Insane Hydraulics

Site theme image

Low Cost Oval Gear Flow Meter for Measurement of Low Flow Rates

In today's article, I want to talk about a handy tool that I have been using for a long time - a (relatively) low-cost oval gear flow meter, which I believe to be the perfect instrument for measurement of low and extremely low flow rates. I will discuss why being able to reliably measure low flow rates is important, why such flow meters are miles ahead of completion when used for this purpose, and I will address the positive and negative features of the model that I use - the K600/3 from PIUSI, primarily intended for controlled dispensing of diesel fuel and lubricants.

When you constantly deal with repair and diagnostics of hydraulic components, being able to measure tiny flow rates with a certain amount of precision is very important, because one of the things that you want to assess is leakage, which can vary from single-digit liters to tens of milliliters per minute.

Leakage is a broad term, but for the purpose of this article, I am referring to the flow from high pressure (P) to low pressure (T) areas of a hydraulic component that happens either through natural clearances of pressure exposed parts (like spools, poppets, pistons, and whatnot) or through unintended paths (damaged seats, missing O'rings, etc). It's very simple - if you have an idea about the normal leakage rate of a component that you are testing, an abnormally high leakage rate will signal you that it is either worn out or damaged and command further investigation or replacement.

Consider the following scenario. You are adjusting an orbitrol steering valve on your bench and you have to set the relief valve. A very simple and straightforward procedure. All you need for this is a very basic fixed-displacement pump test bench and a pressure gauge.

So, you block the A and B lines, back out the pressure setting screw, feed the flow from your fixed displacement pump (say - 16 l/min - an 11 cc pump turning at 1450 rpm) in the P line, provide the steering input in one of the directions with whatever tool you usually use, and then turn the pressure setting screw clockwise till you reach the desired pressure setting, say, 160 bar. You're all done, right? Maybe give it a couple of extra steering jerks in the end just to re-check that the pressure setting is correct. You're surely done now, aren't you? The pressure's set, right? What else do you want?

As it turns out - the pressure is, indeed, set. But how can you be sure that the valve's internal leakage is acceptable? What if there's an O'ring missing under the seat, or the poppet is worn out badly, or there's contamination in the P to T check-valve and it's by-passing. If the undesired by-pass is less than your pump flow at the set pressure (especially with the cold oil you are testing it with) you will not see it unless you read the return flow! The orbitrol will seem perfectly fine on your bench, and yet the steering will be slow and sluggish when it is back in the tractor.

Now, if you had this very handy tool mounted in the return line - such a malfunction would be very easy to detect. All you needed to do would be to set the steering relief first (just as described above) and then lower the bench pressure and raise it back again slowly while providing the steering input (obviously) and monitoring the flow in the T line. Even the smallest leakage would be immediately revealed by the volumetric readings. And if you see the T flow increasing as you raise the pressure, you have an immediate indication that something is "fishy".

Plus - if everything is OK - you'll see at what pressure the relief valve cracks, and at what pressure it lets all of the 16 liters per minute pass through it. If you've never done this before - you'll be surprised to discover the override of certain relief valves you've been using all your life, I guarantee you that! And no turbine flow meter in the world will be able to detect a relief valve "cracking" as good as an oval gear one with a permanent magnet pickup (more on this later).

The same goes for pressure reducing valves. Of all kinds. From pilot pressure joysticks to reducing valves that supply pilot pressure to electrically controlled proportional DCVs. They work, and they do supply correctly reduced pressure as required, but an increased leakage rate is a good indicator of excessive wear. And very often the leakage rate is very small, a couple of liters per minute tops - so you can't reliably measure this with a 150 l/min rated turbine.

These are just a few examples, but there's more, so much more! My point is - a tool that allows you to detect low flow is very handy for checking leakage rates and "cracking" pressures of hydraulic components.

But what about the good old stop-watch and bucket test? Eh...?

It is unsafe, messy, and limited in time - and such a tool allows to perform the good old bucket test in a safe and scientific manner, and with a far greater precision. So - ditch the bucket.

Let me summarize the main advantages of this tool for the purposes of hydraulic diagnostics:

Its type (oval gear) and pick-up (permanent magnet + reed switch).

It is an oval gear type flow meter, which means that the flow measurement is volumetric, and this makes it extremely suitable for the detection of very low flow rates. Then - the pick-up is done via a magnet that is placed on the face of the gear and a reed switch. And this means that even if the speed of the gear is very slow, it will still generate a reliable pulse.

