TPLINK this XDR3230 wireless specifications is AX3200, 3200 very odd is not, other wireless routers rarely have this specification, because its 2.4G is WIFI5, 4x4MIMO, the highest rate of 800Mbps, 5G is also 4x4MIMO, support WIFI6, in 80Mhz bandwidth the highest rate of 2402Mbps, so it adds up to 3200. So it adds up to 3200. It doesn’t support 160Mhz, and because of the 2.4G parameters, I mistakenly thought I was using MT7615N or MT7611N.
XDR3230’s outer packaging.
The package comes with a network cable, super Category 5 twisted pair, 26AWG wire gauge, short line does not matter, just run 10,000 megabit.
Power supply specifications 12V/2A, very good, and a relatively large head, I think it is very good. Power consumption is not determined by the power supply, but by the power consumption of the router itself. 12V/2A does not mean that the router power is 24W at all times, do not misunderstand. The power supply rated output power is larger, so that the router can easily cope with the power supply when there is peak power
The XDR3230 looks like a scaled-down version of the XDR6060, with a total of eight external antennas, four 2.4G and four 5G.
The back of the interface: four full Gigabit network ports, regardless of WAN and LAN, support blind plug, RESET key to the right is the “Mesh” network button.
“To ensure a good experience of using Mesh network, currently up to 4 Mesh routers can be supported to form a network, more Mesh routers are recommended to form a new Mesh network.”
At first I thought it was recommended not to exceed 4 units, it turns out that the rules are dead not to exceed 4 units, why is the same Qualcomm platform velop but not this limit?
Disassembly began, there is a screw under the label at the bottom of the fuselage, just take it off, but I can not do it intact to tear off the label paper, it is easy to break:
Can not do a non-destructive disassembly, even if you can tear down intact, the back of the demolition of the shell will break a few jaws, because more than half of the jaws point glue 。。。。。。。 It would have been very difficult to open the shell, alas. It is easy to put back together, not 5 seconds, but it takes 5 minutes to take apart.
Bayonet with glue.
There is a black aluminum heatsink on the motherboard and a ceramic heatsink above the heatsink, could the chip underneath be MT7621?
8 antennas soldered on the motherboard, not good to flip the motherboard, but it was difficult to find a suitable angle, finally able to turn over to see the back of the motherboard:
The top picture has a small area of aluminum plate at the bottom left, usually this small area of aluminum heat sink is to help the front of the motherboard amplifier chip heat dissipation. The top right of the aluminum heat sink has a flash memory chip, model EN25QH128A-104HIP, capacity 16MB. quite large ah, SPI 16MB, brush the third-party firmware is expected, right?
All the screws on the back of the motherboard unscrewed, the front of the motherboard heatsink can also be removed. And this little piece of aluminum after taking off looks like this.
There is thermal grease, in order to keep it from dirtying my background cloth, I put the aluminum piece back on. Now turn over and look at the front of the motherboard, it has been removed from the heat sink looks like.
See two pieces of shield, and between the heatsink and the useful thermal conductive silicone pad, thickness 1mm. this time it is much cleaner, unlike the XDR6060, XDR6030, XDR5430 as in the shield is coated with thermal grease, dismantling the machine made very dirty.
Remove the network transformer near the piece of ceramic heatsink, not the CPU MT7621 ah, but the gigabit switch chip RTL8367S, then there it is certainly no MT7621, because the MT7621 itself integrated gigabit switch. So what CPU is inside the shield? I will pry open the shield to see an understanding.
There is a small chip on the motherboard attracted me, because the previous disassembly has not seen, as follows, just above the two shields, belonging to the MEDIATEK MT6380N, from the location of the circuit, very much like a power management chip, high-fashion called PMIC.
So I looked for information, found the MT6380 point diagram, looking at the diagram below, my mind sought to see if I could lead a few lines through it, to the USB port, and then there was MTK PEP fast charging? After I woke up from my dream, I counted the MT6380N pins on the upper board, yes, also 10 pins per side, a total of 40 pins, also QFN closed, the right head! It is the power management IC. is also responsible for the power supply to the CPU.
I’m going to prepare to remove the two pieces of shielding, the preparation of the work as follows.
Who said that the shield can be easily removed with a suction cup ah? No L use. I was in a dream, Archimedes once said to me, as long as he was given a fulcrum, he could pry the earth. Then find a fulcrum on the board, to pry open the shield is not a problem. But the motherboard usually has many small components and fragile alignments on it, so use this.
This will be easy and not afraid to damage the board. The two shields are removed, the shield a total of four chips, only the memory chip without thermal conductive silicone gasket, the other three chips are also used on.
First look at what the CPU is, the model is MT7622B, I first saw MT7622B, dual-core 1.35GHz, the frequency is significantly higher than the MT7621 880MHz. Memory chip model is M15T2G16128A, capacity 256MB, DDR3 memory.
