The recession has not hit that hard that we are rid of the traffic jams that plague us. And although we make progress in getting more efficient cars, they still use a lot of gas. Public transport, although I use it, still is, as someone once called it, a 19th century solution for a 21st century problem. It leaves from a place where your are not, before or after you get there and arrives at a place where you don't quite have to be, too early or too late. Not just that, but the seats are not as clean as you might like, there is no privacy etc.
The next idea is an idea to get you out of the jam in an energy efficient way that has none of the disadvantages of public transport.
Segway is developing the Puma This is a very cool vehicle that let's you 'drive' fully on electricity. It is much smaller than a car, but does keep you dry. The small size makes it perfect for the city. But there is a big disadvantage. For any distance over 20 km or roughly 12 miles it can't do that without charging and it is also a bit slow for long distances.
My idea is to combine this with technology we know from ski lifts. Many of the ski resorts have detachable chairlifts. Very comfortably, when you arrive, the chairs move at a slow pace. But when you're settled, it is dropped on the cable and you move at a much faster paste.
Image if if we build cable systems that do roughly the same, but pick up these Puma's rather than detachable chairs. The Puma would have to be equipped with a connector, so that you drive in a small lane where the connector is picked up, just like the ski lifts do. Then you're dropped on the cable and travel a large distance to for instance the inner city.
Take for instance New York. What if you live across the Hudson. You could use your Puma to drive to a cable station. You drive in and get picked up and dumped on the cable. The cable then transports you at high speed straight over the Hudson (no need for bridges!) to a station in the city. There you are released and travel the last bit using the Puma independently.
Advantages
- low power usage and no CO2
- In the city you can move much faster due to the small size of the Puma
- much cheaper to park
- leaves when you are ready
- leave from home
- arrive directly at work
- no standing in a traffic jam
- no polluting of the cities air.
This cannot replace all car traffic, but for commuters, it would get them to work faster, cleaner and cheaper and more comfortable then by train. If this is done by enough commuters, it will also relieve the traffice for others that need bigger vehicles (i.e. trucks, vans etc.)
Wednesday, April 28, 2010
Tuesday, March 24, 2009
Working on the road
They take good care of roads in Europe. If you've ever driven through Germany you know that the roads are in great condition. If you've ever driven through Germany you also know that it seems that they are always working on the roads. There is no way to drive through that country without passing large stretches of road work. And that's annoying, you will loose quite some time because of this.
But to keep the roads in good condition it needs to be done. Roads wear pretty quickly. The asphalt wears out, especially in the busy areas. And then the road works start again. Is there no other way?
Once I saw an episode of Thunderbirds where they had this massive machine that just drove over area, burned down anything in it's path and behind it let out a new road. Nice, but not really feasible.
Here's an other idea. Why not replace the top layer, the part that wears, with something like a carpet. Of course not a regular carpet, but a 3-5 cm flexible layer that can be rolled up. The material could be a bit rubber like, and very open so that water (from rain) would drain through. This may wear even faster, but there replacing it will take a lot less work. In one weekend you can do miles and miles of road. Just roll it up and roll out the new layer.
This would still be a big logistical operation. And the rolls would have to be several lanes wide. But the rolls could be transported to a storage area somewhere besides the road. During the transport they can transported 'oblong'. During roll out it is turned 90 degrees to be multiple lane wide.
Get some giant nails in akin a huge nail gun to staple it and you're done...
I know this is wild and bizar idea, but nonetheless, I do feel that there has to be a better way than scraping of the asphalt every 5-15 years and laying a whole new layer.
But to keep the roads in good condition it needs to be done. Roads wear pretty quickly. The asphalt wears out, especially in the busy areas. And then the road works start again. Is there no other way?
Once I saw an episode of Thunderbirds where they had this massive machine that just drove over area, burned down anything in it's path and behind it let out a new road. Nice, but not really feasible.
Here's an other idea. Why not replace the top layer, the part that wears, with something like a carpet. Of course not a regular carpet, but a 3-5 cm flexible layer that can be rolled up. The material could be a bit rubber like, and very open so that water (from rain) would drain through. This may wear even faster, but there replacing it will take a lot less work. In one weekend you can do miles and miles of road. Just roll it up and roll out the new layer.
This would still be a big logistical operation. And the rolls would have to be several lanes wide. But the rolls could be transported to a storage area somewhere besides the road. During the transport they can transported 'oblong'. During roll out it is turned 90 degrees to be multiple lane wide.
Get some giant nails in akin a huge nail gun to staple it and you're done...
