question

Pulling 300A the shunt will have a voltage drop of 30mV so a power generation of 9W. If the shunt is located in free air with the shunt horizontal and the vanes vertical then it should be able to convect that heat away. Are you absolutely sure that you do not have a poor connection causing the heat or is your shunt shrouded or not ideally oriented.

@ChrisPI
When you 500A Shunt get's hot at 300A,, then please double-check your cables and lugs. The shunt is so massive, that it might get warm after several hours, but never hot.

Any bad connection, even oxidation on the lugs will cause a resistance which creates heat. So please check you connections, the lugs, crimping and cable-diameter. For 300A I would recommend at least 100mm2, or better 120mm2.

thank you both for the answers.

there are hall sensors which are quite equal to shunts, not the cheap alibaba stuff for 10Euro. understand that for easy cheap solution like 130Euro smartshunt 500A the shunt is the way to go but if you get in higher current areas the shunts is becoming the problem and you have to massivly oversize is so its resolution and therfor precisnness also goes down. But in a 350Euro class for A BMV712 with a 1000A shunt yes that gives the budget to get the right hall sensors that are definitly on par with a 1000A shunt.

the shunt in my case is directly connected to the terminal of my 1088AH DIY Lithium bank which is a silver plated 7mmx30mm busbar with 99.8% cooper that is rated and officially certified for 1500A continious AC and DC, best you can get from eaton for grid power stations. So 210mm2 more cooling and better connection you cannot get, massivly oversized for 300 and even 500A continious.

on the output its also connected with the same busbar to the NH3 500A fuse. no lugs, crimps or wire here. connection busbar to shunt measured 0.07mohm, thats not the issue.

after the fuse main positive load bus bar also 7mmx30mm like above with 6xNH2 fuses 200A, each on the output a 70qmm cable...2x70qmm per multi 3000/12/16 inverter (2x), one legacy house loads and one new house loads. All installed in a catamaran running on 12V, and no not 24V as all other equipment incl. 2 engines is 12V too and new or freshly overhauled.

at 300A the 50MV shunt will generate around 30 W of heat, quite significant and introduses this heat into 99.8% cooper busbars which get to 40 degress after 20min of continious 300A draw.

Shunts are rated to 66% of their contious draw means a 500A shunt 300A is at 60% already, means close to their max rating and yes will get hot.

in my case Its a 700A 50mV Rideon shunt biefier then 500A victron, so 300A is 42% and continious draw limit is 466A. my install is speced for 500A continious draw which the 700A shunt can do too.

Eledtrodacus BMS which has load (700A rideon shunt) and charge shunts (500A Victron shunt) on postive side. the 1000A victron shunt is too large to fit on load, also on the negative terminal with the BMV712. so have the 700A shunt there too connected to BMV 712 (just take the small board of the 500A and convert it to the 700A and modify the resistance setting in the BMV712) and same issue as on postive terminal. the 500A victorn shunt in charge also heats equally up with 200A charge from 2xMulti.

The busbars get to 27 degrees when pulling 400A for 20min straight, the same temp the Lishen cell have. The busbars are not the issue. I done a lot high current installations and things get different if you exceed 300A continuous draw and I always had trouble with the shunts heating up and grid stations have all hall sensors because of that while Victron brought a 2000A shunt out. 500A-700A is already mad and precaution and checks I need to do is insane but necessary, above 1000A install I done 3 and at the end used hall sensor battery monitor as shunts were heaters and current+voltage was the only information we needed (car hifi completion car)….cannot image 2000A using a shunt, never ever…

To discover issues I run it at 25% then 50% then 90% load and immediately you see with a flir camera for your phone and a if temp gun where you have issues and fix them by cleaning, slightly sanding and applying no-Al-ox or other dielectric grease, retighten with torque spec (and even a bit above if still getting hot) or cut of crimp and recrimp till all stays cool. Where possible I use flange nuts, if not then after tightening lock it with 2 drops of nail lacquer on the corner so they are not loosing over time on a boats vibration. Done that and fixed all beside that b… shunt. Ordered the same new, maybe it has some defects…and I need the shunts to get reporting via Cerbo on battery and install.

Yes shunts are easy and versatile but significantly above 300A the selection of install material is extremely small, expensive as only top notch material are sufficient and only available when presenting certification of installer due to that’s dangerous. So you can expect that installer can deal with hall sensors and use them accordingly, and for 300Euro I get super precise hall sensors and I know which I would source but how do I integrate them into Victron environment?

Also, just an idea:

what kind of output (to cerbo) is the shunt delivering - 0(4)-20mA or 0(1)-10V to be recognized ?

What is the output of your Hall sensor. May some interface coupler help. Like the things from Pepperl & Fuchs?

And a hardcore solution, non isolated: measure the mV from the very beginning of the busbar to the first outlet. Knowing mm² and constant of copper you can calculate the A via the voltage drop in mV