PDA

View Full Version : Tapered sheave bushings?



lazlo
06-25-2008, 12:25 PM
Inspired by Macona's success with his 10EE, I'm doing a servo drive retrofit for a Bridgeport J-Head I bought without a motor.

The AC brushless servo I have has a 24mm shaft, and most commonly available sheaves in the 'States have 1" bores. But MSC carries tapered 24mm 'H' and 'B' bushings, and the matching pulleys.

I know tapered bushings are used for high-speed rotation, but is there any advantage of a tapered bushing for a machine tool spindle, where the rotation is ~ 4,000 RPM max?

What's the difference in the tapered bushing styles ('H' and 'B')? Emerson/Browning's catalog is pretty lousy, and there's no product selection guide for the bushings...

By the way, the reason I'm buying the pulley instead of making it is because the Browning's are ~$38 for a 7" cast iron sheave, and MSC is having their bi-weekly 25% off sale, so I can't buy the aluminum or cast iron stock for that price...

wierdscience
06-25-2008, 08:48 PM
H1 ,P1 and Q1 are the most common sizes in the Browning line.The B size is an attempt to introduce a size series that more closely mirrors the QD style bushings that have become increasingly popular over the years.

Browning is the only bushing with an external key which keys it into the mating bore.That is the main advantage the Browning taper bushing has over the QD style.

Specs for the different Browning bushings can be viewed here on page 3

http://www.emerson-ept.com/eptroot/climate/pdfs/HVAC-05%20Catalog%20LR%20p150-157.pdf

If all you have is a 24mm shaft an H1 is all you need,it is smaller,lighter and pulls down to a narrower assembled profile(overall width of the assembled bushing and PT componet),Plus it's cheaper and easier to find matching sheaves.

One thing to check when you order your sheave,make sure it is balanced to your RPM requirement.Not all styles are,some are just a machined casting intended for rough service at lower speeds where balancing isn't required.

I always recomend taper bushings over plain bores when possible,you get better shaft location,a tighter fit and less chance of creeping axially.

One more tip,the bushing instructions say to use no lube on assembly,BS,a light coat of way oil on the shaft and sheave bore means you won't have trouble removing them later.

lazlo
06-25-2008, 09:22 PM
If all you have is a 24mm shaft an H1 is all you need,it is smaller,lighter and pulls down to a narrower assembled profile(overall width of the assembled bushing and PT componet),Plus it's cheaper and easier to find matching sheaves.

Thank you so much Wierd -- that's exactly what I needed!!!
Just put the order in to MSC...

lazlo
06-25-2008, 09:32 PM
By the way, because brushless servos have a flat torque profile from 0 to 3,000 RPM, I'm going to skip the drama of extending the motor shaft to match the stock shaft on the Bridgeport pancake motor. If anyone's interested in doing a similar Vari-speed conversion without modifying a stock C-face motor, here are some measurements I took today:

The spindle pulley stack is 4 pulleys, 7.25", 6.25", 5.25", and 4.25" OD.

I bought the Bridgeport J-Head (just the milling head) without the motor and motor pulleys, but by extrapolating the stock speeds on the Bridgeport motor plate, the motor pulley stack would be 3", 4", 5", and 6".

Each belt section on the pulley is .625" wide. Standard A-33 V-belt.

The aluminum head casting is .6265" thick where the motor mounts.

There's a .125" gap between the bottom of the aluminum head casting and the top of the first spindle pulley.

A standard NEMA 58C/145C C-face motor has a 2.125" shaft. The AC brushless servo I have has a 55 mm (2.165") shaft.

So 2.125" (standard motor shaft length) - .6265" (alum head casting) - .125 (clearance for the spindle pulley) - [thickness of your motor plate] = how far your motor shaft will extend into the pulley chamber.