I know this is wild and bizar idea, but nonetheless, I do feel that there has to be a better way than scraping of the asphalt every 5-15 years and laying a whole new layer.
Wednesday, March 11, 2009
Using ram and batteries to get green and faster Hard Drives
Reducing power usage on hard disks by using ram and batteries
Most computers use hard disks. When it comes to price per Gigabyte that still is the most economical solution. They also use quite a bit of power. Anyone that has ever assembled a low power machine and started adding the watts knows that in particular 3.5 inch HD's use quite a bit. 2.5 inch hard disks (used in laptops) use substantially less, but they are also much more expensive and cannot hold the same amount of data.
The most important reason that these disks use this much power is that the data is placed on so called platters. These platters move around their axis at (commonly) 7200 rpm. So the platters make 7200 rounds each minute.
The disk keep the moving, even if the disk is not accessed. This is partly for speed reasons, spinning up takes time. So if your computer needs to read or write a file, it needs to wait until the disks are spinning. But also this is for durability. Hard disks suffer more wear from spinning up than rotating the platters continuously. Laptop hd's are designed to spin down more often. Therefore they can handle it that power management turns the disk off more often to conserve energy. But that does make them slower as well.
RAM memory is much faster than hard disks and uses a lot less energy. But using RAM is not very smart as RAM is not just much more expensive, it is also volatile. As soon as the power is of, all the data is gone. But RAM is getting cheaper.
If you have a large hard disk, you under most circumstances always use roughly the same set of files. The rest of the files are basically just stored there (hard disks provide storage).
So for a relatively small set of files we keep the disks spinning. We can't use ram for this since RAM is not storage. My idea is to combine disks and RAM to make sure that the disks are spinned down most of the time.
Let's add to a regular HD (say 1 TB) 8 gb of RAM. Now I realise that' s not cheap, but we don't need the fastest and most expensive RAM. Any RAM is already faster then a HD. And now also add a battery.
Now also add to the controller in the disk logic that does the following:
- it keeps every file read in RAM until that RAM is full, when files in RAM get flushed based on how much is needed and the priority a file has. The priority is really a combination of when files are last accessed and how often they are usually accessed. (this sounds like cache, but there is more)
- every file that is written is also written to RAM first. The files don get written to disk yet, the write to the disks is delayed.
- the files that are in delayed write are written to disks when:
- the RAM is getting full (i.e. when the delayed writes > 3.5 Gb)
- on set times during the day (i.e. twice a day)
- when the computer is turned down.
That's where the battery is for. When the computer is turned down, or for any other reason the HD no longer gets power, it flushes the delayed writes to disk. The battery provides the power to complete this task when the power to the HD is cut. Also it keeps the files it has read in memory, for as long as it has power.
That means that if you reboot and there was still enough power in the battery to keep the RAM up, the files it needs to reboot are likely in the RAM part of this disk.. As result, considering that RAM is much faster than disks, the whole boot part gets to be much faster.
The result: in most use cases your computer get's a much faster hard disk that uses a lot less power. The disks are spinned up only when accessing rarely used or overly large files and a few times a day for the scheduled writes and writes of larges files.
As a result you get a hybrid between RAM and storage. It should be much faster and also use a lot less energy. And yet you still get the terabyte of storage.
I think that even with just 4 GB you can achieve this. Of course 8 GB would be more effective.
Most computers use hard disks. When it comes to price per Gigabyte that still is the most economical solution. They also use quite a bit of power. Anyone that has ever assembled a low power machine and started adding the watts knows that in particular 3.5 inch HD's use quite a bit. 2.5 inch hard disks (used in laptops) use substantially less, but they are also much more expensive and cannot hold the same amount of data.
The most important reason that these disks use this much power is that the data is placed on so called platters. These platters move around their axis at (commonly) 7200 rpm. So the platters make 7200 rounds each minute.
The disk keep the moving, even if the disk is not accessed. This is partly for speed reasons, spinning up takes time. So if your computer needs to read or write a file, it needs to wait until the disks are spinning. But also this is for durability. Hard disks suffer more wear from spinning up than rotating the platters continuously. Laptop hd's are designed to spin down more often. Therefore they can handle it that power management turns the disk off more often to conserve energy. But that does make them slower as well.
RAM memory is much faster than hard disks and uses a lot less energy. But using RAM is not very smart as RAM is not just much more expensive, it is also volatile. As soon as the power is of, all the data is gone. But RAM is getting cheaper.
If you have a large hard disk, you under most circumstances always use roughly the same set of files. The rest of the files are basically just stored there (hard disks provide storage).
