Piston vs Gear Pump 5

[lukin1977]

We have in operation one axial piston pump in an extruder press. Every 6 to 8 month we have to replace the pump because of internal damages. The pump is correctly selected for pressure, flow, etc. We are planning to change the pump and install one from another brand (better one) and we are thinking about changing the type of pump too. Maybe a gear pump

So my questions are: When it is necessary to use a piston pump and when a gear pump? Is a gear pump more resistant to sudden pressure changes?

Regards

lukin1977

 

[Oldhydroman]

Give us a clue - what is your fluid (and temperature), nominal pressure, drive speed and flow rate. Is the pump handling the stuff you are extruding or is it the hydraulic pump driving the hydraulic motor operating some sort of screw which is driving the stuff you are extruding.

Do you suspect that there are some pressure surges in your system that are causing the pump to exceed its peak pressure capability? If so, have you managed to measure these pressure peaks? (don't try to do it with a simple boudon tube pressure gauge - you need a digital gauge with a peak pressure measuring facility).

DOL

 

[lukin1977]

Fluid: Hydraulic oil Grade ISO 68
Operation Temp: 55 to 70 ºC
Nominal System Pressure: 1500 psi
Pump Nominal Pressure: 2500 psi
Drive speed: 1450 rpm electric motor
Flow rate: Pump fixed flow is 28 l/min

The pump is uses to move a cylinder that pressed the extruded material.
The pressure surges that I refere to are the ones created on DOL start and sudden machine stops. And also I think that a piston pump has many small moving parts inside that usually break like rings, little bolts, springs etc.
Gear Pump are more robbust?

The question is basically the same: Piston Pump vs Gear Pump, when to use?

lukin1977

 

[Oldhydroman]

You might want to consider moving this question to the fluid power forum to get a greater spread of opinion from the oil-hydraulic fraternity, i.e., the fluid power specialists.

My own opinion is that piston pumps would be the preferred choice for higher pressure applications and when there is a need for greater efficiencies, but in your application (running at 1500 psi) a gear pump should be quite alright - if not significantly cheaper and more readily available. Do expect it to be noisier, dirtier (creates more debris in the fluid) and less efficient (higher electricity usage, greater heating of the oil). However, here's a warning: you might be solving the symptom but not the problem!

You could say that a gear pump is more robust than a piston pump because it is more tolerant of fluid contamination - but that's because the volumetric efficiency isn't too good to start with so doesn't seem to degrade much when the pump wears a little. A gear pump has far less to go wrong with it but cannot be repaired economically - so you treat it as consumable and replace the whole thing when the time comes.

If you are looking for about 28 L/min at 1500 psi and 1450 rpm you would expect to use a pump with a displacement of about 22 cc/rev. This would be a [mid range] group 2 external gear pump or, for a higher pressure capacity, a [shortish] group 3 external gear pump. You might also want to consider an internal gear pump - supposedly quieter and robust.

I would, however, be very surprised if the piston pump you were using couldn't survive a pressure peak of 2500 psi - so I suspect that your pressure peaks (if that's the problem) are considerably higher than this, or that the problem is nothing at all to do with pressure.

The normal start up and shut down pressure surges are nothing to worry about. The run up time of your electric motor is relatively slow in the world of hydraulics. Similarly, the same rotary group in your pump is probably also used in that particular manufacturer's range of hydraulic motors - and they survive applications where they start and stop every machine cycle. You can be a little kinder to your pump if you want by making sure it is unloaded (outlet diverted to tank) whenever you start and stop the electric motor - but it still doesn't count for much.

If your measured pressure peaks are MASSIVE then they will also damage a gear pump. You can "live with it" and treat the cheap gear pump as consumable (renew it every three months) or you could try to do something about the pressure peaks which might be generated by your other components in the circuit or the poor control of them. You could investigate fitting accumulators, slower switching valves, pulsation dampers, critically placed hoses etc. I suppose you could do with a data-logger connected to your system and try to identify which part of your machine cycle is giving the pump the most grief. Then check which valve just switched at that part of the cycle - and there's your culprit.

Keep your mind open to the other (usually more probable) causes of failure - contamination of the fluid. What sort of filtration system do you have? Where are the filters? What grade of element do they have in them? What type of element media are you using? Do you have any clogging indication? Does anyone take any notice of the indicators?

