The below question went unanswered so I am trying to figure it out with simulation.
https://forum.allaboutcircuits.com/...rectional-circuit-breaker-application.205326/

This circuit should give an overvoltage threshold of 58.4V and undervoltage threshold of 43.2V, but it doesn't seem to be following the theoretical thresholds. Does anyone see what could be wrong with the design?

The sim and LTC4368 datasheet and calculated values for R1-3 are below.

Parameter	Value	Units
V_OSUV_Target	0.003000000	V
I_UV_Target	0.000000010	A
UV_TH_Target	42.000000000	V
OV_TH_Target	58.000000000	V
R1+R2_Target	300,000.000000000	Ohms
R3_Target	24,900,000.000000000	Ohms
R3_Selected	25,600,000.000000000	Ohms
R1_Target	223,275.862068966	Ohms
R1_Selected	218,000.000000000	Ohms
R2_Target	82,000.000000000	Ohms
R2_Selected	82,000.000000000	Ohms
UV_TH_Actual	43.166666667	V
OV_TH_Actual	59.403669725	V

 

Replying to myself lol, but just in case anyone is curious or reviewing this in the future, I discovered LTspice doesnt allow the capital M for megaohms, and then converting to kiloOhms I entered the wrong value. 25.6MOhms is 25600kOhms. Now the simulation is working nicely and doesnt seem to care much about R4 (results are roughly the same from R4 50k to 800k).
I'm still insterested in how to determine R4 if anyone knows, but for now I think the solution is accepttable and Im going to order a PCB to test it.

 

I discovered LTspice doesnt allow the capital M for megaohms

LTspice uses meg for the 1 million prefix.
It doesn't differentiate between capitals and small letters, so M is interpreted as m (milli)

I'm still insterested in how to determine R4 if anyone knows,

It's arbitrary.
You can pick a convenient value for R4 and then calculate the rest of the resistor values for the voltages you want.