Pulley taper – Herbert B drill

[gerry maddenParticipant @gerrymadden53711]

The drive pulley on the motor of my mid 60’s Herbert B drill sits on a tapered sleeve, as you can see above. I want to make/find some larger pulleys so have measured (using gauging balls) the taper (total angle) to be 16 deg. 42mins., or 3.52″/ft.

There may be some error in my measurement of course, but from the application and my approximation, can anyone deduce the actual taper spec this component is likely to be please?

Gerry

 

 

[David SeniorParticipant @davidsenior29320]

I don’t know whether there is any connection but the taper on an R8 spindle is apparently 16 deg 51 mins

Dave

[FulmenParticipant @fulmen]

You did all the math and still missed it. It works out to a ratio if 0,300014. The accuracy is impressive.

[gerry maddenParticipant @gerrymadden53711]

Fulmen, it seems you are right! But I’m less clear on the significance of 0.3 in the engineering world. What am I missing?

Gerry

[FulmenParticipant @fulmen]

Nothing, except that whole numbers tends to be intentional. So it stands to reason that the taper is in fact 16°42′ or 3/10.

 

[gerry maddenParticipant @gerrymadden53711]

So Dave Seniors R8 with its 16deg.51mins, or 0.3029, is a dreadful abberration on all levels then 🙂

[FulmenParticipant @fulmen]

Grmf. You might be right, it’s awfully close. And the 0.3 ratio is just “off” compared to most standard tapers, both metric and imperial.

[not done it yetParticipant @notdoneityet]

Questions:  How are you driving this motor? If 3 phase mains or converter, run it from a 220V VFD after setting the motor to delta configuration

A VFD should likely provide an adequate increase in drive speed.  Might even exceed the design limits of the drill?  I’m assuming, here, the motor spindle is running on ball bearings, not sleeve type.

[gerry maddenParticipant @gerrymadden53711]

NDIY    I’ll be running it from a rotary converter. This drill is designed to run up to 18000RPM. It should come with a set of pulleys enabling this but unfortunately mine has only the smallest which will only give me approx. 2800RPM. As I already have a mill that can do 3600RPM I need to make a large pulley so I can get the benefits from this unit.

I intend to drill small holes in bearing rings for lubrication feed, so high speed will cut down the time spent doing this…but of course the real pleasure is just bringing and old machine back to a full life 🙂

Gerry

[noel shelleyParticipant @noelshelley55608]

Put different size pullies on the motor shaft which will be parallel, may be 5/8″, fix with a grub screw. Noel.

[Michael GilliganParticipant @michaelgilligan61133]

I just stumbled across this, Gerry … from 2020

https://www.mig-welding.co.uk/forum/threads/pollard-bench-drill-refurb.92809/page-3

Check post #60

MichaelG.

.

Hint:  http://www.herberttooling.com/?page_id=12

[gerry maddenParticipant @gerrymadden53711]

Noel, I did wonder about doing as you suggest, it would certainly be an easy-to-do solution. But to run at 18k I think the taper fitting will best for minimising eccentricity and vibration in the driveline. I already have the AEI motor stripped and can see that its build quality is excellent for what appears to be just an ordinary 3ph industrial motor. When I see this sort of thing, I don’t like to compromise on other aspects, unless I really have to 🙂

Michael, thanks for the link on the drill overhaul. I hadn’t seen this one! Mine is similar to the green one so I will enjoy going through it later. Re the ‘herberttooling’ link, I had already been in contact with them. Sadly they informed me that they no longer sell any Herbert m/c tool parts, only general engineering components.

 

[gerry maddenParticipant @gerrymadden53711]

After some internet trawling on the likely standard that Herbert have used for this pulley taper, it seems that Fulmen was half right. From some time ago it would seem that industry has generally agreed that the standard for “non-stick” tapers should be 3.5″/foot. This is close to a ratio of 0.3 but not quite.

When it comes to the angle subtended by this specific gradient, there is lots of variation in articles on the net. I ‘think’ this could the result of different ways of measurement leading to people to use Tan instead of Sine and vice versa.

My gauging ball method determined and angle of 16deg 42 mins on my pulley. The 3.5″/ft spec gives a subtended angle of 16deg 36min. This looks surprisingly close but may be more from luck than anything else. However, this raises another question in my mind which time-served designers and machinists may be able to answer. What would be a typical angular tolerance for manufacturing the taper used to couple two parts in a machine?

Gerry

[Michael GilliganParticipant @michaelgilligan61133]

Good work so far, Gerry

Whilst not being a trained fitter … I regret to say that ‘near perfection’ will be the grade of fit that you are seeking: get the blue out and start chasing those ‘tenths’

MichaelG.