Then look at your housekeeping practices: Does the system get topped up frequently? If so how is this carried out? If "Oily Joe" regularly pours in a pail of oil through an old traffic cone then expect trouble. Let's cut to the chase - do you have a measure of your oil cleanliness?

Also check what's happening on the suction side of the pump - piston pumps don't like too low a suction pressure (gear pumps will tolerate it better but if it's a problem then it still needs to be solved). Check that your suction pipework is big enough (work out the pressure drop at minimum temperature / maximum viscosity). Also check that you haven't got a hose with an internal collapse in the suction side, or that you haven't got a clogged suction strainer, or that the suction isolation valve isn't still partially closed when you think it's open.

DOL

 

[lukin1977]

We have a filter on the suction side of the pump. Inside the reservoir. The pump is located under the reservoir. The filter is change every month for preventive porpouses
We always fill the tank with new oil directly from it´s conteiner

Hydraulically speaking the machine is very simple.
Reservoir - Pump - Relief valve - Check valve - Directional valve - Manually Pressure regulation valve (on forward side of cylinder) - Manometer - Cylinder

I think that our problem is the quality of our actual piston pump. That is why I said that we are thinking to buy another brand.

Please recommend me some good Pumps brands

lukin1977

 

[Oldhydroman]

Well, as a direct response to a direct question, here's some excellent brands for piston pumps: Bosch Rexroth, Eaton Vickers, Parker Hannifin, Linde, Samhydraulik (part of Brevini), Kawasaki (KPM), Dynex-Rivett, Sauer-Danfoss, Moog, Oilgear, Rotary Power, Atos, Hawe, Duplomatic, Continental, Yuken ...

Here's some names that my colleagues might consider to be a little bit "second division" rather than "premier league" but I'm not sure that would always be the case: Huade, Metaris, Cassapa, Bondioli & Pavesi, Hydro-leduc, Hystar, Tokimec, Sunfab, Daikin, Nachi...

That's just off the top of my head and I'm not suggesting that each of these companies would have exactly what you want - BUT my "expert" advice to you is that you're barking up the wrong tree.

If the only filtration you have is your "suction filter" then that won't do much for you at all. The available pressure drop across a suction filter is so small that the filter has to be the size of a fridge in order to pass the required flow through an element fine enough to give the pump the proper protection. Chances are that what you do have is a "strainer" - something that could just about stop a wrench inadvertently left in the tank. If you are actually managing to clog this "strainer" then you must be pumping some extremely dirty fluid and your very short pump life is about right. If your strainer has a bypass valve then when it opens you have no filtration at all. Some piston pump manufacturers will void their warranty if you use a suction strainer - the reduction in the suction pressure at the inlet to the pump can cause a lot of damage (not just cavitation but also stresses trying to pull the slipper pads off the pistons in swash plate pumps).

If you are changing the element every month as a matter of course then that shows that you have no monitoring of the filter and that your maintenance efforts are somewhat unfocused and misdirected.

New oil "direct from the container" is not clean oil - it MUST MUST MUST be properly filtered prior to use.

Before you do anything else you should get your fluid analysed - your local hydraulic equipment supplier will be able to take a sample (there are special techniques involved in order to ensure the sample is properly representative of the fluid in the system). The particle count will tell you how clean your oil is - compare this with the pump manufacturer's recommendations. If the oil is too dirty then it doesn't matter how magnificent your new pump is, you will still get a short service life.

Or you could put on a cheap and cheerful gear pump and continue to run with dirty fluid. But it's not just the pump that suffers; you will find problems with your other valves and with the cylinder seals. The reliability of the whole machine will be less than it could be and you will be wasting energy as well. It's basically an economic argument: if the pump (and labor) is cheap and the unreliability of the machine doesn't cause any problems then run at the dirty end of the spectrum. The hydraulic components have no divine right to be run on clean oil, you own them so you can do what you like with them.

Remember that the original pump manufacturer sells to more customers than just you, if the problem really was with the quality of the pump then all of their customers would be having the same problem and the manufacturer would soon go out of business or shut down that product range because of poor sales or unprofitability. By the way, whose pump are you using now?

DOL

 

[lukin1977]

Oldhydroman- first of all. Thank you very much for taking the trouble of answer me. Your help is much appreciated.
Can you tell me what will be a correct setting of streiner and filter for my case?