Looked for information on the MT7622. As follows.
MT7622 is fabricated with advanced silicon process and integrates a Dual-core ARM® Cortex-A53 MPCoreTM operating up to 1.35GHz and more DRAM bandwidth. This SoC also includes a variety of peripherals,including SGMII, RGMII, PCIe2.0, USB2.0(Host) ,USB3.0(Host) ports, and 5-port 10/100 switch. To support popular network applications,
MT7622B also implements 2.5Gbps HSGMII and 1Gbps RGMII Ethernet interface, embedded a 5-ports 10/100 switch and 802.11n 2.4GHz 4T4R WLAN radio, 802.11ac WLAN connection can be supported thru its PCIe port.
The MT7622B has HSGMII interface, which supports a transfer rate of 2.5Gbps, and then I checked the information of RTL8367S “The RTL8367S also supports one Ser-Des interface that could be configured as SGMII/HSGMII interfaces. The RTL8367S also supports one Ser-Des interface that could be configured as SGMII/HSGMII interfaces.” There is also HSGMII, so the bandwidth between the CPU and the network port can be 2.5Gbps. (The WAN port is connected with RGMII at 1Gbps rate)
I say this because the maximum 5G rate of XDR3230 is 2402Mbps, if you use 2 LAN ports to transfer data to 5G at the same time, can it reach its maximum speed. If the CPU only supports SGMII, the transfer rate is only 1.25Gbps.
The MT7622B has an integrated 5-port 100GbE switch, so you have to plug in the RTL8367S to support the Gigabit ports.
MT7622B integrated 4x4mimo 2.4G, support 256-QAM, the highest rate of 800Mbps. (Asus PCE-AC88 connected XDR3230 2.4G, the display rate is 800Mbps)
MT7622B integrated 2.4G wireless chip, there must be gains and losses, gigabit switch chip MT7530 can only be discarded. Otherwise, this CPU is too hot and may not have space to plug in. So to external RTL8367S.
CPU integrated 2.4G, with an external independent amplifier chip, the chip above the screen print 7626H, the model is RTC7626H, is a 2.4G FEM chip.
The parameters of the RTC7626H are as follows.
Frequency range: 2.4 – 2.5 GHz
3 V to 5 V single supply voltage
Integrated high performance PA, LNA with bypass function, and SP3T switch
Input & output fully 50 ohm matching
Integrated positive slope power detector
+21 dBm linear output power at 3% EVM, 802.11g 64 QAM, 54 Mbps, 5 V
Small 16L QFN 2.5×2.5×0.75 mm 3 package
RoHS / Halogen Free Compliant
Moisture Sensitivity Level : MSL 3
Displays high power in general.
Now look at the 5G chip, the following chart:
5G radio routing two chips, respectively MT7911AN and MT7975AN, the past MTK AX1800 program is MT7905DAN + MT7975DN, “D” means that at the same time support 2.4G and 5G. fiducia K2P with MT7615DN is the same The MT7615DN used in the Fusion K2P is the same principle.
It can be understood as follows.
MT7905DAN+MT7975DN=MT7915D, supporting dual-band 2x2MIMO WIFI6.
MT7911AN+MT7975AN=MT7915A, supports single-band 4×4 MIMO WIFI6.
From e.g. a quote from mediatek’s website: MediaTek MT7915 integrates a simultaneous dual-band architecture that delivers multi-gigabit Wi-Fi speeds via either 4×4 (MT7915A)
or 2×2 (MT7915D) antenna options.
I also found the MT7915 field in the TTL readout of the XDR3230. It is as follows.
role_selection_wlan_ready_handle(472). [role selection] start bh apvap s
[ 27.983590] mt7915_apply_dpd_flatness_data: eeprom 0x62 bit 0 is 0, do runtim
[ 27.991182] MtCmdChannelSwitch: control_chl = 48,control_ch2=0, central_chl =
48 DBDCIdx= 0, Band= 1
[ 28.000314] BW = 0,TXStream = 4, RXStream = 4, scan(1)
[ 28.019624] *MtCmdChannelSwitch: control_chl = 4,control_ch2=0, central_chl =
4 DBDCIdx= 0, Band= 0
[ 28.028705] *BW = 0,TXStream = 4, RXStream = 4, scan(1)
[ 28.147606] mt7915_apply_dpd_flatness_data: eeprom 0x62 bit 0 is 0, do runtim
[ 28.155195] MtCmdChannelSwitch: control_chl = 149,control_ch2=0, central_chl
= 149 DBDCIdx= 0, Band= 1
[ 28.164500] BW = 0,TXStream = 4, RXStream = 4, scan(1)
[ 28.191593] *MtCmdChannelSwitch: control_chl = 5,control_ch2=0, central_chl =
5 DBDCIdx= 0, Band= 0
[ 28.200646] *BW = 0,TXStream = 4, RXStream = 4, scan(1)
[ 28.315598] mt7915_apply_dpd_flatness_data: eeprom 0x62 bit 0 is 0, do runtim
Split a MT7915 into two chips, one is to solve the problem of heat dissipation, the second is flexible with.
