Poor quality gears

I mentioned the advantage of a true universal mill to some one recently on here who was lusting after a Bridgeport. Trouble is though finding one that has the gear with it and is in good order. Some of the smaller ones are also surprisingly heavy and are likely to be odd not well known makes with what seem to be rather large motors.

The brown and sharp book I have and mentioned is also on the archive, several versions of it. One thing it adds is the best route to designing a set up and why. For other aspects the book I linked to in my opinion is a lot clearer. Just search gearing.

John

-

Your normal home workshop is a bit bigger than mine . A milling spindle on a swiveling vertical slide provides the same functionality:

Neil's Potts could have a use!

A pair of trial 32 DP 60 and 30 degree gears in ally (actually 31.75DP = 0.8MOD)

Cheers,

Rod

Very possibly, but I suspect that my universal mill and dividing head may have cost less than a Myford swivelling vertical slide and milling spindle judging by some of the prices on the internet.

Your gears look good. Presumably the milling spindle is swivelled to the helix angle and the leadscrew gear train is set up to provide the appropriate rotation of the spindle for a given saddle movement. Is the cut applied by turning the leadscrew handle? When it comes to indexing each tooth presumably the gear train is disconnected, the spindle moved an appropriate number of teeth and the gear train reconnected? Always assuming of course that the number of teeth required is a factor somewhere in the gear train.

I don't follow the 30º and 60º helix angles for the same OD. Presumably if the OD of the two gears is the same then the PCD will also be the same for both gears? I am assuming that the number of teeth is in the ratio 2:1. The PCD is given by the number of teeth divided by the product of the normal DP and the cosine of the helix angle. So if I've done my maths correctly twice the cosine of the larger angle (on the gear with fewer teeth) should equal the cosine of the helix angle on the gear with the smaller angle. That does not seem to be the case for 30º and 60º. I've no doubt that I am making a prat of myself over this by missing something obvious, but hey, I should learn something from it.

Even if it is only shutup and don't post nonsense.

Andrew

Andrew,

Sorry, I have misled you slightly - the angles for identical PCD (2:1 ratio) are 26.6 degrees and 63.4 degrees but 60 and 30 is as close as you can (sometimes) get from stock gears. PCD is, as you say, number of teeth N / (DP x COS (a)). so for 32 DP 16T 30 degrees PCD is .577" and for 8T 60 degrees .500". Quite close: They are identical for the "magic" angles.

You are quite correct about the technique for machining except that I mounted the blank in a collet in the 3 jaw with a dividing plate mounted on it and an indexer fastened, lathe carrier style, to the blank.

I'm sure you are correct about the relative equipment costs but possibly nullified by the need to move to a bigger house - my traction engine is going to be just the one MInnie

Cheers,

Rod

Rod: Ah, that's neat putting the dividing head on the spindle. I'm relieved that my maths isn't completely out of order. While waiting for supper to cook earlier I thought I'd have a go at solving the problem analytically. For those that can't be bothered to do the maths here is my solution:

I make the PCD of the gears, with the precise angles, equal to 0.559". That gives leads of 3.5123" (89.217mm) and 0.8781" (22.303mm). From the tables for my dividing head the nearest leads with standard gears are 89.30mm and 22.32mm - it's a metric mill. I reckon that gives errrors of 0.9µm and 0.7µm per millimetre of face width. That's pretty small and certainly good enough for skew gears. In practise these relatively short leads would require a significant step up ratio in the gearing, so the dividing head spindle would be better driven directly from the table rather than through the worm reduction in the dividing head.

To be precise the universal mill was £145 and the dividing head £125. One would need a pretty solid workshop for the mill, it weighs around 3500lbs, has 5hp motor and knocks the Bridgeport into a cocked hat when it comes to rigidity and simply ploughing through metal like it wasn't there. I've learnt the hard way that you really need to clamp the work down properly.

Andrew

Well done. Nice to see it done properly! I'm afraid I resort to a spreadsheet driven iterative process for this sort of things these days.

Rod

Andrew - you can cut out the stuff in the middle and combine lines 1 and 2 directly, giving:

2 Cos(90-a1) = Cos(a1)

but Cos(90-a1) = Sin(a1)

So line 1 becomes 2Sin(a1) = Cos(a1),

ie Tan(a1) = 1/2.

(The mixed blessing of having children of secondary school age)

Still don't understand the gear stuff though!

Murray

Edited By Muzzer on 16/07/2015 22:29:58