Radical HP spike at a particular speed

[skgcog]

Hi everyone,
We have a very strange phenomenon occurring here at our pulp mill on a new pump we have just installed. The pump is a Trashex 30 which is comparable to Voith's Contaminex CM30/31. The layout is such that it can be used to pump a batch pulper to a tank, or switched to recirc the contents of the tank to keep it agitated. We have just installed the system and are trying to run it on water to begin commissioning. The installed HP is 200 with a 1185rpm motor and a 14" DriveR sheave with a 28" DriveN sheave (2:1 reduction in speed). Initially, we tired to start it with it's across the line starter in recirc mode. The amp draw was over 500 on a motor with an FLA of 243A. The initial thought was we were simply trying to move more water than we had HP, so we connected a VFD. The Suction line comes from the tank (22'Diameter by 30' high with a usable of 25' at 100% full). The suction line is approximately 2' from the bottom of the tank and the discharge into the tank (tangentially) is at approximately 9'. With the tank full,it is essentially a closed loop with the only discharge head being line loss. What we have found is that we have a HP spike between 31% and 32% pump speed. At 31% we are drawing 170A and at 32% the amps are 270. This is with a 6rpm increase in pump speed. The crazy thing is that this same spike occurs at the same speed whether we are pumping from the pulper (which is above the pump and effectively will gravity drain until we reach about 60% tank level), in recirc with the tank at any varying levels from 20% to 100% or with the discharge valve closed (dead head). In any case the same problem occurs at 31% to 32% speed increase. Anyone seen anything like this or have any ideas? Sorry for the windy post, just trying to give as much info as I can. Thanks in advance.

 

[skgcog]

Also, just to help the process, we have already tried another motor. Also, running the pump dry with just seal water one for the packing it runs up to 100% speed in 5% increments and the amp draw tracks with speed nicely up to a full draw of only 100 amps. Nothing mechanical or drive related that we can see.

 

[TenPenny]

Is the tank (and the discharge line) full of stock when you're sspeeding up the pump?

Also, at what speed does the pump generate enough head to overcome the friction, and start moving fluid?

Without any real thought to it, I'm wondering if that jump in motor load corresponds to fluid starting to move, or something along that line.

 

[skgcog]

Thank you for the kind reply TenPenny,

1) the tank has water in it. We are still trying to get the pump to run properly on water and have not put stock to it yet.

2) Unfortunately, during our haste, we have not determined at what speed it does begin to move water, but at 20% to 30% speed (12 to 18.6 Hz) we know it is pumping, simply because with the levels equalized in the pulper and tank, we can lower the pulper level and raise the tank level by running the pump, it is pumping at some low rate which we did not quantify.

3) This spike occurs at a fixed speed regardless of pumping conditions, even with fluid moving, when we cross the magical boundary of 31% (18.6 Hz) the current goes extremely high and the motor actually begins to slip.

Also, contacting the vendor, he recommends an impeller trim, they inadvertently shipped an 830mm and recommend 750mm. I haven't looked at the difference yet, (just got the info) but even so, the speed seems awfully slow and the spike seems awfully high for a 6 rpm change in speed, but that is why I am asking for help. I surely don't understand this contraption.

 

[BigInch]

Do you get the spike without the VFD, or just a high starting current?

Got an accurate diagram of the pipe and fitting configuration?


**********************
"Pumping accounts for 20% of the world’s energy used by electric motors and 25-50% of the total electrical energy usage in certain industrial facilities."-DOE statistic (Note: Make that 99% for pipeline companies)

http://virtualpipeline.spaces.live.com/

 

[Artisi]

Can you post the performance curve together with relevent detail.
With such low head have you looked at where the pump is on its curve - you are possibly beyond end of curve and overloading.

 

[BigInch]

I wasn't thinking that would happen at 30% rpm.



**********************
"Pumping accounts for 20% of the world’s energy used by electric motors and 25-50% of the total electrical energy usage in certain industrial facilities."-DOE statistic (Note: Make that 99% for pipeline companies)

http://virtualpipeline.spaces.live.com/

 

[Artisi]

Just a quick thought as there is virtually no head on the pump - maybe the pump is oversized by a factor of 2 or 3 - who would know with the limited detail we have.

 

[BigInch]

I have my suspect. Suspect #2 is a hard hat in the tank.
OK, OK. I'll be patient.

**********************
"Pumping accounts for 20% of the world’s energy used by electric motors and 25-50% of the total electrical energy usage in certain industrial facilities."-DOE statistic (Note: Make that 99% for pipeline companies)

http://virtualpipeline.spaces.live.com/

 

[IFRs]

Could there be a resonance being set up somewhere?

 

[BigInch]

I think its a bit like Artisi says and a bit like I'm thinking. The pump is 200 HP, too big for just pumping to the tank (any valve in the discharge?), so they didn't run past their starting current draw and shut it down. Then they decided to drop in a VFD to slow things down, so I just bet they didn't do any power conditioning, but I'm trying to be patient.

**********************
"Pumping accounts for 20% of the world’s energy used by electric motors and 25-50% of the total electrical energy usage in certain industrial facilities."-DOE statistic (Note: Make that 99% for pipeline companies)

http://virtualpipeline.spaces.live.com/

 

[skgcog]

First of all, I would like to thank all who contributed and give you all an update. We opened the pump and found the impeller to be 830mm rather than the spec'd 750mm. Nothing else unusual except the vortex breaking blocks seemed to be excessively large. Upon removing them and replacing them with a piece of square stock rolled into a helix, also called a spoiler bar. After this modification we were able to run the pump dead head up to FLA without overcurrenting. We then changed the impeller to the smaller 650 on the tank to screen pump and ran it starting at 20% and incremementing up by 5% until we went overcurrent at 70%. We backed off to 65% and increased 1% at a time until we got to 68% motor speed and FLA. From this we used the affinity laws to calculate a trim for the impeller and sent it out to be trimmed. We should have it back tomorrow to try again.

Looks like the vortex breaking blocks along with the oversize impeller was the culprit. Again, thanks to all who contributed and hopefully anyone else who has to deal with one of these out of the ordinary pumps can gain a little from it.

Best regards,

Steve

 

[skgcog]

Sorry, I wasn't clear on the HP spike. After removing the vortex breaker blocks and installing the spoiler bar, the HP spike at 31%-32% disappeared. We still had an oversized impeller to deal with, but at least we could begin to collect data to deal with it.