• Welcome to engineeringclicks.com
  • Gear pump design

    Discussion in 'The main mechanical design forum' started by NovaNath, Feb 15, 2014.

    1. NovaNath

      NovaNath Member

      Joined:
      Feb 2014
      Posts:
      6
      Likes Received:
      0
      Hi, hopefully someone with experience of gear pumps can give me some advice!

      We've designed and manufactured a pump with mod 6 gears, for high flow rather than high pressure.
      However, when we increase the speed beyond 300 rpm the pump seizes and galls the housing.
      There's not a tremendous amount of backlash between the gears, and the theory I'm working on is that the oil can't escape quickly enough from the closed pocket created as the gears come into mesh. There are wells in the closing/bushing plates that leave a narrow piece of land just where the gears mesh (copied from an automotive oil pump), but my question really is, how wide should this land be in relation to the gear mod?

      Any help most gratefully received! :)
       
    2.  
    3. Erich

      Erich Well-Known Member

      Joined:
      Feb 2012
      Posts:
      223
      Likes Received:
      2
      Could the problem be related to temperature? Differential expansion of parts may be the cause.
       
    4. NovaNath

      NovaNath Member

      Joined:
      Feb 2014
      Posts:
      6
      Likes Received:
      0
      It runs and pumps ok at 300rpm, but when we start to ramp it up (1200rpm target) it seizes.
      We have hardened steel gears in a mild steel housing and brass bushings so hopefully things will grow at a similar rate; the application is to pump 200°c oil!
       
    5. Lochnagar

      Lochnagar Well-Known Member

      Joined:
      Feb 2011
      Posts:
      157
      Likes Received:
      0
      When something spins - the radius of the gear grows in relation to the square of the rotational speed - so I would expect the growth on each gear (in a radial direction) to be 16 times greater at 1200rpm - than 300rpm. Bear in mind you have two gears - "growing towards" one another - then this maybe something you haven't factored in??
      See page 704 in Roark's Formulaes for Stress and Strain - 6th Edition.
      Hope this helps.
       
    6. Erich

      Erich Well-Known Member

      Joined:
      Feb 2012
      Posts:
      223
      Likes Received:
      2
      The gears will not be at the same temperature as the housing. They will be hotter.
       
    7. NovaNath

      NovaNath Member

      Joined:
      Feb 2014
      Posts:
      6
      Likes Received:
      0
      Roark's Formulaes for Stress and Strain; what a phenomenal resource:
      http://materiales.azc.uam.mx/gjl/Clases/MA10_I/Roark's formulas for stress and strain.pdf

      I think the short answer is that I have gears which are too close fitting with minimal backlash where they aught to be loose; and insufficient accuracy in the housing where we really require it.

      Erich: will the gears be hotter due to immersion in the hot fluid or because of the work being done moving the oil?
      By the way, we are testing the pump at ambient temperature before trying the oil heater.
       
    8. Lochnagar

      Lochnagar Well-Known Member

      Joined:
      Feb 2011
      Posts:
      157
      Likes Received:
      0
      Cheers NovaNath for the pdf link to Roark's Formulae For Stress and Strain. It is a more up to date version than my book:)

      I ran some numbers on some 50mm gears (just to get a "ball park" feel for the problem) - both for temperature and for rotational speed - and I was surprised at how small the numbers actually were. So whilst it is true that the gears do grow under both temperature and rotational speed as previously discussed - I am not persuaded that the numbers are significant in this particular application.

      Maybe as you say - if the gears have minimal backlash to begin with - and maybe if the limits and fits between the gear and shaft and shaft and bearing are "sloppy" - that might be another issue - or misalignment between housing bearings and cover bearings.
       
    9. NovaNath

      NovaNath Member

      Joined:
      Feb 2014
      Posts:
      6
      Likes Received:
      0
      My colleague's thoughts were that the gears would need to be massive for significant growth due to rotation speed.

      Anyway, we have finally resolved the issues.
      After false starts where we were basically just polishing high-spots off the gears and housing, we had a precision engineer check bushing and housing concentricity and the error was greater than the design clearance between gear and housing. Block and lower plate were remachined when assembled and block/top plate were faced.
      Simultaneously we got the gears groud as they were running out a bit.

      It's surprising it even pumped at all! :)

      It ran so much more smoothly after this work and we gradually took it up to 1200rpm, at which point it ran for about a minute and then seized!
      Stripped down and found the faces had picked up, so surmised we had insufficient oil film between faces and plates; machined oil grooves into plates to flush oil across face from high pressure side to low pressure side. Repeated test; failed in exactly the same manner as before (1200rpm). Strip down indicted same problem as before; my colleague had the theory that the gears were warming up due to doing work to the oil, expanding a little till there was insufficient oil film to keep the gear faces away from the plates, at which point they seized.
      Face clearance measured as 0.05mm overall, and we calculate an 80°C increase in gear temperature would make the gears expand this amount. We shimmed the top plate by 0.1mm to give 0.15 overall face clearance. Retested and boom...up to speed, no problem!

      Adding heat hasn't affected the pump... so far!
      We should be delivering to the customer early next week :)
       

    Share This Page