Continue to Site

Eng-Tips is the largest engineering community on the Internet

Intelligent Work Forums for Engineering Professionals

  • Congratulations IDS on being selected by the Eng-Tips community for having the most helpful posts in the forums last week. Way to Go!

Pump protection against dead heading 3

Status
Not open for further replies.

eeprom

Electrical
May 16, 2007
482
Hello,
I am an EE, and I'm working on a control system to protect centrifugal and vortex pumps from catastrophic failures due to dead heading. I am going to use temperature as a leading indicator of failure, so I intend to install a temperature switch onto the housing of the impeller. From a control standpoint, this is very simple: it is just a thermostat. But given the huge variation in pump sizes and shapes, selecting and mounting a good switch is not so simple.

The best solution I have come up with so far is to use a thermally conductive epoxy to mount to the pump casing. But this would be a bit of a pain to install, and it may not hold up as well as a bolt. Tapping into the casing is out of the question. So I am stuck with the question, "how to mount a switch to the casing of a pump?"

Does anyone know if a switch is already made for this application?
If not, does anyone have any ideas for mounting a switch?

thanks
EE
 
Replies continue below

Recommended for you

BigInch,
I cannot tell if you are being straight or sarcastic. Of course there is a temperature rise. The temperature rise continues until the fluid reaches saturation temperature. At that point the heat input is used to overcome latent heat of vaporization. While in saturation there is no temperature rise with an increase of heat energy.

Anyway, I have not been very clear. This concept is not for one application. This system is a design for any pump regardless of size and fluid type. The only common is that all pump fluid runs normally below 140F, and all pumps are vortex or centrifugal.

Using current to monitor where the pump is operating is not reliable. There are too many variations. The current changes very little at no load. What changes is power factor. And that would be expensive to monitor on a wide spread application.

This is a catastrophic insurance policy. It has nothing to do with saving the pump. This is to shut the pump down in order to protect personnel.
 
Then maybe ask me again when you have a real-life problem. These solve-all problems for pumps that don't exist, that have flat pump curves, that have abrasive fluids, that don't have relief valves ... well... I'll see if there's a baseball game on. Good luck.

What would you be doing, if you knew that you could not fail?
 
PD pumps use both a dry running temperature sensor as well as Overpressure Protection device.

Overpressure Protection device is a diaphragm contact pressure gauge with an adjustable contact for maximum pressure.

The Dry running protection device TSE measures the temperature between rotor and stator is permanently sensed thermoelectrically via a temperature sensor integrated in the stator and compared with
the limit value set at the TSE control unit.

When the pump runs dry, the temperature will rise due to the increased friction between rotor and stator. When the set limit value has been reached, the TSE control unit switches off the pump drive and triggers a fault message.



Tayco manufactures surface mount temperature sensors for many applications including pumps. General purpose metal or ceramic encased elements can be laminated in polyimide, silicone, or epoxy.

 
You might want to look into the ITT Goulds 'PumpSmart' products; they use VFD technology, but they have an internal algorithm to allow them to be configured to protect centrifugal pumps against deadheading and cavitation.

I'm not plugging this product, but it sounds like you're looking to do what they have already done, so it's worth a look.

I think, and look at the PS200 PumpSmart product.
 
Measure temperature increase across the pump with surface mounted thermocouples on suction and discharge piping?

One of the first relevant links in a google search for "differential temperature thermocouple" is:

 
An ultrasonic flow switch would cost less than a $1000.

The variable is "flow" so detect "flow" not a symptom of low flow.

works on sewage and slurries as well as clean fluids.

Cost of the instrument is less than the engineering.

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

 
Thanks for all the recommendations. I think I may include the ultrasonic flow switch as an option. Flow switches can be so squirrely, especially ultrasonic ones. You have to make sure the sonic goop stays on the sensors, and that doesn't work so well outdoors.

This is supposed to be catastrophic protection only. Nuisance trips will upset the process, and that will be expensive.
 
What is the catastrophic failures you are expecting?

I really don't see what the problem is, there are 1,000's of slurry pumps operating round the world on 100's of different applications with some installed and working under atrocious mis-matched aplications - how many have such sophisticated protection -I would suggest nearly none.
Have you undertaken any research to establish what the likelihood is and how many have suffered catastrophic failure when operating at shut head, again I would suggest very very few?



It is a capital mistake to theorise before one has data. Insensibly one begins to twist facts to suit theories, instead of theories to suit facts. (Sherlock Holmes - A Scandal in Bohemia.)
 
I didn't create the problem, I was hired to analyze and solve it.
 
If a catastrophe can occur then do the risk benefit analysis before speaking in terms of expense. A magflow meter or Coriolis meter would be the choice if the cost of a catastrophe, in money, life or environmental terms is high.

Define catastrophe? Loss of production, environmental spill, loss of life???

The petrochem industry does a SIL analysis to determine the risks involved with control systems. en.wikipedia.org/wiki/Safety_Integrity_Level

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

 
Apparently very much less than the cost of a strap-on flow meter.

What would you be doing, if you knew that you could not fail?
 
I think your best bet would be something like a paddle flow switch, kind of like this.