Do not blame me for being too verbose, if the Internet can find the details of these chips I do not have to speculate so hard, and I do not know if it is right.
From the above figure on the 5G circuit, and did not see an external amplifier chip, a little disappointed. According to the previous XDR1860 5G real-world test, the signal is not strong, the speed is not good. Redmi’s MT7615N can also be stronger than K2P’s MT7615DN, so let’s see what happens after the actual test.
There is a large capacitor, 16V 1000 microfares, and three 16V 470 microfares, sufficient power supply capacity.
There is an empty pad above the CPU, which I assume is the SPI flash memory.
TTL pin definition of XDR3230: (baud rate 115200, Mercury is 57600)
Finally, a chip model summary chart.
So far, TPLINK’s XDR3230 disassembly is complete, but the article is not yet finished.
Although 2.4G is not used on the WIFI6 chip, but then, now 2.4G most of the smart devices left to use, as long as to ensure that the 2.4G signal is stable and strong signal is OK, and it has a separate FEM chip, even if the PA is not strong, but there is LNA to increase sensitivity. Practical! The phone wants to be fast or to be connected to 5G WIFI6 to do so.
The following is a simple but complex test. To carry out two tests, one is to test how much it can reach the maximum wireless speed of 5G, and the second is to test whether its 5G signal will be stronger than the Mercury X188G.
Only a gigabit port, so to use 2 network ports concurrently, the receiving end to find a support 2.5GbE port, TPLINK’s XDR6060 is the most suitable, ASUS AX86U can also be, but its 2.5GbE network port problems. The approximate connection diagram of each device is as follows.
XDR6060 wireless bridge XDR3230 5G, the highest wireless connection rate between the two is 2402Mbps, and then a computer connected to the XDR6060 2.5G network port, then the XDR6060 acts as a wireless network card role, Intel X550 is a 10 Gigabit network card, can identify the 2.5Gbps rate network port, this is the client side, the server side is equipped with I350 dual-port Gigabit network card, connected to the XDR3230 two LAN ports, that is, (LAN1 + LAN2) to 5G test. The server side is equipped with I350 dual-port gigabit network card, connected to the XDR3230’s two LAN ports, that is, (LAN1 + LAN2) to 5G test. IxChariot test results are as follows.
The total speed of the two network ports is 1768Mbps, measured a few times to take the maximum this time, sometimes the total speed is just a little more than 1000. I do not know if it is not stable wireless or wireless bridging, and then I used the Asus AX86U wireless relay XDR3230, this time you can not use the Intel X550 network card, you have to use the AQC107 to use its 2.5GbE port properly. Then the test is similar.
How about the result of 1765Mbps? Normal
From here, we can prove that the HSGMII interface is used between RTL8367S and MT7622B.
I added 2.4G together with the test to see what the total dual-band wireless throughput rate can be? Open one more computer, use ASUS PCE-AC88 to connect XDR3230 2.4G, the server side with the motherboard integrated network card connected to the third LAN port, and 5G wireless bridge together to run up. The result is as follows.
The total wireless throughput rate is…. Wait I first took out the calculator and added it up, and the result was 1959 Mbps. not bad, I don’t know what the result is, because there is no reference.
Followed by to test the 5G signal strength, and then take the results with the results of the test Mercury X188G a few days ago to do a little reference. Also in C and D point with Xiaomi 4C for a simple speedtest speed test (broadband 300 megabytes). By the way, also test the power in various states.
I list some hardware parameters and test results of XDR3230 as follows.
The 2.4G test speed is about the same, so let’s leave it at that. How to say, XDR3230 has the shadow of Redmi AC2100, there is potential. This is only a preliminary simple test to a detailed review shall prevail. This article is only a disassembly article, disassembly I can do very subjective, but the review I have to do very objective, in order to be objective and narrow the error will take a lot of time and effort, so I am here again for the delay in some reviews and in finding excuses.
The disadvantages, does not support 160 bandwidth, if you need 160 bandwidth can buy XDR5430, plus more than a hundred, but this XDR3230 good chance to brush third-party firmware. mt7622 is not a new CPU, in 2018. If you can brush the firmware, WIFi6 has a stronger signal for 5G and 2.4G has a separate amplifier, both very practical.
It gave me two surprises, one is that the CPU is different from the expected, but this many people have guessed that it is MT7622, and the second is that the actual test 5G is significantly stronger than expected, only blame the initial XDR1860 and X18G to give me a bad impression of the 5G signal.