How much of an effect does the cable choice have on the signal strength

[tonydobbs]

I'm trying to optimize my LTE setup in the mountains of rural CA, and I'm running into an issue. I mounted this 8db antenna outside my window (~20feet above ground) with mostly unobstructed LoS (2 small mostly dead trees about 50 and 100 feet out). Shockingly my signal strength seems worse than with the 2 omni's that came with the device. I've adjusted the antenna every way that I could think of, and now I'm wondering if the issue might be with the cables I selected. I'm using this cable which is only 6.5ft. Is it possible that my cable selection is the issue? If it's unlikely to be the cable, I suppose the next step is to try other antennas and consider picking up more testing equipment.

[JimHelms]

It is easy to lose more thru cable loss than is otherwise gained from an external antenna.

With that said, I would not think you would lose a lot at 6.5 feet.

Loss per foot (per 30 cm) at 900 MHz: 0.14 dB
Loss per foot (per 30 cm) at 2500 MHz: 0.24 dB

RG58 cable is best suited for applications below the 2.5GHz (2500HMz) frequency.

What router, firmware, modem, and carrier are you using?

[tonydobbs]

I'm using a WE826-T2 + MC7455 from you guys with the R00ter firmware on ATT band B2. I had another thought... my antenna is attached to my house instead of a pole, so the grounding strap isn't attached, could that be the issue?

[JimHelms]

The ground wire does not make a difference in its reception. It is to ground the antenna to the ground to help protect during a lightning strike.

Is the antenna pointed directly at the tower. Those antennas are very directional and, much like a satellite dish, a little off aim can have a profound affect on signals, completely missing the tower.

What does it look like when you run the AT Command:

Code: Select all

AT!GSTATUS?

[tonydobbs]

I'm thinking it must not be aimed well enough. I only know generally where the tower is, so I aimed it by recording the signal strength 3 times, moving it 2 degrees, and repeating the process.

Here is the current status (not too bad currently)

Code: Select all

Current Time:  18260		Temperature: 43
Reset Counter: 1		Mode:        ONLINE         
System mode:   LTE        	PS state:    Attached     
LTE band:      B2     		LTE bw:      20 MHz  
LTE Rx chan:   800		LTE Tx chan: 18800
LTE CA state:  NOT ASSIGNED
EMM state:     Registered     	Normal Service 
RRC state:     RRC Connected  
IMS reg state: No Srv  		
PCC RxM RSSI:  -77		RSRP (dBm):  -106
PCC RxD RSSI:  -78		RSRP (dBm):  -109
Tx Power:      22		TAC:         8B06 (35590)
RSRQ (dB):     -10.0		Cell ID:     0A15ED09 (169209097)
SINR (dB):      9.6
OK

EDIT: I'm not seeing SCC listed... does that mean I'm only connected to the primary?

[tonydobbs]

So this is interesting... I've been swapping antennas and messing around all day and nothing changed much. Just now I tried seeing how things performed with one SMA plug attached, then the other, then both... Now I'm on B12 with better values, and my speedtest results are much lower than before... But check this out... SCC appears.

Code: Select all

Current Time:  19164		Temperature: 44
Reset Counter: 1		Mode:        ONLINE         
System mode:   LTE        	PS state:    Attached     
LTE band:      B12    		LTE bw:      5 MHz   
LTE Rx chan:   5145		LTE Tx chan: 23145
LTE CA state:  INACTIVE    		LTE Scell band:B2     
LTE Scell bw:20 MHz  		LTE Scell chan:800
EMM state:     Registered     	Normal Service 
RRC state:     RRC Connected  
IMS reg state: No Srv  		
PCC RxM RSSI:  -70		RSRP (dBm):  -96
PCC RxD RSSI:  -67		RSRP (dBm):  -88
SCC RxM RSSI:  -88		RSRP (dBm):  -110
SCC RxD RSSI:  -86		RSRP (dBm):  -110
Tx Power:      0		TAC:         8B06 (35590)
RSRQ (dB):     -11.6		Cell ID:     0A15EA10 (169208336)
SINR (dB):      3.4
OK

[JimHelms]

Your signal stats are not that bad.

Yes, aiming an antenna can be a tedious process. I prefer to use the screen you reference above for the final tweaking.

Also pay attention to your RxM (main) and the RxD (div) RSRP while you are fine tuning. You want little difference between them.

