Shopsmith Forums

dusty wrote: ↑Sat Jul 30, 2022 7:46 pm ...
The only problem I have with anything stated herein is belt length. I have three (two new) 504193 belts and their ODs are all a tad over 27". I also have the very first belt that came with my 510 when purchased new. It is a Perelli carrying the 504193 PN. It is only .48" wide but it too is 27".

You are getting a consistent measurement over multiple samples, as am I. The most likely explanation is that we are measuring differently. In measuring circumference of a belt, how we measure introduces an error. I'm using a thin flexible measuring tape normally associated with sewing. If I use a metal tape that I would use for just about anything else, I get a tad over 27". Both are actually a little bit higher than actual circumference, but a thin flexible measuring device comes closest.

BUT. I still do not achieve anywhere near 700rpm.

Nor did Dennis. What do you think is wrong? At the low speed stop, your belt positions in the sheaves should be similar, or not, to the pics I gave of mine. Measuring ratios, calculating ratios, and measuring RPM should all give the same approximate answer or we should be able to explain why. So far we seem to have resolved why we were getting different answers between the first two.

Charts will not convince me, even my own. I can make the numbers work by manipulating ratios but that doesn't guarantee a functional machine that runs the speed chart. Tell me you have one that does this and you became convinced by seeing the numbers on a tach and I'll become a believer.

Two things I know of are going to give us variation:
1) Actual motor speed Vs. rated. Test conditions are about as close to no-load as we can get on the Mark V headstock. Motor speed should be much closer to 3600 RPM than to 3450 RPM. Tachometer measurements are likely to be about 4% high compared to calculated.

2) Distance between shafts varies based on Idler eccentric adjustment and is not consistently 7.250". The effect will be the same as small variation in belt length. This will cause variation, one machine to the next, in how close we get to the theoretical minimum achievable speed.

Taken together, I'm not seeing how those get us anywhere near the 1055 RPM number that Dennis reported. As one example.

- David

I do understand the variability between machines but (for me) that is not a factor. I am comparing my results with SS documented numbers (the chart).

Belt pitch length = belt outside length - (belt thickness X 3.14). The actual belt thickness (not belt width) comes into play.

So for example if anyone has a belt with a 27" outside length and a 5/16" thickness, the pitch length of that belt is 26".

Traditionally the "centroid" of the belt section has been used to define pitch length. What I have been using is not the "centroid" point but rather the "geometric" point. The geometric point is easier to locate and differs only 0.016" from centroid point. Attached shows these two points.

Using centroid point instead of geometric point does slightly increase min RPM and slightly reduce max RPM from my table values. So how we measure the belt does matter.

The centroid is where the "neutral plane" of the belt section is located. Neutral plane is where, when the belt bends, it sees no stress, either tensile or compressive. Now when a belt is in operation and under tension, there really is no zero-stress plane but for speed calc purposes we are concerned about the geometry of the section not the stress distribution across the section. The centroid location is based on section geometry.

If I had 2 belts, identical in geometry but different in reinforcing-wire distribution - would they differ in pitch length? Or pitch diameter when on a sheave?

I have looked at the Gates material discussing this issue, and per the above I am not convinced that relocating the belt length-measuring point to near the belt top surface (where the steel wires are concentrated nowadays) would be correct for speed ratio calculation purposes.

Gates reading material attached. Gates is basing everything on fixed- diameter sheaves, and they do not apply to Shopsmith sheaves.

SteveMaryland wrote: ↑Sun Jul 31, 2022 10:59 am Belt pitch length = belt outside length - (belt thickness X 3.14). The actual belt thickness (not belt width) comes into play.

So for example if anyone has a belt with a 27" outside length and a 5/16" thickness, the pitch length of that belt is 26".

According to some sources, the pitch or datum line is much closer to the outer diameter of the belt, depending primarily on the location of reinforcement. This has changed over time, so for any given belt it evidently depends on actual belt construction. Which, for the modern SS belt, we don't really know.