Most (if not all) turbine type flow meters detect the movement of the steel turbine with an inductive sensor. Inductive pickup is cheap, it's easy to seal and isolate from a high-pressure area, and it's virtually bulletproof, but it also needs the turbine wheel to move at a certain speed to be able to generate reliably detectable pulses. This is one of the reasons common turbine sensors max/min flow rate is in the order of 10-20 at most.

This tool is unbeatable in this regard. The PIUSI manual states the flow rate of 10 - 100 l/min for this model (6 - 60 for the oil version) but I assure you that it can do a million times better. Have a look at this small test that I ran:

I took a liter of oil, and pored is very-very slowly through the meter. It took me a couple of minutes - and it still detected every drop.

You can see that it reads 983 ml, which is very precise (considering that I also spilled some of the oil).

So - it can detect minute amounts of oil that pass through it very reliably. And the resolution is great as well. These are the numbers that I wrote down, as the volumetric reading was increasing:

1 014 11 163 21 313 31 123
2 029 12 178 22 327 32 477
3 044 13 193 23 342 33 123
4 059 14 208 24 357 34 506
5 074 15 223 25 372 35 521
6 089 16 238 26 387 36 536
7 104 17 253 27 402 37 551
8 119 18 268 28 417 38 566
9 134 19 283 29 432 39 581
10 149 20 298 30 447 40 596

As you can see - the resolution is about 15 cc, which gives about 66,6 pulses per liter. (The manual actually says the pulse version gives 35 pulses per liter. I'm guessing that this one uses two reed switches and the pulse version uses only one. I think I'll gut it and confirm this as soon as I have time. In any case, the point stands - it can read virtually zero flow, which makes it a perfect scientific replacement for the bucket/stopwatch combination.

Flow rate display.

The Second advantage - while being able to show you, with great precision, the amount of oil that passed through it, it also gives you the immediate flow rate reading. And while it can't measure hundreds of liters per minute, it can still, very respectably, go up to 100 liters per minute. Mine did 70-80 l/min for extensive periods and hasn't exploded - so I am very happy with its performance. Its enormous ratio between the max and min measurable flow is unparalleled.

Very effective learning aid.

It teaches you stuff. It really does. If you get used to using it all the time as you test hydraulic components, being able to perform an adjustment while looking at the T line flow rate will teach you a lot about pressure override and leakage rate of certain valves and components. This practical knowledge is invaluable, and you won't find it in manuals and catalogs. It's almost like it will give you the "sixth sense" on adjusting components. By "sixth sense" I mean the ability to quickly detect a fault.


It is relatively cheap. At least in comparison with turbine meters or high-pressure capable oval gear meters. This one cost us less than 200 bucks. I've had it for seven years now, working problem-free every day. A good investment.


It's pretty portable, there's no wires, and the batteries last for a long time. Definitely not the 18 months PIUSI promises in the manual, but still a long time. I actually had to hack mine, because it uses a weird battery size that is kind of hard to find, so I made a longer battery compartment to house two triple-A batteries:

But, of course, not everything is perfect, and this gadget has its drawback as well.

It doesn't work with oil at high temperatures.

Being able to reliably measure low flow rate makes this meter a perfect tool to measure case drains of hydraulic pumps and motors - but there's a very big catch. Since the pick up is based on a permanent magnet tripping a reed switch - when the measuring fluid gets hot - the magnetic field becomes too weak to trip the switch, and the meter, unfortunately, stops counting.

In high-performance hydraulic systems, where hydraulic motor drains can easily reach temperatures of 80 C and more, this meter, sadly, will not work.

It can't take high pressure.

The diesel model goes up to 30 bar and the oil one up to 70, according to the manual, but I wouldn't test it. The aluminum walls are thin, especially the backplate. So you can't insert it in a pressure line, which means - returns and drains only!

Slow update speed and low resolution of the flow rate display.

The instant flow rate reading is usable and relatively precise, but it's very slow and has limited resolution. For example, when it measures a fixed flow of 16,5 l/minute, its reading oscillates between 16,3 and 17,8, and the update rate is about once a second.

Its speed counting routine is very basic, it could be much better. Physically it is definitely capable of it. For its defense - it's not primarily intended to measure flow. Maybe I'll hack mine in the future. Just don't expect it to be lighting fast - and you'll be good.

The interface is very basic.

You obviously can't expect much from a cheap battery-powered device. So, no fast switching between units and whatnot. All you get is the volume and the flow rate, that's it.

In all - despite all the drawbacks, it's been a very useful tool. It's always sitting on one of the return lines of the test bench, and whenever I test something - it always "participates".