LTE Band specific network corruption 20 Nicholson Road, Subiaco Western Australia


Tower at 20 Nicholson Road, Subiaco Western Australia has failed LTE band 3 equipment. It allows devices to connect (with good signal strength) but websites either don't load or only partially load. Fails with Optus AC800s modem using APN connect/connectme/yesinternet, iphone 6, Lumia 640LTE w/windows 10 and Android phone with either of three Optus network SIMs but all work again after cycling the APN until LTE band 7 or LTE band 28 become the primary carrier. Any OS, browser or device then suffers partial internet access when handed over to LTE band 3 again. Get a LTE band that is not 3 and it works again. Has been like this since PM Wednesday 25 October. I reported it 26th and 27th through chat and 131344 on the 27th. Despite explaining in detail that this isolated to Band 3 and not device/browser specific 131344 tried to fob me off with 'no other issues reported' and "I've run [obviously worthless] diagnostics, there is nothing wrong". The site is still listed on the outages page as 'No issues reported'. Liars.

The chat consultant on the 26th tried to phone me, despite having the phone in my hand with full reception the phone never rang and she was diverted to voicemail, got through on the Telstra number. Received text message regarding missed call on the optus number later that night. 131344 promised to email through confirmation the issue would be looked into in 'five minutes' but that was a lie too.

Optus, stop lying about 'no issues reported', I can have 100% reproducability on this from various devices and can temporarily restore functioning internet by cycling APNs until I get off band 3 or travelling 1km up the road and getting band 3 from another tower. If I could blacklist band 3 I would, but most devices don't give that functionality and the AC800S has it locked behind the ENTERCND password.

Re: LTE Band specific network corruption 20 Nicholson Road, Subiaco Western Australia


Thanks for sending this through @Squizz, I'm sorry no one has been able to help previously. I'll pass on the information to our networks team to see if someone can look into this for you. Please send me a private message with your account number, full name, DOB and preferred contact number.

I’m part of the Yes Crowd team, employed by Optus to help run our online community. This guide explains how everything works on here and you should also check out our Community Guidelines.

Did we answer your question? Please mark it as a Accepted Solution and be generous with that Kudos button Smiley Happy

Re: LTE Band specific network corruption 20 Nicholson Road, Subiaco Western Australia


I'd tried multiple times to get Optus to acknowledge (then hopefully fix) this issue by chat and 13 13 44.

Within 12 hours of my post above the tower was finally marked faulty on the outages page.

I received a forum reply from Toomey, to which I replied and provided contact details. I also subsequently received a PM from

Gen_R "thanks for the extra details. I've had a look and can see that there is an outage on the tower which is investigated at present. We'll pass your feedback on though.".


The tower's listing on the outages page has been reverted to 'no issues reported' but of course the problem is not resolved. No one contacted me and clearly no effort was made to replicate the issue.


I suspect that most people are being defeated at the level 1 helpdesk. 'Clear cookies, reboot computer & modem' - 'Oh it works now', call closed. Unaware that rebooting the modem triggers a new connection which is likely to be band 28, ~20 minutes later the modem moves over to band 3 and the internet is unusable again.


Let me be very clear, the fault is band 3 routing in Optus' equipment. Connection strength is great, and some data can be loaded from some sites. All data can be loaded from all sites on the other bands, this occurs on multiple devices, OS and browers.


I myself have had fully working internet, but I expect the diagnosis and work-around is beyond the average user. I consider it to be unreasonable, unadvisable and unnecessary for the customer to need to take the following actions but it does demonstrate that the issue avoided by removing Optus band 3.


With March '17 firmware (Revision: NTG9X40C_11.14.08.17 r5297 ntgrbc-fwbuild4 2017/03/02 18:39:45) QPST DM Port VSP is open at allowing backup of NVRAM.

Reviewing the NVRAM manually is a nightmare as it is littered with Telstra, AT&T and Sprint settings (post-paid and pre-paid) in addition to the Optus settings. Fortunately Optus has a pretty unique combination of LTE bands.

Bit masking the 44 possible bands for the AC800s operating on 40, 28, 7 ,3 & 1, gives us:

1000 0000 0000 1000 0000 0000 0000 0000 0100 0101

Corresponding to nibbles:

8 0 0 8 0 0 0 0 4 5

Reordering to the way UINT64 would be stored in memory:

45 00 00 08 80

Searching the NVRAM backup with hex editor of choice finds the two Optus records (one for the capability of the device and the other for the preferred bands).

Changing the 5 nibble to 1 will remove band 3 from the device capabilities and preference and thereby unavailable irrespective of the AT!BANDS specified in the Android OS and SIM configuration.



Re: LTE Band specific network corruption 20 Nicholson Road, Subiaco Western Australia


Still no resolution on this.

Can anyone enlighten me on the geographical locations of: (MCC 505, MNC 2, TAC 59502)

LTE Cell ID: 27426059 PCI: 314

LTE Cell ID: 27426099 PCI: 314

LTE Cell ID: 44096530 PCI: 313

LTE Cell ID: 27426098 PCI: 313


I seem to be between LTE Cells that are using PCIs common to their neighbours.

