Slow Speeds on RM502Q-AE

[Haze69]

Hello,

I am hoping someone will be able to assist me or troubleshoot the slow speeds on my RM502Q-AE modem. I am consistently getting slower speeds with the modem vs inserting the sim in an iPhone.

Short description of my setup, recently moved into an RV to travel full time.

• 2 Quectel RM502Q-AE inside USB Enclosures (Visible/AT&T Tablet Plan)

• 2 Raspberry Pi 4's running GoldenOrb

• 1 Linux SFF PC with Speedify installed

• 2 Parsec Husky Pro 7 antennas installed on the roof of the RV ( https://www.mobilemusthave.com/Parsec-H ... _1408.html)

• Linksys WRT 3200 (To pull in the campground wifi)

The two antennas are each connected to a modem. The Wifi antenna cables are connected to the Linksys. Each modem is connected to a Raspberry Pi 4. Both Raspberry Pi 4's and Linksys are all connected to the SFF Linux PC running Speedify to bond the three connections. The bonded connection is shared to my desktop computer (wifi router/switch in the future once I am fully satisified with the setup)

The problem I am currently having is that I believe I should be seeing much higher speeds with this setup and I am currently only averaging 10-20mbs per connection. I understand these plans are lower priority and will vary depending on where I am however when I insert these sims into an iPhone/iPad and run a speed test - I am getting double or triple the numbers I am seeing on the modem. I am getting 2xCA and 3xCA on the modem (depending on the time of day) on GoldenOrb. I have the correct TTL setup for Visible to bypass the 5mbs restriction and the ATT modem has magic. Both plans are 4G only however I have had Visible hit 180mbs on an iPhone and the AT&T plan hit 100+ on the iPad.

Some of you may be wondering why I purchased 5G modems for 4G plans and the answer is because I wanted to future proof the setup and have the ability to upgrade to 5G plans. I also liked the 7xCA specs however I don't see the modems hitting anything past 3xCA.

Please let me know if you have any suggestions or thoughts on how to improve the speed.

Thank you

[mtl26637]

Are your 'per connection' results from a direct connection to the modems or are the results based on what you are seeing through speedify?

If you know how to send AT commands, can you post your results of AT+QENG="servingcell" ? First thing I would do if not done already is eliminate the aggregation for now and get some direct test speed results from each modem and take a look at your signal information.

[Dr-BroadBand]

WoW cool setup!!

If I had to guess the bottleneck is with The USB connection.

Would get with the boys at the The Wireless Haven Store and find a good router for that killer Modem you have.

Maybe something like this


https://thewirelesshaven.com/shop/routers/lte-rou ... 8ghz-wifi/

[Haze69]

Are your 'per connection' results from a direct connection to the modems or are the results based on what you are seeing through speedify?

Yeah, the per results were from connecting the ethernet directly to the Raspberry Pi's and running speedtest from:
-Google's Speedtest
-Speedtest.net
-Speedof.me

If you know how to send AT commands, can you post your results of AT+QENG="servingcell" ? First thing I would do if not done already is eliminate the aggregation for now and get some direct test speed results from each modem and take a look at your signal information.

Visible
AT+QENG="servingcell"
+QENG: "servingcell","NOCONN","LTE","FDD",311,480,2173102,406,5230,13,3,3,8906,-79,-12,-50,11,0,-160,-

OK

AT&T
AT+QENG="servingcell"
+QENG: "servingcell","NOCONN"
+QENG: "LTE","FDD",310,410,C92D217,473,66711,66,4,4,700A,-95,-13,-62,12,0,90,-
+QENG: "NR5G-NSA",310,410,65535,-32768,-32768,-32768

OK

[Haze69]

WoW cool setup!!

If I had to guess the bottleneck is with The USB connection.

Would get with the boys at the The Wireless Haven Store and find a good router for that killer Modem you have.

Maybe something like this


https://thewirelesshaven.com/shop/routers/lte-rou ... 8ghz-wifi/

Thank you I appreciate it, I was just trying to find the best way to create a stable network to work while on the road. Regarding the USB, I was under the impression that the Raspberry Pi 4's had full gigabit and running over USB 3.0 also allowed for gigabit connections. I initially had both modems plugged directly into the SFF Linux and ran them directly but I prefer having them seperate so that I can access them individually and see their stats. When they were connected directly the the Linux box, the OS seems to jumble their connections together and it made troubleshooting them a bit more difficult. If I need better hardware or a proper router for them to operate properly, I dont mind going down that route.

[mtl26637]

Signals look okay but was hoping it would show the CQI values but I forgot you have to run an AT command to first enable that. On another note, if you won't be using any 5G connections, I would switch the modem to run in 4G mode so its not searching for any 5G bands on any of the multiple antenna ports. You can do this via at command or easiest way is to enable it through the /modem/miscellaneous/Cellular Mode. At least it might help on the ATT setup as it shows its searching for the 5G signal.

Also, if possible, might take the setups close to the tower and run some tests. Would be the easiest and quickest way to rule out signal quality / antennas.

