The charge pump is a small pump, normally pumping about 10 percent of the gpm's of the VSP pump. The main purpose of the charge pump is to keep fluid at some pressure, about 320 psi if I remember correctly supplied to the VSP. because some fluid is lost through hydraulic wheel motors, and is also used to help cool the case side of the VSP pump. I believe the motor case lines come back to the pump case drain, or meet somewhere and then goes to the cooler. The charge pump doesn't need to be large, because it is only a makeup pump, making up what is sent out of the wheel motors case drains and pump case drain. Another thought here is that not all wheel motors have case drains, so maybe the charge pumps on that system might be smaller. My main thought here is that it is not pumping a lot of fluid all the time, like when sitting still, and therefore it is pumping less, and filtering less, just enough to keep the charge pump side pressurized.
In my system, the PTO and the steering pumps are always running. They will build up heat just sitting there with the engine running. If relief valves come on a lot they generate heat. Attachment motors working hard, generate heat.
Someone mentioned about putting a high pressure filter in the attachment return line. I suggest a better place would be in the return outlet on the PTO block, and that would catch any debris from any attachment. There should be a gage in the line before the filter to give you an indication when the high pressure filter might be getting clogged. If you do this , I would note down the normal operating psi of the attachment. If it starts to run higher, it might be time to install a new one. Yes, you could put one on each attachment, but it will get expensive.
If the 2 gpm charge pump was pumping fluid at full capacity, it could filter 20 gal of hydraulic fluid in 10 min. This can not happen in our system, because we are not filtering fluid in a linear fashion. Contaminants are not injected in a linear fashion, but random, like when plugging in the quick disconnects, or opening the tank cap. Otherwise, we don't filter the entire tank of fluid and then we quit, what we are doing is to keep mixing filtered fluid in with unfiltered, and I don't know the math to compute the total filtering sequence since the oil is circulating in a continuous loop. A visual concept would be to put some kind of dye in the hyd fluid, and run it through a filter to remove the dye, and see how long it takes to completely filter the dye from the 20 gal.