[Soekris] [soekris] net4826 + two 400mw miniPCI?
broustis at cs.ucr.edu
Sun Jan 28 02:22:25 UTC 2007
We have been using these cards for about a year now in our testbed in
UCR. Even though they are quite stable, as of now we haven't been
able to transmit with more than 19 dBm. We are PXE booting Debian 3.1
and we use the madwifi drivers, on net4826. We haven't tested 2 cards
in the same box. This email is just to let you know that I have not
been able to use the max power advertised. If someone manages to
transmit with more than 20 dBm, i would be really grad to hear how he/
she did it.
On 27 Jan 2007, at 16:15, Michael Bell wrote:
> I have had this card in for extensive testing (laboratory and
> real-world) so I am quite familiar with it. I have one personally
> in my
> net4521 however I did not choose it because of the 26dBm TX capability
> (I have access to 20W amplifiers if I really need power) but rather
> because of the chipset.
> Please keep in mind that TX is only one half of the equation. Receive
> sensitivity plays a big role in how well the link will establish and
> maintain communication. With proper antenna selection and alignment,
> even lower power (high receive sensitivity) cards can perform
> That being said, your question regarding power consumption. Testing
> showed the card at peek *output* to draw between 692mA and 752mA
> a 24 hour stress test reaching a maximum operating temperature of 47
> degrees fahrenheit above room temperature.
> 2 of these cards operating at peek *output* and assuming the worst
> power consumption observed would mean that the net4826 would need to
> provide 1.5A @ 3.3V or 4.95W. I haven't tested the net4826 boards
> so I
> don't know what it's voltage regulators are capable of providing to
> Mini-PCI slots however since the board cannot draw more then 12
> watts of
> power you may encounter difficulty with them cranked up that high (let
> alone FCC violations depending upon your antenna selection).
> Not sure that helps you but maybe somebody on the list has already
> tested the net4826 and can answer the power supply question.
> I specifically used output, and not throughput above. The transmit
> power of the card is reduced by nearly 4dBm @ 54Mbps 802.11g and
> 12dBm @
> 54Mbps 802.11a (5.7ghz)
>> Thanks for reply.
>> I'm using TriBand Rubber Duck Antenna(omnidirectional antenna). I
>> couldn't find any better gain than that of this antenna which I can
>> use a/b/g together.
>> On 1/27/07, Egbert Jan <egbert at vandenbussche.nl> wrote:
>>> Are you using this config as repeater in some way? When you need
>>> (bundeled?) power, use antennas with a higher gain. You will be more
>>> efficient with your input power. An antenna with 3 dB more gain
>>> will give
>>> what you need. If you need omnidirectional radiation you might
>>> need more
>>> Egbert Jan NL
>>> -----Original Message-----
>>> From: soekris-tech-bounces at lists.soekris.com
>>> [mailto:soekris-tech-bounces at lists.soekris.com] On Behalf
>>> Of Jeongki Min
>>> Sent: Saturday, January 27, 2007 3:43 AM
>>> To: soekris-tech at lists.soekris.com
>>> Subject: [Soekris] [soekris] net4826 + two 400mw miniPCI?
>>> I have a question about using Soekris net4826 with two 400mW
>>> Currently I'm using two 200mW MiniPCIs, but I think I need more
>>> tx power.
>>> I'm considering to buy product like below;
>>> I really don't know about hardware.:(
>>> It might be stupid thing, but I heard that I have to be careful
>>> electric power consumption when I choose MiniPCI.
>>> Do you think there is no problem on using two 400mW MiniPCI on
>>> And if I use 400mW MiniPCI, could I can see better tx power?
>>> I would really appreciate it if you give me any information.
>>> Thanks for useful information always,
>> Soekris-tech mailing list
>> Soekris-tech at lists.soekris.com
> Soekris-tech mailing list
> Soekris-tech at lists.soekris.com
More information about the Soekris-tech