Belt Tension

[Bob La Londe]

On some things you have a specification. On a lot of modern automobiles you simply relieve pressure with the belt tensioner slap the new belt on and let the tensioner set the belt tension when you release it.

When you are working on something that you don't have the specification for, are designing or building something new, or the specification simply isn't available to you are there any rules of thumb you use to guide you to that sweet spot?

Tight enough not to slip for a long while, but not so tight as to overload the components.

Recently when I replaced the motor on my air compressor I struggled with getting the belt tight enough. Finally I came up with a solution of snugging down one bolt and using a hammer and drift to shift the base of the motor over to straighten it and tension the belt. As you might have guessed this resulted in a way too tight belt. My mistake. It was a little tighter than I thought was right but it seemed to run okay so I ran it.

The circuit breaker tripped a couple times so I checked my wire size and my distance thinking maybe because I'd gone to a shorter body motor the starting torque wasn't as high and the starting current was too high. My wire size would allow for the next larger breaker so I installed it. That was the wrong solution. No I didn't burn anything up and I didn't violate any electrical codes other than not paying the local bribe to hire somebody else to do it.

The very next day my compressor had shut off again but the breaker hadn't tripped. The thermal overload on the motor had tripped. Some little nagging thing in the back of my mind told me that the belt tension was too high. I was still concerned with getting the belt tight enough, and I remembered how I struggled with that when I first installed the motor. Instead of loosening the bolts trying to position it I only slightly loosen the bolts. Then I marked the motor position on the top plate of the compressor with a marker and gently move the motor over with a 3 lb hammer and a drift.

I tighten down the mounting bolts for the motor and powered it up. I haven't had an issue since.

Oh, I'm not looking forward to the roasting I'm going to get for this question, but I'm sure some of you have some great insight on this topic.

[J Tiers]

No roasting here. I just go by the deflection of the belt when pressed in the middle of the "span". For most that are a foot or so span, that's about a half inch or a bit less for v-belts.

I usually do not measure, but go by the pressure to deflect.

They will probably roast both of us.

[BCRider]

No roasting here. I use the same "feel good" method. But I read somewhere a long time ago that V belts did not need to be and in fact should not be tight like guitar strings. And I'd used that idea all along.

Consider that the V belts are somewhat self tightening once given their initial tension. The V shape under load wants to self engage deeper into the groove. So all we need to do is have enough slack out of the setup that they self engage.

Since it's easy I just checked my lathe, mill and drill press. Using easy one finger pressure I can easily deflect all my belts 1/2" per foot on the tightest setup (mill) and a little more on the other two. So I'd say I'm not even running them as tightly as you describe. Yet I never hear or see any signs of slipping.

Pushing on a scale with roughly the same force I seem to be pressing with about 2 to 3 lbs to do this.

My air compressor uses a poly V belt. That one is tighter and more like the tension used for polyV belts on cars. So not the same thing.

[Bob Engelhardt]

If you want to be Doozery about it, here is the procedure from my IR Type 30 manual:

When I got the compressor I did indeed want to be Doozery about the tension and I made a gauge to do it. I wrote it up for the RCM newsgroup, but it's a PDF, so not uploadable here.

[metalmagpie]

Well, there are belts and there are belts. I am going to be using Habasit fabric belts on my Champion Blower and Forge, and those have to be joined by scarf/glue joints and thus will be endless belts. Running endless belts on a machine with no belt tension adjustment is a challenge. For me, I think it's going to be "as tight as I can make it without it becoming overly difficult to change the belt from one step to another". On the other hand, such belts do not need to be detensioned after use like leather belts.

The very latest camelback drill presses used vee belts which did have belt tensioning capability. I'm guessing I'll set those the old way, by checking belt deflection mid-span.

metalmagpie

[A.K. Boomer]

So much depends on belt design too, serpentine or standard, If not auto-tensioned I just put my two cent's on it - I always error to the side of less due to it being just fine and most thing's handling it and the bennies of longer component life,

however - you can actually go too loose even if slipping does not occur, you don't want the belt going through traumatic fluttering stages destroying itself...

When I took the A/C system out of the Peashooter the AC shared the same pulley runs as the water pump, it was a 5 rib serp. after the compressor was removed all that had to be driven was the water pump. and the belt option they give is of course way shorter, yet still 5 rib, to further save fuel I sliced the 5 rib lengthwise into a 3, and ran it very loose, however - upon giving the engine a few revs noticed the belt going into drastic flutter mode at certain RPM's so snugged it back up some - still loose cuz alls it's doing is driving a water pump but tight enough to stabilize it...

[Glug]

Those acoustic belt tension measurement gizmos look neat. The twang of the belt determines the tension, like a guitar tuner, but way more expensive.

[BCRider]

....I always error to the side of less due to it being just fine and most thing's handling it and the bennies of longer component life,

however - you can actually go too loose even if slipping does not occur, you don't want the belt going through traumatic fluttering stages destroying itself.....

That sounds like it's just right. It's pretty obvious when they are too loose by the flutter. So tight enough to not flutter and that's dandy.

[55fairlane]

A belt tension gauge is well worth the money, but most service guys won't know what to tell ya......some wire possessing machines you use a tuning fork to know if the belt is exseive loose or tight...PIA

[The Artful Bodger]

We usually run the threshing mill with a straight belt but in windy conditions we twist the belt.

[Ian B]

Bodger,

Doesn't twisting the belt make the thresher run backwards and mix the wheat back in with the chaff?

Ian

[Robg]

Deflection of the belt in the middle of the pulleys of about the width of the belt.

[A.K. Boomer]

Deflection of the belt in the middle of the pulleys of about the width of the belt.

Deflection middle of the pulleys but how much pressure? and width of the belt of what ? a multi-rib serpentine?, some are 3 or 4 or even 6 or more ribs, and in the middle of the pulleys is a pretty broad statement depending on machine/engine design and span...

You can find Info in most cases per exact piece of equipment because they are taking all of it into consideration, another factor? overall belt length due to belts having elasticity so drive components putting demands on the drive side of an extra long belt can really have a loosening effect on the other side...

[Bob Engelhardt]

Deflection of the belt in the middle of the pulleys of about the width of the belt.

According to IR, the deflection is proportional to the span (1/64" per inch), with the specified force. For V belts.

[The Artful Bodger]

Bodger,

Doesn't twisting the belt make the thresher run backwards and mix the wheat back in with the chaff?

Ian

There is a lever on the traction engine where we can select 'grain' or 'chaff' to meet market needs.