By Ryan Rice, contributing writer
My company has been making hydraulic valves for more than 50 years. A fork truck customer in a cold climate discovered that when lowering a heavy load, the valve’s spool would stick and the load would keep lowering. The operator had to physically pull the lever back to neutral to get it to stop, and the effort required was quite significant. The heavier the load on the truck, the worse the problem got. This could be replicated in any truck, and the problem went away as the truck warmed up. After a few minutes, the condition was gone, but returned upon startup after the truck was allowed to cool.
In very simplified form, a hydraulic valve consists of a cast-iron housing with a honed bore, into which goes the spool. The spool is moved to meter oil flow to a given area, and it is spring-loaded to return to neutral upon release. To control leakage, the spool to bore clearance is a few ten thousandths of an inch — a tight fit. We had seen spool bind like this before, and the typical cause is one area of the casting under high pressure, while the area right next to it is at atmosphere, and the differential causes the iron to move enough to grab the spool. The remedy is to undercut the spool in the affected area by a few ten thousandths of an inch to give a little extra clearance.
The problem here, though, was that the conditions under which the pressure differential was greatest, which is before the load actually starts to move, didn’t cause any spool binding. Not until some flow started, and the differential was reduced, did we see binding. Why?
Another phenomenon in hydraulic valves is what we call flow forces. This is when a large mass of fluid goes so fast that the pressure differential from one point to another causes a spool to move or stop. Could this be the problem?
This didn’t seem likely. I made several spools with small flow notches to cut the flow down to test this theory and the problem remained. Again, I was at a dead end. Most hydraulic problems have either pressure or flow at their source, and it seemed like neither was the culprit.
After more experimentation, I wondered if the spool might be heating up when the flow started, causing it to expand and stick in the bore. I set up a valve on a test stand and lowered a full simulated load. The spool stuck. Then, I quickly cut off the supply flow going into the valve. At this point, nothing was going through the valve, so there couldn’t be any flow forces. Pressure transducers confirmed that there was no differential between areas of the housing, so there wasn’t any housing deflection, either. The spool remained stuck for a few seconds after I cut off the flow then popped back to neutral.
The only reason it could have remained stuck was because it was momentarily larger in size due to localized heating. Heavier loads applied more heat to the spool-bore junction. The housing is a big mass of metal and can dissipate the heat easily. The spool, being a small cylindrical rod, couldn’t. So, for a brief moment, the spool was hotter than the housing and it expanded to cause the sticking.
A new housing design was needed to fix the problem permanently.