October2001
Success in the Field -- Pennsy Realizes Big Efficiencies with Fleet Management System
Plant Sense -- Essentials of Plant Component Selection
Maintenance Matters A Systematic Approach to Seal Failure Analysis
Gravitas Superabit--Roof Control in Stone Mines, Primarily with Roof Bolts
Pennsy Realizes Big Efficiencies with Fleet Management System
In 1995, Pennsy Supply, Inc., a division of Oldcastle Materials/CRH
plc, implemented Command Data fleet management software and signaling
technology for its ready-mixed concrete trucks as well as its on-road
aggregate and asphalt delivery trucks.
At the beginning of the project, Mike Jenkins, Pennsy Supply’s vice
president-customer support, admits that Pennsy was only looking to install
a management system for its fleet of 65 ready-mixed concrete trucks at
the time.
“In the definition process of the project, we asked, ‘Why shouldn’t
we be looking at this technology for our on-road haul trucks, too?’”
said Jenkins. “Our first answer was that not many people were doing
it. But, we concluded that if we can save time and money controlling mixers,
we could do the same with the 130 dump trucks that we were running at
the time.”
Pennsy Supply looked to incorporate the technology into its own fleet
as well as what it calls its “inner core” of contract haulers
which are used on a daily basis. In 1995, the company had 64 on-road haul
trucks and used about 70 contract haulers. Its inner core of contract
haulers at the time was about 45 to 50 trucks.
“By involving contract haulers very much like a Pennsy truck, we
are not degrading our service to our customers by using contract trucks,
as everyone is on the same system,” said Jenkins.
Above are screen captures taken at Pennsy Supply of two of the several screens offered by Command Alkonıs CommandAggregate software. The top screen graphically depicts truck demand for all sites to help maximize truck utilization. The bottom screen offers an order entry template, which enables the system to automatically calculate and charge the customer.
Pennsy equipped these haul trucks with two-way radios and
signal status boxes, which were controlled by Command Data’s ControlAgg
DOS-based system.
“We calculated the pay-back period for the project to be four to
five years, but to make the project fly, we calculated a three-year period,”
said Jenkins. “We thought we were being incredible risk takers dropping
the pay-back period to three years, but after assessing operations, we
concluded that the real pay-back was less than one year.”
The assessment was based on truck loads per day. Pennsy set a target to
increase delivery by two loads per truck per day—average truck capacity
is about 22.5 tons. With the system, Pennsy was able to achieve that goal—an
additional 45 tons of product delivered per truck per day. Pay-back was
only calculated on Pennsy trucks, not the contract trucks that also benefited.
CES
Wireless Technology signaling box enables GPS tracking and allows truck
drivers to send ³canned² messages to central dispatch and to receive messages.
Other pay-backs included the benefit the additional hauling
capacity allotted to Pennsy’s own paving crews as well as the customers
it served.
“Toward the end of the 1995 season, I had drivers coming to me concerned
that our business had fallen off,” said Jenkins. “They were
used to working until 5 or 6:30 p.m., and suddenly they were parking at
3:30 and 4 p.m. They were actually delivering the same, or in most cases,
more product in less time.”
A long-term benefit has been the ability of Pennsy to lower its own fleet
of haul trucks and utilize more contract haulers, which was one of the
goals from the onset.
“We had 64 dump trucks and today we are operating 39 trucks. In the
same time, we doubled our volume of asphalt produced and increased stone
production by 15 to 20 percent. That’s about 25 less tri-axle trucks
we don’t have to maintain—and we didn’t calculate that
into the pay-back estimation,” said Jenkins.
Jenkins added that the company was sensitive to the drivers: “Early
on, we divulged that one of our long-term goals was to reduce our fleet
size. We also said that no driver needed to worry about layoffs as the
company would reduce the fleet size through attrition.”
Though Jenkins says it’s hard to estimate the technology investment
for the aggregate and asphalt haul trucks, as Pennsy bought the software
and equipment as a package with the concrete truck management system,
he puts it in the range of about $225,000 in 1995.
“There is no question that a significant portion of the savings was
purely the investment in radios. But, we also believe that the tracking
and signaling technology we invested in accounted for at least half or
more of the realized savings,” said Jenkins.
2001 system upgrade
Today, Pennsy Supply consists of five ready mix sites, six
asphalt sites, a sand operation, two high calcium limestone operations,
eight crushed stone operations centered around Harrisburg, Pa., plus a
newly acquired aggregate operation in Hazleton, Pa.
This past year, Pennsy decided it was time to upgrade from its DOS-based
fleet management system.
“The system was still working fairly well, but we started to have
reliability issues with the computer and the communication gear, plus
the status boxes were almost six years old, and they were starting to
have some problems,” said Jenkins.
When Command Alkon, the company recently formed by the merger of Command
Data and Alkon, announced that it was no longer providing modifications
or upgrades to the ControlAgg program, Pennsy management decided it was
time to upgrade the whole system.
Pennsy went with the Command Alkon CommandSeries system, which uses a
Windows-based server. The CommandSeries includes Command Aggregate, CommandConcrete
and CommandAsphalt.
Pennsy purchased the Command Concrete software to manage its ready-mix
fleet, and the CommandAggregate system to manage its aggregate and asphalt
fleets. CommandAsphalt is designed strictly for asphalt producers.
“Our aggregate and asphalt people work very closely together because
they are using the same trucks, so we don’t split the system by commodity
there,” said Jenkins.
In the new system, the status-only boxes were taken out of the trucks
and were replaced with boxes (costing about $1,300 each) that enabled
messaging and the use of a global positioning system (GPS).
Pennsy integrated the CommandSeries fleet management software system with
a signaling, messaging and GPS system made by CES Wireless Technology.
“We know the pay-back with this system won’t be as dramatic
as the first system,” said Jenkins. “By adding the GPS and messaging
capabilities to the system, we can generate additional savings which helps
to justify the investment better than if we stayed with just signaling
boxes.”
The CommandSeries Windows-based system allows a single dispatch operator
to keep track of several trucks at different sites on a single screen.
The program enables quick verification that the right trucks are going
to the right places. It also enables quick, efficient rerouting of trucks
to take care of unforeseen developments, such as traffic problems, truck
breakdowns, job delays, etc.
Adding the CES Wireless Technology system enables Pennsy to trace any
truck on the system, pinpoint its location in real-time and provide visual
recognition of the truck position on a computer map. A dispatcher can
quickly see from the computer screen where the truck is in relation to
the job site, nearest aggregate or asphalt plant and what roads and highways
are available to move the truck where it needs to be.
The messaging system enables Pennsy to program a series of “canned”
messages such as “yes”, “no” or “involved in
accident” that the truck operator can send to dispatch by simply
typing a number into a keypad on the truck’s signal box. Likewise,
dispatch also has a series of canned messages it can send back, such as
“stop at dispatch” or “dump it”. The system not only
expedites communications, but it greatly lowers unnecessary chatter on
the radio. Pennsy currently uses about 30-40 canned messages each for
truckers and dispatch. At this early stage, Pennsy is still adding and
dropping messages to see what works best. As of this writing, the company
will attempt to move exclusively to messaging and will pull the radios
from its contract haul track fleet.
With less radios, the number of canned messages will inevitably increase,
but Pennsy hopes the pay-off will be in improved efficiency. Dispatch
also has the capability to type in custom messages to the trucks as well
as use the canned ones.
In addition, the system can be used by human resources for truckers to
electronically punch-in and punch-out of work to further improve efficiencies.
Pennsy recently purchased an aggregate plant with a small fleet of on-road
trucks in Hazleton, Pa. At this site, the company is beta-testing a signaling
system using Nextel digital phones.
“You basically plug in a $30 Nextel phone into the truck’s cigarette
lighter, and you have voice and signaling capabilities."
Jenkins says that the start-up cost is dramatically cheaper, but factoring
in the cost of monthly charges, the cost of both systems equal out at
about four years in service.
“If you are starting like we did in 1995, with no radios in trucks,
the Nextel system is a slam dunk,” said Jenkins.
At this stage, Nextel has no messaging or GPS capabilities, but the company
told Pennsy that they are only 12 to 24 months away from adding these
type of capabilities. In addition, as part of the package, Nextel is developing
a tracking screen, a rudimentary version of the Command Alkon tracking
functions. Jenkins says the tracking system should work with the small
fleet at Hazleton, but it would be unworkable to handle a larger fleet
size or a large volume of orders.
At this point in technology, truck signaling, messaging and GPS systems
seem to be the battlegrounds for market share. Command Alkon also offers
its own signaling system. In addition, Command Alkon announced in a recent
newsletter that a partnership is being developed with Nextel.
However technology moves, Pennsy Supply can only see its efficiencies
going up.
Essentials of Plant Component Selection
By Ed Hayes
Editor’s Note: This is the fourth of a series of articles examining common plant problems and how they can be fixed.
When designing a new plant or an upgrade to an existing plant, aggregate producers have many options to consider for different sizes, models and manufacturers of equipment components. Many factors have to be analyzed in order to choose the best combination of equipment components to maximize production rates, minimize downtime, and as sales demands change, to produce the required tonnage of each and every product to specification gradation. To ensure that any plant system will operate efficiently, at capacity, with production and product flexibility and at the lowest possible cost-per-ton, every aggregate producer should perform a detailed analysis of the following issues.
Overburden Removal
Since raw-material deposits can vary extensively across their width and depth, a detailed mine plan and study is required to determine all of the short-term (one to five years) and long-term (five-plus years) mining and plant operational parameters. Overburden removal, mining and reclamation sequences need to be analyzed, as well as any changes in the raw-material characteristics such as compressive strength, toughness, hardness, friability, abrasiveness, and in addition, any materials that will have to be removed in the production process such as contamination from overburden and/or clay seams.
Market Needs
A detailed short-term and long-term market study is essential
to determine the potential market and sales projections for each specific
type and size of product, and the extent to which these sales will vary
over time. These percentages are important to determine the total tons-per-hour
of plant capacity required and especially the fine crushing and screening
capacity essential to meet the demand for the finer products. The tertiary
crushing circuit should be the limiting tons-per-hour production factor
for the total plant, so if the plant isn’t designed with the capacity
to produce the required market quantities of the finer products, it will
have to operate extra over-time hours or extra shifts at a considerably
higher cost-per-ton to meet the finer-product demands.
Crushing Options
Crushing equipment options should be examined very carefully because there are many manufacturers, sizes, models, configurations and wear-part options from which to choose. Since the crushing characteristics of various types of rock vary so greatly, representative raw-feed samples should be sent to crushing laboratories to replicate and predict actual crushing plant results. In addition, there are many positive and negative factors to consider when choosing impact crushing versus compression crushing in any primary, secondary or tertiary crushing circuit. In most applications, the following comparisons are valid, however actual short-term versus long-term cost comparisons have to be analyzed on a per-circuit basis:
- Compression crushers cost more than impact crushers for the same tons-per-hour capacity.
- Impact crushers produce a higher ratio of reduction than compression crushers, which often reduces the number of crushers required for the total plant system.
- Wear-part costs and change-out time and labor increases significantly for impact crushers as compared to compression crushers as the abrasiveness of the feed material increases. The savings realized from the lower investment in an impact crusher can often be lost in wear-part costs and change-out downtime and labor in a matter of weeks or months.
Foundations, weights, dimensions, dynamic forces, horsepower and feeding equipment required can also vary considerably for the different types and sizes of crushers. These capital and operational costs also have to be carefully factored into the total crushing circuit costs in addition to the cost of the crusher itself.
Screening Options
Screening equipment options should be examined very carefully
because there are many manufacturers, sizes, models, configurations and
wear-media options from which to choose. In addition, there are many positive
and negative factors to consider when choosing a horizontal screen versus
an inclined screen for any application:
- Horizontal screens are more expensive and draw slightly higher horsepower than inclined screens.
- Horizontal screens require less height, reduced steel requirements for structure and chute work, and a shorter feed conveyor with less horsepower.
- Most horizontal screens have adjustment features to easily change
RPMs, angle of stroke and amplitude of stroke for more efficient screening,
especially in the finer-size ranges.
- Horizontal screen media change-out is faster, easier and safer than
a 20°-slope.
- Horizontal screen media openings are true openings when viewed vertically as compared to inclined screen media openings, which “lose” a percentage of their opening dimension on the 20°-slope.
- A horizontal screen does not rely on gravity to assist the particles across the deck. Gravity can accelerate particles across a lightly loaded inclined screen deck, drastically reducing the number of openings the near-size particles have to fall through, resulting in increased near-size particle carryover. This carryover can contaminate products or send products back into a crushing circuit instead of a stockpile.
Open-area for each screen deck has to be analyzed very carefully. Computer simulations used to determine the size of a screen during the plant design process don’t always properly calculate the required screening area because they don’t adequately allow for the loss of each screen deck open-area from clamp bars, bucker rubber strips, center hold-down strips and the substitution of screening media with less open-area. Substitution of heavier screening media can reduce open area up to 30 percent, and certain urethane and rubber media can reduce the open area up to 50 percent. Therefore, any screen that was marginally sized in the computer simulation will be seriously undersized in actual production with even a 10-percent to 20-percent loss of open-area. This loss of screening efficiency will send undersize material back to crushers that should be sent to stockpiles or it will send undersize material to product stockpiles contaminating the product.
Other Design Considerations
- The primary feed hopper should be designed with enough live capacity so there is always a sufficient bed of protective material covering the feeder when the haul truck dumps. In addition, the feed hopper should also be sized for the additional live capacity required if higher-capacity haul trucks are planned for the future.
- The vibrating grizzly feeder should be sized not only for adequate feeding capacity, it should include a grizzly section of adequate open area to allow most of the raw feed, which is smaller than the closed-side setting of the jaw crusher, to bypass the jaw crusher. This finer material, if fed directly into the jaw crusher, reduces its crushing capacity, decreases the crushing chamber voids, increases horsepower draw and increases jaw liner wear.
- Conveyor designs should be checked carefully to make sure that adequate safety margins are included in all components to allow for any temporary overload. In addition, the motor horsepower and drive assembly should not only be adequate to operate at full design tonnage, but also to have the capability to restart the fully loaded conveyor after a shutdown or power failure. This is certainly more economical than sending a crew of men with the resultant plant downtime to shovel off the conveyor (or conveyors) so it (they) can be restarted.
- A fractionated plant should be seriously considered by producers located near state lines or near customers where gradation specifications may vary considerably. In many market areas, producers can quickly recover the added investment for a fractionated plant because they can acquire a considerably larger market share when they have the ability to blend basic sizes into a multitude of specification products on a load-by-load basis.
Obviously, there are many other factors that have to be
considered in any plant design that can’t be covered in a brief article.
It is important for any producer to carefully study all areas of any plant
design and to seek opinions from his operations personnel and management
team. If issues and questions remain unresolved, the producer should seek
outside opinions from unbiased sources. This is more prudent and economical
than proceeding to build a plant or system that fails to meet production
and cost-per-ton expectations.
Ed Hayes is the president of H&H Solutions, Inc., in Gettysburg, Pa.
A Systematic Approach to Seal Failure Analysis
Editor’s Note: This monthly column is supplied exclusively for AggMan by The Equipment Maintenance Council (EMC).
The most common type of dynamic seal in use today is the
oil seal or rotary shaft seal, and, while its initial cost is minimal,
its impact on maintenance, time and labor can be significant. As many
maintenance managers have experienced, an early seal failure can derail
even the best maintenance program.
Whenever fluid is found leaking from a vehicle or a piece of machinery,
the programmed response is to question the integrity of the seal. Initially,
fluid leakage might seem like an easy problem to correct; seals are inexpensive
and simple to replace. Sometimes seal replacement may be the answer to
fluid leakage because the seal has reached its maximum life expectancy.
