Hi,

Just in here observing what is being said and I agree with you on not be putting in any more refrigerant into it unless it's absolutely necessary.
Especially, if you turn on the A/C and the condenser fans instantly kicks on and stays on as the refrigerant maybe stacking up in the high side.

Wide temperature swings and driving speeds can drive one crazy trying to "Dial" in the proper charge.
It sounds like it's very close and 28 to 40 grams moves the "capacity" or "capability to cool" around a lot!
A stuck or partially plugged orfice could be causing all of this. It can starve the evaporator and push the high side up and overcharging will just confuse things worse.

Here is a You Tube for A/C for house use with a R-22 example but it will help give you an idea of what happens with temperature change on the flooding of an evaporator. It's at the end things can go bad with slugging.
https://m.youtube.com/watch?v=AR3xe-VorMw
With capillary or orfice style systems you are using a "compromise" that "cost savings" does not always raise up a red flag too other issues.

We think these cars have better A/C but it's really only slightly bigger for R12, but a "have to do" for R134 to work decently ahead. Shame they went cheap on us thermostatically.

If You need to get accurate weight readings when charging. The best device to use is a cylindrical charging station that shows the liquid inside the vessel with a scale on the outside. It takes away the weight of hoses and any liquid in the hoses can be milked right back towards and into the charging cylinder.
These were used years ago before digital scales to charge domestic refrigerators. This is where a few tenths of an ounce is critical for proper operation in different parts of the world. They are sold locality specific due to large temperature fluctuations. If you ever move from Maylasia to a colder climate or vice versa, leave your fridge as it won't work right due to its charge and capillary tube.

When it comes to A/C, it's humidity that has to be calculated into the heat loading.
With newer high efficiency units, the location of the appliances is even more crucial. They now void warranties, if you shove them into hot garages, small unventilated rooms or outside on a back porch during a winter.

The next best thing is to put gauges everywhere in the car. Not just pressure gauges but temperature gauges to find what the coils are doing.
An Infrared meter might work on metallic parts of high side lines and for sure the low side to measure for super heat under stabilized cabin temperatures.
The evaporator needs to be flooded as much as possible up to the accumulator inlet and then show "superheating" in there and out going back to the compressor to prevent slugging.

The condenser need to show some cooling from the bottom of the condenser and amount of sub-cooling in the area ahead of the orifice. This will gage that high side is to be full up to the device but not too full.

The same thing is done for all other expansion devices but for a fixed capillary and orifices the charge sets the whole tone! No moving sensing tubes or adjustment. You work with what you got.

Now a variable orifice, is just what it is, another variable to deal with! It recalibrates to the variable heat load and that can affect the charge pressure ratios!
Best to be in a laboratory of under controlled conditions if you want to get that one caught right on target because it's going to keep moving. That is, until it gets stuck with trash, wear or whatever. Very delicate when compared to a TXV's as some are those are serviceable.

Luckily, on cars it is a high temperature application and not for food preservation!
Shooting for 35 to 40 degree vents is really overkill for sweat removal.
IMPO.

Phil