[NEBULA] NWA1123-AC-PRO Poor Throughput with even a Basic Config

Gi11yGi11y Member Posts: 10
edited April 21, 2020 9:36AM in Nebula AP

Hi All,

Hoping for some help on this.

Setup:

  • NWA1123-AC-PRO
  • GS1920-24HP


No rate limiting, captive portal, bandwidth management etc. Basic network set up on switch and basic set up on WAP.


Internet speed at router: 120 down / 40 up (average)

Internet speed plugged in to the switch: +/-5 same as speeds at router

Internet speed when connected to wireless: 35 down / 25 up (average)


Has anyone experienced any similar issues?


As a test I have not plugged the WAP directly in to the router as of yet but will be in due course.

«1

All Replies

  • Zyxel_PandaZyxel_Panda Zyxel Official Agent Posts: 88  mod
    edited February 3, 2020 9:33AM

    Hi @Gi11y

    What's the test client? laptop, mobile phone or else?

    Is that issue happened in 2.4GHz , 5GHz or both radios?

    Would you enable Zyxel support in Help> Support request for us?

    Please share me the Org./site name information.

    We'll help to check the setting and log for this issue.

    Thanks!

    Zyxel Panda

  • Gi11yGi11y Member Posts: 10
    edited January 30, 2020 2:16PM

    Hi,


    Thanks for your reply. This happens on both 2.4 and 5GHz.


    I have been testing with both a laptop and mobile device.


    Org name is "Home".


    Thanks for your help. Look forward to hearing from you.

  • Zyxel_PandaZyxel_Panda Zyxel Official Agent Posts: 88  mod

    Hi @Gi11y

    Is your site name: Kąty Wrocławskie ?

    Because there are other Org the same name: Home.

    Thanks!

  • Gi11yGi11y Member Posts: 10

    Hi,


    Very sorry, my mistake.


    Org name: Scott Gilliland

    Site: Home


    Thanks

  • Nebula_FredaNebula_Freda Zyxel Official Agent Posts: 241  mod

    Hello @Gi11y,

    Thanks for your information and I remove the test device MAC address because of the security issue. We'll help to check and update for you asap.

  • Nebula_FredaNebula_Freda Zyxel Official Agent Posts: 241  mod
    edited January 30, 2020 4:55PM

    Hi @Gi11y,

    After checking the log of the AP, here are some questions need your clarify.

    1. Would you share the test time for us and we can check the status between the AP and station at that time.

    2. Is there any other client using wireless when you did the test?

    3. Did you note about the connected signal strength and Tx/Rx rate? If you didn't check that, would you help to do the test again and check it in Current stations of AP information page (Access point> MonitorAccess> point> GIL-WAP1).


    4. We also noted from the log that the station will roaming between 2.4GHz and 5GHz, so would you open another SSID for 2.4GHz or 5GHz only to do the test for avoding any interference by station roaming.

    It would be appreciated if there's more information to share with us about the test.

    Thanks.

  • Gi11yGi11y Member Posts: 10

    Answers to your questions as number:


    1. I have ran fresh tests as part of your advice. Test times as per below.
    2. I was the only person in the premises at the time of test, any other devices would have been IOT which would transfer next to know data and require little to no bandwidth.
    3. TX rates noted in tests below.
    4. 2.4/5 network tests as per below:


    Test on normal "WIFI" SID

    • Time: 2030
    • Download: 30
    • Upload: 26
    • TX rate: 72
    • RX rate: 120
    • Signal strength: 60


    Created a SID called "2.4", set to only use 2.4GHz

    • Time: 2033
    • Download: 37
    • Upload: 27
    • TX rate: 72
    • RX rate: 143
    • Signal strength: 60


    Created a SID called "5", set to only use 5GHz

    • Time: 2035
    • Download: 63
    • Upload: 27
    • TX rate: 292
    • RX rate: 526
    • Signal strength: 72


    Little to no other devices on the "WIFI" SID at the time of testing. Only device with previous mac address noted was used on the test SID's.

    In comparison, I performed a speed test from a windows desktop pc, connected to the same switch as the WAP, results:


    • Time: 2040
    • Download: 132
    • Upload: 31


    I also plugged the WAP directly in to a port on the router and the speeds were much the same. These tests were completed at approx 2050.


    I have the left test SID's in place for you to review and I look forward to hearing more from you.


    Thanks

  • Nebula_FredaNebula_Freda Zyxel Official Agent Posts: 241  mod

    Hi @Gi11y,

    Thanks for your quickly reply for this test.

    After we checked your test result, here are conclusion for your reference.

    As the result shown, the upload should be no problem that they are almost the same as wired test result.

    Basically, the throughput result should be 50% of the data rated if the physical speed supported, but it would be impacted by interference, signal strength, or the test environment. So, the 2.4GHz test result is reasonable that it's about half of Tx rate. However, the 5GHz is result is only 63Mbps, so we checked the log of AP and found that there's interference of the CH36. We suggest to change the Channel width to 40MHz or 20MHz for 5GHz and it can be avoid interference from CH36 after doing DCS. And, the signal strength is not good for 5GHz test, some devices start to scan for a better signal when the RSSI is lower than -70dBm, so if you want to do the test again, please make sure the AP is line of sight and station can get RSSI better than -60dBm.


    BTW, may I know what's the test tool you used?

  • Gi11yGi11y Member Posts: 10

    Hi, Thanks for your reply.

    I dont necessarily agree with your conclusion.

    You say it is normal for the WAP to only get 50% of the rated speed on the line? This is not true, I work in the IT sector and although I wouldn't expect to see the exact same speed I would expect it to be very close. Usually 70-80% depending on size of underlying infrastructure. As this is in my home and a small network I would expect to have no more than a 10% loss.

    I have used this WAP for over 1x year previously and always had the same speed over wireless as I would have had wired. I believe it was the recent update to the access point which has caused this as I had not used it for approx 6x months while moving premises and in plugging it back in it updated.

    Is it possible to roll back as a test?

    I was less than 10ft from the WAP when I ran all of these tests. If you are telling me signal strength is poor then it would indicate a fault with the access point. I ran tests on two device at the same time also to ensure it was not the mobile device I was using for testing.

    You can see there is little to no interference on the wireless channels being used, especially on 5GHz and there was no brick wall/any object between me and the WAP.

  • RichardHanRichardHan Member Posts: 29  Freshman Member

    @Gi11y I think the interference that Freda means is from wireless part and not caused by the wall or any physical object. You can try if changing the using channel improves the wireless performance.

    Also, since the Ethernet only provides around 100 Mbps, you can simply use 20/40MHz channel width on 5G radio, which can provide better wireless transmission quality.

    There are also some IoT devices in my house(google dot, home, etc...), from my experience some devices (which don’t support new wireless standard)surely downgrade the wireless performance even they’re just simply connecting. Therefore in my house I left 2.4G for those IoT devices, and use 5G for my laptop/phone use.

    Sharing my experience with you.

Sign In to comment.