EAP225-Outdoor antenna position and/or replacement
Hello,
I bought the EAP225-Outdoor mainly because the antennas can be moved and/or replaced but I am using it indoors.
I have I mounted in a wall in a closet and the only areas I want to cover are:
- the closet
- the same closet on the floor above
- the same closet on the floor below
I don't need (nor even want) to have the signal go far horizontally. This WiFi network is for specific devices only and those don't come with any wired connectivity option.
(Unfortunately, the horizontal floors are much thicker than the vertical walls... but that's just the nature of a building I guess ;)
Now, the user manual doesn't say much about antenna placement/orientation...
So my questions would be:
1) What are the best positions for the antennas if I want to have mainly vertical coverage?
2) Are there other kinds of antennas I should try instead of the ones that come with the device?
3) Loosely related question I am curious about: if I go with one EAP per floor, what kind of antenna should I use to get extremely SHORT coverage (Ideally I'd like 2.4 GHz to stop at the first wall), which means I need lower power than the minimum I can set in software (Omada doesn't allow to set below 6 dBm for 2.4 GHz) .
3b) Just for general understanding: could I use paper clips ? ;)
4) What would you recommend to rationally measure the signal quality in the building after changing antenna positions or antenna types?
5) In case you know a good tutorial for learning about antennas / radiation patterns / signal propagation, I am interested in a link. I am coming from an engineering background (but not really related to radio) and I am willing to invest several hours to a few days into understanding it.
Thanks!
- Copy Link
- Subscribe
- Bookmark
- Report Inappropriate Content
Ok, I'll take a crack at these :)
1) What are the best positions for the antennas if I want to have mainly vertical coverage?
This is exactly the kind of antenna research area that has never had much practical application...horizontal gets the investment. That's not to say you cannot get narrow beam antennas (down to about a 30' cone)
2) Are there other kinds of antennas I should try instead of the ones that come with the device?
The dipoles that come with the device are close to omni directional, their coverage being roughly shaped like an apple, with the antenna being the vertical core. You can get patch (flat panel) antennas which create a narrow cone-shaped beam down to about 30', but they are unidirectional...ie think searchlight vs lighthouse. Yes the AP device has 2 ports, but they are not equivalent, they are Main and Diversity and are expected to be matched both in pattern and direction and gain if the radio is to be able to properly figure out the spatial path between itself and the client (so you cannot just use one port to point 1 flat antenna up and the other down). If you can put a small hole through the ceiling/floor, then 'leaky coax' might be worth looking at.
3) Loosely related question I am curious about: if I go with one EAP per floor, what kind of antenna should I use to get extremely SHORT coverage (Ideally I'd like 2.4 GHz to stop at the first wall), which means I need lower power than the minimum I can set in software (Omada doesn't allow to set below 6 dBm for 2.4 GHz) .
I think you might want a PAD (inline passive RF attenuator device that only permits a certain percentage of power to actually pass through to antenna) rather than some esoteric antenna. You can buy them off the shelf in calibrated attentuations like -6dB, -12dB etc. in suitable SMA type form factors.
3b) Just for general understanding: could I use paper clips ? ;)
Absolutely and they work shocking well, not just for wifi, but cellular as well! Probably way to well for what you are trying to do.
4) What would you recommend to rationally measure the signal quality in the building after changing antenna positions or antenna types?
I've found 'Wifi Analyzer' an Android app by A.M Sid to be sufficient for most stuff I've done. I would recommend trying to get your RSSI down to between -85dBm and -90dBm as sort of the hairy edge beyond which communication typically becomes unreliable. If you have a lot of local interference/competition, then adjust upwards as necessary.
5) In case you know a good tutorial for learning about antennas / radiation patterns / signal propagation, I am interested in a link. I am coming from an engineering background (but not really related to radio) and I am willing to invest several hours to a few days into understanding it.
Nothing off the top of my head, but keep in mind that radio properties aren't that much different from sound (penetration) or light (reflection).
The big problem you are going to have is that you are going to reflect a very large percentage of the RF energy you are trying to push through the thick floor and that's all going to go somewhere else. So if you can't physically 'pipe' it through...and I guess the answer is no or you'd probably just use CAT6 cable, then the spray and pray approach is what's usually done. However, it looks like you don't want to excessively interfere with the local environment. One suggestion I have is to 'back to back' flat panel antennas, so a flat panel on the floor above pointing straight down, and one on the ceiling below pointing straight up, using narrow beam panel anteannas and PADs to further reduce the RF power as needed, you should be able to launch just enough RF power through the concrete to establish a radio link, without excessively spraying it all over the place, and the antenna itself should act as a shield to limit the amount of RF that escapes. It is then up to the receiving device to redistribute, either via wifi or ethernet to the local devices that need connection. If you use EAP225-outdoors, then a mesh link is possible at 5.8G, and user access can be achieved at 2.4G, otherwise you can experiment with Wifi repeaters/boosters
- Copy Link
- Report Inappropriate Content
Ok, I'll take a crack at these :)
1) What are the best positions for the antennas if I want to have mainly vertical coverage?
