WiFi Card upgrade testing

[ChrisM101]

The 1W card won't fit with that heat sink. As a matter fact the mmcx connections on it may have posed a challenge with size once the antenna side was connected. When I ordered the 800mw cards they sent me the 1W cards, more expensive part but I had to exchange them because it just wouldn't fit.

Thanks,
any other cards that you considered?

Also Would you see any benefit to changing both to a dual band card in your opinion?

 

[NYG]

Thanks,
any other cards that you considered?

I searched for a few days looking at options. Most cards don't list TX-Power. There was another card from Ubiquiti but it was an older chipset, lower power (400mw) and more expensive so i passed on it.

 

[NYG]

Also Would you see any benefit to changing both to a dual band card in your opinion?

That's a good question. I can't see why 5.8Ghz wouldn't work. That frequency would have a harder time penetrating obstructions but it's a less used band so less interference. I think you would also need to modify /etc/hostapd.conf on both the bird and controller since that's at least one area where it tells it to use the 2.4ghz band.

 

[ALLA]

That's a good question. I can't see why 5.8Ghz wouldn't work. That frequency would have a harder time penetrating obstructions but it's a less used band so less interference. I think you would also need to modify /etc/hostapd.conf on both the bird and controller since that's at least one area where it tells it to use the 2.4ghz band.

You mentioned that you were fine for now with the "possible" restriction to 27db, but might try later to go to 29. Please let me know if you do try and if it had any effect. I just ordered two of these cards and will join the test as soon as they are here. Your findings so far look very promising. I personally might be interested in trying 29db settings, if possible at all, as well as maybe 5.8GHz down the road.
Thanks again for sharing this!

 

[Earldgrayjr]

Yesterday I posted on another thread about testing I'm doing with WiFi card upgrades in the Solo.
The stock cards are mini PCIe and based on the Atheros AR9382 chip, both controller and Solo have these cards.
I've got a pair of MikroTik R11e-2Hnd (800mw) cards that I've been wanting to test, they're based
on the Atheros 9582 chip. The Solo uses the ath9k driver that is compatible with a wide range of
Atheros based cards.

I've been regularly doing my testing at a local middle school/high school facility. It's not the cleanest
area for WiFi, especially with the surrounding neighborhood but it's only a few minutes from leaving home to flying.

With the stock wifi cards and stock antennas the max range I got was 600'-700' before RTH. This was pretty consistent.
With just the new Mikrotik wifi card in the Solo my range increased to 1500' before RTH.
Today with the controller also upgraded with a Mikrotik card I went over 2500' before RTH.

With a pair of router antennas I was getting 1300'-1500' before RTH with the stock wifi cards.
With just the Solo upgraded I reached 2300' before RTH.
Today with both the Solo and controller upgraded I reached over 3700' before RTH.

I also have a homebrew high db panel antenna that I reached just under 4000' with today but I'm finding it very difficult to aim when i get out that far.

Today is also very humid and cloudy day. Looks like rain soon.

I had a couple of concerns going into this:
1) would the new cards generate more heat and would that be a problem? So far this looks to be OK
2) would the newer cards eat more battery? This is still an open question but I haven't noticed any difference just on normal flying habits with the limited tests done so far.

Also of note I don't know the TX power of the stock wifi cards and I have yet to find a compatible linux command to display it on the Solo or controller (certain commands are missing). I also saw in one of the startup scripts that it *may* limit max TX power to 27dbm. I also don't know the exact TX power of the new Mikrotik cards, based on the results though the Mikrotik cards are significantly better.

More research needs to be done on this. I also need to do some testing from the command line to see if i can boost power from the Mokrotik cards.

So far so good. I'm pretty damn impressed right now.

I also planned on testing a 1W mimo booster I had but decided not to do it yet. I just don't feel like carrying around the booster, extra battery, cables etc...

I'll keep you all posted as this venture continues.

Really nice work... Thank You

 

[NYG]

An update for those following this discussion. I've gotten some very good help on the Solo FB mod page. I was told that in my older version of Linux the TX-Power being reported back to me for the Solo cards was not accurate, I needed a later version.

I booted under Ubuntu 15.04 and it shows the Solo card having a TX-Power of 21dBm. If that number is accurate then we're looking at 126mw for the Solo cards

27dBm would be 500mw.

 

[ChrisM101]

An update for those following this discussion. I've gotten some very good help on the Solo FB mod page. I was told that in my older version of Linux the TX-Power being reported back to me for the Solo cards was not accurate, I needed a later version.

I booted under Ubuntu 15.04 and it shows the Solo card having a TX-Power of 21dBm. If that number is accurate then we're looking at 126mw for the Solo cards

27dBm would be 500mw.

Thanks for the update.

 

[ChrisM101]

So im thinking perhaps they made the same mistake when they set it up, perhaps their version reads incorrectly also...

Have you by chance figured out the type of original card that they used>?

Also is the limit 27 still in solo? meaning theres still 300 mw available on the mikrotik card?

 

[NYG]

This is some code that's in the sololink_config file. If this code is being executed on the Solo and controller you can see they've set a limitation of 27dBm. We can try changing this file to increase the limit to the MikroTik card max of 29dBm, or we can create our own scripts that execute on startup to override what this is doing. Do we know this is doing what we want it to do? I'm not sure because I haven't found a way to list the actual TX-Power on the Solo or controller.

One of the things I should have done before installing the new cards was to run them in a laptop to test how warm they get running at 800mw.

