There is often a perception of stability issues when using different types of hydraulic calibrators. This is typically most noticeable when switching from a traditional deadweight tester (DWT) to a more static based system. Some of these truly static systems include our E-DWT, MPG-2, and comparison test pumps.
In most cases the perceived stability issue is actually related to adiabatics. When you generate a high pressure into a small closed volume there is increase in fluid temperature as well as some swelling of the plumbing which creates a volume change. It takes some time for the temperature to return to room temperature and for the plumbing to stabilize. Since both temperature and volume are major components of the basic pressure formula this has a big impact on the measured pressure. See this Excel(r) file that shows a typical drop in pressure when a closed, liquid filled system is brought up to 5000 psi from a vented condition, Stability Test 5000 psi. After about two minutes the pressure is fairly stable but even at 10 minutes the pressure is still on a slight downward trend but very close to fully stabilizing or where the true leak rate is obtained if that rate of change is not continuing to decrease with time.
DWT's will generally appear more stable as these systems are not truly static. Pressure is applied to the bottom of a piston which floats within a cylinder (a volume) when the applied pressure equals the same force as the mass load sitting on top of the piston. The floating of the piston allows it to move up or down (a volume change) to compensate for the adiabatic effects that are seen in all hydraulic systems when changing pressure. As a result you don't "see" a lot of the adiabatic effects but they are still there. This is the reason why sometimes you do have to re-float the piston if the effects are extreme enough to cause the DWT to go outside its float zone. A DWT is basically a really good regulator by the nature of its design so the measured pressure will generally appear very stable as it is not truly static in its mechanics.
Our E-DWT, MPG-2, and comparison test pumps (P5514, P5515) are truly static. Pressure is generated by compressing a volume of fluid with a screw/piston type mechanism. There is nothing to compensate for the adiabatic effects so it will be observed in the measured pressure and are often perceived as poor stability or a leak (since in most cases pressure will be dropping) as they settle out. The benefit to these type of calibrators is that you don't have to load heavy masses, they are much easier systems to calibrate, and digital standards can be incorporated to modernize, etc., so there are some good trade offs once it is learned how to deal with the adiabatic effects. Small variations in the design of these static systems as well as the volume connected, and amount of air in the system can also make significant differences in the amount of effects that will be observed.
You first want to ensure as much air as possible is removed from the system by proper priming and purging. In a truly static system/fully enclosed hydraulic system the air has no place to escape so it will cause issues setting the pressure as it takes much more effort to compress it and it will move around in the system which can cause some pressure fluctuation. See this article for tips to remove air from a gauge, Filling test devices (gauges) with liquid before calibration
To reduce the adiabatic effects it is good habit to to set pressures slowly or at reduced increments when practical. If you slowly increase to your set point the pressure will naturally only drop a little whereas if you do this rapidly will just have spend more time adjusting the pressure which is typically more frustrating. If go slow will typically just need to bump the pressure back up a few times until it stabilizes to an appropriate amount. It could still take a couple minutes per test point to become stable and lock on to your desired pressure. Patience is your friend!
Make sure you are using the display resolution that is truly needed. We often see that the resolution is set to say three digits out when the device being tested cannot come close to reading this, so it is thus overkill. The rapid movement of too much resolution seems to make the human brain think there is a bigger issue than there really is.
To essentially prove all the stuff mentioned above about adiabatics and to ultimately determine how stable a static hydraulic system is, you want to run a proper leak check to quantify the stability. Easiest way to do this is set full scale pressure, do not adjust the pressure back up to the original target, watch the pressure rate of change value and you should see it becoming smaller with time as the adiabatic effects settle out. Let it sit for 10 minutes or longer and then come back later to see what the rate of change is. You will likely see the measured pressure very stable and this confirms it was adiabatic effects. If you take the current pressure and the current rate of change after the 10 minutes this will allow us to quantify the actual stability/leak rate. See this article for a general leak test procedure, Leak Test, Pressure - General Procedure