It is not receiving the Secondary Carrier used for Carrier Aggregation. This could a tower limitation. Have you every noticed the secondary before.

[JimHelms]

Your slower speed is probably due to B12 only being 5MHz where B2 is 20Mhz.

[JimHelms]

And, the tower you are pointed towards, is an AT&T tower?

[tonydobbs]

Your signal stats are not that bad.

Yea they're not terrible by any means... I just was expecting the directional antenna to perform better than the tiny omnidirectional ones that came with the router.

Yes, aiming an antenna can be a tedious process. I prefer to use the screen you reference above for the final tweaking.

What do you use for your primary aiming? Do you just drive to the towers to confirm which one is for the provider and then use a compass to determine the heading?I had thought about that, but my cell provider is Verizon and my LTE home setup is ATT... I'll have to see if I can pop this SIM into a phone and get readings from that.

Also pay attention to your RxM (main) and the RxD (div) RSRP while you are fine turning. You want little difference between them.

Thanks for the tip!

It is not receiving the Secondary Carrier used for Carrier Aggregation. This could a tower limitation. Have you every noticed the secondary before.

I have, but mostly with the omni's. Now I'm seeing it with my directional, but speeds are down again.. in the 8mbps down territory now.

[tonydobbs]

Your slower speed is probably due to B12 only being 5MHz where B2 is 20Mhz.

That's what I was thinking, probably makes more sense to stay on B2.

And, the tower you are pointed towards, is an AT&T tower?

I believe so, but my methods were exceptionally crude. Do you have any recommendations on how to be sure?

[JimHelms]

I generally use antennasearch.com and download the "antenna" spreadsheet.

I then locate the tower/antenna that I am interested in, and the plot the Lat/long on all the antennas on google earth along with the location where my antenna will be mounted.

I then connect them with lines that reference the degree heading of each tower.

I then use a compass to aim the antenna at that heading, and use the AT!GSTATUS? to tweak them.

Here is a simple example:

Tower.jpg

Also notice the elevation readings which may favor using omni-directional antennas over a directional panel.

[tonydobbs]

Oh that makes much more sense than what I was doing... and just to be clear you're downloading the antenna spreadsheet not the tower spreadsheet? I was checking towers before...

Selecting antennas seems to give me almost 60 results, most of these I'm assuming are not LTE towers for me to connect to.

[tonydobbs]

I'm going to have to play with this some more later... I plotted my location to the tower and realized my antenna was off by ~8*, so I adjusted that and my stats have not changed outside of typical variance... The only other thing I could think of is if this is not my tower, and I'm picking up some tower in the distance that is not appearing on the antenna search... that seems crazy though.

Edit: After watching the stats for a few minutes they did get slightly better! But then I did a speed test 3 down .7 up... it's black magic, must be!

[tonydobbs]

Final update of the night... I can definitely say that both my average signal values and peak values have gone up since the aiming adjustment, albeit only slightly. With that being said I used the FindTower app on my phone, and it says that I'm aiming at a Verizon tower, and that the AT&T tower is in a completely different direction. I guess the next step is to take the antenna down, go into the middle of the field where there are 0 obstructions of any kind, and take some measurements to confirm which tower is correct.

[JimHelms]

Oh that makes much more sense than what I was doing... and just to be clear you're downloading the antenna spreadsheet not the tower spreadsheet? I was checking towers before...

Selecting antennas seems to give me almost 60 results, most of these I'm assuming are not LTE towers for me to connect to.

I generally have better luck with using the antenna results. It is not uncommon for two carriers to lease space on the same tower which is owned by a third party.

The only downside to using antenna search is their database if often not current.

[JimHelms]

Final update of the night... I can definitely say that both my average signal values and peak values have gone up since the aiming adjustment, albeit only slightly. With that being said I used the FindTower app on my phone, and it says that I'm aiming at a Verizon tower, and that the AT&T tower is in a completely different direction. I guess the next step is to take the antenna down, go into the middle of the field where there are 0 obstructions of any kind, and take some measurements to confirm which tower is correct.

That may well be the issue. There are also a few more tools mentioned on the STORE that may assist in locating the tower.

[tonydobbs]

That may well be the issue. There are also a few more tools mentioned on the STORE that may assist in locating the tower.

I'll have a play with some of the other tools and see what shows up. If I've been pointing to the wrong antenna that would be
pretty damn funny... Probably am.

