces on this PCB. As long as the PCB is flexible you can use one side PCB and to bend it this way in order to provide some space for the parts:
Some very small parts have to be used, but considering your other questions on electronics.stackexchange.com you already have most of them.
BTW, if it is too expensive, or too hard to put everything in a SD card package, there is another solution, accessible even for an amateurs - put everything in small box that ends with SD card-like connector. This way, the thin part of the design will be only the connector - a piece of PCB with contact tracks: