Average Railcar Heel? 2

[TiCl4]

We top off-load railcars of monomers and have been experiencing variances in our inventory management system on these materials. We recently compared tank level via gauge as well as stick measurement before/after emptying a railcar and found the storage tank to be light 700-1,200 gallons.

I'm not that knowledgeable about railcar design, but the top-offload configuration has a well at the bottom that should allow the dip tube to remove most of the material.

Compared with the other offloading station, this station usually takes much longer to offload; 4-6 hours as opposed to 2.5-3 hours.

I've been told by maintenance that the gear pump was replaced in the past with no discernable improvement. We'll start looking for vacuum leaks, especially around the swivel joint in the unloading arm.

However, 700-1,200 gallons of heel is a lot more than I'd expect so I wanted to get other's experience on railcar emptying and issues with overly large heels. Is that a normal heel in a railcar for top-offload, or am I looking at a likely vacuum leak?

 

[LittleInch]

Monomers?

Can you be more specific in terms of viscosity, density, "stickiness"?

Is this stuff heated?
Can it be heated?

Rail car floor angles can be quite low - what does yours look like? Can you post a cross section of a typical rail car?

Are you sure your dip tube is going into the well? Do you have a depth monitor on it? Does the well get filled up with debris / solid material?

What's that as a percent of the load? "Rail car" is a bit of a moveable feast. It does sound excessive, but if this stuff is high viscosity it might be all you can get.

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

 

[MJCronin]

... and, more specifically... is your sticky substance being shipped in right kind of rail car ?

Or, at the last moment... was another type of car substituted ... What type of car does the maker of the chemical recommend ?

MJCronin
Sr. Process Engineer

 

[TiCl4]

Sorry on the non-specificity on liquid type. The monomers are usual and typical monoemrs - styrene, acrylic monomers, etc, have a viscosity in the 0.5-2 cP(ish) range, have a s.g. of ~0.9-0.95. Monomers are not heated.

I'm unfamiliar with what "type" of railcars come in. Normal load limits marked on the sides are usually in the 180,000-215,000 lb range, so we might be losing 3-4% on each unload. We are being charged the full PO amount, so this is a significant loss.

I will have to assume the railcars are appropriate, MJ. They are shipped directly from the manufacturers, who are large and reputable companies (ala Dow, BASF, etc etc)

The railcar type doesn't seem to matter too much - we always have the longer unloading issues and inventory issues with a single unloading spot (which handles two different monomers).

Pictures below of two different rail types. Slopes are fairly low, but I don't have any cutsheets of any of the cars. I have no idea as to the internal configuration of the dip tubes.

 

[pierreick]

Hi,
Normally your PO is based on Weight not volume! Weird to me.
Based on my experience of unloading different chemicals from trucks, including monomers.
You may want to install a mass flowmeter on your discharge line.
Good luck
Pierre

 

[MJCronin]

Also, you could consider putting your storage receiver on load cells and get a more accurate number

Insist on paying only for what you receive ...

Also, rail car weigh cells exist .... for your consideration

https://www.walzscale.com/product-category/rail-scales/rail-car-weigh-scales/

Please complete this thread and tell us what you finally decide

Best of Luck to you sir !!!

MJCronin
Sr. Process Engineer

 

[LittleInch]

A DOT 111 railcar seems to come in different guises.

Not all of them have sloping floors and very few have any defined sumps. Even the slopes tend to be really quite small.

This issue seems to be one station and not another?

What does the dip tub look like? Photo?, drawing? sketch of what this looks like in section?
Is it able to reach right to the bottom?
Maybe it needs a smaller diam stripper arm?

Why is this not connected to the nozzle on the base of the rail tanker for unloading??

If this is a 4" arm with a tee at the base, then you will end up with min 4-5" of product in the base and possibly more as is cones down



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

 

[TiCl4]

Pierreick, the PO is in pounds, yes. The quantity lost appears to be ~6,000-9,000 lbs of material.

MJ, I've looked at railcar weigh cells, but they are prohibitively expensive for what would be essentially inventory tracking and management. The storage tanks are large, flat-bottom tanks, so load cells are not an option. Current level measurement is via dP meter.

LI, the unloading arm is 3". Where would you suggest I find drawings of the railcar? Some appear to be UTLX cars (rental fleet), so even the manufacturer may not have ready access. Again, the long off-loading time persists between different railcars and even different materials. We've only tested 1 railcar for heel, but based on past inventory cycle counts this problem is system to that particular unloading station.

We do not bottom offload due to the hazard potential of the bottom connection breaking and spilling the contents of the railcar. Top offload is inherently safer.

My first action will be to leak check the offloading arm and pump suction piping to make sure we aren't drawing air during off-load.

I guess one thing I was looking for was if anyone had experience with how much heel a railcar with similar material would normally leave. Does this 6-9,000 lb heel sound excessive?

 

[MJCronin]

"We do not bottom offload due to the hazard potential of the bottom connection breaking and spilling the contents of the railcar. Top offload is inherently safer."

