Author Topic: Serpentine pulley rib dimensions: CAD specs needed  (Read 31056 times)

jtodd

  • Full Member
  • ***
  • Posts: 190
    • View Profile
Serpentine pulley rib dimensions: CAD specs needed
« on: September 03, 2008, 04:07:27 AM »
I'd purchased one of the Allmand pulleys a while back when I was working on my (then) 12/2, but now I have a 16/2 with different specs.  I talked to George a few months ago, and I'm sure his pulleys are the best I can get.  However, the pricing made me do one of those cartoon routines where my eyes bounce out of my head on springs.  After that price realization, I figure this gives me a good excuse to become more familiar with the CNC components of my lathe.  :-)

I'm drawing up my own pulley on the ole'  CAD platform here, and I'll try to get my hands on a ductile iron round from one of the local foundries or maybe I'll use a steel round after some experiments with wood or plastic.  However, before I can embark on this adventure of Learning The Hard Way, I've got some questions that perhaps can be answered by the wisdom of the crowds:

1) What are the exact dimensions of a serpentine belt groove configuration?  I mean the depth of the grooves, top-of-groove width, angles of groove, etc.  (see below)  I could guess, but I'd prefer real data instead of squinting-and-measuring.

2) My measurements of the Allmand pulley seem to show that the "outside" diameter of the pulley is measured at the top of the "valleys" on the grooved area.  Anyone disagree?

3) I've got SK bushings as 3/4" per foot of taper.  Right?

SK dimensions:
http://www.maskapulleys.com/propro.htm  (click on "QD bushings" in the menu)

My findings thus far:

The SK bushing and primary hole part were easy, even though I'm learning a bit about this CAD program as I go.  Measure the two holes on the existing pulley (diameter, distance apart) then create two circles with those measurements and create a skinned solid which automatically handles the tapered cone portions (excellent!  no math required.)  Delete this cone shape from the pulley cylinder.  Measure the head offset cut; create cylinder; delete it from the pulley cylinder.  Add chamfers.

Now, the ridges are a little more complex.  It appears (though this is just from careful measurement with inadequate tools) that the grooves are 0.10 deep, the tops of the grooves are 0.035 wide, 0.105 "valley" width at the top (distance between the ridges on the pulley.)

Measuring a belt, it seems that the "top" of the belt ridges are 0.070 wide, which would mean that when they are riding in the pulley they should be encountering a constriction between the two "walls" of the pulley groove when the valley between the walls gets to something around that width.  The grooves on the belt only seem to be about 0.065 deep, so that means that the grooves on the pulley are deeper than the matching ridges on the belt, which is expected - you'd want to have the constriction be the reason that the ridges don't quite mesh exactly perfectly with the pulley, otherwise you'd be losing that friction point.  So that means that there needs to be a taper on the pulley grooves such that at around 0.060 (depth) the width between the two walls is 0.070.

All this fussing about the groove angles and stuff might actually be a moot point if there is enough surface friction just with straight-cut grooves on the pulley.  But I'll try to be a perfectionist in design, if not construction. :-)

JT

jtodd

  • Full Member
  • ***
  • Posts: 190
    • View Profile
Re: Serpentine pulley rib dimensions: CAD specs needed
« Reply #1 on: September 03, 2008, 05:20:33 AM »
This is a page with some info on GoodYear products :

http://www.goodyearep.com/uploadedFiles/Products/Power_Transmission/Banded_Belt/poly-v.pdf

OK, thanks for confirming a few things and the details  you sent.

Interestingly, the belts that I was using (NAPA 8-rib, model 25-080968) on the 12/2 don't match any of the specs in that Goodyear document you forwarded.  The "height" of this belt is 0.170 (+/- 0.002) and the valley-to-valley distance is 0.150 (+/- .005).  This seems to match pretty well the pulley's dimensions.  It appears that Goodyear site is for non-automotive belting, so I'll clarify my request and say that this is for 8-rib automotive (General Motors - GM) belting standards.  I'm betting that both Ford and GM have almost identical (or completely identical) standards, as I've seen belts interchange between late-model engines (length and rib count seem to be the only issue that matters.)

I figured as much for the grooves needing to be tapered - that seems to make the most sense when I thought about what was happening under load - torsional force translates into friction on a wider and increasingly higher resistance area.

