Pump no flow protection.

[fornhamspark]

I have an application where i need to detect no flow or a dead end condition. As the heat builds up in the pumped fluid could give rise to an explosion, so iam told.
The fluid is liquid fertilizer and quite corrosive, pipe size 6inch stainlesss steel flow rate 1000l/min.

Iam looking for instruments to use.
1. Paddle switch probably not last long as imersed in fluid.
2. Thermal dispersion needs to be wet all time
3. Mag flow very expensive.
4. Current, amps only change by 2%

Any suggestions please also what about looking at this from a temperature point of view.

 

[Skogsgurra]

Red herring?

Gunnar Englund

http://www.gke.org

--------------------------------------
100 % recycled posting: Electrons, ideas, finger-tips have been used over and over again...

 

[danw2]

Gunnar, your command of idiomatic English is amazing.

I can't answer for the status of durian. My food importer friend is a 'Lars' by name and deals only with that which the Swedish expatriate community craves: Vasa bread, Kalles kaviar and the like.

Dan

 

[DaveScott]

I think with the information provided, I would prefer to opt for temperature detection, since this is the direct problem that is to be avoided.
A further problem that I noticed is that the pump self primes every time. If doesn't, then it will also get hot, but outlet pressure will be low.
The most simple way of avoiding the problem, instead of detecting it, is the one that has already been suggested twice: make a small bypass from the pressure side back to the underground tank.

 

[Skogsgurra]

Dan,
I think he should bring in some Leksands Brungräddat and also some Vikabröd. They are really something, if you are into rye crisp.

Dave,
That bypass needs to be more than "small" if it shall have any effect. Can you justify the extra energy losses?

Gunnar Englund

http://www.gke.org

--------------------------------------
100 % recycled posting: Electrons, ideas, finger-tips have been used over and over again...

 

[jmw]

The usual way would be to control the by-pass with a pressure relief valve and he energy associated with operating this would be irrelevant to the pumps normal operation.
The pressure relief valve would be set to operate if the pressure exceeded the pressure when the pump is delivering into the open ended system. This is normally done to protect the pump and this approach presumes the pump was not supplied with an internal bypass.
From what has been said, that wouldn't be appropriate anyway since local internal circulation would probably also generate excess heating in the small amount of fluid being recirculated.
So: in this case an external bypass that feeds back to the tank would be necessary.
This means installing a T and some return pipe work with a pressure relief valve. I suspect that this would be more expensive than simply installing one of the sensor options and using them to shut down the pump.
Once tripped, the problem will disappear and thus you might need to latch the switch off until the cause has been investigated and it is safe to restart the pump.
Much depends on how troublesome it is to break into the pipeline to install various of the sensor options and hence the ultrasonic flow switch might prove a useful alternative.
One assumes that once you propose breaking into the pipeline sensor costs may be the least of the costs if you have to do a Haz-Op review and if the installation is going to be expensive whatever the sensor chosen.

JMW

http://www.ViscoAnalyser.com

 

[Skogsgurra]

Temperature sensors can easily be mounted outside pipe/pump. Surface temperature rises almost as fast as internal temperature.

Gunnar Englund

http://www.gke.org

--------------------------------------
100 % recycled posting: Electrons, ideas, finger-tips have been used over and over again...

 

[DaveScott]

//Can you justify the extra energy losses? //
In a way, I think so.
If we consider a 30kW pump, then roughly speaking, it would heat 1L of water by 7C per second. So if the bypass is 1L/s, then I suppose that the temperature rise is not dangerous.
A 30kW pump can pump roughly 750L/s over a 5m head. So the efficiency loss is quite small (1 in 750).

That is, assuming my scrap of paper calcs are not full of holes.

 

[JimCasey]

I don't like temperature sensors for this application. With the associated time delay, mostly they tell you that the pump got hot, not that the pump is about to get hot. Since the stuff in these pumps is explosive, if you live lone enough to get a HI temp alarm,...RUN!.

That's why I proposed a completely passive strategy to ensure pump cooling; and always-on bypass line. Sure it is not as energy efficient as some other scenarios, but it has no moving parts and can't fail.

 

[itsmoked]

I'd run open loop and temp sensors.

If the product isn't being in actively pumped to the 'user' then it's being fully or majorly bypassed back to the tank.

An 'always on' bypass can still be subject to heating when it clogs, or the drunk operator passes out and the whole unmonitored tank starts to heat up. (stretching?)

Keith Cress
Flamin Systems, Inc.-

http://www.flaminsystems.com

 

[ScottyUK]

Jim's idea of making the process self-protecting when all elements are behaving normally is a good idea. On a hazardous process I would expect at least one backup layer of protection to cover contingencies, and a temperature measurement with a tripping function designed to take the process to a safe state seems a reasonable second layer protection. Most temperature loops can be configured to give a digital output signal on a sensor fault which allows the control system to either warn the operator or automatically initiate an orderly shutdown on loss of the protection system. In Europe the current industry 'best practice' would be to carry out a SIL study on the protective elements in a hazardous process.


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

Sometimes I only open my mouth to swap feet...

 

[JimCasey]

Thanks for embellishing, guys. I certainly have no reservations about temp sensors except as a single and primary fault protection. Normally I'm resistant to suggest a continuous bypass because of the waste in energy , but with a dangerous fluid as this has been described it is prudent to have a totally passive protection system. Additional active systems such as temp monitors only contribute to additional safety.