Having said all that I've found the change in connectivity between the Solo and controller to be so dramatic that I'm not sure I need to change anything.

get_wifi_power() {
if [ -z /sys/kernel/debug/ieee80211/phy0/power ]; then
echo "error: unable to read power level"
return 1
fi

power=`cat /sys/kernel/debug/ieee80211/phy0/power` <- The output of this is 30
if [ $power -gt 27 ]; then power=27; fi
echo "$power"
return 0
}

set_wifi_power() {
desiredPower="${1}"
if [[ $desiredPower =~ ^-?[0-9]+$ ]]; then
if [[ "$desiredPower" -lt 0 || "$desiredPower" -gt 27 ]]; then
echo "error: wifi power level must be between 0 and 27"
return 1
fi
else
echo "error: attempting to set wifi power to non-integer"
return 1
fi

echo "$desiredPower"
iw phy0 set txpower fixed $((desiredPower*100))
return 0
}

 

[ChrisM101]

well 21 certainkly gets them under the 1 watt limit, but the max is 1w (30) plus 6bdi of antenna right?
Legally 29 plus up to a 7dbi antenna would be the ideal setup to max it out.

Anyhow, do you think you will get out to some farmland anytime soon to do a longrange test?

This is absolutely an no brainer for me to do, but im still on the fence about the d%^* gimbal

I would do it in a heartbeat if i had the gimbal already and it is what its supposed to be.

 

[NYG]

Anyhow, do you think you will get out to some farmland anytime soon to do a longrange test?

I'm going to do my best to get to some open space this weekend.

 

[Rich]

To those that asked about using a dual band radio, don't forget that this would require new antennas for both the controller and Solo.

 

[ChrisM101]

To those that asked about using a dual band radio, don't forget that this would require new antennas for both the controller and Solo.

Yeah I would adapt rpsma connectors to the exterior of solo and screw on some dual band rpsma antennas and just screw them onto the controller as well. If you could keep it on the 5.8 band there would be virtually no interference

 

[NYG]

Found an area to fly that has a good coverage of trees. I went out after a few thunderstorms passed this morning. Much less WiFi interference in this location.

I used my old pair of router antennas. They're dual band but do a better job on 2.4Ghz than the stock antennas.

I went out 1.1 Miles during this test.

 

[ChrisM101]

Found an area to fly that has a good coverage of trees. I went out after a few thunderstorms passed this morning. Much less WiFi interference in this location.

I used my old pair of router antennas. They're dual band but do a better job on 2.4Ghz than the stock antennas.

I went out 1.1 Miles during this test.

Are you still line of sight? Most of my tests lose at 1.2 miles except the alfa directional 7dbi. They go 1.5 to 1.6

 

[NYG]

Are you still line of sight? Most of my tests lose at 1.2 miles except the alfa directional 7dbi. They go 1.5 to 1.6

I was standing in a parking lot back below a tree line. I was aiming in the direction of the bird but I don't know how close it would come to that tree line being that distance out. I'm definitely trying to maintain line of site but I couldn't see the Solo anymore. Thank goodness RTH works well.

Are you getting 1.2 miles with a Solo on stock cards? That's pretty damn good if you are. I have a set of the Alphas but have yet to use them. I've been trying to use the same antennas that I did from the start but I'll certainly try them soon.

I went out 1 mile with a set of quad patch antennas. The RSSI remains much better than the router antennas but it becomes hard to aim and the RSSI turns bad on me before I can re-align in time.

I only had an hour window to fly today and those were my two flights I got in.

On a side note I just had a tube of electronics grade silicon come in the mail. when I install the RP-SMA connectors outside the Solo I'm going add some silicon to the ufl connectors.

 

[ChrisM101]

Yeah I was able to get 1.2 stock. But I just flew 9dbi omni from Alfa same distance. Who knows. I will go try them in the farmland and see. They're literally the size of drum sticks.

 

[NYG]

Yeah I was able to get 1.2 stock. But I just flew 9dbi omni from Alfa same distance. Who knows. I will go try them in the farmland and see. They're literally the size of drum sticks.

I'll have to head out east to get to full open space eventually. Still some farmland out that way.

 

[NYG]

Ok. I think I found a launch point in the same vicinity that I flew today with a better shot dealing with the tree line. I'll check that tomorrow morning. Plus it puts me further away from the buildings I was at today. I'll see how it goes.

If that goes well tomorrow I may try increasing the TX-power of the cards. But I'm nervous doing that.

 

[NYG]

Went out for a bit this morning. I found a much better launch point at a High School that put me at higher point and further back from the trees. It was a good area but hot today and a bit windy.

1st flight out i did a poor job of direction. I ended up flying over shopping centers after clearing a section of forest and RTH triggered 1.3 miles out.

2nd flight had me concerned. I swapped batteries out and the new battery was draining fast. I was about a mile out and brought it back to land. I'm not sure what's up with the battery or why that happened. I landed with about 25% remaining.

I have a 3rd battery and swapped to that one. I took a different angle out and monitored the battery much closer along with the RSSI. I ended up over a neighborhood 1.5 miles out. Video was getting choppy RSSI was bouncing around and I was paranoid from the previous battery so I hit RTH and just let it come home.

The bird fly's a hell of a lot faster coming home than I do going out. Maybe I should be flying faster than I have been with these longer runs. All those flights today were strictly with the router antennas.