What about other customers from your supplier ?.... Others must be purchasing this "stuff" and performing railcar offloading... What is their opinion and safety preferences ?

Who else buys these monomers and what do they do ? ... Informally contacting these other customer and asking your ultra-specific question may be fruitful

Please respect us and keep us in the loop .... Tell us of your final decision and the reasons for it..

Best Regards

MJCronin
Sr. Process Engineer

 

[3DDave]

Yes, 4.5 standard tons of liquid is more than one would expect to remain. That's pushing 10% of the tare weight.

If you are using suction then the maximum dP is 14.5 psi. An air leak will reduce that; while it's a lower SG material than water, you need to overcome friction losses so a plain ~30ft lift isn't going to be fully available. That should be detectable as a continuous and severe drop in flow rate.

Assume a 45 foot long by 9 foot diameter tank.

You have 9000 lbs * (1/.9) / 62 lbs/ft^3 or 161 ft^3 or so of material; 3.5 cu-ft/foot of tanker = 3.5 square feet section.

That works out to a segment of 500 square inches, or about 12 inches deep.

The 1200 gallons mentioned earlier would be nearly 200 cu-ft, so even deeper. If the tank is longer, then a bit shallower, but doubling the length to 90 feet leaves a 7 inch depth.

 

[georgeverghese]

Since this long unloading time is peculiar to this unloading station, like you say, the unloading arm and suction hoses may be leaking in air, resulting in loss of suction lift. Also check the shaft seals on that gear pump.

 

[LittleInch]

Dimensions do vary but this looks about right.

9' 10 OD, 53' long cylinder.

Does the flow rate slow or is able to be reduced at the end of the offloading?

Any sort of decent flow will just cone down, so maybe need a stripper pump for the last little bit.



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

 

[LittleInch]

TiCl4 - I meant a drawing or sketch of your unloading system, arm size and length, max height of the arm, length of pipe work to the pump, flowrate of pump, etc

How many railcars do you unload at the same time?

Is this just this particular station with one arm? If it is literally just that one, then there seems to be something amiss with that location. Leaking valve, leaking joint?, some sort of obstruction in the arm or pipe??

Do you have any pressure readings? flowrates etc over time during the offloading?

There have been a few posts here with bits of information which really needs bringing together into a single cohesive post in order to help further.

Can you not unload the majority via your arm and then connect a stripper line to the central valve?

But it looks like something not right with one particular location so trouble shoot that locations would seem to be the answer to me.

I can understand your issue to do with bottom loading, but I would have thought a hard arm would reduce the risk to acceptably low to connect to the bottom loading point. I hadn't heard of anyone trying to unload from a dip tube before now.

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

 

[pierreick]

Hi,
Take a look at this brochure for acrylic monomers:

https://www.petrochemistry.eu/wp-co...uidelines-final-01-2016-blocked-version-2.pdf

and styrene:

https://www.saferack.com/bulk-chemical/styrene-monomer/

As suggested you should contact your supplier(s)or other users for more input and experiences sharing.
My 2 cents
Pierre

 

[TiCl4]

Thanks for you input all. There is a good bit to test on the existing system, so I'll let you know when we've eliminated some of the potential issues.

 

[danw2]

>styrene, acrylic monomers, etc, have a viscosity in the 0.5-2 cP(ish) range, have a s.g. of ~0.9-0.95

>Current level measurement is via dP meter.

DP level error is proportional to specific Gravity deviation from the assumed SG.

The DP transmitter reports the hydrostatic head in pressure units. Somewhere, usually in the controller/HMI, pressure units are converted to level or volume using an assumed SG for the material.

If any given tank's contents are assumed (for units conversion purposes) to be an SG of 0.92, but the actual SG is 0.90 due to temperature or chemistry, the conversion is in error by 2.2%.

How well do the reported readings track the real S.G.?

 

[pierreick]

Hi,
"I've been told by maintenance that the gear pump was replaced in the past with no discernable improvement"
I don't think this is a right technology for a like water product.
Pierre

 

[3DDave]

Pierre - the OP claim is that recovery on other stations at the same company for the same/similar product is better than this one. He's looking to diagnose a problem with an under-performing station while providing insufficient information to do so. The "heel" appears to be a red herring. No car would be designed to allow that amount of material to remain.

It's really not clear why this station would continue to be used rather than diagnosed and repaired.

 

[pierreick]

Hi, to be frank, I'm puzzled with the Raw materials management, you cannot rely only on PO without reconciliating the bill of material with a meter or load cells systems. To me you have no guarantee that any of your unloading stations are working well.
NB: Those raw materials are expensive, adding meters should be a piece of cake with a nice ROI.
Just my view, but your operation team and finance controller are not doing very well in term of stock taking/yield performance (monthly report). No one is challenging the supplier....
BTW you may have a hole in the upper part of the dip pipe (anti siphon) which explains some product in the railcar at the end of the operation.
Pierre

 

[danw2]

If the tanks are not vented to atmosphere and the DP level transmitter connection to the top of the tank is a dry leg, then an accumulation condensed liquid in the dry leg could give a lower-than-expected level measurement error.