Tekmos Talks
A Newsletter for the Semiconductor Industry
 February 2014
|
|
|
|
|
|
| 
Welcome to Tekmos Talks February. The new year is promising to keep us busy with our customers and ongoing innovation. This month's newsletter will talk about Flash Memory Emulation, introduce another Tekmos Executive, and an update on TK68020.
|
From the Desk of the President
|
 | Lynn Reed, President
|
Flash Memory EmulationIn addition to ASICs and microcontrollers, memories also become obsolete. One of our projects is a replacement for an obsolete Intel 28F010 flash memory. This is a 1 MB, 5 volt, flash memory, and does not have any sectors. Unlike our other products which use standard processes, flash memories use a dedicated process. And once that memory has become obsolete, the wafer fab is usually converted over to a newer process so that it can make newer flash memories. This causes customer problems since the newer flash memories use lower voltages, and frequently have different programming algorithms. As a result, the customer is faced with a system redesign because the older parts are not available. We can't recreate the old process. What we can do is to use one of our ASICs as an interface between the old application and a newer flash memory. We then buy flash die, and assemble both the flash memory and our ASIC in a single package that now works in the old application. This approach has several interesting engineering issues. First, we have to design a voltage regulator so that the external 5 volt supply is stepped down to the 3.3 volts required by the memory. And it has to have a fast response to changes in load current, because when a flash switches to write mode, the current can increase from 10ua to 50 mA in a few microseconds. Another problem is the write algorithm. Flash memories have a write algorithm that is used to prevent accidental writes to the memory. For this chip, we have to detect when the write algorithm occurs, and replace it with the write algorithm required by our chip. And since there is no clock available, we have to turn on a ring oscillator and make our own clock. A third problem involves the use of high voltages. The Intel part uses ~13 volts to indicate the programming and to enable a special device mode to read the manufacturer's ID. Our part will self-destruct if a voltage greater than 7 volts is applied to it. To get around this, we have to put an internal resistor between the VPP pin and our internal detect circuit that will limit the voltage that the part actually sees. A fourth problem occurs when using our device in some programmers. They will immediately remove the power from our part after the write is finished. Unfortunately, we require a little time to finish our own write algorithm, and so we will fail in some programmers. I don't think that we will be able to solve that problem, and so we will have to provide our own programmers, or we will have to ask the various programmer companies to alter their algorithms for us. And that is how we make Flash memory replacements.
For more information on inventory or orders, please contact Bob Abrams, Vice President of Sales & Development or call 512-342-9871
|
|
|
|
Meet Tekmos
|
 Jon Gehm, Director of Operations How long have you been at Tekmos? I am happy to say that I have been with Tekmos for 3 years and am looking forward to many more. What are your responsibilities here at Tekmos? While at Tekmos, we all wear multiple "hats", I focus on operations and to ensure the quality system continues to improve and respond to customer needs. What are your favorite tasks or projects to work on here at Tekmos? My personal favorite tasks at Tekmos, oddly enough, are following through with corrective actions. I really enjoy the process of finding a problem, analyzing it, and watching the system improve through the corrective process. What project are you working on now? We are currently going through a review of the entire Quality Management System document structure, to ensure that it is as current and efficient as possible. On the operations side, we are continuing to improve the test floor capability and streamline the production flow. Can you offer any suggestions to Tekmos clients to make their experience more efficient and successful? We strive to have all customer experiences be as successful as possible. However, the customer relationship can always be enhanced. I believe that improvement starts with the simple act of communication. A clear understanding of the customer needs may begin, but does not end, with the initial email or phone call. The customer relationship is ongoing and dynamic. The better the communication, the better the chance Tekmos has to achieve a successful experience for our customer. I would encourage all customers to keep those lines of communication open. If there is ever a question or a concern that is not being addressed, bring it to our attention. That is what we are here for. Generally, the earlier we know of a concern, the easier it is to address. Please tell us about your professional and educational background. It is said that your past always seems to catch up to you. I believe that the saying was meant to have a bit of a sinister connotation, but in this case, things seem to be working out for the better. I have degrees from the University of Texas, as well as a graduate of the Ranch Management program at TCU. Originally finding my start in agriculture, fifteen years managing farming operations, through multiple counties, has served me well in helping juggle the complex production and test flow challenges Tekmos offers. My persnickety side was developed through my experience as a managing principal in investment banking operations, as well as taking an interest in and overseeing mortgage banking opportunities. Attention to detail is crucial when dealing with financial concerns. Luckily, this compulsion has found a new challenge in developing the quality management systems of Tekmos. In addition, in an effort to further my understanding of quality theory, I was able to fulfill the requirements to become an American Society for Quality (ASQ) Certified Quality Engineer in 2013. While traditional education and on the job experience is invaluable, I find that the experience of extensive travel also rewarding. I was fortunate in having visited over twenty countries prior to joining the Tekmos team. That experience of developing relationships with people of different countries and different cultures has positioned me well for the challenges of overseas business development. The Tekmos business and customer base is becoming much more global in nature. With major projects underway with companies in Asia and parts of Europe, the ability to breach differences in communication style and quickly resolve misunderstandings is a value our customers appreciate. I look forward to bringing my past to our customer's present, in helping to develop long-term, strong and rewarding opportunities, with every project we win and every part we make.
|
Tekmos TK68020
| 
TK68020 Update
It has been a while since I have talked about our 68020 development so it is time for an update.
The basic design flow in this project was to extract a transistor level netlist from the layout, convert it into standard, synthesizable logic, verify it in an FPGA, and then make it. We are now in layout, though like everything else in this project, layout has also been a difficult task.
Creating a hierarchical transistor level netlist from the layout started out tough, until we developed some procedures that greatly simplified the task. Next we developed software that identified the basic gates and replaced the corresponding transistors. The complex logic gates were next, followed by the dynamic flops and latches.
However, the design didn't simulate in Verilog. So we used whole chip Spice simulations to debug the Verilog. This pointed to issues in our interpretation of the clock drivers, and then the chip began to work. Eventually, we had the chip passing all 51 Freescale test vectors.
But, and this is a big but, the netlist contained a lot of dynamic logic. So our next step was to convert all of the dynamic logic into static logic. This was a difficult task. In particular, the data buses were frequently used as dynamic latches, and proved to be difficult to convert into normal logic. We also had a lot of problems with dynamic multiplexors, which was a type of latch with 3 to 4 inputs and the same number of enables. Eventually, these were resolved, and the netlist was released to layout.
And this is where the next set of problems occurred. The design has a lot of 32 bit busses; at least 10 that I can think of. The buses are driven by a small tri-state driver that occupies one block, and has 4 signals going to it. This is a high density of signals, and can cause routing problems. This is aggravated by the tendency of the autorouter to place all of the tri-state drivers next to each other, producing an unroutable ball of logic.
So we have been doing a manual placement of the data path. There are a lot of cells, and so we have been placing the data path for a month now. When that finishes, we will auto-place and route the rest of the circuit, tape out, and have prototypes 8 weeks later.
It has been a long project (2 years), and we are looking forward to the end.
For more information on manufacturing, contact our main office 512-342-9871
|
|
|
We appreciate our customers. Tekmos is meeting innovation every day for tomorrow's challenges.
Sincerely,
Lynn Reed, President
4120 Commercial Center Drive, Suite 400Austin, TX 78744Phone: (512) 342-9871Fax: (512) 342-9873
|
|
|
|
|
