Posted: Mon Dec 03, 2007 20:39 Post subject: GW2348-4 current @ 12 VDC?
If anyone here has a GW2348-4 with four DCMA82 installed and powered via 12 VDC, can you measure the electrical current in the 12 V circuit (NOT in primary 230 VAC circuit!) ? For a non-power-supply-installation with solar and wind energy generators, i need more informations about power consumption of a GW2348-4. Idealwise, the radios are in TX stage and normal EIRP output setting (17 dBm or so). I need real, practical values, not typical values from the manual.
form the avila man
"The DC/DC has a minimum input voltage of 9VDC and a
maximum of 48VDC. ..... The input power is supplied through
a power connector or through the RJ45 Ethernet connector in a power over
if u want to power the board with solar and wind energy u should use hight voltages for power supply. higher voltages with lower current can be more easly produced then the 12v...especially when u will use all of the 4 mini-pci slots. u will need aprox 5w for the board and max 18W for all mini-pci slots (i think ~ 2,5W per minipci card will fit for u) = 15W for the unit.
with 12V u will need 1.25A + losses in the avila dc/dc converter, cabeling and the solar/wind power converter.
with, lets say, 40V u will need only 0.375A + losses
for night operation u will need a buffer batterie,too. the job will be done by 2 or 3 12v accu´s in series connection = ~36V.
btw higher voltages are better to transmit over cat5 (lower loss).
Posted: Wed Dec 05, 2007 0:00 Post subject: GW2348-4 current @ 12 VDC?
My concerns are similar to those expressed above. 12 volts is not a very good idea for running the 2348 when using 4 high power cards. I seriously doubt that you will be able to get the board to behave in a stable manner on 12 volts especially if all 4 cards are active in transmit at the same time. I realize that solar/wind systems are difficult to engineer at other than 12 volts, however it can be done and I would suggest that you will be much happier with the results if you engineer the system to operate at 48 volts. The Avila board will be much more likely to be stable and you run less of a chance of overloading the power supply if you get a burst of activity on multiple cards at the same time. It should also be pointed out that the power measurements that you are asking for are probably not possible unless we know something more about the utilization of this system. Transmit vs receive duty cycle has a marked effect on power consumption as does data rate. Counter-intuitively, lower data rates generally increase the power drain during transmit by a substantial amount (sometimes more than 30 percent) The combination of higher current draw and increased transmit duty cycle from the lower data rate can easily double the power required to transmit a given amount of data. In my setup, I run Ubiquiti XR5 and XR2 cards which have similar specs to the DCMA82 though they are single band cards. I have seen a 4 card system spike up to 22 watts though in my setting they usually run on 12 or less. If you are mounting this in an outdoor enclosure, especially one without a fan, give careful consideration to thermal issues. In my systems I have found that solar heating can be a big problem even though the 2348-4 allows the cpu temp to go up to 80C without damage. You will probably need to heatsink the cpu for good results.
Goodl luck and keep us posted on how your system does.
Berrien Springs, MI USA
The only cause i want to use DCMA82 is its temp spec, *NOT* their highpower capability.
High power cards are a little bit stupid in directional wlan installations with high gain antennas and shot low loss cables when the system should be comply with *german* regulatory rules.
What i need is: A system what a) can be run with solar/wind power, b) works stable when the board is mounted in a outdoor case and the sun shines on it and c) has a economical power consumption.
@Economical: A solar system which delivers more than 12 volts are more valuable than standard solar installations @ 12 volts. Beginning at the solar panels over the charging regulator within to the batteries, a 48 volt system is difficult.
Posted: Thu Dec 06, 2007 1:09 Post subject: GW2348-4 current @ 12 VDC?
I agree that high power is generally not necessary in directional installations such as you describe. That fact was not in your original post and the only information I had to go on was the specifications of the card and the fact that you intended to run 4 of them on one GW2348-4. Now I understand your situation a little better. I am still concerned that at 12 volts the GW2348 may not be stable with 4 cards installed. The 24 volt option suggested by another poster may be enough to overcome that problem though. If solar heating is your major concern then I would suggest the following practices will help you. First, be sure to heatsink the processor of the GW2348 directly to the metal of the enclosure. This will provide the maximum heat dissapation. Second, paint the enclosure WHITE. Don't rely on the reflectivity of bare metal or of a grey or off-white enclosure. Paint it white as snow. You will be surprised how much you can reduce the solar heating effect that way. Third, fabricate a sun shade for the enclosure. The sunshade can be a folded piece of sheet metal attached to the same support structure that holds the enclosure. It should allow plenty of airflow around the enclosure and should be several centimeters away from the enclosure. If it must contact the enclosure at all it should only do so at a small point of attachment. In our installations we use a small enclosure from mini-box (www.mini-box.com) which we paint white. We then fold a piece of sheet metal to fit around the top of the enclosure allowing about 5-7 cm space between the sheet metal and the enclosure surface. The enclosure is mounted in a north facing orientation with the sunshade covering the top and south sides. What we have found is that the temperature rise inside the enclosure is about 18-22 degrees C over the ambient outdoor air temperature. As long as the ambient doesn't go much above 58 Degrees C it should not burn up I hope this helps.