I made a 3/4" thick motor mount, so my 55 mm motor shaft will extend .6625" into the pulley chamber. Since each pulley stack is .625" thick, that gives .0385" of tolerance. If I need more room to play with, I can drop the motor plate to as thin as 5/16".

wierdscience
06-25-2008, 09:56 PM
Wish I had known that,in your case the Taper-Loc bushing design would have been a better choice.The H series will work fine,just the taper-locks are a bit neater install and even more compact than the H series.

See page 3

http://www.dodgept.com/pdf/catalog/pt_components/2004_pdf/bushing_taper_lock_spec.pdf

Oh and you could have ditched the vee-belt pulley on the quill and went with a poly-rib belt,but to late:D

lazlo
06-25-2008, 10:03 PM
Doh. I just checked MSC, and they don't seem to carry Taper-Loc bushings. In fact, MSC doesn't appear to carry Dodge at all.

Do the Taper-Locs need special Dodge sheaves?


Oh and you could have ditched the vee-belt pulley on the quill and went with a poly-rib belt,but to late

I thought about doing that, or a timing belt, but then I'd have to replace the spindle pulley stack, and the Bridgy spindle brake is buried in there.

wierdscience
06-25-2008, 10:18 PM
The Taper-lock sheaves to match the bushing,they are made by Dodge,Martin sprocket ,Fenner and a few others,but not by Browning.

See it's like this-

"Taper bushing" is a general descriptive term of the device and it's function.

"Taper bushing "is the term which Browning uses for their product specificly,the H,P,Q bushings.

Then there are "QD bushings" which is another type similar to the Browning,but made in a wider variety of sizes and styles.

Both the QD and Taper bushing styles have a tapered section with a bolt flange.QD's are made by practically everybody in the PT business and follow industry standard specs.Browning taper bushings are pretty much only made by Browning and won't interchange with QD bushings.

Now we have the "taper lock" bushing.It was originated by Fenner IIRC ,but also mfg by Dodge,TB Woods,Martin,Link Belt etc.

It has no flange and uses setscrews or bolts which thread into the bushing/bore interface.

Now that your thourghly confused:D Some tips on identifiing them.

Browning Taper bushings usually will have for the H series two threaded holes and two clearance holes in the flange,the P and Q series will have three clearance holes.but only two threaded holes.

The QD bushings will have three threaded and three clearance holes and will also have a stubby looking taper depending on series.One other feature will be a setscrew in the edge of the flange over the keyway.

The taper-locks ,well they are unique,no flange and two half blank holes along with one half threaded hole.

Of course everything revolves around numbers which are usually stamped somewhere on the product,usually on the face.

It's a good idea to learn the diffrences,it will save you some trouble and money later.

wierdscience
06-25-2008, 10:27 PM
I thought about doing that, or a timing belt, but then I'd have to replace the spindle pulley stack, and the Bridgy spindle brake is buried in there.

Did one already,the spindle stack is just a keyed bore,the replacement sheave with the addition of a spacer and machined up brake drum is all it takes.

oldtiffie
06-25-2008, 11:01 PM
Excellent thread and posts, but could not resist the link:
http://en.wikipedia.org/wiki/Bringing_in_the_Sheaves

For those (others) that are old enough to remember it, I can still see "Ma and Pa Kettle" in one of their movies singing it - great comedy movies.
http://en.wikipedia.org/wiki/Ma_and_Pa_Kettle

lazlo
06-26-2008, 08:54 AM
The Taper-lock sheaves to match the bushing,they are made by Dodge,Martin sprocket ,Fenner and a few others,but not by Browning.

QD's are made by practically everybody in the PT business and follow industry standard specs.Browning taper bushings are pretty much only made by Browning and won't interchange with QD bushings.

It's a good idea to learn the diffrences,it will save you some trouble and money later.

Thanks again Wierd. I've got plenty of room inside the Bridgeport head for the flange, and the MSC's 25% off is pretty compelling, so I'm going to stick with the Browning H bushing.


Did one already,the spindle stack is just a keyed bore,the replacement sheave with the addition of a spacer and machined up brake drum is all it takes.