So for a relatively small set of files we keep the disks spinning. We can't use ram for this since RAM is not storage. My idea is to combine disks and RAM to make sure that the disks are spinned down most of the time.
Let's add to a regular HD (say 1 TB) 8 gb of RAM. Now I realise that' s not cheap, but we don't need the fastest and most expensive RAM. Any RAM is already faster then a HD. And now also add a battery.
Now also add to the controller in the disk logic that does the following:
- it keeps every file read in RAM until that RAM is full, when files in RAM get flushed based on how much is needed and the priority a file has. The priority is really a combination of when files are last accessed and how often they are usually accessed. (this sounds like cache, but there is more)
- every file that is written is also written to RAM first. The files don get written to disk yet, the write to the disks is delayed.
- the files that are in delayed write are written to disks when:
- the RAM is getting full (i.e. when the delayed writes > 3.5 Gb)
- on set times during the day (i.e. twice a day)
- when the computer is turned down.
That's where the battery is for. When the computer is turned down, or for any other reason the HD no longer gets power, it flushes the delayed writes to disk. The battery provides the power to complete this task when the power to the HD is cut. Also it keeps the files it has read in memory, for as long as it has power.
That means that if you reboot and there was still enough power in the battery to keep the RAM up, the files it needs to reboot are likely in the RAM part of this disk.. As result, considering that RAM is much faster than disks, the whole boot part gets to be much faster.
The result: in most use cases your computer get's a much faster hard disk that uses a lot less power. The disks are spinned up only when accessing rarely used or overly large files and a few times a day for the scheduled writes and writes of larges files.
As a result you get a hybrid between RAM and storage. It should be much faster and also use a lot less energy. And yet you still get the terabyte of storage.
I think that even with just 4 GB you can achieve this. Of course 8 GB would be more effective.
Monday, February 23, 2009
Effective way to harvest power of ocean currents
Using strong ocean currents for power generation
Anyone that ever played on the beach with one of those plastic mills knows that water is a much better medium than air to make it go. A little bucket of seawater made that thing go. Blowing didn't do much. Also it's quite common in the world to generate power by building dams with large artificial lakes behind it and at the bottom of the dam where the water is released you generate power.
This in fact was already economically viable for years. Even when not many cared for CO2 emissions. Only downside is that you need to create these large lakes and that means sacrificing large areas of often build up land.
So to me it seems more obvious to use the currents in the ocean to generate power than to build windmills in that same ocean.
There are even some initiatives in doing this. They include....submerging windmills. To me that seems a bit strange. I guess if you ask a carpenter to build a car you get a wooden car.
One of the reason that the hydro-dams were and are successful is that they concentrate the power. As a result you don't need a large amount of turbines to generate the power. One will generate enough. Building a dam in the ocean, however, is not such a great idea.
There is a much simpler idea that only a tent-like construction, anchored to the ocean floor on one or two places.
A picture explains much more:
I know I'm not good at drawing. But this about the principle. One cable is to keep it in place. The other is to raise and or deploy it. Just like with a sail on a sailing boat. And the whole system hangs on large buoy.
By making the part that catches the water wide and going narrow in the end (close to the water turbine) you get a very strong current. All you need now is the water turbine to turn this into electricity.
This entire system can be build on land, transported to the right place and then simply deployed. If due to the strong currents that gets a bit difficult, the concentrator could have on the seems something like a big zipper. That way the deployment can be done with little pressure. And when its in place, close the seems and the pressure will build.
The OCS site talks about maintenance. My idea is really cheap in maintenance. All you need is a ship that docks the floating part, and does the hauling up. Release the turbine and load it into the ship. And replace it with a 'new' turbine. Then inside the ship or on shore, the maintenance and repairs on the turbine can take place in a dry environment. Same for all the other parts. The most difficult part to deploy is the anchor.
I believe this system is simple, easy to build and deploy, and generates much more power.
I do think though that it is necessary to place a net in front of the concetrator to prevent sea animals to end up in the turbine. The smaller ones get through but they are likely to just wash up at the other side. Maybe they end up a bit confused.
Advantages:
- lightweight construction
- nearly fully build on shore (except for the anchors)
- easy maintenance
- can be made mobile (i.e. for military purposes)
- high power yield
- cheap
Disadvantages
- has a bigger impact on the currents and water flows then the other solutions.
Of course there are more ways to do this. You can even use multiple generators that don't even surface or hang on buoys, such as in the example below:
That way, entire farms can be build below the water, not interfering with shipping etc. In the above example hauling the generators up for maintenance gets a bit more work though.
Subscribe to:
Posts (Atom)