This is exactly the kind of antenna research area that has never had much practical application...horizontal gets the investment. That's not to say you cannot get narrow beam antennas (down to about a 30' cone)
2) Are there other kinds of antennas I should try instead of the ones that come with the device?
The dipoles that come with the device are close to omni directional, their coverage being roughly shaped like an apple, with the antenna being the vertical core. You can get patch (flat panel) antennas which create a narrow cone-shaped beam down to about 30', but they are unidirectional...ie think searchlight vs lighthouse. Yes the AP device has 2 ports, but they are not equivalent, they are Main and Diversity and are expected to be matched both in pattern and direction and gain if the radio is to be able to properly figure out the spatial path between itself and the client (so you cannot just use one port to point 1 flat antenna up and the other down). If you can put a small hole through the ceiling/floor, then 'leaky coax' might be worth looking at.
3) Loosely related question I am curious about: if I go with one EAP per floor, what kind of antenna should I use to get extremely SHORT coverage (Ideally I'd like 2.4 GHz to stop at the first wall), which means I need lower power than the minimum I can set in software (Omada doesn't allow to set below 6 dBm for 2.4 GHz) .
I think you might want a PAD (inline passive RF attenuator device that only permits a certain percentage of power to actually pass through to antenna) rather than some esoteric antenna. You can buy them off the shelf in calibrated attentuations like -6dB, -12dB etc. in suitable SMA type form factors.
3b) Just for general understanding: could I use paper clips ? ;)
Absolutely and they work shocking well, not just for wifi, but cellular as well! Probably way to well for what you are trying to do.
4) What would you recommend to rationally measure the signal quality in the building after changing antenna positions or antenna types?
I've found 'Wifi Analyzer' an Android app by A.M Sid to be sufficient for most stuff I've done. I would recommend trying to get your RSSI down to between -85dBm and -90dBm as sort of the hairy edge beyond which communication typically becomes unreliable. If you have a lot of local interference/competition, then adjust upwards as necessary.
5) In case you know a good tutorial for learning about antennas / radiation patterns / signal propagation, I am interested in a link. I am coming from an engineering background (but not really related to radio) and I am willing to invest several hours to a few days into understanding it.
Nothing off the top of my head, but keep in mind that radio properties aren't that much different from sound (penetration) or light (reflection).
The big problem you are going to have is that you are going to reflect a very large percentage of the RF energy you are trying to push through the thick floor and that's all going to go somewhere else. So if you can't physically 'pipe' it through...and I guess the answer is no or you'd probably just use CAT6 cable, then the spray and pray approach is what's usually done. However, it looks like you don't want to excessively interfere with the local environment. One suggestion I have is to 'back to back' flat panel antennas, so a flat panel on the floor above pointing straight down, and one on the ceiling below pointing straight up, using narrow beam panel anteannas and PADs to further reduce the RF power as needed, you should be able to launch just enough RF power through the concrete to establish a radio link, without excessively spraying it all over the place, and the antenna itself should act as a shield to limit the amount of RF that escapes. It is then up to the receiving device to redistribute, either via wifi or ethernet to the local devices that need connection. If you use EAP225-outdoors, then a mesh link is possible at 5.8G, and user access can be achieved at 2.4G, otherwise you can experiment with Wifi repeaters/boosters
- Copy Link
- Report Inappropriate Content
@d0ugmac1 Thanks a lot for all this precious information!!
Things I had no concept of before today but that change my whole view of the WiFi world:
- Leaky coax (will look into it)
- Inserting attenuators before the antennas (definitely will try)
- Thinking about RF as light and the fact that a significant part will reflect off walls (makes me wonder if there is the equivalent of black paint for 2.4 and 5 GHz wavelengths, I don't mind if it absorbs and heats up a little ;)
Very useful!
About this though:
> Yes the AP device has 2 ports, but they are not equivalent, they are Main and Diversity and are expected to be matched both in pattern and direction and gain if the radio is to be able to properly figure out the spatial path between itself and the client (so you cannot just use one port to point 1 flat antenna up and the other down).