Yeah, the Bridgey brake was a lame excuse ;) Plus you don't need a mechanical brake when you're driving the head with a servo drive or a VFD -- the DC injection braking will stop the spindle a lot faster than a mechanical brake anyway, and with an external braking resistor you can stop the mill nearly instantaneously. That should be fun with a heavy milling cutter in the spindle :)

So is a Poly-V a lot smoother than a conventional V-belt or a timing belt? I know there's a bit of torque cogging that goes on with a V-Belt as it contracts and expands around the pulley diameter, which is the reason for cogged V-Belts like the Gates PowerTwist belts ...

wierdscience
06-27-2008, 12:32 AM
Vee belts get lumpy for starters,they also are less efficent than poly-vee even if it's only a couple percent.

Where they pay off is vibration and noise.Poly-vees run smooth,much smoother than vee's.There is a SFPM limit on vee belts also,that is one reason the automotive industry began using them.

Plus using the Stevenson-O-matic machining method you can make your own sheaves quick.

lazlo
06-27-2008, 01:46 AM
Where they pay off is vibration and noise.Poly-vees run smooth,much smoother than vee's.

Plus using the Stevenson-O-matic machining method you can make your own sheaves quick.

OK, you shamed me into the Poly-V's Wierd. :) I actually have the Kennametal 40 TopNotch inserts, so turning Poly-V sheaves is no problem.

Is there a Poly-V calculator somewhere that I can calculate how wide a belt/sheave I need? My servo is 1.5 KW (just over 2 horsepower) in continuous duty, and servos run at 150% in intermittent duty, so I'm guessing I should make a sheave wide enough for 2.25KW = 3 horsepower.

John Stevenson
06-27-2008, 04:06 AM
Need to know max revs and smallest pulley size, plus what section J or L [ J would be best ]

Forget it just read 3" and 4,000

Service factor = 0.89

So number of ribs = power over service factor 3 / 0.89 = 3.37

So 4 grooves will handle this.

.

lazlo
06-27-2008, 09:55 AM
Forget it just read 3" and 4,000
Service factor = 0.89

So number of ribs = power over service factor 3 / 0.89 = 3.37
So 4 grooves will handle this.

Thanks John! A servo is designed/rated to run at ~ 3x rated torque intermittent, so wouldn't the Service Factor be ~ 1.5? A standard NEMA motor (not your typical Chicom treadmill motor) has a service factor of 1.0...

Also, don't you need to factor horsepower and especially torque into the belt width? My servo is 1.5 Kw, and 16 Ft/lbs (21.6 Nm) maximum torque (!), so I would think you'd need a wide Poly-V to keep it from slipping under worse-case conditions?

MickeyD
06-27-2008, 10:42 AM
I dug around in my junk box and found an old set of pulleys that came off of a 1.5hp table saw. It was setup for a 1 to 1 drive and the pulley was a six groove approximately 15mm wide (no idea where the original belt is). If you look at page 1028 in the online mcmaster catalog, they list a 10 groove j section belt and bushing style pulleys that should easily handle the load. Also, from looking at the way they were made, they should be pretty simple to make since you have the inserts.

lazlo
06-27-2008, 11:22 AM
If you look at page 1028 in the online mcmaster catalog, they list a 10 groove j section belt and bushing style pulleys that should easily handle the load. Also, from looking at the way they were made, they should be pretty simple to make since you have the inserts.

Thanks Mike! The 8- and 10-groove V-belts are only $15 each, they're made of cast iron, and they take the low-profile TaperLoc bushing that Wierd recommended. I'm going to order those from McMaster.

Wierd: if I'm reading the McMaster catalog correctly, the TaperLoc bushings handle a range of bore sizes? McMaster's part number for the bushing specifies 1/2 - 1" bore???

You guys are a bad influence -- now I'm going to have to take the head apart to get the spindle pulley stack off... :p

lazlo
06-27-2008, 11:55 AM
Ah, I found the answer to my first question about John's numbers: according to Machinery's Handbook, you need to calculate service factor times horsepower -- that makes a lot more sense.