Often, however, replacing the seal does not eliminate any problems, but
rather raises more maintenance questions.
For example, what if a newer unit is leaking fluid? What if leakage continues
after the seal is replaced? Seals on similar operating units have not
failed, so why did this one? Perhaps the problem is not a bad seal. All
these questions and many more need to be addressed and answered to solve
a persistent fluid leak.
A Little Detective Work
Equipment managers typically wear more than one hat in their
daily job. They often function as a manager—handling items such as
human resources, budgeting and cost management—and might often pitch
in to help operators, technicians or maintenance personnel complete various
jobs. According to Joseph Lang, director of advance quality planning for
SKF/Chicago Rawhide in Elgin, Ill., when diagnosing fluid leaks, the equipment
manager needs to don yet another hat—that of “detective.”
“As with all good detective work, the time to begin accumulating
clues is before anyone has disturbed the evidence,” he said. “When
a leak has been detected, there is a vast amount of information that can
be garnered by simply surveying the equipment and its environment. Assessing
the amount of dirt, grit or frequency of washings can tell a manager a
great deal about the possible causes of failure.”
Next, applying a systematic, step-by-step analysis process will reveal
more possible causes for a leak. Some clues might be subtle, while others
might be quite obvious. Armed with this information and several key analysis
tools, a maintenance manager should be able to unravel the mystery of
the fluid leakage problem. Once a true root cause has been identified,
corrective action typically is easy.
In order to properly assess a system leak, remember that the seal is only
one of six components within the sealing system. To resolve a leaking
problem, it is necessary to focus on the entire sealing system, rather
than simply the seal itself.
The sealing system consists of the following six components:
- Rotary lip seal;
- Housing;
- Shaft;
- Lubricant;
- Internal environment; and
- External environment.
Any one of these components could be causing a leak in the sealing system. Taken individually, each of these components might seem fairly simple and straightforward, but in reality, the sealing system and the interactions between these components are deceptively complex. An equipment manager might actually find more than one root cause to a system leak.
Gather Critical Information
Just like any other component failure, information is the key to diagnosing the problem. It is important to gather as much data as possible in the following four areas:
- Operating environment—Try to determine the operating conditions at the time of failure.
- Time of failure—Record the time of failure in hours or miles. There is a different analysis method for short-term failures and long-term failures.
- Type of leak—Note where the leak is originating. Is it coming from the inner diameter of the seal lip and shaft interface or is it originating from the outer diameter, where the seal outer case presses into the housing bore? Some seals may be reversed as they press onto the shaft and the seal lip contacts the housing bore surface. Describe the leak as static, dynamic or both. A static leak continues when the unit is not running. A dynamic leak occurs only during operation. It is possible to have both a static and dynamic leak.
- Severity of leak—Record whether the leak is intermittent or consistent in trace or heavy amounts.
Once an equipment manager has gathered this information,
several resources are available to help diagnose the seal problem. An
invaluable analysis tool is the Rubber Manufacturers Association (RMA)
OS-17 System Analysis Guide. This guide provides detailed analysis procedures
and also documents many known failure modes. It gives some corrective
actions and is the standard for establishing seal performance. Chicago
Rawhide uses it regularly for training. For more information on this publication,
visit the RMA web site at www.rma.org.
A video camera can be helpful in solving seal failures. If possible, videotape
the entire unit to record its condition at the time of the leak. This
helps to visually identify the unit and its operating environment. Lang
said customers have sent videos to him in the past requesting help in
diagnosing seal problems.
If a video camera is not available, the next best diagnostic tool is an
analysis diagram. Lang suggests simply drawing the sealing system components
and using a color-coded key to show where the leak is occurring and other
pertinent pieces of data such as where dirt and sludge have formed. Remember
to include the unit number, location and hours/miles at the time of failure.
Ultraviolet dye can be beneficial in determining the exact origin of a
system leak. UV dye is placed into the operating fluid and the unit is
then run for a short period of time. The unit can be inspected with a
black light to pinpoint leaks.
