μblog: engineering from the trenches

The AD639 is a general trigonometric function generator where the output is a scaled ratio of two sines, where each sine has tunable parameters. The package on this one was another ceramic case with a gold cover that was held on by solder. Heating the case and pushing over the solder revealed a very clean chip with a seemingly small number of transistors. No markings were found on the chip other than the model number and Analog’s logo.

( ad639.pdf )

del.icio.us | digg

Diesel Sweeties by R. Stevens.

xkcd by Randall Munroe.

Toothpaste for dinner by  Drew.

If you have others that are both funny and safe for work, please post links in the comments.

del.icio.us | digg

I am coming to the realization that the PLL is one of the most under-appreciated components used in almost all aspects of electrical engineering. In many PLL applications, such as clock generation, the ratio of the output frequency to driving frequency is below ten, but a well designed PLL can push this ratio to several orders of magnitude. Finally, very special regard must be held for designers of very low frequency PLLs since keeping track of things for tens of milliseconds or more is very difficult in the analog domain. This is not to discredit designers of very fast (>500MHz) PLLs where the high frequencies tend to bring out non-ideal effects. First, here is a guide from Freescale Semi. that outlines the design of PLLs through a control theory approach based in the s-domain. Next, here is an application note written by Bob Pease in 1979 that describes some non-standard uses for PLLs as well as low (~1-10kHz) frequency design. Finally, here is report from MOSIS outlining a physical implementation of a given PLL design. (MOSIS tends to publish the projects that it funds internally for universities.)

( an535.pdf ) ( an-210.pdf ) ( digital_pll_report_1.pdf )

del.icio.us | digg

Only days after the European PLAYSTATION3 launch and people are already taking their systems apart. Unfortunately, I didn’t find any easter eggs on the PCB. The biggest problem should be in the absence of a hardware PlayStation2 emulator resulting in some controversy. I am not sure if it is apparent from these pictures that the chip is missing, so I will try to post pictures of the U.S. or Japanese mainboards if I can find them.

( Via Engadget )

ps3_motherboard_1.JPG ps3_motherboard_2.JPG ps3_motherboard_3.JPG

ps3_motherboard_4.JPG ps3_motherboard_5.JPG ps3_motherboard_6.JPG

ps3_motherboard_7.JPG ps3_motherboard_8.JPG ps3_motherboard_9.JPG

ps3_motherboard_10.JPG ps3_motherboard_11.JPG ps3_motherboard_12.JPG

ps3_motherboard_13.JPG ps3_motherboard_14.JPG ps3_motherboard_15.JPG

ps3_motherboard_16.JPG ps3_motherboard_17.JPG ps3_motherboard_18.JPG

[WordPress didn't thumbnail these correctly, sorry!]

del.icio.us | digg

A couple times a year, I have to reconfigure my Linksys wireless game adapter. Every time I figure that the setup utility is online and realize that it is nowhere to be found. Consequently, I spend about an hour every time trying to find my original install CD and cursing Linksys. I am guessing that some other people might be going through the same process a couple times a year, so for you, here is the setup utility.

( wga54g-setup.zip )

del.icio.us | digg

Here we have a 4K SRAM from Cypress. Looks like the original design is from 1985, however, the date of manufacture is unknown. This chip was in a skinny DIP package but it was ceramic instead of plastic. I carefully opened the package using a chisel and hammer and was surprised that the die had a bubble of soft, clear material over it. It might be due to different mechanical properties of the ceramic case and the silicon. Regardless of the reason, it made it much easier to recover the intact die.

(CY7C172-35PC datasheet still missing in action.)

del.icio.us | digg

Bob Pease is regarded as Mr. Analog by many people and he continues to explain misunderstood phenomena in a wide range of applications.  The first edition of this note on capacitor soakage was published in 1982 in EDN, the year that I was born, and is still one of the few decent articles I can find covering the subject. This should prove useful to designers of sample-and-hold and integrator circuits along with those who are interested in a deeper knowledge of capacitors.

del.icio.us | digg

Since the last post about the rotovaporator on a budget, we have made some improvements.  First, we added a transmission to lower the power delivered to the motor and avoid the need for any kind of cooling. Second, we added a higher stand to allow us to sonicate the flask while rotating it.

P.S. We also have a DIY centrofuge.

del.icio.us | digg

In case you need to create a monolayer of lipids inside a spherical flask and don’t to spend $3000 on a commercial rotary evaporator, this is the design you need. Justin Ingram and I set out to build such a device from things found around the lab for a small fraction of the cost.

The first step was to create a commutator that could support the flask and allow for the evacuation of the solvent vapors from the flask. We used a piece of Teflon coated stainless steel that was held in a piece of Delrin with a pair of Delrin spacers. We also used a planar ball bearing to lessen the friction between the stainless steel and the bushings. A piece of a 5ml syringe was attached to the other end of the tube and was then was covered in parafin. This piece was just the right size so as to fit into the opening of the spherical flask and be able to support its weight.

The next step was to provide the rotary action. We found a DC motor around the lab which was connected to a standard variable power supply. We transferred the force from the motor to the flask by a belt system implemented using a standard rubber band. Although the motor drew almost 1.5A while in operation, it did not get too hot over a period of half an hour. We are working on a thermocouple to allow us to attach a peltier device with a  heat sink or just a heat sink to allow for more efficient cooling.

The main purpose of the vacuum system was to collect the solvent vapors, not to reduce the pressure inside the flask. For this reason, we were able to loosely connect a pair of luer-lok adapters together to allow for them to rotate and did not worry about the poor seal. We turned on the vacume system and verified that the vacuum was good enough to take in smoke from the surrounding area and assumed that it was sufficient. The vapors that we are working with are that harmful, but it is also possible to run this system inside of a fume hood. The commercial system also has a water-cooled condenser for collecting the solvent, but it is snowing outside, so we can put something out there to act as a low-cost condenser.

( evap-action.AVI )

del.icio.us | digg

Something to do on the day after St. Patrick’s day? Look at application notes. By way of looking for some video encoders to generate all of the timing signals I came across an app note for increasing the longevity of CMOS devices and another regarding appropriate selection and use of passive components. They are both a little dated but still have applicable information to ever-shrinking devices.

( an-348.pdf ) ( an-349.pdf )

del.icio.us | digg