OpenWrt Forum Archive

Topic: Update on Linksys WRT1900AC support

The content of this topic has been archived between 16 Sep 2014 and 7 May 2018. Unfortunately there are posts – most likely complete pages – missing.

Hi,

I've a small problem with TFTP Firmware flashing.
I've got two WRT1900AC v1 where at both the power lamp was blinking, so I tried to TFTP flash them as described at the wiki.
With the first everything was ok, but the second one will not flash.
The router connects to the tftpd32 server but the transferrate stands at 0 B/s and nothing happens. After a while I get an 'Undefined error code' (0). Afterwards the NAND is erased, which is "ok" and then the NAND is written again which also leads to an "ok".

Can someone give an adwise how to flash the router now?
Thanks!

plamka00 wrote:

Hi,

I've a small problem with TFTP Firmware flashing.
I've got two WRT1900AC v1 where at both the power lamp was blinking, so I tried to TFTP flash them as described at the wiki.
With the first everything was ok, but the second one will not flash.
The router connects to the tftpd32 server but the transferrate stands at 0 B/s and nothing happens. After a while I get an 'Undefined error code' (0). Afterwards the NAND is erased, which is "ok" and then the NAND is written again which also leads to an "ok".

Can someone give an adwise how to flash the router now?
Thanks!

If only the WAN light and power light comes on, with the wan blinking, the bootloader is corrupted and the router is hard bricked.  If this is indeed what has occurred, you can try the bootloader recovery in the wiki.

You can verify this by using a usb - ttl cable, opening a serial terminal, and then watching the output.  If everything appears as it should, it's not hard bricked.  I've had a WRT1900 with the same issue [prior to nitroshit's bootloader recovery contribution] and it was still within the return period, so I exchanged it.

(Last edited by JW0914 on 7 Oct 2015, 02:01)

Hello,
I am still running stock firmware on my WRT1900AC V1.  I've flashed Gargoyle in the past on a Netgear router but never OpenWRT.  I've found a few screenshots to get an idea of what the interface is like but I think some showed LUCI, etc.  Gargoyle was very usable but it has not been updated since August and does not have the CC final build as the base.  Mainly I would like SOME decent logging to determine if when I have issues, is it the modem/isp connection or the router is having problems or the device I was on at that time...  The stock firmware is "pretty" but I'm amazed how limited it is.  Would the CC final build be generally as stable as stock 1.1.10.167514 release?  I am technical but I don't have lots of time to tinker with a million settings.  I would like to be able to flash it, configure some basics - maybe QoS for VoIP but leave it at the defaults to the extent possible.  I have a fast connection (COX 150 Mbps) so I am mostly concerned with keeping a reliable connection and a strong signal (transmit power, etc.) rather than tweaking the router to the max.  If I flash CC final, is the web interface decent or do I need to install another 3rd party UI so I don't have to drop to a command line except in a worst case scenario?  I ordered a USB to TTL cable for $6.99 off Amazon just in case.  Any input would be appreciated. smile

RichMD wrote:

Hello,
I am still running stock firmware on my WRT1900AC V1.  I've flashed Gargoyle in the past on a Netgear router but never OpenWRT.  I've found a few screenshots to get an idea of what the interface is like but I think some showed LUCI, etc.  Gargoyle was very usable but it has not been updated since August and does not have the CC final build as the base.  Mainly I would like SOME decent logging to determine if when I have issues, is it the modem/isp connection or the router is having problems or the device I was on at that time...  The stock firmware is "pretty" but I'm amazed how limited it is.  Would the CC final build be generally as stable as stock 1.1.10.167514 release?  I am technical but I don't have lots of time to tinker with a million settings.  I would like to be able to flash it, configure some basics - maybe QoS for VoIP but leave it at the defaults to the extent possible.  I have a fast connection (COX 150 Mbps) so I am mostly concerned with keeping a reliable connection and a strong signal (transmit power, etc.) rather than tweaking the router to the max.  If I flash CC final, is the web interface decent or do I need to install another 3rd party UI so I don't have to drop to a command line except in a worst case scenario?  I ordered a USB to TTL cable for $6.99 off Amazon just in case.  Any input would be appreciated. smile

I'd recommend RC3 over CC Final until a few bugs get ironed out

Hi,

Which OpenWrt Firmware can I use for
Linksys WRT1900ACS-EU?