[swwifty]

I have a few thoughts, and a few questions.

The antenna you're using has a really wide beamwidth, so to be honest, I doubt making slight little adjustments in the aiming will make a big difference. Should you still aim it as accurately as possible? yes, but the different will likely be subtle unless you are totally pointing the antenna 90 degrees off from the tower, or even 180 degrees off.

As Jim stated, the speed difference is due to band 12 having such low bandwith, vs band 2.

Here's the questions I have:

1. Did you lock your modem to band 2 only? I see in the first signal stats you shared, that there was no carrier aggregation (CA), and only band 2. I would unlock if you have locked it to only band 2, to see what signal stats look like, and share those results.

2. Can you share the results of the modem command 'at!lteinfo' ? This shows all the cells in your area, and we can see if there are more around, and if they are interfering and reducing your SINR (LTE is a interference limited system).

3. Can you share the signal stats of when you were using the omni antennas? Also, were you using the omnis inside or outside?

[tonydobbs]

I have a few thoughts, and a few questions.

The antenna you're using has a really wide beamwidth, so to be honest, I doubt making slight little adjustments in the aiming will make a big difference. Should you still aim it as accurately as possible? yes, but the different will likely be subtle unless you are totally pointing the antenna 90 degrees off from the tower, or even 180 degrees off.

I had thought this as well, which is why I was so surprised to be having issues aiming it, although now I'm starting to think that's because I'm trying to aim it at the wrong tower.

As Jim stated, the speed difference is due to band 12 having such low bandwith, vs band 2.

Yea, that part makes perfect sense to me, which is why I'm now wondering if I should have gotten an antenna optimized for a narrower range of frequencies. I didn't notice any speed gains from having CA between 12 and 2 as opposed to just band 2.

Here's the questions I have:

1. Did you lock your modem to band 2 only? I see in the first signal stats you shared, that there was no carrier aggregation (CA), and only band 2. I would unlock if you have locked it to only band 2, to see what signal stats look like, and share those results.

For all the stats I posted the modem was unlocked. I only locked it to band 2 afterwards because it was trying to stay on band 12 exclusively for some reason.

2. Can you share the results of the modem command 'at!lteinfo' ? This shows all the cells in your area, and we can see if there are more around, and if they are interfering and reducing your SINR (LTE is a interference limited system).

Sure, here are the current results. (not too bad all things considered, a bit better than last night)

Code: Select all

at!lteinfo
!LTEINFO: 
Serving:   EARFCN MCC MNC   TAC      CID Bd D U SNR PCI  RSRQ   RSRP   RSSI RXLV
              800 310 410 35590 0A15ED09  2 5 5   1  56 -10.7 -106.4  -75.1 --

IntraFreq:                                          PCI  RSRQ   RSRP   RSSI RXLV
                                                     56 -10.7 -106.4  -75.1 --
                                                    421 -16.7 -113.0  -87.0 --

InterFreq: EARFCN ThresholdLow ThresholdHi Priority PCI  RSRQ   RSRP   RSSI RXLV

GSM:       ThreshL ThreshH Prio NCC ARFCN 1900 valid BSIC RSSI RXLV

WCDMA:     UARFCN ThreshL ThreshH Prio PSC   RSCP  ECN0 RXLV

CDMA 1x:   Chan BC Offset Phase Str

CDMA HRPD: Chan BC Offset Phase Str


OK

3. Can you share the signal stats of when you were using the omni antennas? Also, were you using the omnis inside or outside?

Sure, here are the omni stats. They are the tiny ones that come with the device and are indoors, which makes their performance against my panel, all the more impressive, although right now they're

Code: Select all

at!lteinfo
!LTEINFO: 
Serving:   EARFCN MCC MNC   TAC      CID Bd D U SNR PCI  RSRQ   RSRP   RSSI RXLV
              800 310 410 35590 0A15ED09  2 5 5  -10  56 -17.0 -111.3  -68.4 --

IntraFreq:                                          PCI  RSRQ   RSRP   RSSI RXLV
                                                     56 -17.0 -111.3  -68.4 --
                                                    421 -16.9 -110.8  -85.3 --

InterFreq: EARFCN ThresholdLow ThresholdHi Priority PCI  RSRQ   RSRP   RSSI RXLV

GSM:       ThreshL ThreshH Prio NCC ARFCN 1900 valid BSIC RSSI RXLV

WCDMA:     UARFCN ThreshL ThreshH Prio PSC   RSCP  ECN0 RXLV

CDMA 1x:   Chan BC Offset Phase Str

CDMA HRPD: Chan BC Offset Phase Str


OK

Thanks again for your assistance, I still have a lot to learn!