And finally, an ultrasonic leak detector can identify very slight air
leakage paths, which would pinpoint areas of possible fluid leaks.
Lang suggested remembering three things when a unit is leaking fluid:
- Keep in mind that the seal is only one part of the entire sealing system.
- Take a systematic approach to determining the proper root cause or causes for system failure.
- Take advantage of the diagnostic tools available to you. The RMA Guide, visual identification and UV dye can be worthwhile tools when assessing complicated leaks.
Editor’s Note: Portions of this article were taken from material copyrighted by SKF Industries, with permission.
The Equipment Maintenance Council (EMC) is an individual membership organization comprised of equipment maintenance professionals. Its members are responsible for the purchase, maintenance, employee training, shop facilities, and parts management of leading corporations and government entities that utilize heavy, off-road equipment. Its members also represent the major manufacturers and suppliers of the heavy equipment industry. EMC provides end users with cutting-edge education, and it is the only organization to offer a certification program for the industry, the Certified Equipment Manager (CEM). For more information, contact Stan Orr, CAE, EMC executive director, at (970) 384-0510, e-mail at ceo@equipment.org, or visit EMC’s web site at www.equipment.org.
Gravitas Superabit--Roof Control in Stone Mines, Primarily with Roof Bolts
By Jack Parker
Why discuss roof control? Because we have talked about going underground
in general terms, about roof design and pillar design factors, and the
next topic I usually run into is “How do we control the roof?”
For producers who have not worked underground before, this can be a new
problem as well as a significant cost.
Why is roof control primarily achieved by use of roof bolts? Because in
stone mines, with a product selling for only a few dollars per ton, we
cannot afford the time, money or space needed to install timbers, steel
sets or other systems which might be used in coal or metal mines.
A Philoisophical Observation
Sad, it is, but true, that most of our roof-control problems could be
diagnosed most can be corrected by the men already on the job. The clues
are in front of our noses, but we are reluctant to believe them and, perhaps,
even to see them. We have come to rely on “experts,” who, if
they are any good, have seen these clues often enough to believe what
they see.
Most of the important observations and interpretations have been written
and published, often more than once. Every generation produces a bright
young person who rediscovers the obvious, and his light shines for a while—but
will be forgotten. Then, as a wise man said: “History is bound to
repeat itself.” And we pay again.
Forgive me if I quote again from our friend Winston, who wrote: “Man
will occasionally stumble over the truth, but most of the time he will
pick himself up and continue on…”
Isaac Newton observed the fact that apples always seem to fall downward
and uncovered the law of gravity. H.G. Wells went a step further and invented
anti-gravity paint, which enabled his first men on the moon to get there
without a rocket assist. When I retire soon, I plan on marketing that
anti-gravity paint in five-gallon pails, to be sprayed on the roof and
so eliminate the need for supports.
End of philosophical observations. Let us now look at the realities of
roof control problems and remedies.
First, Define the Problem
The phone rings. After the preliminaries, a gentleman on the other end
says, “We gotta problem. A big fall of roof last night at our Mine
A.” That defines the problem as he first sees it. Gravity will prevail.
(“Gravitas superabit” is how Julius Caesar, a gentle fellow,
said it when he fell off the couch on the third day of his inaugural banquet).
OK, so we talk about the problem, and the prospect of retiring and going
fishing recedes like a mirage in the thirsty desert. But it’s interesting
work, so let’s get on with it.
“Can you come and look at it?” he asks. That, you may remember,
is the first and great requirement—to go look at it. To observe,
eyes and ears open, mouth shut. The general idea is to be there with an
open mind, well rested and ready to soak up all the information your senses
can feed into that old-fashioned computer of yours.
Eyes, ears, touch, even taste and smell all contribute. Those bits of
information will ferment and be sorted out upstairs as moments, days or
weeks go by (your necktop computer is processing them). We try not to
draw conclusions on the spot. We must look for patterns of roof behavior
and for departures from patterns—so we ask to see more than this
one roof fall. Ask to see best, average and worst conditions. Look at
all of the roof falls because the clues are in the failures.