I found the formula to calculate the horsepower ratings of Poly-V belts on 2419 of the 27th Machinery's Handbook. What you do is calculate the horsepower rating for a single V, and then pick the number of V's to accommodate the horsepower.

The problem is that I'm getting 1.6 HP per rib, which doesn't sound right. Hmmm....

pcarpenter
06-27-2008, 12:06 PM
I am pretty fond of the Fenner Power Twist belts for anything that sits under tension. You would be surprised how much vibration is induces by an egg shaped traditonal v-belt from forming to the shape of the pulleys around which they are wrapped. I am sure, however, that there is probably a little "give" in a link belt like that though.

By all means don't give up the brake even if your DC controller lets you do braking. I don't use it much for stopping the mill....I am not in a hurry and there's enough driveline friction in the variable speed heads I am used to that the spindle stops quickly. On the other hand, I don't know how you would ever tighten a drawbar properly without the brake (short of building a power drabar with an impact wrench)

Paul

wierdscience
06-27-2008, 01:52 PM
Wierd: if I'm reading the McMaster catalog correctly, the TaperLoc bushings handle a range of bore sizes? McMaster's part number for the bushing specifies 1/2 - 1" bore?

That's one of those things where you call them up ,give them the part number and they ask you what bore size you want in that range,saves catalog space.

John Stevenson
06-27-2008, 05:51 PM
Robert,
The service factor came out the table in the Polyvee book and relates to the belt, not the motor.

.

lazlo
06-27-2008, 06:14 PM
Robert,
The service factor came out the table in the Polyvee book and relates to the belt, not the motor.

Now I'm thoroughly confused :rolleyes: Machinery's Handbook shows a table of "Design Horsepower" which is Horsepower x Service Factor.

Does your Poly-V book indicate how many ribs would be needed for 3 horsepower? I'm guessing somewhere in the 8 to 10 rib range, since Mike's bandsaw had 6 ribs for 1.5 HP.

I'm going to make a little spreadsheet of the Machinery's Handbook Poly-V formulas. The J-Belt horsepower formula has 13 terms in it :(

lazlo
06-27-2008, 06:20 PM
By the way John, I'd rather err on the side of too many ribs in the V-belt, because I want virtually no slip.

The biggest reason I went with a servo drive is because I've finally collected all the parts to make the servo gear hobber I've been mentioning for awhile. Rather than have a rotary encoder on an outrigger like you did on the Victoria, I'm going to pull the mill spindle quadrature off the servo controller into the gear hobber divider circuit.

lazlo
06-28-2008, 11:40 AM
A bunch of searching turned up several belt calculators, but none that calculated Poly-V belts. To complicate matters, "Poly-V" is apparently someone's trademark, so the various vendors call them different names. Machinery's Handbook calls them "V-Ribbed" belts.

I called Gates yesterday, and they turned me on to their "DesignFlex" software, which calculates a custom drive configuration given the physical drive requirements (horsepower/torque, service factor, drive ratio, desired RPM's, optimal center distance, ...). The software is a free download, but you have to register.

The software is limited on drive motors, so I specified a 2 HP, NEMA 146 motor with 3450 RPM max. That's 450 RPM faster than my servo is rated, so I modified the desired top-end RPM by 3450/3000 to get the correct final RPM for the Bridgport head: 4,000 RPM. The servo is 2HP continuous, 3 HP intermittent, so I specified a service factor of 1.5. The center-to-center distance on the Bridgeport pulley stacks is 8.25" +/- ~ 10% (the amount you can wiggle the motor plate for belt tension).

In any event, after plugging the data for my servo into the Gate's calculator, they come up with a recommendation for a 240J6 belt/sheave, which is a 6 ribbed, J cross-section belt. Notice that they give you a complete parts list of belt, sheaves and bushings, and even tell you the "optimal" belt tension for a new belt and a used belt. Very nice!

