Actuator sizing

[ScottyUK]

Not my normal field, but I've been asked to review valve & actuator sizing. Actuators are Rotork, valves are a well-known Western manufacturer.

One of the actuators is sized at 100% of predicted valve operating torque. To me this instinctively feels incorrect - I made a few calls to friends elsewhere who both suggested actuators are normally capable of 150% - 200% of valve rated torque, without eating in to the actuator short-term stall capability.

Are there any industry standards which give support to these rule-of-thumb figures - an API document perhaps? Any assistance with supporting references would be really appreciated.


----------------------------------

If we learn from our mistakes I'm getting a great education!

 

[ione]

The valuable Emerson control valve handbook treats this topic (pages 129-133)

http://www.documentation.emersonprocess.com/groups/public/documents/book/cvh99.pdf

 

[bcd]

Frequently, the valve manufacturer adds a safety margin to their recommended valve torque. Check with the manufacturer to determine how much margin there is. Is the manufacturer giving you a recommended actuator torque or the just valve torque. I would be very leery of any manufacturer that is only giving you the valve torque because dimensional tolerances, packing adjustments, the media being handled, debris, etc. can cause the the torque to vary widely.

 

[steamdog]

There are many variables, and you should consider conditions years out as the valve may age. I size actuators all of the time and use a 25% to 30% safety factor depending on conditions. I think a 100% (2x required force) may be overkill. Also be careful, valves have a maximum force rating that will be the maximum allowed from the actuator.

 

[JLSeagull]

Many prefer actuators selected with 40% or 50% safety factor based upon the minimum supply pressure etc. I agree that 200% is wrong - and have seen that in one client specification.

 

[stanier]

The break out torque for a butterfly valve can exceed the duty requirements. Actiuator sizing should be based on overcoming the initial "stiction" and process conditions.

I could not see the type of valve you were talking about?

I had a poor experience of a well known western butterfly valve manufacturer who didnt take this into account. 270 valves had their actuators replaced.

It wasnt just a case of the valve not working. These were pneumatically operated valves. the pressure would build up until the appied torque overcame the stiction of disc on seat. the valve would then slam shut or open. This casued a waterhammer event that fractured the pipe.

With an electric actuator the drive will just overload and stop.

"Sharing knowledge is the way to immortality"
His Holiness the Dalai Lama.

http://waterhammer.hopout.com.au/

 

[rmw]

It also depends on the characteristic of the actuator. Rotork has some in their stable that are the old Jordan Controls designs that utilize a Scotch Yoke mechanism that gives them torque multiplication at the ends of stroke. I always liked them for BF's.

You didn't mention whether these are direct mount or linkage connected valves. If linkage connected, you can bias the linkage so that you get some mechanical advantage at the part of the stroke that you need it.

You didn't mention the valve type either. Different valves have different characteristics and torque can vary throughout the range depending on the valve type - BF vs Ball vs plug, etc.

Also, valves can develop more torque requirement after being in service for a while. I had to replace the actuators on a whole bunch of valves (not quite as many as above) when river silt got in behind the seats of some ball valves and the operating torque requirement greatly exceeded the capabililty of the actuator.

To me Rotork says electric and electrics can break gears when they get into a bind. See if the actuator can be funished with some kind of torque switch to take it out when it won't move as commanded.

Electrics aren't fast either. Trying to select one that will get the job done faster will cost you in the long run. One of the most successful brands out there, Beck is one of the slowest actuators made and there are lots of them made because people prefer them. That and their low current synchronous motors.

Post back what type of valves and service - in general terms if necessary - and maybe we can zero in on your application.

rmw

 

[ScottyUK]

Thanks guys.

Turned out that the valve manufacturer had made a mistake on their data sheet and I'd picked up that there was something wrong without being 100% sure what the right value should be. The proposed actuator has about 160% capability.

Ione - I should have said this is an isolating valve and the Rotork is an IQ series operating between limits, so the modulating actuator information didn't directly apply. Nice link though - thanks for the input.

bcd - I agree. It was poor data from the vendor which initially caught my eye. I asked for the maximum withstand torque for the valve to compare with the nominal value and the actuator capability and the vendor realised that they had made a typo or two.


----------------------------------

If we learn from our mistakes I'm getting a great education!

 

[gerhardl]

Everything seems OK and solved then, just let me add a couple of points only briefly mentioned above:

a) Electrical actuators have an inbuilt gearing down from a fast running electrical motor, where motorsize and gearing might have several choices within same physical size of actuator, or an added outside gearing between valve and actuator could be supplied. Result: wattage and closing time could vary, and also torque, depending on the sizing of part of the equipment for a single actuator.

It is of this reason very sensible to start with the knowledgde of how slow you can accept the closing and opening time to be, keeping the general cost (small actuators larger gearing) and electrical max effect down.

b) A good electrical actuator is not good if limit and torque switches are not included as standard inbuilt and supplied equipment with the actuator, along with overheating protection.

 

[Drexl]

gerhardl,

In your point b), generally speaking, how big a valve/actuator would you say should have torque switches??

 

[ScottyUK]

A few replies must have sneaked in while I wa distracted. Thanks to everyone who replied - I wasn't intending to pick out ione and bcd over the rest of you.

Valves are Cameron Twinseals, which I understand from a distinctly non-expert perspective to be a modified plug design. Actuator is an IQ series electric type. Should have mentioned the types earlier - apologies. The brand name is too well known and would have given away the manufacturer which I didn't initially want to do but as the problem is resolved it won't do any harm.

I'm not actually speccing the valves, but the actuator information kinda got dumped on my desk because no one else would take it.


----------------------------------

If we learn from our mistakes I'm getting a great education!

 

[gerhardl]

Drexl:

On my point b)

This could be a lengthy discussion, but let me try to put down some of my own general viewpoints for electrical actuators:

The philosophy is generally that if you can accept failures and lack of signals for failure from actuators, eg. valves not closing properly or not opening, you can also accept actuators without signal givers or alarms.

(In some cases you can do without if you have visual access and control to the valve positions with persons present all the time, and small consequences if failure)

For 'larger actuators' you have the additional problem that the actuators could give damage to the valve, eg actuators able to give such a force to the valve (perhaps helped by existing additional gearing mounted between valve and actuator) that the valve or seating or a weaker spindle construction might be damaged.

All this is of course argumenting for additional control, at least limit switches. In general good electrical actuators built for long lifetime (cost/lifetime is low) will either have standard inbuilt limit and torque switches, or have this as an option, but generally at a higher cost (mechanical alterations might be necessary) if as later ad-ons.

To put the border for yes/no at size rather than application would be difficult, but you could say that you for lower pressure classes, simple applications, and sizes 50mm up to 100mm (150mm?) or more (up to 250mm?) see that solutions without torque switches are relatively common, more seldom also without limit switches.

Praxis might vary from country to country.

In my opinion there is no other reason than keeping the buying price low (and not consider cost/lifetime and dependabillity) to select low-quality electrical actuators without the option for theese ad-ons.

(A remark is that solenoid operated valves are an exception, and only in cases are given limit switches (the solution is available!), and no torque swiches, as the movement is axial)