JT

adhall

  • Full Member
  • ***
  • Posts: 236
    • View Profile
Re: Serpentine pulley rib dimensions: CAD specs needed
« Reply #2 on: September 03, 2008, 05:43:34 AM »
I found the infomation you seek on the Gates web site (www.gates.com). You will have to do some poking around to find the "K" series Micro-V belts, but I know I found it eventually. I've got the files somewhere, but they elude me temporarily...

Best regards,
Andy Hall
JKSon 6/1, 5 kW ST Head, 1992 Dodge RAM Cummins 5.9L Turbodiesel, 2001 VW TDI 1.9L Turbodiesel, 2006 Jeep CRD Turbodiesel, Yanmar FX22D Diesel Tractor

mobile_bob

  • Administrator
  • Hero Member
  • *****
  • Posts: 2940
    • View Profile
Re: Serpentine pulley rib dimensions: CAD specs needed
« Reply #3 on: September 03, 2008, 07:04:54 AM »
and i am almost certain that the included angle will change as the pitch diameter increases or decreases
just as V section belts do
and good luck finding the info on that one!

you might well be served to find a pulley/sheeve with the approx pitch diameter you want to make and measure the included angle.

i cant remember which way it goes,, but i think the angle gets larger as the diameter gets smaller,, but i may be wrong on the direction/relationship
but i am 99.99% certain that the included angle is different as the diameter changes.

bob g
otherpower.com, microcogen.info, practicalmachinist.com
(useful forums), utterpower.com for all sorts of diy info

LowGear

  • Casey
  • Hero Member
  • *****
  • Posts: 2130
  • What? My diesel had fries for lunch?
    • View Profile
Re: Serpentine pulley rib dimensions: CAD specs needed
« Reply #4 on: September 03, 2008, 07:59:37 AM »
Have you asked George?

Casey
NPR Tipper/Dump Truck
Kubota BX 2230
Witte BD Generator
SunnyBoy 6000 + SolarWorld 245

jtodd

  • Full Member
  • ***
  • Posts: 190
    • View Profile
Re: Serpentine pulley rib dimensions: CAD specs needed
« Reply #5 on: September 03, 2008, 10:58:01 AM »
The automotive standard is either J or K series. I forget which one. The K section shows a 0.14" distance between valleys and you measured 0.15". Your previous comment talked about 'squinting' while measuring so I am assuming it is quite possible you have a 'K' section setup there. You measured 0.17" for the belt 'height' and the 'K' section is spec'd at 0.165.

So, to sum it up, I am virtually certain that you have a 'K' section belt there. The spec's I referred you to do not include groove angle but as far as I can tell it is 40 degrees inclusive from one face to the opposing face or 20 degrees to the center line.

Jens

Great, thanks!  Given that I'm measuring rubber with somewhat of a grippy surface on it, the measurements are within tolerance for the K series, and I've based my other assumptions on the .140 inter-ridge widths shown in that PDF.

More data here, mostly so that I have a single place to put it where others might find it and hopefully learn from my successes/mistakes:

On the pulley:
 Â - valley floor widths: .046 (assuming a 0.105 top width and a 0.070 width 0.060 from the top, and a 0.10 total ridge height, some quick ratios give me 0.046 width at the bottom)
 Â - ridge distances: .105 between ridges, and .035 ridge width at the top (0.140 total distance between two "valley" midpoints)
 Â - ridge dimensions: .035 on the flat "top" of the ridge, and .094 at the "base" of the ridge (subtraction got me here)

I think that's all the data I need to create the ridge profiles.  I don't need angles, since just calculating the top and bottoms of the ridges can automatically create the angular profile via the CAD program.  I'll check to see if it's the degrees you mention. (answer: no, it's not - it's ~33 degrees with the measurements above... so something is still probably amiss.)  So, the CAD drawing is done, and I'm heading to bed.  I found several "quirks" in the CAD software (aka: crashes) but surprisingly it did what I wanted it to.  I've now got a 10.333" pulley for my downrated 16/2 running at 800 RPM.  Now, to find some ductile rounds >10.5"...