[JimHelms]

You might also try some coax cables like the LMR200 or LMR400 that are designed for this type of application.

[tonydobbs]

You might also try some coax cables like the LMR200 or LMR400 that are designed for this type of application.

I was thinking that originally, but you're out of the LMR400 and the 5ft LMR200

[swwifty]

Based on the signal stats you shared (granted this was only a moment in time when the AT command was run, and not an average) with the antenna outside, I have a few observations, and ideas.

Observations.

1. The panel antenna is better. You can see this under the SNR (Actually, SINR in LTE, which is signal to noise plus interference) value. The little omnis have an SNR value of -10 (how they work like this is always amazing to me), and the panel has a SNR of 1. I suspect that the SINR is fluctuating a lot, if you run the at!lteinfo command a few times, or even at!gstatus.

2. Why is SINR so important in LTE? (Well really in any wireless system) It's because LTE has adaptive modulation rates, so the higher your SINR is, the higher modulation rate you can get, and hence more data throughput given the time frame.

3. SINR in LTE is: Signal (Desired Signal) divided by Background Noise + Interference (Interference in this case, is other cells broadcasting on the same frequency. LTE is as spectrally efficient as possible, and re-uses the same frequency as much as possible without causing too much of an impact to SINR, if possible.). Like I said earlier, LTE is a interference limited system, meaning that other cells will interfere with your desired cell signal, and cause your mobile/modem to have a poorer SINR ratio.

4. I explained all of this, because based on your signal stats from at!lteinfo command, it's clear that there is two cells that your antenna detects. The one your connected to which has a PCI (Physical Cell Identifier) of 56 and there's another that has a PCI of 421. You can see that RSRP is very close (This and SINR is the value you should care about most) and RSSI is also pretty close. This means basically that you are in the middle of two cells. It's possible, both of these cells are on the same tower, and you're just on the edge of two sectors, but it's tough to say. Based on your omni stats, and the PCIs though I suspect that one is physically closer than the other.

Ideas

You could try some high directional antennas specific to band two. This might help as the beamwidth of high gain antennas is much more narrow, so it will bring the desired signal in stronger, and weaken the signal that is coming from another direction. But then again, it's possible both of these cells are on the same physical tower, and by getting a higher gain antenna, it only gets worse. There's only one way to know for sure, and that's to try it, unfortunately.

Screen Shot 2019-02-11 at 9.37.31 AM.png

Above is a screenshot of my signal stats this AM. You can see there's a secondary cell (when I say secondary I mean, one I'm not connected to) on band 2 (PCI of 113) and a secondary cell on band 12 (also with a PCI of 113). You can see i'm connected to a serving cell of 213, and have a CA cell of 213 on band 12 (even though its only listed under InterFreq) Also notice that my SNR of 14 is the difference in RSRP between the two band 2 cells I detect. This means that interference from another cell is my limiting factor (this is a limited interference) as opposed to just background noise limits.

[swwifty]

I was thinking that originally, but you're out of the LMR400 and the 5ft LMR200

I don't think the cable here is an issue, 5ft of what you have already should cause minimal signal loss.

[tonydobbs]

Thanks for the info Swwifty! It seems that there was an aiming issue... I went out to the field with all the gear and tested different positions, only to discover that the signal became much stronger if I aimed the antenna in the opposite direction. I don't believe this tower showed up on my maps, but who am I to argue with signal strength values. Once I remount the antenna for this aiming I will consider switching the antenna out for the yagis or the 15dBi panels that are still out of stock

[swwifty]

Thanks for the info Swwifty! It seems that there was an aiming issue... I went out to the field with all the gear and tested different positions, only to discover that the signal became much stronger if I aimed the antenna in the opposite direction. I don't believe this tower showed up on my maps, but who am I to argue with signal strength values. Once I remount the antenna for this aiming I will consider switching the antenna out for the yagis or the 15dBi panels that are still out of stock

Did that translate to better performance with these antennas at your house? Once you knew the true location of the signal?