No, the SDA should go to SDA on RPI and SCL (what you call SLK) should go to SCL on RPI.
Check my comment on the build thread: https://forum.chordata.cc/d/71-mini-k-ceptor-design/64
You seem to have a shortcircuit on your U2 LTC4316 IC, and the solder job on the IMU looks a bit messy. Can you recheck how the pads are soldered?
valor The short-circuit on U2 doesn't matter. Those two pins are GND.
It looks messy because of the amount of flux I applied. I just cleaned it.
I followed the instruction on this thread(https://forum.chordata.cc/d/34-soldering-question) to solder IMU. It was harder than I expected. I may overheated the chip. Not sure how to check the solder under the chip though.
The problem here is my hub board, not the K-Ceptor(yet). I can't expect reliable results with it.
I will post the result once I assembled another Mini Hub board. Currently, I am waiting for the delivery of new components.
ChangyeobOk
That IMU is sitting way too skewed. Even if it's not the case, it takes time to eliminate each possible problem.
Usually it's always a lot easier if you build a few of each board. This way it might be easier to troubleshoot.
ChangyeobOk When I said all get positives, it meant the result of "i2cdetect -y 1"
Is it normal?
No, it's an error on the bus. Most probably the SDA line is shorted with GND, but could also be something else.
Are you getting that result with no KCeptor connected? If so then you should focus on finding some soldering issue in the Hub first. Do you have a tester? you can test continuity between SDA and SCL, and also btw them and ground.
What happens if you unplug the Hub and run i2cdetect?
The solder bridges in the KCeptor @valor is pointing out will probably give you communication errors as well, but if understood the situation correctly they are not the cause of your current issue.
daylanKifky I assembled another Mini-Hub 
and it worked until I plugged a KC into #5. I think it was my KC that made my hub undetectable. But I don't exactly know how it happened.
I ordered my PCBs from JLC PCB and it came like this. See the two pads on U2 valor pointed out. They are already shorted before I solder them. Will it be a problem? Then I should ask them to send them again...
@valor Do your boards have shorted pads on U2?
valor I resoldered the IMU with hot air and it's sitting on the right spot. Still both 5v 3.3v red LEDs are on.
Question. If you connected a 6p6c cable into OUT on your KC, would it make your hub undetectable?
I checked my cable with a multimeter. All lines are good and it looks the same as your drawing. https://forum.chordata.cc/d/71-mini-k-ceptor-design/21
ChangyeobOk yea, those two pads on U2 are actually supposed to be shorted. Sorry I draw attention to that, as it should not be a problem on your board.
Connecting a 6p6c cable (with nothing on the other end) into the OUT connector on Mini K-Ceptor (or IN for that matter) should not have any effect at all.
Connecting the Mini K-ceptor to a hub via IN or OUT connector should just work, since it's a shared bus. Direction labels "IN" and "OUT" are made for convenience and to ensure that the k-ceptors face the same direction thus the directions of the IMU data are consistent.
On the Mini HUB, check the resistance between the VCC and all 6 pairs of SCLx and SDAx lines and make sure that you get around 10Kohm.
valor Thanks for the clear explanation about IN/OUT connector! I was wondering about it.
Only SCL3 reads 10.8k ohm and SCL6 is fluctuating between 68k. No readings for other points.
I think:
The chip looks it's soldered well but it is not.
and/or
I may overheated the chip while I was soldering with hot air.
and/or
I used old flux(it's brown color looks nasty) and the remaining flux affects the signal? I cleaned it with 91% isopropyl alcohol with a brush and a cotton swab though.
Starting to feel frustrated...
ChangyeobOk
I assume you are measuring with the Mini Hub powered off.
Do you have a microscope to check the solder joints thoroughly? You can always check for shorts on the neighboring pins which are not supposed to be shorted.
Ugly old flux should not be a huge problem if it works. But if it's way past its age, it might not be wetting the pads well enough for the solder to pull.
If you feel like the frustration is starting to build up, it might be a good idea to take a step back, maybe sleep on it. Tomorrow will be a fresher day for troubleshoting.
valor My bad. The power was on.
SDA1: 10.8 K ohm
SCL1: 2.9K
2: 10K
2: 2.6K
3: 9K
3: 9.7K
4: 1.9K
4: 5.4K
5: 1.8K
5: 9.4K
6: 9.3K
6: 9.3K
As I clean the board again and again, the value gets closed to 10k. Strange huh?
Thanks for your advice but it's 4pm here. I will try to check the solder joints again. 
ChangyeobOk
Okay, we might be on to something. Those readings are definitely off. Each of those should read as 9-10K.
Your U5 might be defective.
ChangyeobOk After I touched the joints with a fine tip soldering iron, every joint is around 10k. Bravo!
SDA1: 10.8
SCL1: 10
10
10
10
9.7
9.9
9.9
9.9
9.8
9.5
9.6
The SDA1 is a little high and SDA6, SCL6 are a little lower but other points are almost 10K.
I think I overestimated my soldering skill. :p
ChangyeobOk This time hub doesn't show up on i2cdetect -y 1 command. Which pin is related to hub detection?
ChangyeobOk
SBC_SDA and SBC_SCL
valor I wish I had a microscope. I am relying on the magnifying glass on my helping hands. 
ChangyeobOk well the best tool is the tool you have
daylanKifky
While I am trying to fix the soldering issue on my hub board, I get different result for i2cdetect -y 1.
My first mini-hub appears on 0x71,
My second mini-hub appears on 0x73 at first but after I tried to fix soldering issues(retoucing the joints between chip and the board with soldering iron), it appears on 0x70
Do you have any idea what's happening? Why does it need to be always 0x73?
ChangyeobOk I have assembled my 3rd mini-hub and it's working well. However, I would like to fix the 1st and 2nd boards for later projects.
ChangyeobOk sounds great! On the 2nd board: try replacing the U5 first.
ChangyeobOk I get different result for i2cdetect -y 1.
My first mini-hub appears on 0x71,
My second mini-hub appears on 0x73 at first but after I tried to fix soldering issues(retoucing the joints between chip and the board with soldering iron), it appears on 0x70Do you have any idea what's happening? Why does it need to be always 0x73?
@daylanKifky Do you have any idea by chance?
Hi!
The address of the Hub is set at U5's pins A0 and A1 by the resistors R2 and R3. With the configuration as in the scheme below the address should be0x73`
Check these resistors on your first Hub.
With that said, the notochord can be configured to read another address for the Hub. You know that it reads an XML file upon startup? it is normally called Chordata.xml but you can also pass another file like the default_biped.xml. Take a look at the <Mux> tag on those files: it contains the address 0x73. If you are getting a (consistent) different address you can use it this way