JT


mobile_bob

  • Administrator
  • Hero Member
  • *****
  • Posts: 2940
    • View Profile
Re: Serpentine pulley rib dimensions: CAD specs needed
« Reply #6 on: September 03, 2008, 02:47:44 PM »
i think you may still need the specific angles

because when you fold the belt the bottom lands get wider by a bit.

depending on how small a pulley you need  such as a small one for a 3600rpm gen
you might be off a degree or two

my hats off to you for giving this a crack, but in the end if you figure your time worth much you might not save much over
buying one already done. the result of buying one is you keep this almand guy in business making them and in doing so keep them
available for those that cannot make them themselves.

unless of course you plan on mass producing them? then you will find that with the cost of material, shipping, tooling etc
there isn't much fat for profit,, unless you work for very low wages.

personally i wouldn't go to the trouble to make what i can purchase,,( my time is better spent doing stuff that pays me i guess.)
unless i had a real need for an odd size that no one else offered.

bob g
otherpower.com, microcogen.info, practicalmachinist.com
(useful forums), utterpower.com for all sorts of diy info

MacGyver

  • Sr. Member
  • ****
  • Posts: 452
    • View Profile
Re: Serpentine pulley rib dimensions: CAD specs needed
« Reply #7 on: September 03, 2008, 06:01:05 PM »

Just as an additional comment, the 'K' profile that you seem to have concerns me. It is shown as a special purpose profile in a number of places I looked. I am pretty sure that automotive applications use J profile.

Jens

No, "K" series is correct for the automotive belts.
I'm using a Gates K060910 with my Allmand pulley.
My Subaru uses a K series belt also.
Steve

JKson (PS) 6/1 'roid & ST 7.5

jtodd

  • Full Member
  • ***
  • Posts: 190
    • View Profile
Re: Serpentine pulley rib dimensions: CAD specs needed
« Reply #8 on: September 03, 2008, 06:48:42 PM »
i think you may still need the specific angles

because when you fold the belt the bottom lands get wider by a bit.

depending on how small a pulley you need  such as a small one for a 3600rpm gen
you might be off a degree or two

my hats off to you for giving this a crack, but in the end if you figure your time worth much you might not save much over
buying one already done. the result of buying one is you keep this almand guy in business making them and in doing so keep them
available for those that cannot make them themselves.

unless of course you plan on mass producing them? then you will find that with the cost of material, shipping, tooling etc
there isn't much fat for profit,, unless you work for very low wages.

personally i wouldn't go to the trouble to make what i can purchase,,( my time is better spent doing stuff that pays me i guess.)
unless i had a real need for an odd size that no one else offered.

bob g

I don't know if it's actually Allmand that is making them still, or if George is getting them done elsewhere.  Anyway, I'm making this because:

  1) It's a custom size
  2) I'm learning more how to use the CAD/CAM programs
  3) I'm learning more how to use my CNC lathe
  4) It probably will be cheaper (possibly not by much unless the iron is really inexpensive)

I'm not looking to mass-produce anything; this is for my own use.  I think George and the fellow on the East Coast (whose name escapes me now) are the best sources for the pulleys ready-made and well-tested, but sometimes doing it yourself is the right way even if it isn't the best way.  If I only did the things that paid me, I wouldn't be puttering with Listers at all in my spare time, would I?  :-)

Jens - I'll try to revise my figures to match the 40 degree angle and see what it gets me.  However, I have the day job to tend to at the moment so I'll do that later tonight.  I'm pretty sure this is a K belt; I'm not off my measurements by THAT much.

JT

mike90045

  • Mendocino Metro
  • Hero Member
  • *****
  • Posts: 1594
  • Mmmm BBQ
    • View Profile
    • Mikes Solar PV page
Re: Serpentine pulley rib dimensions: CAD specs needed
« Reply #9 on: September 06, 2008, 06:42:48 PM »
> The tops and bottoms of belts/pulleys should never meet.

So, you are saying that the rib sides is where the friction is ?   I thought the serpentine belts relied on the flat of the belt, and the ribs were to "guide" the belt, and this was where the advantage over a V grove belt being wedged into the pulley, and yanked back out.
 My truck (dodge) uses loads driven from both sides of the belt, as it winds it's way along the path.  The heavy loads, and the crank, drive the ribbed side. I see shiny wear marks all over the pulleys, so I know the belt is hitting all areas.

I'm just asking to understand how the flat belts work.

As to absolute diameters, if you are "rolling your own" then I'd play with the engine RPM's a bit to find a sweet spot (625, or 683) where the engine is happy, and then size the pulley for that RPM.