Thanks Wierd, John and Mike!

Mike -- can I come by this afternoon and take a look at those Poly-V sheaves you have? There's beer in it for you :D

http://i164.photobucket.com/albums/u15/rtgeorge_album/PolyVGates.gif

ckelloug
06-28-2008, 01:47 PM
I've put on my asbestos suit in case of a serious flaming but I have to ask a retarded question;)

What on earth is a tapered sheave bushing (or any kind of sheave bushing) and what is it for?

--Cameron

lazlo
06-28-2008, 02:30 PM
I didn't know what they were either Cameron -- it's literally a tapered plug with a locking taper that cinches up the sheave onto the motor shaft. Apparently a tapered sheave has a superior fit and less vibration that a conventional straight bore sheave:

http://www.mcmaster.com/catalog/114/gfx/large/57095kc1l.gif

John Stevenson
06-28-2008, 07:33 PM
True story about poly vee belts.

I originally got involved with these working on some conversions for some Taiwanese wood lathes. They had double expanding pulleys to get the speed down for large bowl turning but the drives slipped like crazy so I looked at fitting a 4 stage poly vee drive before the last expanding pulley. This way it was possible to get from 80 up to 4,500 for small work.
I contacted the main people in the UK at the time TBA belting and got a design handbook.
Brilliant book gives centre distances, power belt load per groove, ratio's etc but the one thing I wanted was missing.
As this lathe had fixed centres, I knew the belt length and the centre distance and also knew one set of pulleys but I had to have the other set exact as I had only the tiniest amount of adjustment with an idler roller.

The book gave everything but so i rang them up and explained the problem but couldn't get it over to them what i wanted. I knew what I wanted and it meant transposing the formula for the belt length knowing centre distance and both pulleys but I'm crap at algebra .

Rang this professor up I know but he was on holiday so I wished I had payed more attention when I was at school.

School ????

Right laid it all out on a sheet of paper and headed it as follows.

" I was born in 1948, the year Aneurin Beven promised the people of the United Kingdom we would be looked after from the cradle to the grave [ a remark about our health service ] so seeing as I never managed to understand algebra when I was at your school from 1959 until 1964 could you please transpose the following formula to give me the values for d' and D'"

This was faxed off to my old school.
Following day came back the two formula's with the note.
"Pay more attention in class !! "

.

oldtiffie
06-28-2008, 09:13 PM
I've put on my asbestos suit in case of a serious flaming but I have to ask a retarded question;)

What on earth is a tapered sheave bushing (or any kind of sheave bushing) and what is it for?

--Cameron

Hi Cameron.

Much as lazlo says.

It is really just a variation on the principles of the "ER" series of collets where the "collet" is "pushed" into engagement where-as the "C" and MT (and the like) collets are "pulled" into engagement. The same rules apply to the taper bushing as regards accuracy of concentricity, taper and "round/parallel" as does the ER (or any) collets.

There is not or should not be any radial clearances (on diameters) or "slack/back-lash" as might be the case with key-ways either.

They are a very elegant and practical solution to positive location/alignment and drive.

There are a lot of variations of them in use in many fields.

lazlo
06-29-2008, 10:21 PM
I contacted the main people in the UK at the time TBA belting and got a design handbook.
Brilliant book gives centre distances, power belt load per groove, ratio's etc but the one thing I wanted was missing.

One of the members here (I don't know if he wants me to use his name :) ) pointed me to the EPT/Browning Poly-V design handbook, and a hour of reading and some simple calculations answered several questions.


I found the formula to calculate the horsepower ratings of Poly-V belts on 2419 of the 27th Machinery's Handbook. What you do is calculate the horsepower rating for a single V, and then pick the number of V's to accommodate the horsepower.

The problem is that I'm getting 1.6 HP per rib, which doesn't sound right.

So it turns out that my calculation was correct. Poly-V belts are fascinating little beasties! :)

It turns out that the horsepower rating on a Poly-V belt is proportional to the belt friction. You can create that friction either with a wide belt (more V's), or with a bigger pulley, which has more belt in contact with the sheave.