Lukin1977

 

[micalbrch]

lunkin1977: I have a question. What does the piston pump do when the cylinder reaches its end position (pressure stroke)? Does it reduce the flow down to zero while holding the pressure? If yes, be careful with a gear pump. It cannot do the same. The gear pump will release the full flow over a pressure relief valve. That causes heat.

Oldhydroman's advise concerning "new" oil is very important. The oil from the barrels must be filtered when filled into your hydraulic tank. There are drum pumps available with integrated filter or - if that is possible - you fill the tank through your filter. What is the grade of filtration of your filter? It should be 10 microns at least for a piston pump.

 

[lukin1977]

micalbrch

when the cylinder reaches its end position the whole flow flows back to the reservoir through the PRV (pressure regulation valve). The piston pump is a fixed flow type. The PRV is used to manually set the desire extrution force of the cylinder during its stroke. Excess of flow always goes back to reservoir through the PVR

Apparently we dont have a filter in our system. Just a streiner which is just a metallic mesh for big particles (0,50 mm aprox). Where do we need to install the filter?

The pump is Hidrobom - Type: BGF

http://www.hidrobom.com.ar

 

[lukin1977]

Type: BFG

 

[micalbrch]

You can install the hydraulic filter in your system either in the return line or you install a separate filter circuit. Probably the first option is easier for an existing system.
Defintely the strainer does not help and this fact (no filter) might be the cause of the trouble.

 

[Oldhydroman]

The Hidrobom product range looks very similar to the [now long superseded] range of A2F, A2V and A7V pumps and motors from Rexroth. As long as the quality of the build is correct then there's no reason to suspect the pump of being fragile - this design of pump has many millions of running hours under its belt.

So I think we are all agreed that it's a fluid cleanliness problem. There are basically four places you can put a filter in a hydraulic system:

1) On the outlet of the pump - this gives good filtration for the oil because a high pressure differential is available across the filter. A filter here [let’s call it a "pressure filter"] protects the downstream system from debris created by the pump. The filters can be quite small for the flow rate involved and very fine (because of the high differential available) but they can also be expensive and heavy (because they are effectively demountable pressure vessels). A high pressure drop will increase your energy consumption (hope your electric motor is big enough) and this will decrease your efficiency (meaning your coolers work harder or your system temperature rises - or both). Pressure filters will eventually clean up the whole oil fill - but it depends on the other things happening in your plant.

2) On the inlet to the pump. This is the so-called “suction filtration” and it sound like it should be great but it isn’t because the pump can’t suck hard enough to give any worthwhile filtration action. The decrease of pressure at the pump suction port damages piston pumps – and Hidrobom specifically advise that you do not install any suction filters on their products (kiss goodbye to the warranty).

3) On the return back to the tank. This is “Return Filtration”. All the oil going back to the tank has to pass through the filter – so the tank contents are clean which means ALL the oil going to the pump inlet is clean (but is it really?). The problem with this approach is that the return circuits of the hydraulic system often can’t stand much back pressure without things starting to go wrong (cylinders start to creep out, manual valve actuators start to become too stiff to move and so on.) So to keep the back-pressure down the return filters have to be much bigger for the flow rate than the equivalent pressure filter. And there can be a flow amplification issue as well – if you have cylinders with a 2:1 area ratio then you get 56 L/min back from the full bore end for every 28 L/min you push into the annulus end – so the return filters have to be even bigger again. But I agree with Micalbrch that a return filter will probably be the easiest to install as a retro fit. This will have to be a “line mounted” filter – when installed as part of a new system the return filters are often fitted through the lid of the tank, it’s very convenient for a new build but you haven’t got that luxury.

The problem with relying on “return filtration “ is that it’s not quite true that filtering the return oil ensures the oil in the tank is clean. You must have a working/good quality/properly sized air breather filter as well. And you must have no other open ports in the top of the tank.

If you are using a “filler-breather” consider changing this for a “breather only” and then fit a quick release coupling for filling. This means that the only way new oil can be let into the system is for it to be pumped in – so also invest in a transfer pump with a built in filter no coarser than 10 micron. Then instil a huge amount of respect for fluid and component cleanliness with all the people that work with the system.

Unfortunately, even then you can’t guarantee that the oil in the tank will be clean because there will be some case drain lines and leak lines from external components (and from the immersed pump itself) that connect to the reservoir without any filtration at all because these lines cannot stand the slightest back pressure. Nevertheless, if you can afford it, a combination of pressure filtration AND return filtration provides a good and workable solution to the problem of oil cleanliness. Make both filters the same grade (otherwise the finer one will try to do all the work) and do not cut corners on the flow capacity of the filters – be super generous.