It has 1.6Ghz and 512MB RAM

Thanks

JW0914 wrote:
plamka00 wrote:

Hi,

I've a small problem with TFTP Firmware flashing.
I've got two WRT1900AC v1 where at both the power lamp was blinking, so I tried to TFTP flash them as described at the wiki.
With the first everything was ok, but the second one will not flash.
The router connects to the tftpd32 server but the transferrate stands at 0 B/s and nothing happens. After a while I get an 'Undefined error code' (0). Afterwards the NAND is erased, which is "ok" and then the NAND is written again which also leads to an "ok".

Can someone give an adwise how to flash the router now?
Thanks!

If only the WAN light and power light comes on, with the wan blinking, the bootloader is corrupted and the router is hard bricked.  If this is indeed what has occurred, you can try the bootloader recovery in the wiki.

You can verify this by using a usb - ttl cable, opening a serial terminal, and then watching the output.  If everything appears as it should, it's not hard bricked.  I've had a WRT1900 with the same issue [prior to nitroshit's bootloader recovery contribution] and it was still within the return period, so I exchanged it.

Hi! The WAN is not blinking at all - its dark. When I start the device only the power LED is blinking. I've checked again the booting log and it is almost the same like that shown in the wiki. The biggest difference is at the "auto_recovery" section:

BootROM 1.20
Booting from NAND flash
Step 1: First phase of PEX-PIPE Configuration
Step 2: Configure the desire PIN_PHY_GEN
Step 3 QSGMII enable
Step 4: Configure SERDES MUXes
Step 5: Activate the RX High Impedance Mode
Step 6: [PEX-Only] PEX-Main configuration (X4 or X1)
Step 6.2: [PEX-Only] PCI Express Link Capabilities
Step 7: [PEX-X4 Only] To create PEX-Link
Steps 7,8,9,10 and 11
Steps 12: [PEX-Only] Last phase of PEX-PIPE ConfigurationSteps 13: Wait 

15ms before checking resultsSteps 14: [PEX-Only]  In order to 

configureSteps 15: [PEX-Only]  In order to configureSteps 16: [PEX-Only] 

Training Enablestep 17: max_if= 0x7
step 17:  PEX0  pexUnit= 0
PEX0 : Detected No Link. Status Reg(0x00041A64) = 0x00000002
step 17:  PEX1  pexUnit= 0
step 17:  PEX2  pexUnit= 0
** Link is Gen1, check the EP capability
 --> 0040
mvPexConfigRead: return addr=0x%x0040
 --> 5001
 --> 5001
 --> 7005
 --> 7005
 --> 0010
 --> DC12
Gen2 client!
step 17:  PEX3  pexUnit= 0
PEX3 : Detected No Link. Status Reg(0x0004DA64) = 0x00000001
step 17:  PEX4  pexUnit= 1
PEX4 : Detected No Link. Status Reg(0x00081A64) = 0x00000001
step 17:  PEX5  pexUnit= 1
PEX5 : Detected No Link. Status Reg(0x00085A64) = 0x00000001
step 17:  PEX6  pexUnit= 1
PEX6 : Detected No Link. Status Reg(0x00089A64) = 0x00000001
                                                            DDR3 

Training Sequence - Ver 4.5.DDR3 Training Sequence - Static MC Init
DDR3 Training Sequence - HW Training Procedure
DDR3 Training Sequence - Switching XBAR Window to FastPath Window
BootROM: Image checksum verification PASSED

 __   __                      _ _
|  \/  | __ _ _ ____   _____| | |
| |\/| |/ _` | '__\ \ / / _ \ | |
| |  | | (_| | |   \ V /  __/ | |
|_|  |_|\__,_|_|    \_/ \___|_|_|
         _   _     ____              _
        | | | |   | __ )  ___   ___ | |_
        | | | |___|  _ \ / _ \ / _ \| __|
        | |_| |___| |_) | (_) | (_) | |_
         \___/    |____/ \___/ \___/ \__|
 ** LOADER **


