The following warnings occurred:
Warning [2] Undefined variable $search_thread - Line: 60 - File: showthread.php(1617) : eval()'d code PHP 8.1.27 (Linux)
File Line Function
/inc/class_error.php 153 errorHandler->error
/showthread.php(1617) : eval()'d code 60 errorHandler->error_callback
/showthread.php 1617 eval




Thread Rating:
  • 0 Vote(s) - 0 Average
  • 1
  • 2
  • 3
  • 4
  • 5
Diff mesh pattern puzzle
#1
Many aspects of motor vehicles defy immediate logic. For example top rings wear on the top surface not the bottom. Seven cranks often fail at the flywheel where there is little theoretical bending.

Older books which show diff cw mesh patterns often mildly state “with the cw rotated”. However in a recent book I found this very much emphasised. On a diff with good bearings I cannot see how the pattern would alter whether pinion or cw rotated. Any theories? 
Has anyone experimented and what did they notice? I sometimes wonder if it was meant to emphasise the cw pattern rather than the pinion pattern and  the editor got confused, but I have now read it in so many places this cannot be the case.
Reply
#2
Greetings Bob

Here's what my 1953 edition of "Modern Motor Engineering" has to say:

"When a bevel drive gear is dismantled, cleaned and examined, the condition of the mating faces of gear teeth present a clear indication of how the gears have been meshing.  Similarly, by using engineer's marking paste on newly assembled gears one is readily able to check these for correct meshing.

In this connection it should be mentioned that when the engine is driving the rear wheels, the driving faces on the pinion and crown-wheel teeth are different from those in driving contact when the rear wheels are driving the car, as during coasting operations.

Referring to Figure 35, which shows tooth contacts under different conditions, Fig 35 (2) shows the ideal meshing contact for well-made new bevel pinion and crown-wheel gears.  To check this contact, paint the faces of the teeth of the crown-wheel at three or four different places around the circumference with white lead, using this sparingly. With the pinion in position relative to the crown wheel so that its heel is flush with the outward circumference of the crown-wheel, the latter should be adjusted so that the white lead will be wiped off its teeth, as shown in Fig 35 (2).  This may be done by rotating the crown wheel several times in a direction opposite to normal rotation on forward speeds. (My Italics)  This will cause the white lead to be wiped off the gear teeth sufficiently to indicate the contact region"

I can't easily reproduce Fig 35 (2) here, but suffice to say it shows an ideal line of contact which sits evenly between the face (outward extremity) and the flank (root) of the tooth, and not biased to either the heel (large end) or toe (small end) of the tooth.

My interpretation of this is that you could either rotate the pinion in the normal direction for forward drive, or rotate the crown wheel in the opposite direction.  However, given the natural friction in the assembly, rotating the crown-wheel might produce higher pressure between the teeth, and thus give a better marking ?
Reply
#3
Bob no component is perfect, I suppose they are covering the possibility of some variation at different points on the wheel due to run-out, etc. As you observe, this should be tiny IF the components are good and IF they are correctly assembled... No point getting it all set up lovely at 12 o'clock if there's a tight spot at 6 o'clock.

My own experience is limited but I found it all much less clear than pictures in books!
Reply
#4
Fig.35 from ‘The modern motor engineer’


Attached Files Thumbnail(s)
   
Reply
#5
These diags and better appear in many books, many with advice on which way to move the gears to correct. 
But why does it matter which gear is turned to obtain contact? A slight loading is applied either way.
If anyone has a diff on the bench perhaps they could experiment.

 I suspect many Austin mesh patterns are dubious. Tend to be altered whenever an axle is replaced.The few failed ones I have seen only did so when the teeth wore entirely through! I dunno what they sounded like? The diff oil, if any, must have been about 30% metal.

The observation that diags apply to new gears is important. Can waste hours with worn gears. Have to settle for the best, usually as previous .If there is a wear step the teeth can ride entirely on it so should be carefully removed. (The later pinions were shortened presumably to avoid this. A lesser step forms on the cw.)
Reply
#6
Bryan,

I found this on YouTube it is quite a good explanation.
https://youtu.be/nVb5WRUfM7Y



The lower two diagrams on your post  show high and low contact areas on the tooth, said to be noisy. 
This has been my problem although the wear pattern on both gears looks even along the length of the tooth. 
How does one eliminate this, as moving the crown wheel in or out affects the backlash, does this also then affect the height of the contact area?
Could someone please explain how to achieve the best height?

Roger
Location:- Haverfordwest, Pembrokeshire.
Reply
#7
In assessing high and low note that this is relative to the possible extreme contact area; there is considerable air space in the valley floor. The backlash figure is a vital guide with new gears. Minimum must be maintained always but if significantly worn ithe ideal will likely be increased; the pattern is first factor.
The Youtube videos adopt remarkably large adjustment increments. I dunno just how applicable hypoid diags are to spiral bevel.
The diagrams above curiously show the drive side of cw teeth as concave?
Note that 42:8 diffs teeth do not "hunt", every pinion tooth does not contact every cw tooth. Can get different wear in the two sets of mesh which can be very confusing!

The various Youtube videos are entertaining but the maker's enthusiam often exceeds knowledge or skill. One on a Mustang diff, they use a bearing crudely relieved to simplify repeat assembly, but, except for final pattern, do not check that its assembled dimension same as the new one used.
Reply


Forum Jump:


Users browsing this thread: 1 Guest(s)