[Michael GilliganParticipant @michaelgilligan61133]

I was just digging-around on the forum, and came across this post:

https://www.model-engineer.co.uk/forums/topic/alfred-herbert-high-speed-bench-drill/#post-575090

which could potentially be very helpful

But, unfortunately the author @davidgurney1 has not been active recently.

MichaelG.

[gerry madden]

On gerry madden Said:

… However, this raises another question in my mind which time-served designers and machinists may be able to answer. What would be a typical angular tolerance for manufacturing the taper used to couple two parts in a machine?

Gerry

Not time-served, but I don’t think tapers were made by measuring angles, so no need for an angular tolerance.

I believe they were cut by measuring linear distance, not by setting machines to a measured angle.   Inches are easy to measure very accurately, angles aren’t.  Herbert would have set up to cut a 7:24 ratio and not measured an angle of 16°35’16”.   Like as not the tool-room made tenths accurate gauges for the purpose, so shop floor machinists didn’t have to do anything complicated.

However, the angle is handy for identifying an unknown taper in the workshop because it’s tricky to recover the ratio from a real taper; tapers are an awkward shape.  Fortunately not too difficult to measure the angle well enough to tell the difference between a 3:12 taper (14°15′), a 7:24 taper (16°35’16”), and a 1:3 taper (18°55’28”)

Even though our forefathers took a long time to standardise and there are potentially several unusual tapers lurking in junk boxes it’s fairly safe to assume an approximate 16deg 36min pulley has a 3.5:12 ratio taper.

Dave

[gerry maddenParticipant @gerrymadden53711]

 

Not time-served, but I don’t think tapers were made by measuring angles, so no need for an angular tolerance.

Dave, surely every feature on a good quality engineering drawing has to have a tolerance?!

You are right that in the process of manual machining the angle is generally achieved by ratios, but I cannot recall a ratio being specified on an engineering drawing. It’s usually specified in degrees and then converted by the machinist, process engineer or cnc programmer for production?

Gerry

 

 

[gerry maddenParticipant @gerrymadden53711]

Thanks Michael, that was a really useful link. I should just get on and start machining now 🙂

[Howard LewisParticipant @howardlewis46836]

The BS tapers used on the arbors on milling machines specified the taper as 3.75″ per foot

In the same way the taper used on the camshafts for the gear on In Line Fuel Injection pumps , and their drives, was, from memory, quoted as 1 in 6.

In more modern times (With Metrication?) ER collets are specified as having 16 degree , and 60 degree, tapers.

Howard

[gerry madden]

On gerry madden Said:

 

Not time-served, but I don’t think tapers were made by measuring angles, so no need for an angular tolerance.

Dave, surely every feature on a good quality engineering drawing has to have a tolerance?!

You are right that in the process of manual machining the angle is generally achieved by ratios, but I cannot recall a ratio being specified on an engineering drawing. It’s usually specified in degrees and then converted by the machinist, process engineer or cnc programmer for production?

Gerry

 

 

Well, I said no need for an “angular tolerance”, not that there’s no need for a tolerance at all.   A linear tolerance might well be specified to the ratio, maybe  3.5000 : 12.0000″ (implying tenths accuracy), or explicitly ± somefink.

Not much advantage in specifying an angle by ratio on a drawing unless it’s based on whole numbers, and most angles aren’t.  Tapers may be a special case because they are ratio derived.  This drawing defines MT tapers with linear measurements, not angles:

That the drawing specifies a centre tip angle of 60° implies anything between 29 and 31° degrees will do.  It’s not critical.  Conversely, the taper is critical, and the necessary precision is in table defining A,B,C,D,E and L.

Conversions are certainly done to convert drawings into reality, machinists have always complained about drawings that are unnecessarily difficult to make and then cannot be assembled due to other parts getting in the way.  In my workshop it’s all done by me, but firms like Vickers had large teams.  Plenty to go wrong between Chief Engineer and the machinist, many of whom were unskilled.  Tool-rooms and production engineers were essential!

In this example I’ve drawn a 3½:12″ taper both ways, and I think it shows, in this case, that ratio is ‘better’, not least because it make it easy to set up a sine bar.

Conversions are best avoided because they’re extra work and error prone!    The common 4-figure trig tables introduce error and slide-rules are even worse!  At school, the 5-figure trig tables were kept locked up, too expensive for children, and only brought out by sir on special occasions,  Yobs were allowed to look but not touch!   Modern calculators do better (mostly!), but – for making tapers – I suggest it’s hard to beat specifying the angle as a ratio.

This table from 1947 is useful:

Sorry about the blur, the taper per foot fractions are:   1/8, 3/16, 1/4, 5/16, 3/8, 7/16, 1/2, 9/16, 5/8, 11/16, 3/4, 13/16, 7/8, 15/16, 1, 1 1/4, 1 1/2, 1 3/4, 2, 2 1/2, 3, 3 1/2, 4, 4 1/2, 5 and 6

Dave

 

 

[Author]

Posts