U-Boot 2011.12 (Feb 06 2014 - 17:14:13) Marvell version: v2011.12 

2013_Q1.2

Boot version:v1.3.25

Board: RD-AXP-GP rev 1.0
SoC:   MV78230 B0
       running 2 CPUs
       Custom configuration
CPU:   Marvell PJ4B (584) v7 (Rev 2) LE
       CPU 0
       CPU    @ 1200 [MHz]
       L2     @ 600 [MHz]
       TClock @ 250 [MHz]
       DDR    @ 600 [MHz]
       DDR 32Bit Width, FastPath Memory Access
       DDR ECC Disabled
DRAM:  256 MiB

Map:   Code:            0x0fea7000:0x0ff5e2d4
       BSS:             0x0ffefd80
       Stack:           0x0f9a6ef8
       Heap:            0x0f9a7000:0x0fea7000

NAND:  Spansion 1Gb(ID=F101) 128 MiB
MMC:   MRVL_MMC: 0
Bad block table found at page 65472, version 0x01
Bad block table found at page 65408, version 0x01


#### auto_recovery ####
[u_env] get auto_recovery == no
[u_env] get auto_recovery == no
[u_env] get boot_part == 2
[u_env] get boot_part_ready == 3
auto_recovery enabled:0, boot_part:2, boot_part_ready:3

PEX 0.0(0): Detected No Link.
PEX 0.1(1): Root Complex Interface, Detected Link X1, GEN 1.1
PEX 0.2(2): Root Complex Interface, Detected Link X1, GEN 2.0
PEX 0.3(3): Detected No Link.
PEX 1.0(4): Detected No Link.
PEX 1.1(5): Detected No Link.
PEX 1.2(6): Detected No Link.

boot_end Offset: 0x100000
u_env_off Offset: 0x100000
s_env_off Offset: 0x140000
devinfo Offset: 0x900000

===================
total_badCount: 0
boot_badCount: 0
u_env_badCount: 0
s_env_badCount: 0
buff_badCount: 0
===================

FPU initialized to Run Fast Mode.
USB 0: Host Mode
USB 1: Host Mode
USB 2: Device Mode
Modules Detected:
mvEthE6171SwitchBasicInit finished
Net:   mvSysNetaInit enter
set port 0 to rgmii enter
set port 1 to rgmii enter
egiga0 [PRIME], egiga1
modify Phy Status
auto_recovery_check changes bootcmd: run altnandboot
Hit any key to stop autoboot:  0

NAND read: device 0 offset 0x3200000, size 0x400000
 4194304 bytes read: OK
Wrong Image Format for bootm command
ERROR: can't get kernel image!
deniz2304 wrote:

Hi,

Which OpenWrt Firmware can I use for
Linksys WRT1900ACS-EU?

It has 1.6Ghz and 512MB RAM

Thanks

No Way! Is there a version of the WRT1900AC with a 1.6Ghz CPU now? Where did you see the clock speed of the processor is it on the packaging/box?

And does it support DFS??? (Channels above 48) on the 5Ghz band in EU?

Could you post the results of http://<IP-Adress of you router>/sysinfo.cgi

(Last edited by Nijntje on 7 Oct 2015, 08:51)

Nijntje wrote:
deniz2304 wrote:

Hi,

Which OpenWrt Firmware can I use for
Linksys WRT1900ACS-EU?

It has 1.6Ghz and 512MB RAM

Thanks

No Way! Is there a version of the WRT1900AC with a 1.6Ghz CPU now? Where did you see the clock speed of the processor is it on the packaging/box?

And does it support DFS??? (Channels above 48) on the 5Ghz band in EU?

