liquid ring vacuum pump

[DenSocling]

We build vacuum dry kilns for the forest industry. We have been building them for about 20 years. These machines usually run 24/7. We use liquid ring vacuum pumps. We have recently started having trouble with one that was installed about six months ago. Sometimes it trips the VFD. The operator says it won't turn unless you use channel locks. This pump is relatively small at 7.5 HP. The operator says that after it "loosens up", it starts and runs with no problem. In this process, the pump is started and stopped about four times an hour. It will start over and over then it will need the pliers. Does anybody have any idea what is going on? It's not flooding and the coupling is in alignment.

 

[MikeHalloran]

Does it have packing glands on the shaft(s)? ... correctly adjusted?

Could it be full of sawdust?

... fouled with wood resin?

What about the seal water? Correct flow? Correct temperature? No particulates?




Mike Halloran
Pembroke Pines, FL, USA

 

[DenSocling]

Mike,
We always use mechanical seals.
We always have strainers in the seal water supply.
We always use condensers before the pump.
We have a flow meter in the seal water supply coming from a cooling tower.
But thanks for your suggestions. And like I said, this kiln has been running about six months but our first kilns from twenty years ago are still running.
Den

 

[Artisi]

After 20 years and after how many units (I assume many), is this the first instance of a problem?
Have you thought about changing out this particular unit for another?
Have you have a close inspection of the v/pump,motor and VFD after a stoppage?




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.)

 

[DenSocling]

Yes this is the first time we have ever seen anything like this. We probably have installed 80 to 100 pumps.
This pump is relatively small but it's still $5000.
Unfortunately, the installation is several hundred miles away. We are connected through a supervisory computer and input from the operator.
Yesterday the operator had to reset the freq drive repeatedly. Today it cycled on and off all day with no problem. It just doesn't make any sense.

 

[MikeHalloran]

It makes sense if there's a stick or a board or something similar wandering around in there, too big for the pump to pass, flexible enough to rub on the rotor without making a lot of noise, and small enough to get lodged in something when the pump knocks it back up the intake pipe.

I'd suggest a trip to witness, in person, disassembly and inspection of at least the adjacent piping if not the pump itself.



Mike Halloran
Pembroke Pines, FL, USA

 

[DenSocling]

Tomorrow I will post a couple pictures. Thanks for the suggestions tonight.

 

[DenSocling]

This is the kiln that is giving us trouble. We have built the same system many times. The blue vac pump is on top.

The tank at the end is temporary seal water for testing. Here's another.

The grey shell and tube is the condenser. The condensate falls into the SS tank. Everything has strainers.

 

[MikeHalloran]

Perhaps that particular VFD is misprogrammed or defective.

Mike Halloran
Pembroke Pines, FL, USA

 

[lukin1977]

Is it tripping for overcurrent or what? Does the VFD gives any useful information once is tripped?

 

[DenSocling]

Yes the VFD overloads but I think I have an idea. This has never been seen but there is a first time for everything. Following are three screen shots from this kiln. The green line is chamber pressure. The black, stepped line is the PLC dumping the condensate collection tank. When that black line goes flat, there is something wrong with the level sensor. I wonder if there could be so much water backed up that it is starting to flood the condenser. Vacuum pumps don't like to lift water!

 

[LittleInch]

I'm glad you understand those traces, because without a P & ID and a GA with knowledge of which instrument is which line, it means very little to most of the rest of us.

If the inlet line is becoming blocked or full of water then yes, the vacuum pump won't like it.

For any intermittent fault you just need to look at potential causes and gradually wipe one off after another. Start with easy things like checking all the sensor inputs - impulse lines can block and calibration of instruments can drift. If the control system is getting duff info it will make bad decisions.

The key one will probably be a need to strip down the vacuum pump or at least remove all the connections and have as good a look inside the pump and piping as possible.

A good log of what was happening each day and trip will help as operators have a habit of not telling you the full story and blaming the pump is an easy out. Getting access to the control system historian is a good idea to see what was actually happening as opposed to what should happen.

Let us know what you find, but clearly something is not right on that bit of kit and my feeling is that it is not the vacuum pump....

Remember - More details = better answers
Also: If you get a response it's polite to respond to it.

 

[Artisi]

Can you monitor power input of the v/pump, this would be a start to problem solving.

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.)

 

[DenSocling]

I agree that the graphs don't mean anything if you don't know the assignments. The point was the green line. Look at how many times the darn thing can start without a problem.

We are trying to be methodical. I'm trying to stay out of direct contact with the customer and a technician is doing the talking. We don't want to overwhelm the customer. The technician and I have regular meetings.

We used to connect the freq drive to modbuss but they are too noisy. We have asked the customer to give us readings.

We asked the customer yesterday to help us log problems and readings.

Thanks again.

 

[LittleInch]

Oh wow, do you mean this thing starts dozens of time a day?

Is that normal?

Why have got a VFD if it starts and stops so much?

what are the complete loss of vacuum - when they load the chamber with more wood?

Remember - More details = better answers
Also: If you get a response it's polite to respond to it.

 

[DenSocling]

Using a VFD does two things. It stops wear and tear on the coupling and it also makes control of the seal water easy. We ease it open while the pump is starting and ease it shut as it comes to a halt.

When drying wood, we use a chamber pressure set point to set the boiling point of water and therefore the temperature of the wood. Then we operate over a deadband. While pressure is going up, the wood is heating. When pressure is pulled down, evaporation is rapid. On/off is part of the process that has given us phenomenal results over the years.

Now you see why this intermittent problem is so perplexing?

Complete loss would indicate the chamber was opened.

 

[LittleInch]

Sorry, re-reading the post makes it clear you've said it was on/off - I just didn't pick it up.

Only other thing I can say looking closely at the photos is whether the condenser piping works well. Looks to me like the air inlet is at the bottom fighting with the condensate drain down to the stainless steel tank and also the line from the top of the condenser doesn't look like it has a slope on it back to the condenser. If you're getting start / stop then is there possibility of water collecting in it over time which then hits the pump as a slug?

Try angling all the inlet pipe back to the condenser as an easy win to eliminate this.

Otherwise there must be something different in this kiln than others. Type of wood? wetness of wood ( is that the correct term?)
relative number of cycles?

intermittent problems are the worst because you are never really sure if you fixed it....

You may just have a badly set up pump which heats up and then jams.

Keep at it and let us know if you discover anything.

Remember - More details = better answers
Also: If you get a response it's polite to respond to it.

 

[DenSocling]

Something must be different but we can't see it.

We have dried many species of wood around the world. This customer is drying common North American species. Wetness of wood is called moisture content. Again, nothing out of the ordinary.

Since some of you are interested and have shown an interest, here's a piping diagram.

[URL unfurl="true"]https://res.cloudinary.com/engineering-com/image/upload/v1466013866/tips/Cooling_Water_Schematic_2_anujko.pdf[/url]

 

[MikeHalloran]

It's starting to smell like this particular unit is different from all of the others you have built, in a way that you are not likely to detect by remote control, using someone else's eyes.

It's the sort of thing that, when you personally, in person, arrive, and start inspecting the unit, you will say to yourself, "How the hell did THAT get put on the truck?", probably within a minute of arrival.

Sometimes there is no substitute for being there.


Mike Halloran
Pembroke Pines, FL, USA

 

[DenSocling]

My wife used to say that I should give every new operator a hat with a camera on it so I can see what he is looking at!