Silicon Investor (SI) -- The First Internet Community

STOCKTALK

We've detected that you're using an ad content blocking browser plug-in or feature. Ads provide a critical source of revenue to the continued operation of Silicon Investor. We ask that you disable ad blocking while on Silicon Investor in the best interests of our community. If you are not using an ad blocker but are still receiving this message, make sure your browser's tracking protection is set to the 'standard' level.

Technology Stocks : Energy Conversion Devices -- Ignore unavailable to you. Want to Upgrade?

To: Krowbar who wrote (5190)	9/5/2000 12:15:24 AM
From: Krowbar	Respond to of 8393

I think that it is worth noting that WalMart now carries Energizer and Ray-O-Vac NiMH batteries, and I couldn't find any NiCads in either the store in Rockport or Irving Texas. The only other rechargeables were Alkaline. DEl

To: Krowbar who wrote (5190)	9/5/2000 2:06:15 AM
From: wily	Read Replies (1) \| Respond to of 8393

Del, here's my picture of dividing the smooth part of the I-R curve into 4 levels for 2 bits of storage: frontiernet.net What works best for me is to compensate for temperature at the time of writing to the cell. You want to program the cell to a specific resistance for each of the 4 levels, independent of what temperature the chip is at. Since resulting resistance is highly dependent on chip temperature, you have to find a way of reliably compensating for temperature fluctuations. The method I propose is to have a cell or a group of cells that are dedicated test or reference cells (actually you would want a cell or a group of cells for each programming level) Every second or so (chip temperature is not going to change THAT fast) you send a programming pulse to the test cell and see what the resulting resistance is. The first pulse you send will have a current right in the middle of the test range. Referring to the picture, for the #3level, your first pulse would be at 78.4uA. If the resulting resistance was, say, 250,000 Ohms, you would try a higher current, and the one you would pick would be at the high end of the range for level 3, or I2 = 80.6uA. Suppose this produces a resistance of 150,000 Ohms. This is too low, so you go halfway back in the other direction, and so forth, doing a binary-feedback search for the right current. I'm not sure how practical this would be from a manufacturing cost standpoint but it probably wouldn't be too bad, and it would give you the most precision possible in writing. This would in turn make the reads more reliable, and you possibly wouldn't have to temperature-compensate on the reads. wouldn't the whole chip be about the same temperature, or is that much variance from cell to cell? I don't know -- I offer the 16 segment idea just in case there is enough variation to make a difference. I think it's possible there is. If unwritten cells are at the same state, wouldn't reading their resistance give their temperatures? There's no such thing as an "unwritten" cell. In single-bit mode you have set and reset currents and the resulting resistances, and you can see from the charts that the resulting resistances are dependent on the temperature of the chip at the time of programming. In the multi-bit scheme, you have a multitude of different programming currents and a set of 4 resulting temperature-compensated (or pre-determined) resistances. The dynamic range for OUM is so large that it gives you a huge amount of margin between the 4 levels. As you can see from my picture, the gap between each level is 326%. In actual practice, the dynamic range is going to be smaller, because you have to go by the IR curve for the highest temperature in the specified operating range (say, 80C for commercial applications). But, still, the margin is very generous. 4 levels/2 bits is a gimme, IMO. wily

To: Krowbar who wrote (5190)	9/5/2000 9:43:48 AM
From: Futurist	Read Replies (2) \| Respond to of 8393

Can't convince GM: EV Demand very high in California September 5, 2000 Californians Show Potential Demand For Electric Cars at the Right Price By JEFFREY BALL Staff Reporter of THE WALL STREET JOURNAL As California environmental regulators prepare this week to decide whether to water down a requirement that auto makers start selling thousands of electric vehicles there in 2003, a new study predicts strong consumer demand for the environmentally friendly cars. The poll, conducted in July and designed and funded by California environmental groups, concludes consumers in the state would buy between 151,200 and 226,800 electric vehicles annually if they were "available at reasonable prices." That market would amount to between 12% and 18% of all new cars and light trucks sold each year in the state. The survey's organizers say the poll should persuade the California Air Resources Board that there is more than enough consumer demand to keep its current zero-emission-vehicle requirement in place, despite the opposition of the auto industry. The board will hold a public meeting Thursday to consider whether to stick with its requirement that, by 2003, 10% of all cars and light trucks sold or leased in the state by big manufacturers emit no pollution. That mandate would translate into about 22,000 battery-powered vehicles each year, plus, through a system of partial credits, additional high-tech cars and trucks that emit only small amounts of pollution. The poll is likely to carry weight with the California board, said Jerry Martin, a board spokesman. "No one else has done this type of analysis," he said, speculating that the auto industry may have avoided doing such a study itself because it was "afraid of what the answer would be." Auto makers so far have put only about 2,300 of the so-called zero-emission vehicles on the road in California, all of them during the past four years, and have been arguing that the 10% mandate should be withdrawn because there isn't a market in the state for anywhere near 22,000 electric vehicles a year. They dismissed the poll as wishful thinking. "That's a nice theoretical study, but we have actual experience in trying to [lease] these vehicles," said Samuel Leonard, director of mobile emissions and fuel economy for General Motors Corp., which has stopped building its EV1 electric car because of tepid consumer demand. Write to Jeffrey Ball at jeffrey.ball@wsj.com