Could you post the results of http://<IP-Adress of you router>/sysinfo.cgi

https://www.secomp.de/linksys-wrt1900ac … 82448.html

http://www.amazon.de/Linksys-WRT1900ACS … B015E1CLHE

Hello guys.
I installed OpenWrt (Mamba) onto my WRT1900AC v1 a few days ago and it worked great but right now my 5GHz radio won't work no matter what. I even tried to reboot the router (remotely) and restart the "network" startup script and it still won't work.

I am not sure but if it does matter I installed luci-app-qos a few minutes before I found out, but I wasn't connecting to my 5GHz wifi so I am not quite sure.

I appreciate any help. smile

On a side note, I found that there's two version of OpenWrt for WRT1900ac v1 (http://wiki.openwrt.org/toh/linksys/wrt1900ac), Caiman and Mamba. I chose Mamba because I didn't see Caiman on the rest of the page. Should have I installed Caiman, or Mamba is better?

(Last edited by ngbeslhang on 7 Oct 2015, 10:12)

ngbeslhang wrote:

Hello guys.
I installed OpenWrt (Mamba) onto my WRT1900AC v1 a few days ago and it worked great but right now my 5GHz radio won't work no matter what. I even tried to reboot the router (remotely) and restart the "network" startup script and it still won't work.

I am not sure but if it does matter I installed luci-app-qos a few minutes before I found out, but I wasn't connecting to my 5GHz wifi so I am not quite sure.

I appreciate any help. smile

What channels are you using on 5Ghz? I have experienced that  the 5Ghz radio refuses to work when using channels other than 36, 40, 44 or 48.


On a side note, I found that there's two version of OpenWrt for WRT1900ac v1 (http://wiki.openwrt.org/toh/linksys/wrt1900ac), Caiman and Mamba. I chose Mamba because I didn't see Caiman on the rest of the page. Should have I installed Caiman, or Mamba is better?

DO NOT INSTALL THE CAIMAN BUILD ON YOUR WRT1900AC YOU WILL BRICK IT!!!

Caiman is the codename / build for the WRT1200AC
Mamba is the codename / build for the WRT1900AC-V1

If you've installed the Mamba build on you're WRT1900AC-V1 you've installed the right build.

deniz2304 wrote:
Nijntje wrote:
deniz2304 wrote:

Hi,

Which OpenWrt Firmware can I use for
Linksys WRT1900ACS-EU?

It has 1.6Ghz and 512MB RAM

Thanks

No Way! Is there a version of the WRT1900AC with a 1.6Ghz CPU now? Where did you see the clock speed of the processor is it on the packaging/box?

And does it support DFS??? (Channels above 48) on the 5Ghz band in EU?

Could you post the results of http://<IP-Adress of you router>/sysinfo.cgi

https://www.secomp.de/linksys-wrt1900ac … 82448.html

http://www.amazon.de/Linksys-WRT1900ACS … B015E1CLHE

I'm having a really hard time believing Linksys has released a third version of the WRT1900AC with a 1.6Ghz CPU, on the other hand the Marvell Armada 385 CPU is capable of running on that speed and the  "WRT1900ACS" seems to have EAN codes that differ from the "WRT1900AC:
WRT1900ACS    EAN code     0745883701797, 4260184666171

WRT1900AC      EAN code        0745883598649, 4260184663101, 0745883598601

Which OpenWrt Firmware can I use for Linksys WRT1900ACS-EU v1 or v2?

deniz2304 wrote:

Which OpenWrt Firmware can I use for Linksys WRT1900ACS-EU v1 or v2?

That depends on the CPU/chipset that the WRT1900ACS-EU uses if the WRT1900ACS-EU uses the same components as the WRT1900AC-V2 and the only difference is that the CPU is clocked higher, then i guess you can use the V2 firmware.

You could have a look at the infopage at http://192.168.1.1/sysinfo.cgi to find out what hardware the WRT1900ACS-EU is running on, and maybe post the results in this topic.