4) You can install a separate filtration loop, sucking oil out of the “return” side of the tank, filtering it and returning it to the “suction” side of the tank. This can work really well because you can leave it running 24/7, it will be a low pressure “off-line” circuit so the filters can be inexpensive – so you can afford BIG size and VERY FINE mesh, and the filters work very efficiently because they don’t suffer from the pulsation and surging you get in the main circuit. But you only clean the oil on what is effectively a bypass circuit – any huge contamination events (a big ingress of dirty oil) take several hours to get cleaned up and during all that time you are running on dirty oil and damaging your lovely new pump.

So what will be your plan of action? [everyone else feel free to chip in here]. I would suggest:

1) Order a new pump – same as before will be fine because bent axis piston pumps are actually quite robust.

2) Order a filter-transfer pump and a box of replacement elements (10 micron, inorganic fibre type)

3) Order a return filter – 10 micron, inorganic fibre element with a bypass valve and a clogging indicator. Also get a few spare elements so you can change them immediately they need it.

4) Order a new tank breather (3 micron – generous size) and some quick release couplings (a male half and a female half and their matching dust covers).

5) Drain down all your hydraulic oil (pipes, hoses, cylinders, valves, tank etc.) – be thorough.

6) Open up the tank and clean out all the silt. Remove the filler-breather and replace with the new “breather only”. Fit the quick release coupling (male) to the tank – this will be your fill point from now on. Make sure there are no other holes or open ports in the top of the tank. Fit your new pump (I would say fill its case with oil but it’s an immersed installation so that will happen automatically). Discard the suction strainer but you might to install a generously sized pipe on the suction port pointing towards the bottom of the tank (not too close – leave 2-3 diameters gap between the end of the pipe and the floor of the reservoir).

7) Close everything up, use your filter transfer pump to fill the tank with new oil (your old oil will have suffered a chemical degradation from being contaminated for so long). Ideally you would now flush the system with a larger pump/higher flow rate and with the actuators and DCVs looped out – but it’s difficult to get a larger pump in circuit when you have an immersed pump system – so we’ll be pragmatic and skip that point.

8) Start up at very low pressure and cycle all the operations until it is running smoothly (all the air pushed back to tank) – keep an eye on the fluid level and top up if necessary. Keep the system cycling at very low pressure if you can. The oil cleanliness will be improving all the time and although your new pump will be running on oil that isn’t yet quite clean enough – at least you won’t be running it at full pressure on the dirty oil and this limits the wear it will suffer until the oil becomes properly clean. After a couple of hours gradually bring the system up to the normal working pressure.

9) Monitor the filter clogging indicator very closely for the first few days – some people suggest changing the filter after 100 hours but you could just keep an eye on the indicator and change it when it starts to show that it is clogged.

10) Get yourself and your critical members of staff on a training course all about hydraulic system maintenance.

11) Let us know how it all went.

DOL

 

[lukin1977]

WOW. Oldhydroman. Excellent instructions! Thanks.
I must add some things that are very important and that will probably change the "where to install the filter" problem. I did not mentioned before but the electric motor is actually driving 2 pumps (one on each end of the motor). At one end there is the HPLF (high pressure - low flow: 40 l/min (I said 28 l/min but I double check my data and it is 40 l/min - Hidrobom) at the other end there is a LPHF (low pressure - high flow vane pump: 105 l/min). The Vane pump sends oil to the cylinder for backstroke and forward stroke until the cylinder head pushes the material ant the pressure of the system raise. When the pressure raise (150 psi I think) the flow from the vane pump recirculates back to the tank. So from the cylinder point of view: under 150 psi = 145 l/min, above 150 psi = 40 l/min coming from pumps. This is done for speed on the cylinder stroke on reloading

After reading the different alternatives I have an idea. What if I install a return filter on the recirculating line from the vane pump? So I avoid raising pressure on the main return line.

The system has 2 major places where I am sure that a great amount of dirt is entering to the system:

1st: The piston ram (Im not sure if this is the coorect term) going back with extruder material. The material is flux for welding electrodes. So it is basically abrasive sand

2nd: Top of the tank. Tha tank ... IS OPEN!. yes, believe or not. It only has a metal sheet laying on the top without any seal (maybe I can upload a picture)

I already ordered a new pump from Hidrobom and ask for a quotation of an internal gear pump to Bosch-Rexroth
I will correct the tank top problem ASAP.

