Today I received a sample order from Texas Instruments. Among several ICs for a school project I also sampled some of the TMP006 “Infrared Thermopile Sensors”. I don’t have a PCB to solder them to yet, so I decided to take some macro photos.
I apologize for the quality. The top light in my microscope burned out and so all the pictures are light from the side with a handheld flashlight while trying to hold my camera steady and focused through the eyepiece. After this buying a nice microscope with a built in digital camera seems like a good idea.
The title picture is the top of the chip. It has the part number (TMP006), the alignment mark in the lower left, and what is probably the lot trace code. The interesting part is when the chip is flipped over. This is a chip-scale BGA package, which is convenient in terms of the amount of space it takes up, but almost impossible to solder by hand. To give a sense of scale the spacing between the center of each metal ball is 0.5mm.
I would have liked to get some higher quality pictures, but that will have to wait until I have a better setup .
Today I made a carrot cake, which can be a good change from electronics. If something goes wrong it is usually discovered right away, and most of the time it is fixable. This cake took about an hour to cook which seems like an instant compared to the time it takes to get prototype PCBs.
A lot of circuits can be prototyped on a breadboard and verified almost immediately which is great, but higher frequencies are desired this prototyping becomes more difficult. For a project I’m working on right now I needed to use a high speed serial ADC. On the breadboard I could get semi-reliable data up to around 30MHz, although the signal looked terrible. Pushing past this simply didn’t work, and there was no way I could reach the 48MHz needed to get the full 3MSPS out of the ADC.
Instead I had to lay out my board and hope for the best at the higher frequencies. The turn around on boards for me ends up being a couple of weeks to a month, the trade off being that I don’t have to pay through the nose a couple of small prototype boards. This of course leaves an annoyingly long amount of time being committed to one design, and it is impossible to start a second revision until the 2-4 weeks has passed and the first prototype has been built up and tested.
I feel like I have been lucky with my last batch of boards where everything turned out great. I just sent off a new set of boards including the project mentioned previously as well as my LPC1114 DIP board, hopefully everything will come back and build up perfectly around the middle of the month. I will post the final LPC1114 board then as well as an oscilloscope picture comparing the ADC serial signals on the breadboard and final PCB.