I have the Linksys wrt1900ACS not it will get soon.
If I have it, I'll post the  http://192.168.1.1/sysinfo.cgi

deniz2304 wrote:

I have the Linksys wrt1900ACS not it will get soon.
If I have it, I'll post the  http://192.168.1.1/sysinfo.cgi

Great, looking forward to it  i'm curious as to how this new router differs from the WRT1900AC-V2.

plamka00 wrote:
plamka00 wrote:

Hi,

I've a small problem with TFTP Firmware flashing.
I've got two WRT1900AC v1 where at both the power lamp was blinking, so I tried to TFTP flash them as described at the wiki.
With the first everything was ok, but the second one will not flash.
The router connects to the tftpd32 server but the transferrate stands at 0 B/s and nothing happens. After a while I get an 'Undefined error code' (0). Afterwards the NAND is erased, which is "ok" and then the NAND is written again which also leads to an "ok".

Can someone give an adwise how to flash the router now?
Thanks!

Hi! The WAN is not blinking at all - its dark. When I start the device only the power LED is blinking. I've checked again the booting log and it is almost the same like that shown in the wiki. The biggest difference is at the "auto_recovery" section:

BootROM 1.20
Booting from NAND flash
Step 1: First phase of PEX-PIPE Configuration
Step 2: Configure the desire PIN_PHY_GEN
Step 3 QSGMII enable
Step 4: Configure SERDES MUXes
Step 5: Activate the RX High Impedance Mode
Step 6: [PEX-Only] PEX-Main configuration (X4 or X1)
Step 6.2: [PEX-Only] PCI Express Link Capabilities
Step 7: [PEX-X4 Only] To create PEX-Link
Steps 7,8,9,10 and 11
Steps 12: [PEX-Only] Last phase of PEX-PIPE ConfigurationSteps 13: Wait 

15ms before checking resultsSteps 14: [PEX-Only]  In order to 

configureSteps 15: [PEX-Only]  In order to configureSteps 16: [PEX-Only] 

Training Enablestep 17: max_if= 0x7
step 17:  PEX0  pexUnit= 0
PEX0 : Detected No Link. Status Reg(0x00041A64) = 0x00000002
step 17:  PEX1  pexUnit= 0
step 17:  PEX2  pexUnit= 0
** Link is Gen1, check the EP capability
 --> 0040
mvPexConfigRead: return addr=0x%x0040
 --> 5001
 --> 5001
 --> 7005
 --> 7005
 --> 0010
 --> DC12
Gen2 client!
step 17:  PEX3  pexUnit= 0
PEX3 : Detected No Link. Status Reg(0x0004DA64) = 0x00000001
step 17:  PEX4  pexUnit= 1
PEX4 : Detected No Link. Status Reg(0x00081A64) = 0x00000001
step 17:  PEX5  pexUnit= 1
PEX5 : Detected No Link. Status Reg(0x00085A64) = 0x00000001
step 17:  PEX6  pexUnit= 1
PEX6 : Detected No Link. Status Reg(0x00089A64) = 0x00000001
                                                            DDR3 

Training Sequence - Ver 4.5.DDR3 Training Sequence - Static MC Init
DDR3 Training Sequence - HW Training Procedure
DDR3 Training Sequence - Switching XBAR Window to FastPath Window
BootROM: Image checksum verification PASSED

 __   __                      _ _
|  \/  | __ _ _ ____   _____| | |
| |\/| |/ _` | '__\ \ / / _ \ | |
| |  | | (_| | |   \ V /  __/ | |
|_|  |_|\__,_|_|    \_/ \___|_|_|
         _   _     ____              _
        | | | |   | __ )  ___   ___ | |_
        | | | |___|  _ \ / _ \ / _ \| __|
        | |_| |___| |_) | (_) | (_) | |_
         \___/    |____/ \___/ \___/ \__|
 ** LOADER **


U-Boot 2011.12 (Feb 06 2014 - 17:14:13) Marvell version: v2011.12 

2013_Q1.2

Boot version:v1.3.25

Board: RD-AXP-GP rev 1.0
SoC:   MV78230 B0
       running 2 CPUs
       Custom configuration
CPU:   Marvell PJ4B (584) v7 (Rev 2) LE
       CPU 0
       CPU    @ 1200 [MHz]
       L2     @ 600 [MHz]
       TClock @ 250 [MHz]
       DDR    @ 600 [MHz]
       DDR 32Bit Width, FastPath Memory Access
       DDR ECC Disabled
DRAM:  256 MiB