[Haze69]

Signals look okay but was hoping it would show the CQI values but I forgot you have to run an AT command to first enable that. On another note, if you won't be using any 5G connections, I would switch the modem to run in 4G mode so its not searching for any 5G bands on any of the multiple antenna ports. You can do this via at command or easiest way is to enable it through the /modem/miscellaneous/Cellular Mode. At least it might help on the ATT setup as it shows its searching for the 5G signal.

Also, if possible, might take the setups close to the tower and run some tests. Would be the easiest and quickest way to rule out signal quality / antennas.

Not sure if this helps but in GoldenOrb it shows Visible signal strength at 80% and AT&T signal strength at 87%

Last night around 1AM I was able to pull 112mbs from AT&T and 30mbs from Visible. During this hour I am barely able to pull .5-3mbs out of each. On AT&T I shouldn't be depriorized until 22gb and I am currently at 8gb for the month.

[Dr-BroadBand]

Would be helpful if we had all the number of your signal

[Adm1jtg]

Its not just about being de-prioritized it is also about server utilization. Sounds like at 1am there is low utilization and when you last tested there were many more people utilizing the server.

Also I have found different bands not only work faster and slower but also some are less utilized then others.

On top of that if you use a site like speedtest for your testing different speedtest sites can also give very different results.

Weather and temperature also affect signal

So really not that simple at all

[Haze69]

Would be helpful if we had all the number of your signal

What would be the best way to pull this information?

[Haze69]

Its not just about being de-prioritized it is also about server utilization. Sounds like at 1am there is low utilization and when you last tested there were many more people utilizing the server.

Also I have found different bands not only work faster and slower but also some are less utilized then others.

On top of that if you use a site like speedtest for your testing different speedtest sites can also give very different results.

Weather and temperature also affect signal

So really not that simple at all

I've seen devices on the market that run a speedtest on each band and lock on to the fastest band. Would I need to run the command to limit the modem to only use one band at a time, run a speed test to perform something similar? Also, wouldnt locking my band disable the ability to carrier aggregate to obtain faster speeds as it would only be locked onto the one band I set?

[mtl26637]

His signal was shown in the 'servingcell' AT command response.

Visible
AT+QENG="servingcell"
+QENG: "servingcell","NOCONN","LTE","FDD",311,480,2173102,406,5230,13,3,3,8906,-79,-12,-50,11,0,-160,-

^^^ RSRP:-79, RSRQ: -12, RSSI: -50, SINR: 11

AT&T
AT+QENG="servingcell"
+QENG: "LTE","FDD",310,410,C92D217,473,66711,66,4,4,700A,-95,-13,-62,12,0,90,-
+QENG: "NR5G-NSA",310,410,65535,-32768,-32768,-32768

^^^ RSRP:-95, RSRQ: -13, RSSI: -62, SINR: 12

The signal strengths shown in GoldenOrb main status page give a general idea but not very specific. For instance it may show 80-90% strength but that only on the primary band. You could have a very poor signal on secondary signals and it would still show the above percents.

I wouldn't think prioritization or congestion would be the issue either since you get very different speeds using your phone with the same sim. I guess it could be locking on to different bands and things. I still think a trip to the tower for some testing would be next step. Hate to say it after looking at the price of those antennas, but I wasn't too impressed with the return loss per their specs.

[Haze69]

His signal was shown in the 'servingcell' AT command response.

Visible
AT+QENG="servingcell"
+QENG: "servingcell","NOCONN","LTE","FDD",311,480,2173102,406,5230,13,3,3,8906,-79,-12,-50,11,0,-160,-

^^^ RSRP:-79, RSRQ: -12, RSSI: -50, SINR: 11

AT&T
AT+QENG="servingcell"
+QENG: "LTE","FDD",310,410,C92D217,473,66711,66,4,4,700A,-95,-13,-62,12,0,90,-
+QENG: "NR5G-NSA",310,410,65535,-32768,-32768,-32768

^^^ RSRP:-95, RSRQ: -13, RSSI: -62, SINR: 12

The signal strengths shown in GoldenOrb main status page give a general idea but not very specific. For instance it may show 80-90% strength but that only on the primary band. You could have a very poor signal on secondary signals and it would still show the above percents.

I wouldn't think prioritization or congestion would be the issue either since you get very different speeds using your phone with the same sim. I guess it could be locking on to different bands and things. I still think a trip to the tower for some testing would be next step. Hate to say it after looking at the price of those antennas, but I wasn't too impressed with the return loss per their specs.

I purchased these as they were recommend for "RV's" as one of the "best" you could purchase. I just assumed having something on the roof would provide better signal than something inside. Would you recommend paddles on the usb enclosure vs this antenna or a different type or antenna?

Also, I changed the modems to MBIM from QMI and didnt notice any difference. Would you guys recommend one over the other in this setup?

[Haze69]

Further troubleshooting. Took my T-Mobile sim out of my phone and put it inside the modem enclosure. Running speedtest against my wifes t-mobile phone continue to show the same pattern.

