Generic example I made up to help me fully understand this topic:

Fixed displacement pump delivering 10 GPM, and a manual valve handle is pulled such that the valve's output flow to a cylinder is 5 GPM.  Cylinder load-induced pressure is 500 psi.  Relief valve set to 3300 psi.  Assume 0 backpressure from directional valve T port back to tank.  For simplicity, also assume 0 pressure drops through plumbing and other components from pump to cylinder and back to tank.  Only pressure drop is across/through directional valve.  So in this example, there is an excess of 5 GPM.  Pump discharge pressure = directional valve inlet pressure = 500 psi + _____.

There are 2 places for the excess 5 GPM to go: (A) from P port to T port of the directional valve and back to tank, and (B) over the relief valve to tank.

In my research, I am not clear about which of these 2 things happens.  It is implied in most valve literature that (A) happens, but I've never seen it directly stated.  It's probably obvious to those in the know.

Question #1:  I just want to confirm that it is (A) for most/all manual directional valves.  That the excess flow is going from P port to T port of the directional valve and back to tank, and not over the relief valve.

Question #2:  If it is (A), then the pressure at the inlet of the directional valve would be the pressure drop of the excess flow going back to tank.  So in my example, instead of the excess 5 GPM going over the relief valve at 3300 psi, it would go across the directional valve back to tank at a lower pressure.  How would I determine this pressure?  Valve literature gives pressure drop vs flow, but how does this apply to metered/throttled flow?  Doesn't it just apply to full flow in the 4 main flow paths, P->A, B->T, P->B, A->T?

Based upon your description of the conditions, when you throttle the flow going to the cylinder then you have effectively turned your directional control valve into a flow control as well.  The fixed displacement pump is delivering 10 gpm to the system but only 5 gpm is allowed into the cylinder.  The remaining 5 gpm must go somewhere and while it is possible that you may get the leakage past the directional valve to account for the remaining 5 gpm but that is not generally going to happen and can't be determined by info given.  It is expected that the remaining 5 gpm will be directed back to tank over the relief valve.  There would be a pressure drop across the directional control valve of 2800 psi and system pressure would be at relief setting of 3300 psi.

Question #1 - goes across relief valve and then to tank.   Could possibly go as leakage across the valve but that isn't typical nor realistic unless valve is faulty and can't be ascertained from info given.

Question #2 - can't be determined with info given.  Would need to know what leakage rate exists within the valve (from P to T) if any and then it could be roughly calculated if that was the case.  Not typical.

So any time fixed-displacement pump flow is reduced by manipulating a manual directional valve handle, then the excess has to go over the relief valve?  I have to say that's not the answer I was expecting.  It's what makes the most sense right off the bat when thinking about it.  But then thinking about all the mobile machinery that uses fixed pumps and manual directional valves with the operator only slightly moving the valve handle off center to get slow movement of the cylinder.  If what you're saying is true, then every time that partial shifting happens, pressure on the upstream side of the directional valve (between pump and directional valve) goes to relief valve pressure.  I figured that couldn't be the case, because of the heat it generates.  I figured there had to be a way for the excess flow to pass through the partially-shifted directional valve, going from P to T, at a lower pressure drop than over the relief valve.

I think you may be thinking of a load-sensing valve being used with the fixed displacement pump.  Sometimes the mobile sectional valve banks will have a relief valve in the inlet section that also works with load sense to divert excess flow to tank at a pressure a bit higher than the load pressure.  An inlet relief section will accept all of the pump flow and divert it via low-pressure path to tank when no functions are actuated. There are internal “load sensing” pathways in the valve that communicate the highest demanded actuator pressure to the valve inlet. The inlet of the valve will restrict the flow from the pump until enough pressure is created to match what is required to move the load.  Excess flow is directed across the inlet relief valve at load pressure plus some biased spring value - saves energy rather than dumping across relief valve at full 3300 psi.  In the example you gave, 5 gpm would be going to the cylinder at 500 psi and the remaining 5 would be directed to tank across the inlet relief to tank at say "750 psi" (load pressure plus the biased spring value in the relief).  Actual values vary depending on type of valve and manufacturer.

That's why many mobile systems use a load sense, pressure compensated variable displacement pump as it doesn't require special valving to save energy.  Hope that helps a bit.

Ok, that's more like I was expecting, with the excess being dumped to tank at load pressure plus a little.  Can you post a picture or point me to what the hydraulic symbol would look like for that "load sensing inlet relief" you're describing that dumps the excess flow to tank at load pressure + biased spring pressure?

What wording or symbols should I look for in the valve literature that indicates this type of "load sensing" capability for use with a fixed pump?  I don't know if it's allowed, but could you direct me to an example of a specific directional valve that has this function so I can see what to look for should I ever need to spec it?  We have used a monoblock valve before that has an "inlet relief".  I just looked at the catalog for that valve, and according to the inlet relief symbol and setting range, it functions like a typical main system relief valve, not the one you're describing.  This particular monoblock valve does not appear to have any of this "load sensing" capability.  So when partially shifting the handle to get less-than-pump flow to cylinder on this one, system pressure would go to system relief valve pressure with a lot of heat generated.

(I just realized that I should have posted this thread in mobile, not industrial.  I work mostly in industrial, so for some reason I snapped to industrial).

It appears that I am limited as to posting only text so I'll let you do the legwork on looking at manufacturer's catalog but Salami has some good descriptions and graphical images to assist in understanding as does Eaton, Danfoss, Parker,Bosch Rexroth, etc.  


Salami has some good info in their VDP08 catalog.  No particular reason I grabbed them other than I knew they had some good info and I think they make a good product as do many manufacturers and you'll be able to see what has been described.  Hope that helps.

I'll read through those catalogs.  Thanks for taking the time to reply to my questions.

This is a test email from Adele Kayser - please let me know if you receive an automated email notification. Thanks.

yep