When I calculated the 1.6 HP per rib, I was using the 7.25" of the stock Bridgeport conventional V-Belt sheave. That's a giant sheave in the Poly-V world.

So when I was running the Gates DesignPro calculator, I kept trying to force the sheave diameter to what you'd expect in a conventional V-Belt (5 - 7"), and the software kept telling me to use a 3 or 4 ribbed V-Belt. 3 V-Ribs on a 6" sheave at ~ 1 HP per rib = 3 HP, which is the servo rating at the 150% (intermittent) rating.

So 4 ribs with a 6" pulley has the same horsepower rating as 6 ribs with a 3" pulley. The friction generated isn't linear, so you either have to use Gates/Browning's tables, or calculate the horsepower rating according to the Machinery's Handbook formula.

Now here's where things get interesting: this is a servo retrofit, and as Macona pointed out on his 10EE retrofit, servos have full torque all the way down to 0 RPM. Poly-V belts are really intended for high speed operation, since a lot of the belt friction (and therefore, horsepower rating) is from centrifugal tension. In other words, the motor is pulling the belt around the sheaves at 3,00 feet/min, which generates an immense amount of friction.

At low speeds, say 250 RPM for a big flycutter or facemill, the centrifugal friction drops way off, so there are exception tables with minimum sheave sizes for low RPM. For a 3 HP motor at 575 RPM, Gates indicates a minimum sheave size of 4.8" on the driver sheave.

So for a machine tool (lathe or mill), you really want the largest Poly-V you can fit into the spindle. Plug that sheave size into the Machinery's Handbook formula, or look it up in the Gates/Browning tables, or run DesignPro, and it will recommend a pretty narrow Poly-V belt (3 or 4 sheaves).

The other upside to the big sheave, narrow Poly-V belt configuration is that the bigger the sheave, the less belt tension required, so it's a lot less stress on the motor and pulley stack bearings. And the biggest upside: the noise of the Poly-V system (which is really quiet to begin with) is inversely proportional to the sheave size, so the bigger sheaves are quieter.

So I ended up with a configuration of a 5.3" driven sheave, a 4.0" driven sheave, and a 4-ribbed Poly-V belt. Gates DesignPro tells me that system is rated for 4.3 HP, which is healthy safety factor over the servo's max output.

One last comment: the Poly-V sheaves are surprisingly cheap. The powdered metal Gates sheaves are ~ $30 for a 5" sheave at McMaster, and E.B. Atmus in Springfield, MA has them for at least 15% cheaper than McMaster:

http://www.ebatmus.com/index.html

If this is painfully boring, someone say so and I'll just post some pictures when I'm finished with the servo retrofit :D

MickeyD
06-29-2008, 10:34 PM
The scary thing is how interesting this is. I had never worked through the calculations before so I had no idea how efficient these belts were, all I had was seat of the pants experience with them.. Someday I need to change the belts out on my Colchester lathe, so if I can find a belt that is close to the right size, I will switch it from tripple v belts to a single poly belt. That could be an interesting upgrade...

TGTool
06-29-2008, 11:36 PM
If this is painfully boring, someone say so and I'll just post some pictures when I'm finished with the servo retrofit :D

On the contrary, all this is very interesting. Does that mean you won't post pictures? :(

lazlo
06-30-2008, 01:51 PM
I thought about doing that, or a timing belt, but then I'd have to replace the spindle pulley stack, and the Bridgy spindle brake is buried in there.Did one already,the spindle stack is just a keyed bore,the replacement sheave with the addition of a spacer and machined up brake drum is all it takes.

Wierd, John,

What's the diameter of the adapter on the top of the Bridgeport spindle stack?

From Dave Kamp over at PM:

http://www.practicalmachinist.com/vb/attachment.php?attachmentid=2570&d=1205791818

In other words, what TaperLoc bore size do I order for the Driven (spindle) pulley?