Dear Lyle,

All I am referring to is for fixed fill fluid couplings only. So far we used to consider that starting energy will be same for all loaded and unloaded starting conditions and that torque will be maximum starting torque setting of the fluid coupling. But this software vendor suggest that starting torque will vary depending upon load condition (or effective tension at that condition). This way in many cases where we used to get empty starting slip as dictating factor for take-up selection may not be same now and will eventually lead to lower final take-up tension and lower belt rating. I need your opinion on that. Regards ■

Dear Nordell,

Many thanks for your input. It appears that I could not make my point clear till now. I am referring exclusively about fixed fill fluid couplings and not variable filled scoop type couplings. As application engineers, we choose a type and a model of fluid coupling depending upon our requirement. We know different type of fixed fill fluid couplings can provide a minimum starting torque and a maximum starting torque based upon fill level. Let us consider that the maximum fill level is such that it is below the motor torque line and the motor does not get stalled. I think we got to believe the vendor's curve as authentic and the question of impeller/runner design etc. has been taken care of by them.

According to the software vendor "For example let say the minimum starting torque is set to 100%, and the maximum starting torque set to 140%. Then if the demand power for the empty belt is 15% of the motor nameplate rating, then the starting torque would be 100+(140-100) * 0.15 = 106% for that case. If the fully loaded power was 85% motor nameplate, then the starting torque would be 100+(150-100) * 0.85 = 134% for that case." Till now we used to consider that for both empty and full load start, starting torque will be same and will be 134%. The software vendor states "The idea behind this methodology is that the empty belt starting torque can be reduced from the fully loaded torque to more close match the real behavior of the couplings." Now you can easily observe that 106% starting torque will call for much less take-up tension to prevent drive pulley slipping for empty start than what would have happened with 134% starting torque. This will lead to saving in belt rating. I would like to get your opinion on this aspect, if I am clear. Well the relationship they have proposed may be a point of argument, but whether such lowering of starting torque during empty start is logical or not is the issue.

I also understand that fluid coupling manufacturers are the most authentic agency to clarify/confirm the subject and I am writing to them also for their perusal. ■

Dear Nordell,

The software vendor has given the following explanation:

" The bottom line is if you want to do TV,TVV, or TVVS couplings correctly then you need the manufactures k/lambda skip curves and be able to model both the coupling and the flexible dynamic behavior of the belt. This is indeed what we do with our in house dynamic analysis software. Starting torque is a function of fill rate, time, slip, oil viscosity, and MANY other factors.

Rigid dynamics assume you have a CONTSTAT acceleration during starting (hence the definition of rigid dynamic). With rigid dynamics you need is select a SPECIFIC torque during starting (when in fact it is varying over time).

Let's say you select a fixed 165% starting torque. This very well may be quite accurate for a loaded belt start and the rigid dynamic predictions for belt tensions and time will be correct. However, the empty belt may start VERY quickly and thus the tension will show that slip is a problem. The engineer would then need to INCREASE the take-up mass to account for the empty belt slip condition. However, in the field we have seen that for the quick acceleration time the fluid coupling torque really doesn't hit the 165% instantly, but this develops in a given time period. Over this period the developing torque is sufficient to start the empty belt. For example instead of the empty belt starting is say 0.5 seconds, it might be more like 1 second. Of course this depends on MANY factors but this is the basic idea.

Therefore in order is more accurately model this behavior we have added this simple methodology. Again in our opinion this is more accurate than using a fixed 165% (or whatever) motor torque for all cases. Of course if you are looking for the best accuracy you need to get the real coupling k/lambda skip curves, and several other coupling parameters for the manufactures and then do a full dynamic analysis of the conveyor.

Attached is a perfect example of a loaded verse unloaded start. These are from the manufacture and are theoretical, but we have seen this in our measurements on many conveyors in the field."

I hope now it will now be more clear. Please note that I am no way trying to push the vendor's marketing. i am only trying to come to a consensus so that we can get some agreed benefit in static (or rigid dynamic) analysis. Regards

Originally Posted by nordell

Yes, we understand your comments above. Please note a fixed-fill level coupling still has the capacity to accept many fill levels. Once selected a fixed volume, the coupling will transmit a calculatable torques range associated with that specific coupling design and that volume of oil. Change the oil in the fixed-fill coupling, and you change the peak torque. The fixed fill peak torque is reached long before the belt reaches full speed. It gets quite complicated to give a rule-of-thumb. So long as the motor torque does not cross the coupling output torque, this concept works. We have modeling tools that render the torques values to the belt with any coupling and any fill, for most designs.

With regard to your formula, I doubt it has any validity. If the motor and coupling have sufficient torque to accerate the drive to full speed before the belt reaches full speed, this is the point peak torque is achieved, given the fixed fill. Once peak torque is reached, typically above 160% of motor nameplate, the coupling will output torque along the coupling fluid slip curve, typically dropping torque according to the fluid slip ratio between runner and impelller. Numbers like 140% starting torque are highly misleading. They are often exceeded with fixed fill couplings. That is why delay fill chambers were invented.

Attachments

unloaded (JPG)

loaded (JPG)

■

Dear Nordell,

I request your opinion on my last posting. Regards ■

Many thanks to you deyabhijit2005mech for your valued input and very explanatory graphics. However, I find no reason why the principle can not be applied for fluid couplings without delay chambers. In fact such fluid couplings will also follow a starting curve from zero to peak depeding on the level of filling and will exibhit similar behaviour. But couplings without delay chamber can not provide minimum torque as low as couplings with delay chamber. Hence in such cases effect will be less than couplings with delay chambers. ■