Just my thoughts and questions

M61hops

  • Sr. Member
  • ****
  • Posts: 299
    • View Profile
Re: Serpentine pulley rib dimensions: CAD specs needed
« Reply #10 on: September 06, 2008, 11:43:37 PM »
I'm sure that the sides of the ribs are supposed to transfer torque to drive the most important pulleys.  I had a situation where the rib width had worn down to the point that the micro vee belt would not drive the alternator of a mid eighties camero enough to keep the battery charged when the headlights were on.  Even though I tried to tighten the holy living shit out of the belt, the ribs of the pulley were trying to cut into the backing and there just was not enough grip to drive the alternator.  The belt didn't look that old but it just couldn't get a grip on the pulley.  This problem made for a long slow trip home as nobody had a belt in the wee hours of the morning.  When I finally got a new belt a few days later I could see with my eyeballs that the ribs were fatter on the new belt and that fixed the problem.                       Leland
I pray everyday giving thanks that I have one of the "fun" mental disorders!

jtodd

  • Full Member
  • ***
  • Posts: 190
    • View Profile
Re: Serpentine pulley rib dimensions: CAD specs needed
« Reply #11 on: September 07, 2008, 12:18:19 AM »
OK, so I've finally beaten down several CAD program bugs and I think I've got the design ready to go to the lathe.  Now all I've got to do is figure out how to chuck up this 100 pound round of ductile iron and create a CAM profile for it (ugh!)  The sheaves are now at a 40-degree angle, which still seems to fit the profile of the V shapes without binding or "squeezing" the belt, at least according to my imprecise (+/- .005) measurements on the belt fingers.   I found a 11" diameter ductile iron (65-45-12) round 3.5 wide from Pacific Machinery and Tool Steel here in Portland (http://www.pmtsco.com/ 3445 NW Luzon St, Portland OR USA - +1-503-226-7656) for $138, ready for pickup about 2 hours after I called them. 

Here are the images/drawings:  http://www.loligo.com/projects/lister/pulley/

Note that this is for a 16/2 at 800 RPM, not 650 or 850 so it probably wouldn't be a good model for those of you with 6/1s or other 650 RPM engines.  The DXF file doesn't have dimensions or comments; a bug in the CAD software wouldn't allow me to save it as DXF with that layer enabled.  Go figure.

JT

mobile_bob

  • Administrator
  • Hero Member
  • *****
  • Posts: 2940
    • View Profile
Re: Serpentine pulley rib dimensions: CAD specs needed
« Reply #12 on: September 07, 2008, 12:41:47 AM »
Jtodd:

if you are going to the time and expense please do double check your belt angle
don't trust the machinist handbook!

last thing you want after all is done to find out your a degree or two off

a couple probably won't show up as a problem until there is some wear, and a heavy load on a hot day.

i am simply passing this along because of years of practical experience with serp belt drives
and the fact that that angle "does" transmit power, no matter what the engineering text will tell you.

yes you can tighten the living shit out of the belt to compensate, and it won't hurt the belt, but
why put excess loading on the engine and genhead brgs if you don't have to?

we change out lots of k section drive pulleys that look good on trucks, because they will no longer pull the load.
without undue stress on fan drives, alternators and waterpump brgs.

bob g

otherpower.com, microcogen.info, practicalmachinist.com
(useful forums), utterpower.com for all sorts of diy info

mobile_bob

  • Administrator
  • Hero Member
  • *****
  • Posts: 2940
    • View Profile
Re: Serpentine pulley rib dimensions: CAD specs needed
« Reply #13 on: September 07, 2008, 01:47:05 AM »
you might try the browning site as well
some time ago i called each of the major manufactures, engineering dept on belt included angles
and found that not every engineer knows!
Gates engineer had to get back to me, which he did a few days later.
seems as though they have been making belts the same way so long  that no one knows what the angles are :)

that is when i found out that the belts have a specific angle, but that angle changes with the pitch diameter
iirc it gets larger with a smaller pulley and narrower with a larger pulley

just thought i would kick this forward, because if i were to go to the effort making a pulley i would want it to
perform as well as it could be made to perform.

bob g

otherpower.com, microcogen.info, practicalmachinist.com
(useful forums), utterpower.com for all sorts of diy info

Mack

  • Newbie
  • *
  • Posts: 14
    • View Profile