What do you think about the filter at recirculating line?

 

[Oldhydroman]

I still think it would be a good idea to get a measure of your oil cleanliness in the "before" condition. Then, when you've spent a bit of time and money sorting out the filtration you can take a sample of the "after" condition, compare the two and prove to your bosses (and/or your maintenance staff) that the effort/expense was actually worth it. If you get the analysis done the human way (looking through a microscope) then the lab will be able to tell you exactly what the contamination is - to confirm that it is welding rod flux (you hope).

Sorting out a sealed lid for the tank is a must, then get some decent filtration and, at the same time, ensure that the fluid maintenance procedures and resources improve so that your gains are not subsequently lost.

With regard to just filtering the recirculating flow of the low pressure pump - you will only have filtration occurring when the the LP pump is recirculating. Does this represent a long enough proportion of the cycle to be effective?

You have correctly identified that a significant source of contamination will be that which is dragged in on the cylinder rod. If you're just filtering the LP pump bypass then this [dirty] returning cylinder annulus oil will still be getting back into the tank and your sensitive piston pump will still be sucking it in.

How about this for an idea? Replace your LP vane pump for one with separate outputs: one original size flow for the existing LP circuit, and one new flow giving an extra 60 L/min. This extra flow is bigger than your HP pump (150% of 40 L/min). Pass this new flow through a big, fine, low pressure filter and then into the inlet of the HP pump. Spill the extra 50% (20 L/min) back to tank over a simple check valve (say with a 45 psi spring). This will boost the inlet to the piston pump (it will like that) and only feed the piston pump with pre-filtered oil. Eventually the bulk oil content will clean up but your new [sensitive] piston pump will be saved further trouble.

DOL

 

[lukin1977]

I will pick up a "before" sample.
The filter at LP recirc line will be very efficient because one cycle of the cylinder lasts 15 min= 45 sec for reload and 14 min, 15 sec is the high pressure time so all this time the recirc line is ON. (95% of the cycle). And this is just for our fastest consumed material. For other materials the % is bigger
I think i will do this first because is easy, cheap and fast. It will not ensure 100% clean oil into HP pump but i think it will do a BIG difference to what we have now .
Your suggestion is the best technicaly speaking but I think is:
1. Too expensive
2. Too much production downtime. Time we really dont have now or in the near future
3. I am sure that management wont approve it because is too "risky". It consists in a mayor modification to the system and if we have any trouble it will be very hard to go back
4. Easy, fast and cheap is hard to beat

Probably we can do the installation of the filter in the recirc line first and analize the results and then see if is necessary to improve
Going back to the original problem. I still have doubts about the cause of the damages of our piston pumps because:
1. The nature of the fails of the internal parts after we disassembly and analized
2. We all agreed that we (my plant) have serious dirty oil issue but the oil can only become dirtier and dirtier because the lack of filter. So the frequence of damaged pump should always become sooner each time. This is no true. We have an uniform period of time between HP pump fails along the years

lukin1977

 

[Oldhydroman]

Yes - I agree, 95% recirculation time is good enough to be effective in cleaning the oil.

I also agree that easy, fast and cheap is hard to beat.

You also have a good point about expecting the oil to get progressively dirtier and so the pump service lives should be progressively shorter. And because you're not seeing this it does suggest that somehow the Hidrobom pumps aren't as good as they should be. So a two pronged attack is the way to go: clean the oil and get a better pump.

Good luck with your brave new world.

DOL

 

[lukin1977]

Here is the planned filter instalation
notes:
1 High Pressure pump
2 Low Pressure pump
3 Oil return collector

This is an old picture.
We had intalled guards over the couplings and a check valve after HP hose

lukin1977

 

[Oldhydroman]

Thanks for sending the picture - it's nice to get to see what we've been discussing. I appreciate that in all circumstances we each do what we can with the resources available to us, but do tell me, in which part of the world is this installation?

DOL

 

[stanier]

If the piping is a welded system was the piping cleaned/descaled before being put into operation? This is historically one of the main reasons for cleanliness problems in hydraulic systems.

“The beautiful thing about learning is that no one can take it away from you.”
---B.B. King

http://waterhammer.hopout.com.au/

 

[lukin1977]

Oldhydroman: Paraguay
Stanier: No

lukin1977