With a slurry solution it may not respond immediately to a loss of flow, just because of the higher viscosity...i.e. it may take a few seconds to spring back to it's now flow position, but it would be a lot quicker than detecting a temperature rise of the pump casing.

 
Paddle flow switches last nanoseconds in slurries. By the time you engineer it and have a team install same the cost of the primary element is miniscule compared to the risks.

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

 
There have been some very good ideas offered on this topic. Thanks. But I would like to argue in favor of a temperature switch in that: 1. It is non intrusive; 2. It is cheap; 3. It is easy to implement in a pump starter circuit; 4. A pump with a 50psi dead head pressure has to reach 300F before it starts to build pressure, and so a temp switch set to open at 180F has a large margin for safety.
 
Couldn't a pump operating at dead-head against a closed block valve or pipe blockage downstream start to build some pressure immediately coincident with the increase in temperature due to adding horsepower to the fluid that is otherwise (sort of) going nowhere, or is the thermal expansion of the fluid compensated for by the slip back around the impellers? I suppose it depends on overall head capability, speed and impeller clearances, but I suspect the pressure would rise at least somewhat; perhaps not with a slurry pump (coarser clearances). But, certainly you would not be expecting to allow the temperature of even a warm slurry to rise to T(sat) at P(shutoff) before you tripped the pump, correct? Your selection of 180 F is otherwise OK on first glance, but be careful with other components of the pump that might be destroyed by the pumping temperature irrespective of the pumping pressure, such as elastomers, or such as mechanical seals that might require external quench water in order for the pump to be suitable for hot slurries. Chances are, your temperature setting might be governed more by those limitations than by trying to prevent the pump from exploding due to the pressure rise arising from heat input at deadhead.

While I acknowledge merit in the temperature-based concept, I still think flow is better for you.
 
I would question if a pump developing 50psi at shut valve could ever achieve anywhere near 300F. Anyway, seems the OP has already decided on temp as the trigger so we should save our breath, shake or nod our heads depending on own thoughts on the subject and leave it at that.

It is a capital mistake to theorise before one has data. Insensibly one begins to twist facts to suit theories, instead of theories to suit facts. (Sherlock Holmes - A Scandal in Bohemia.)
 
Let's keep it civil, no pun intended. I'm not making things up for the sake of making my job easier.

I completely agree that flow would be better. But given the design requirements (to prevent catastrophe), the temperature switch will allow the pump to be ruined, but it will protect personnel.

The pressure of the pump dead head is simply it's shut off pressure. The pump cannot gain any more pressure unless the pump speeds up (won't happen), or the temperature of the fluid increases to the point that the fluid begins to vaporize (this will happen eventually). If a pump cavity has water at a pressure of 50 PSI, at what temperature will that water become steam? Look it up in the steam tables. It's about 305F.

 
eeprom,

That is where my reference to T(sat) at P(shutoff) came from.

I will try to put a less electrical spin and more mechanical spin on this topic.

Artisi's point is profoundly all-encompassing. Unless your system is, for all intents and purposes, adiabatic, you might never get to the point where the water will ever boil. It might get hot, though. My point is that hot water in a stagnant column against a blockage will expand with increase in temperature, and in the system that you describe, the only avenue for mitigation of the coincident pressure rise will be in the "slip" of water around the impellers. That avenue may, or may not, limit your rise in pressure as a result of heat input at deadhead, at least to some degree. However, I believe it likely that some other mechanical failure will occur before the pump explodes as this operating condition is permitted to continue. I made reference to failure of elastomers and seals, which may not be perceived to be catastrophic but which will nonetheless render the pump inoperative. Further, depending on the configuration of the pump, prolonged operation to the far left or far right of BEP (in your case, as far left as you can get) can give rise to shaft deflection and premature bearing failure due to unbalanced radial forces. If you have a healthy continuous rise to shutoff in your pump characteristic, you might not see load oscillations, but if you are flat or have "droop" in the characteristic near shutoff, you can end up with unstable operation and cyclic loading that can fatigue parts - like the shaft - to failure. If something like that happens, although you haven't blown up the pump due to overpressure, the failure can be equally catastrophic in nature.

In my opinion, the best indicators of deadhead are, in descending order, (1) low or no flow; (2) differential pressure across the pump; (3) pump discharge pressure; (4) temperature rise per your suggestion. In the event that you choose Option 4, then I believe it would be prudent to use the high temperature set point that is the lower of your number 180 F or the temperature limit of the weakest link component in the pump that is susceptible to failure at elevated temperature. That way you mitigate most of the "oil canning" risks with exactly the same safeguard, and the only thing remaining at issue is its set point.
 
eeprom

I posted earlier about an epoxy mounted temperature sensor. Don't know if you have reviewed that post.

I was thinking that there may be a better alternative than epoxy mounting a sensor. You can actually use a non-contact infrared temperature detector. There is an advantage to it in that you can target the beam on the location of the pump where you expect the temperature rise to occur. The mounting of the non-contact infrared temperature detector should be a little simpler (unless your pump is insulated).

Check this out:

 
Status
Not open for further replies.

Part and Inventory Search

Sponsor