Map:   Code:            0x0fea7000:0x0ff5e2d4
       BSS:             0x0ffefd80
       Stack:           0x0f9a6ef8
       Heap:            0x0f9a7000:0x0fea7000

NAND:  Spansion 1Gb(ID=F101) 128 MiB
MMC:   MRVL_MMC: 0
Bad block table found at page 65472, version 0x01
Bad block table found at page 65408, version 0x01


#### auto_recovery ####
[u_env] get auto_recovery == no
[u_env] get auto_recovery == no
[u_env] get boot_part == 2
[u_env] get boot_part_ready == 3
auto_recovery enabled:0, boot_part:2, boot_part_ready:3

PEX 0.0(0): Detected No Link.
PEX 0.1(1): Root Complex Interface, Detected Link X1, GEN 1.1
PEX 0.2(2): Root Complex Interface, Detected Link X1, GEN 2.0
PEX 0.3(3): Detected No Link.
PEX 1.0(4): Detected No Link.
PEX 1.1(5): Detected No Link.
PEX 1.2(6): Detected No Link.

boot_end Offset: 0x100000
u_env_off Offset: 0x100000
s_env_off Offset: 0x140000
devinfo Offset: 0x900000

===================
total_badCount: 0
boot_badCount: 0
u_env_badCount: 0
s_env_badCount: 0
buff_badCount: 0
===================

FPU initialized to Run Fast Mode.
USB 0: Host Mode
USB 1: Host Mode
USB 2: Device Mode
Modules Detected:
mvEthE6171SwitchBasicInit finished
Net:   mvSysNetaInit enter
set port 0 to rgmii enter
set port 1 to rgmii enter
egiga0 [PRIME], egiga1
modify Phy Status
auto_recovery_check changes bootcmd: run altnandboot
Hit any key to stop autoboot:  0

NAND read: device 0 offset 0x3200000, size 0x400000
 4194304 bytes read: OK
Wrong Image Format for bootm command
ERROR: can't get kernel image!

Rename the firmware image to blk-mamba.128mb.img and TFTP flash it again. 

Some batches of the the WRT1900ac will not accept another name when flashing via TFTP (mine happens to be one of them), and is why the footnote of the the image name was added to the TFTP flash section of the Wiki.  I'd also set the server IP to .20 as I've been unable to get it to flash at .2

Successful TFTP Flash Output

Marvell>> setenv firmware_name blk-mamba.128mb.img
Marvell>> setenv ipaddr 192.168.1.1
Marvell>> setenv serverip 192.168.1.20
Marvell>> run flash_pri_image
mvNetaSpeedDuplexSet
Using egiga0 device
TFTP from server 192.168.1.20; our IP address is 192.168.1.1
Filename 'blk-mamba.128mb.img'.
Load address: 0x2000000
Loading: T #########################################################
####################################################################
#####         ##########################################################
done
Bytes transferred = 33292288 (1fc0000 hex)
NAND erase: device 0 offset 0xa00000, size 0x4000000
Erasing at 0x49e0000 -- 100% complete.
OK
NAND write: device 0 offset 0xa00000, size 0x1fc0000
33292288 bytes written: OK

(Last edited by JW0914 on 7 Oct 2015, 19:47)

ngbeslhang wrote:

On a side note, I found that there's two version of OpenWrt for WRT1900ac v1 (http://wiki.openwrt.org/toh/linksys/wrt1900ac), Caiman and Mamba. I chose Mamba because I didn't see Caiman on the rest of the page. Should have I installed Caiman, or Mamba is better?

