That flakey material on the side certainly does look suspect. But the other pictures indicate it wasn’t heat shrink. So what is it then? If the busbar didn’t still have it’s gold patina I would say that it was a coating that deteriorated.
Is that a real flexible busbar? The type that is made up of multiple layers of thin copper? Looking at the pictures again, I’m not so sure. I don’t think you could make such an odd turn with that type of busbar, it would have to be solid copper to do that.
Suggestions on how to improve the contact surface?
Would you swap to the traditional round (thick) cables?
What do you mean? On which photo? Which residu? Can you point me to what you mean?
I use these here for copper contacts: Aeronix Kupfer-Kontaktpaste 35ml
Just as a hint. But a clean and flat contact area is the basis for a low resistance!
Hi Rene. Your last picture. The bottom part of the busbar, with the torque marker towards the orange heatshrink on it, looks really clean and shiny. Different to the top part. Maybe it’s just the photo. Or the used Aeronix paste 
This would be based on the current needs/capability of the battery. The cross section needs to be sufficient.
The issue with the bus bar is the burr on the holes. That needs to be clean and flat. How that is done would depend on the bus material and finish.
Sorry, still don’t get is @ChiefSolar. The last picture, of the busbars without the burning material, the top one is from when I did the installation of the problem box:
The second picture of the clean busbar (without the problem) is from a 2nd box I installed a couple of days ago.
Are you referring to one of these pictures?
Sorry. Yes.. Last picture.
So the picture without the red marking on the screw is the problem terminal?
Then it was not tighten or torque as the marking is/was missing.
No, the other way around. The one with the red mark is from the 2nd (new) box currently working fine. The red mark is to check if the screws/bolts get losser over time. This is a tip I got after I assembled the/my first box. Therefore the red mark is not on the first screw/picture of the first box I assembled.
Although I’m almost sure that I used the correct torque to screw everything tight, it is probably one of the reasons why this happend. Also I’ve a suspicion that bad quality of the screws could also be part of the root cause. Still planning to buy some new good ones and including washer and spring-washer.
This is the terminal/busbar of the new, 2nd box I assembled last week. All brand new.
Ok, now I got it.
Then the thesis with the bad contact through either not flat or dirty surface nor through objects (like the shrink tube) inbetween remains valid.
You mentioned, that you checked the screw torque and this was fine, but the electrical contact is not realized through the screw, the screw hold the the busbars in place. The screw is some kind of stainless steal and so much worst in electrical conductibility, it is only responsible for maybe 10% of the current path. So you don’t have to take to much effort in the screw. It must just guarantee to hold the e.g. 6Nm contacts in place.
Yeah that was pretty obvious to me. I was just wondering about why there is an apparent different look of the metal surface…nit the same everywhere in that picture.
About Aeronix copper paste.
The data sheet doesn’t tell the specific resistance of the product.
I highly doubt Aeronix reduces significantly the contact resistance.
It surely will avoid corrosion and friction, but nothing more.
Poper torque and clean flat surfaces is the crucial factor.
I would also use spring washers only in mobile or vibrating applications. They are quite weak and often fully compressed by 1/2 of the desired torque, especially on 8.8 or 10.9 screws.
I guess it is hard to tell a conductance/resistance of the material that everyone uses with different thicknesses. A specific conductance/resistnace for a defined thickness neverthelss could be given…
But anyway, the clean and flat surface is the crucial factor, I agree. But isn’t this also valid e.g. for heat conduction? There also a flat and clean surface is the key but still you massive imrpove the heat transfer when using a thermal paste as “gap filler” ? 
At least, in theory, such a conductive past should, applied correctly, not worsen the contact.
Wow, crazy. It has to be a loose fastener no? Can you post here eel’s reply?
Apart from they are going to replace the busbars etc. no other reply. I’m in good communication and they are providing good support. But difficult for them, based on a couple of pictures to determine what the root cause is. Hope to get the replacement parts soon so that the box is back up and running again.
I received new busbars from the supplier and I installed the battery box again.
When I connected the busbar I noticed a not perfect connection, see picture below. It looks like there is some isolation material on the right side of the connection.
Maybe this was the case in the previous connection as well and at the end cause this burning issue.
I removed some part of the isolation material. Next picture is not perfect quality, sorry.
Yes, the heat shrink tube is too long, but besides this, the contact surface does not look like it is completely flat? At least I can’t see on the second picture as it is not sharp…
But anyway, check the contact again, that it is clean and flat, then use the required torque and then it will work.
Thanks for the feedback @Jetlag . The first picture is not how I finally mounted it as indeed it looks like the isolation material in between. The second picture is the final mount. I’ve use 5nm torque to mount it, should be sufficient in my view. I’ve cleaned all the connection before mounting it. Not sure what else I can do and how else I can mount it. This is how the busbar and the negative terminal has been delivered.