RM502Q-AE results: AVG 330mbs
Note 20 Ultra: AVG 415mbs

I am having trouble understanding why this modem is underperforming compared to the mobile phones on 3 carriers. If anyone has any other ideas to try, I'd really appreciate it.

[Adm1jtg]

All I can think of is the phones have a higher cat level then your cat 20 modem (very unlikely) or perhaps the phones are 4x4 antennas where as your modem setup isnt (also unlikely)

Kinda stumped as well

BTW I HATE you my system get 20 meg down on the best days

[mtl26637]

The RM502Q-AE should be just as capable I would assume. Did you try a visit to the tower to run the numbers? About the only difference I can think of is the antenna setups, which a trip to the tower would rule out. Last I looked I wasn't impressed with the numbers on your antennas so I wouldn't be surprised if they are the culprit. Trying to compact 4 antennas into a foot round diameter dome enclosure that spans from 600mhz to 2500mhz isn't ideal so I'd start with checking that. Not trying to slam the product as I don't know much about them but would be my next check.

Assuming your Tmo connection is using the typical B66 and n71 or n41, you might also try individual antennas on each port if you happen to have some cheap ones around. I assume the problem lies in the MIMO part of things with speeds that high.

[Haze69]

The RM502Q-AE should be just as capable I would assume. Did you try a visit to the tower to run the numbers? About the only difference I can think of is the antenna setups, which a trip to the tower would rule out. Last I looked I wasn't impressed with the numbers on your antennas so I wouldn't be surprised if they are the culprit. Trying to compact 4 antennas into a foot round diameter dome enclosure that spans from 600mhz to 2500mhz isn't ideal so I'd start with checking that. Not trying to slam the product as I don't know much about them but would be my next check.

Assuming your Tmo connection is using the typical B66 and n71 or n41, you might also try individual antennas on each port if you happen to have some cheap ones around. I assume the problem lies in the MIMO part of things with speeds that high.

Unfortunately I haven't gone to the tower as it isnt ideal at the moment. I would need to packup the whole RV and store everything travel ready. I will def take it to a tower when we are ready to go to the next RV resort.

For now I picked up a T-Mobile Home Internet as my T-Mobile phone has been picking up 300mbs-500mbs pretty consistently since I got here and the Home Internet has been working amazing tbh.

I also purchased a 5G tablet from AT&T yesterday so that my unlimited tablet line will pick up "5G". Also debating on switching from Visible to US Mobile to pick up Verizon 5G and lower latency. Havent tested AT&T yet but will update this thread once I get the new sim card installed in the modem.

[mtl26637]

I also purchased a 5G tablet from AT&T yesterday so that my unlimited tablet line will pick up "5G". Also debating on switching from Visible to US Mobile to pick up Verizon 5G and lower latency. Havent tested AT&T yet but will update this thread once I get the new sim card installed in the modem.

Careful, just yesterday I went over their 'NEW' (unlimited) cap of 75GB on the vzw side of things and its now slowed to a hault for the next 2 weeks until it renews. Still pissed off about there newest caps, is only reason I recently switched 3 lines to them. It works great and just like actual vzw service if you can deal with their unlimited caps..

[Haze69]

Careful, just yesterday I went over their 'NEW' (unlimited) cap of 75GB on the vzw side of things and its now slowed to a hault for the next 2 weeks until it renews. Still pissed off about there newest caps, is only reason I recently switched 3 lines to them. It works great and just like actual vzw service if you can deal with their unlimited caps..

Oh wow thats good to know. I opened a chat with their support 2 days ago and asked that very specific question "after 75gb do you have a hard throttle or will speeds only reduce on a congested tower" their answer was they don't have a hard throttle, only reduced speeds if the tower is congested after 75gb. If the tower is not congested I should continue to see normal speeds after 75gb.

[mtl26637]

I thought the same but logged in yesterday or day before and I had 'LTE' connection but no packets were permitted past my assigned IP from the provider. Also, its hard to tell exact limit numbers too as their software logs 'per day' so when you look at your used bandwidth numbers it shows the past 30 days counting from the current day and not necessarily from your monthly renew date. Only way to find total from what I can tell is to add up each single day from your renew date which is tedious and probably not very accurate. It could be a fluke on my end and it dropped connection for some other reason but that is too coincidental as I know my used data was up there near or slightly over that 75GB, just fyi.

EDIT: I'm an idiot. Went the easy route and tested in my main phone. Connection is there but speeds went from 100-175mbps down to 250kpbs down (Not a typo either kbps). So at least it does have 'a' connection, but NO WAY is it 'not' throttled as I'm very confident my tower isn't THAT congested, lol.

[Adm1jtg]

In the rooter/wifix firmware there is bandwidth monitor that might help you monitor your usage. It seems to start at the first of every month but you might be able to tweak that somehow.

Its found under services/bandwidth monitoring

There s also an addin called lede traffic usage monitor that also might help

[mtl26637]

Great idea for those that leave their setups alone and not redo / reinstall / reflash / reset things every other day lol. But yes, if I ever get to the point I can leave things alone it would be a great way to track it!