If I'm not mistaken, the EAP manual says you should angle the 2 antennas at 90° from each other. Doesn't that mean it's actually designed to figure out the spatial path even with non matching directions?
Thank you.
- Copy Link
- Report Inappropriate Content
The reason you position the antennas in a V instead of say | | or _ _ is that if you get 90' angle at the vertex, then your antennas are positioned optimally to utilize Horizontal and Vertical polarizations for the established path. If your angle is 0 (| |) or 180 (_ _) then you are either V-V or H-H which offers less spatial diversity than H-V.
- Copy Link
- Report Inappropriate Content
@d0ugmac1 Ok, I assume you could also use an L shape instead of a V shape. Assuming fixed WiFi devices (not mobile phone), does it matter how the device embedded WiFi antennas are oriented in regards to the V or L of the EAP or will any diagonal position work just as well once you have 2 antennas at 90° ?
Also thinking: assuming the recommended V in a vertical pane is optimized for horizontal coverage, would it make sense to rotate the antennas so they make a V in a horizontal pane in an attempt to optimize for vertical coverage ?
Thank you !!!
- Copy Link
- Report Inappropriate Content
Now you're scratching the bottom of my comfort level :)
In theory, you should have the most path diversity options in front of or behind the V, because their side view profile is more like a | than a V. As you will notice with decent signal strength and low obstruction/interference, there's little performance difference between V and | | anyways. It's only when the going gets rough that you notice any advantage of polarized antennae.
- Copy Link
- Report Inappropriate Content
One followup in case s/o else gets into this:
- RP-SMA attenuators are hard to get by (regular SMA is much more common)
- The default antennas of the EAP225-outdoor have a gain of 4 dBi
- It is possible to find pairs of RP-SMA "wifi antennas" on Amazon with lower gains (found as low as 2.15 dBi) for a fraction of the price of a single attenuator (those attenuators are designed for higher power than needed)
- Another useful keyword is "MIMO Mini Antenna"
- Copy Link
- Report Inappropriate Content
Fair points, it may be easier to source the SMA-RPSMA M-F and F-M adapters and then use standard SMA attenuators. Buying all 3 of course is not cheap.
Given that 3dB is a halving of power, going from 4 to 2 with different antenna isn't going to reduce your RF output much at all. Not when the receive has gain multipliers in the thousands. You are probably looking at needing something in the order of 10-20dB of attentuation. For instance at max power the AP can put out about +20dB with the stock antenna (so just over 100x). If you added a -8dB pad, taking the output power to +12dB, you'd only be putting out 16x power, and if you used the 2dB antenna instead of stock, you would take than down to +10dB or about 10x.
- Copy Link
- Report Inappropriate Content
With the little I learnt in between, do you allow me to be pedantic and insist that the max output of the AP (in the 2.4 GHz band) is 20 dBm and not 20 dB ? :))
I can configure it to 6 dBm though (antenna gain of 4dBi included, so 2 + 4).
Now, if I understand all this correctly:
By replacing the antenna (not received yet) I'll go down to 2 + 2.4 = 4.4 dBm. That's already down from 100m W to 2.65 mW.
I also ordered -6 dB attenuators (will take weeks to arrive from China). That should allow me to get to -1.6 dBm. That's about 0.66 mW (if I am not mistaken).
One goal I have is to no longer be able to pick up the signal from the other sides of the walls. I figured that by going from 100 mW to less than 1 mW, I would be able to achieve that... but now, if you tell me the receivers have gains in the thousands... it may not be enough... I will experiment and see ;)
Do you have some sources about minimum signal levels wifi receivers need to be able to pick up the signal?
Hum... now you make me think about it... I think I saw RSSIs of -65 dBm... so yeah, I may be way off! Should probably have ordrered -60 dB attenuators!! (Well at least, I'll be able to serialize the ones that already shipped with some additional ones ;)
On the other hand: there is so much interference from other networks around here that my network may get drowned in the background noise well before -65 dBm...
- Copy Link
- Report Inappropriate Content
You are spot on above. I didn't want to muddy the waters with dBm's and dBi's, at the end of the day, the math is still basically the same. You are going to have to play around with power and attentuation, and you'll be looking at the receive characteristics as well (so most people only think about RSSI...which is essential the sound equivalent of everything hitting your eardrum, the next level analysis looks at metrics like RSRQ where you start to look at things like the gap to the noise floor and the overall quality of the received signal.
- Copy Link
- Report Inappropriate Content
Information
Helpful: 0
Views: 424
Replies: 9
Voters 0
No one has voted for it yet.