Re: LTE Band specific network corruption 20 Nicholson Road, Subiaco Western Australia


I seem to be having similar issues with one of the Footscray (Melboune) towers.


Both my mobile and the AC800s intermittently drop out, It would be good to be able to stop the AC800s from connecting to the band on the tower.


Attempting to report this issue to support resulted in support telling me there were no issues reported against the tower ( no surprise since when you attempt to report an issue against a tower they tell you that no problem is reported against the tower and attempt to fob you off )


How did you manage to narrow down what exact band was causing the issues?


Also are you able to use to find the cell?



Re: LTE Band specific network corruption 20 Nicholson Road, Subiaco Western Australia


On the AC800S you need to telnet via the USB port. If you're using Windows on the PC try the program called putty.

Connect to your modem using the supplied USB cable, then telnet to and specify non standard port 5510.

You'll then be connected to the AT command shell of the modem.

As there is no welcome message, enter ATI and press enter to get some basic information on your modem.

To get the band, cell ID and PCI issue the command AT!GSTATUS? and press enter.

I'll need to know what phone you are using to help with field test mode for the phone. Android usually accesses it through system, about phone, repeated taps on internal memory until the hidden page appears.

If you haven't updated the AC800S to the latest firmware, I can guide you through removing the faulty band from the modems capabilities but it a sledge hammer to a thumbtack approach and isn't for the faint hearted.

After another call to 131344 yesterday and insisting on escalation and report number I've had two calls from the case manager today and further escalation to a specialist. The case manager read out all the symptoms on the parent case from other users including dates, which is pretty damning on the earlier calls claiming there were no other reports of issues with this tower.

Re: LTE Band specific network corruption 20 Nicholson Road, Subiaco Western Australia

[ Edited ]

Started working on this earlier today, but I am yet to go home and test it, It's just a simple nodejs app that pings google (prints if it's up) and logs out the modem status every second ( hoping to find a decent log of what it's doing when it's disconnecting and have some definitive proof of what's happening).


Also might try to find the password through brute force.


Nice job figuring out the NRAM, I already updated the firmware.


On the phone I used network cell info lite, to try to figure out what band it was on, and know what band it was on just after it dropped out this morning while browsing, I might keep monitoring this aswell.


Thanks for your help, fairly damming that they have fobbed you off when so many issues seem to exist.


Re: LTE Band specific network corruption 20 Nicholson Road, Subiaco Western Australia


I received an sms this afternoon "Yes! Your fault should now be fixed. If it is not resolved, reply 'HELP ME' and we'll get back to you!".

Needless to say, it is not fixed.

This means that even though I've provided very specific information that the fault is not experienced on other bands but the fault is continuous with MCC: 505 / MNC: 2 / TAC: 59502 / CellID: 27279370 / PCI: 171 / LTE Band: 3, on multiple devices and multiple SIMs, whoever marked it as fault rectified didn't actually have anyone come out and test the issue with a device connected to this cell on band 3. My devices/SIMS work with nearby Optus PCIs 318, 445, 380 and 148 on band 3 as long as I physically move out of the range of PCI 171.

The case manager summarised the fault reports from other customers regarding the tower over the last month as 'strong signal but no internet'.

Along with your experience, there is an almost identical report here:

This unfortunately means that the issue is at multiple sites across the network and Optus NOC diagnostics are unable to identify the fault.




Re: LTE Band specific network corruption 20 Nicholson Road, Subiaco Western Australia

[ Edited ]

After testing with my script I'm not convinced my issues are related to a similar problem, Testing last night with the modem connected via RNDIS to my laptop the internet appeared stable (according to my script, apart from the random restarts that the AC800s seems to suffer from). I'm going to do some more diagnostics on my end to try to obtain the exact reason I am having these issues.


I'll revert my setup, and check if my script returns the same result, if it does then I can assume that ICMP works while HTTP connections are being dropped.


Sorry for hijacking your thread but your advice was really useful for obtaining a way of testing this scenario, something I can say other avenues were not helpful in. 


It's frustrating because Netgear won't support the device and Optus don't have the capacity to support the device, I feel if you're unlucky enough to buy it, you're seriously on your own. 

Re: LTE Band specific network corruption 20 Nicholson Road, Subiaco Western Australia


Yes, this is a nasty one.

Ping works.

Also, from a working connection, I loaded pages that completely fail to load on band 3 (e.g., using firefox.

Using Firefox page information - media you can find the individual media element URLs. I saved a bunch of these of varying size (3Kb to 4Mb) and then used the URLs from the faulty band 3 - all the ones I tried individually worked!!!

This means the connection is there, but something in the routing equipment is tripping from either simultaneous http connections or resources on https. Perhaps volume of data from a full page is hitting faulty memory in the router or something like SPI is trying to interpret something it shouldn't like SAML / SSL?

If you can't even test using a script to download small image files, that'd mean we'd need the script to try and archive a webpage on a periodic basis.

It would be helpful if Optus would send some techs with proper diagnostic equipment to the field to replicated the issue then identify it.

Post a Reply
Top Contributors
25 Kudos
19 Kudos
11 Kudos
10 Kudos
8 Kudos