Here is a pretty good article on pitch/datum of pulleys and belts, includes "...So the diameter through which the pitch line passes is nearly equal to the outside diameter for most belts...": http://gatesbeltsandapplications.blogsp ... meter.html

So I'm not trying to suggest right or wrong, I don't think we know with certainty. I'm suggesting we are estimating. I can see what I presume is the primary reinforcement band in the side of a lightly worn SS belt, and estimate it is ~0.08" from the OD. (I used a slightly smaller estimate earlier in this thread, but the difference in pulley ratio is trivial.)

- David

DLB wrote: ↑Sun Jul 31, 2022 11:55 am

SteveMaryland wrote: ↑Sun Jul 31, 2022 10:59 am Belt pitch length = belt outside length - (belt thickness X 3.14). The actual belt thickness (not belt width) comes into play.

So for example if anyone has a belt with a 27" outside length and a 5/16" thickness, the pitch length of that belt is 26".

According to some sources, the pitch or datum line is much closer to the outer diameter of the belt, depending primarily on the location of reinforcement. This has changed over time, so for any given belt it evidently depends on actual belt construction. Which, for the modern SS belt, we don't really know.

Here is a pretty good article on pitch/datum of pulleys and belts, includes "...So the diameter through which the pitch line passes is nearly equal to the outside diameter for most belts...": http://gatesbeltsandapplications.blogsp ... meter.html

So I'm not trying to suggest right or wrong, I don't think we know with certainty. I'm suggesting we are estimating. I can see what I presume is the primary reinforcement band in the side of a lightly worn SS belt, and estimate it is ~0.08" from the OD. (I used a slightly smaller estimate earlier in this thread, but the difference in pulley ratio is trivial.)

- David

I too have done this and my amateur observations are the same as yours. The reinforcement seems to be very close to the top of the belt (negligible for my non-scientific analysis). I'm just attempting to determine if the reason for my not being able to attain 700 rpm is the fault of either me or my machine). Not how I got into this in the beginning.

I consider the high speed stop's only function is to preclude the control sheave from interfering with the speed control parts by opening the pulley too far.

There might well be another purpose, but every thing else in the drive train is mushy.(wear, belt stretch. . . .)

JPG wrote: ↑Sun Jul 31, 2022 12:43 pm I consider the high speed stop's only function is to preclude the control sheave from interfering with the speed control parts by opening the pulley too far.

There might well be another purpose, but every thing else in the drive train is mushy.(wear, belt stretch. . . .)

I agree but that does have an effect on speed determination (calculations or otherwise).

I caved on the tachometer. Placed tape on the flat washer that helps retain the motor spring, arbor portion of a sanding disc for the upper, similar location on a 3-3/4" faceplate on the Idler. Raw numbers:
Motor 3580 RPM (about what I'd expect for minimal load).
Idler 1246 RPM
Main 790 RPM.

This is the machine I've provided measurements for earlier in the thread. I'll be repeating some of those, as RPM are higher than I expected even adjusted for actual motor RPM. 790 RPM at 3580 Motor RPM would be 761 at 3450.

I checked the sanding disc and tach on my PowerPro. Everything worked as advertised.

- David

Loaded motor rpm will drop taking the other two down as well.

The loaded at 700 rpm motor rpm is extremely low(<3200 rpm) Not a likely scene.

DLB wrote: ↑Mon Aug 01, 2022 4:18 pm ...
This is the machine I've provided measurements for earlier in the thread. I'll be repeating some of those, as RPM are higher than I expected even adjusted for actual motor RPM. 790 RPM at 3580 Motor RPM would be 761 at 3450.
...

The most suspect measurement was radius of the belt on the Motor sheaves. I made a gauge out of a T-Nut, to rest across the sheave rim, and a length of threaded rod. This allowed me to measure more accurately and at bottom dead center. I had also changed my estimated correction for pitch/datum. Between the two, my estimated pitch/datum radius changed from 0.8" to 0.916". Overall resulting in a calculated Idler speed of 1232 RPM based on my best measured/estimated numbers (now using 3580 RPM) Vs. 1246 RPM measured. A difference of 14 RPM, or barely over 1%. Good enough.

I conclude that this machine, as configured, will not get below 790 RPM unloaded. Assuming the speed chart is based on 3450 RPM (fully loaded), the machine won't go below 761 RPM. Not bad compared to speeds reported by some other members, but not great. The belt already runs slightly proud on the Idler sheaves and with the spring at or near full compression.

- David