Trunk
For initial flashing use
openwrt-mvebu-armada-xp-linksys-mamba-squashfs-factory.img

For sysupgrade-ing use
openwrt-mvebu-armada-xp-linksys-mamba-squashfs-sysupgrade.tar

WRT1900AC(v1) - Mamba
WRT1900AC(v2) - Cobra
WRT1200AC - Caiman

davidc502 wrote:

I've been monitoring Wifi temperature, and noticed 5Ghz really doesn't heat up the hardware. However, when I ran a report on Wifi Temperature for the past 24 hours I noticed a increase in temperature that lasted until 5am. My wifi and I were asleep, but my daughter couldn't sleep, and was watching various shows over her 2.4Ghz ChromeCast.

2.4Ghz out to devices
Mbps

Wifi hardware temperature
Mbps

I think this weekend I'm going to do some wifi download tests (hammer it hard), and see what the temperature does for both 5Ghz and 2.4Ghz.

It just seems wifi hardware heated up a lot just for a tv stream of around 5-6mbps.

I have never opened up my router. However, looking through the ventilation holes in the case it looks like there is a 3 inch (75mm) x another dimension (which I can't see).

The point is this. It looks like that Linksys used a huge heat sink over several IC's. While the heat sink will transfer heat OUT; a huge heat sink can also transfer heat IN.  The heat sink will average the various temperatures across various IC's.

Only when there is a great short term overload will specific chipsets register a temperature rise. Gradual temperature changes will be spread across the heat sink; making it harder to identify the specific IC causing the issue.

My direct point is this. Did the WIFI chip cause the increase or did the CPU cause the increase and transfer heat to WIFI?

Using a huge heat sink over several IC's is really dumb design.

Rick

RickStep wrote:

I have never opened up my router. However, looking through the ventilation holes in the case it looks like there is a 3 inch (75mm) x another dimension (which I can't see).

The point is this. It looks like that Linksys used a huge heat sink over several IC's. While the heat sink will transfer heat OUT; a huge heat sink can also transfer heat IN.  The heat sink will average the various temperatures across various IC's.

Only when there is a great short term overload will specific chipsets register a temperature rise. Gradual temperature changes will be spread across the heat sink; making it harder to identify the specific IC causing the issue.

My direct point is this. Did the WIFI chip cause the increase or did the CPU cause the increase and transfer heat to WIFI?

Using a huge heat sink over several IC's is really dumb design.

Rick

Well, even if they were separate heatsinks, they share a very limited amount of air to cool them. So separate heatsinks would still end up equalizing the temp of the various chips, just not as fast. And the resulting multiple heatsinks would be smaller overall, and probably more expensive (both parts cost and assembly cost)

like everything else, it's a matter of trade-offs.

dlang wrote:
RickStep wrote:

I have never opened up my router. However, looking through the ventilation holes in the case it looks like there is a 3 inch (75mm) x another dimension (which I can't see).

The point is this. It looks like that Linksys used a huge heat sink over several IC's. While the heat sink will transfer heat OUT; a huge heat sink can also transfer heat IN.  The heat sink will average the various temperatures across various IC's.

Only when there is a great short term overload will specific chipsets register a temperature rise. Gradual temperature changes will be spread across the heat sink; making it harder to identify the specific IC causing the issue.

My direct point is this. Did the WIFI chip cause the increase or did the CPU cause the increase and transfer heat to WIFI?

Using a huge heat sink over several IC's is really dumb design.

Rick

Well, even if they were separate heatsinks, they share a very limited amount of air to cool them. So separate heatsinks would still end up equalizing the temp of the various chips, just not as fast. And the resulting multiple heatsinks would be smaller overall, and probably more expensive (both parts cost and assembly cost)

like everything else, it's a matter of trade-offs.

You are obviously NOT a manufacturing type.

Here is how it works.

1.  Device Engineering designs a circuit to do "something".
2.  The device might exceed the temperature rating
3.  The device manufacturer is asked if the IC can be kept cool.
4.  The IC Manufacturer says the device can be kept cool with a heat sink.
5.  Device manufacturer asks how much.
6.  IC manufacturer provides a cost of 20% over bare IC.
7.  Device manufacturer responds with 3% or the device will be designed out of the product.
8.  The device manufacture agrees.

I really can't believe that Linksys does not use its "blackmail tactics" to propel their products forward.

(Last edited by RickStep on 8 Oct 2015, 00:07)

RickStep wrote:
dlang wrote:
RickStep wrote:

I have never opened up my router. However, looking through the ventilation holes in the case it looks like there is a 3 inch (75mm) x another dimension (which I can't see).

The point is this. It looks like that Linksys used a huge heat sink over several IC's. While the heat sink will transfer heat OUT; a huge heat sink can also transfer heat IN.  The heat sink will average the various temperatures across various IC's.

Only when there is a great short term overload will specific chipsets register a temperature rise. Gradual temperature changes will be spread across the heat sink; making it harder to identify the specific IC causing the issue.

My direct point is this. Did the WIFI chip cause the increase or did the CPU cause the increase and transfer heat to WIFI?

Using a huge heat sink over several IC's is really dumb design.

Rick

Well, even if they were separate heatsinks, they share a very limited amount of air to cool them. So separate heatsinks would still end up equalizing the temp of the various chips, just not as fast. And the resulting multiple heatsinks would be smaller overall, and probably more expensive (both parts cost and assembly cost)

like everything else, it's a matter of trade-offs.

You are obviously NOT a manufacturing type.

Here is how it works.

1.  Device Engineering designs a circuit to do "something".
2.  The device might exceed the temperature rating
3.  The device manufacturer is asked if the IC can be kept cool.
4.  The IC Manufacturer says the device can be kept cool with a heat sink.
5.  Device manufacturer asks how much.
6.  IC manufacturer provides a cost of 20% over bare IC.
7.  Device manufacturer responds with 3% or the device will be designed out of the product.
8.  The device manufacture agrees.

I really can't believe that Linksys does not use its "blackmail tactics" to propel their products forward.

It looks like the layout of the WRT1900AC main board is faulty because:

1.  IC's that need to be separated are huddled under a communal heat sink.
2.  NONE of the devices on the Linksys mother board NEED to be clustered under ONE heat sink.

A few years back I worked with a company that provided wireless cable technology.  The boxes were 19" x 24" and worked in the GHz. range.

Today the boxes are half the size of a slice of bread; are efficient; etc.

MANUFACTURING costs WILL be driven down unless you are @dlang who thinks I can't back up my post.

Man I wish this PunBB forum software had filtering settings like vBulletin et al.

gonzlobo wrote:

Man I wish this PunBB forum software had filtering settings like vBulletin et al.

I hear ya.

Nijntje wrote:
ngbeslhang wrote:

Hello guys.
I installed OpenWrt (Mamba) onto my WRT1900AC v1 a few days ago and it worked great but right now my 5GHz radio won't work no matter what. I even tried to reboot the router (remotely) and restart the "network" startup script and it still won't work.

I am not sure but if it does matter I installed luci-app-qos a few minutes before I found out, but I wasn't connecting to my 5GHz wifi so I am not quite sure.

I appreciate any help. smile

What channels are you using on 5Ghz? I have experienced that  the 5Ghz radio refuses to work when using channels other than 36, 40, 44 or 48.


On a side note, I found that there's two version of OpenWrt for WRT1900ac v1 (http://wiki.openwrt.org/toh/linksys/wrt1900ac), Caiman and Mamba. I chose Mamba because I didn't see Caiman on the rest of the page. Should have I installed Caiman, or Mamba is better?

DO NOT INSTALL THE CAIMAN BUILD ON YOUR WRT1900AC YOU WILL BRICK IT!!!

Caiman is the codename / build for the WRT1200AC
Mamba is the codename / build for the WRT1900AC-V1

If you've installed the Mamba build on you're WRT1900AC-V1 you've installed the right build.

Mine seems to refuse to work in any channel. I have tried 20MHz, 40MHz and 80MHz width and still won't work. hmm
Also, thanks for reminding me. smile

EDIT: I decided to try all the channels you mentioned as well, it also won't work. I also tried to restart "network" init script but it still won't work as well.

(Last edited by ngbeslhang on 8 Oct 2015, 03:14)