We are trying to define the problem in detail, because a remedy will then
probably be obvious.
Example: Why is my front right tire flat? Aha! I see a nail in it…then
take out nail and plug hole. But if the pattern persists, we look further
in order to find out how the nails got there.
Poem (I forget who wrote it, but it means a lot to me):
I had six honest serving men.
They taught me all I knew.
Their names were What? and Why? and When?
And Where? and How? and Who?
So ask those questions. Odds are that the man on the job has more of
this kind of information than the boss. Sometimes I get paid for telling
the boss what the scaling crew has told me. But he never asks. We can
all learn from each other.
The open mind is extremely important. A real problem in this business
is that experts develop pet theories and do their best to apply them to
all situations. Like your friend who listens to the TV doctor in the morning
talking about clinical depression becoming prevalent, so he instantly
detects the symptoms in you and me both by evening!
I find that it is not good to discuss a problem much before going to look
at it, or the remedy might be preconceived. Better to check into the motel
early the night before, eat a light supper, then watch some dumb TV until
bedtime. Monty Python works for me, or Benny Hill. Nothing too heavy.
Here is an example of an erroneous preconceived conclusion, at Mine B.
An expert, a nice fellow, had learned correctly that many roof control
problems could be attributed to excessively high horizontal stresses built
into the rock—so he designed mine layouts and supports to combat
those conditions. But that didn’t seem to help, so we took a walk
through the mine with engineers and geologists—all good fellows,
all looking and thinking to some degree.
What did we see? Clearly, when we got around to actually looking and
thinking, the geological structures in this part of the basin were all
related to slumping: mud sliding into the basin (as shown in Figure 1),
being pulled apart—not compressed. Tension, not compression, was
the problem.
So the preconceived diagnosis led to wrong prescriptions, and they were
generally accepted because they came from an expert. I could add that
wooden posts were not able to support the roof either—because they
were shoving down into a wet shale floor instead. That possibility, apparently,
was not in the book.
End of digression, now back to Mine A, still on the phone.
Q: Is this the only roof fall?
A: No sir. We have quite a few every year, but this was in a bad place. (The polite “sir” address suggests a Southern origin).
Q: Do they occur at any special times of the year, maybe spring and summer?
A: Yes sir, most of them do.
Q: When humidity is high? The mine is foggy?
The rocks are wet?
A: Yes sir, most of ’em.
And already we have a fair idea of what is wrong. We relate roof failures
to seasonal high humidity and condensation, which affect shales and siltstones
adversely, often severely enough to cause failure.
Shales may revert to their original state—which was mud.
Siltstones soak up moisture by capillary action—and expand significantly.
The outer, exposed surface expands first, before the inner mass, so it
will probably spall off. Given time the moisture will go deeper. That
rock will be confined so the attempt to expand will create an internal
compressive stress, hundreds or thousands of psi, often enough to cause
roof failure, sometimes violent and noisy.
But this is not news.
Charlie Holland wrote about the seasonal effect in coal mines and the
U.S. Bureau of Mines published findings on the expansion and internal
stresses, both in the middle of the last century. No doubt others have
done so too, and yet roof support systems are still designed without recognizing
this basic problem. It, and others like it, should have been taught in
all courses in roof control, in college and out. Roof bolts are prescribed
like aspirin by a tired M.D., whether to cure a headache or a broken leg.
As we walk around doing the diagnostic work (please note: specifically
not driving around in a pickup truck at 30 mph with a spotlight), we may
see that “weathering” has caused the roof to fritter away at
the bolt plates, leaving gaps between plates and roof. Diagnosis might
read that the roof fell because point-anchored bolts were no longer confining
it or, if the roof did not fall, then the diagnosis could be that bolts
were not doing much work anyway—right? (See Figure 2.)
The best remedy would be to stop the weathering, maybe by coating the
rock or, better yet, by controlling the humidity—the root cause.
A partial response is to switch to fully grouted bolts, cement or resin,
which would be partially effective even if a gap develops between plate
and rock.
(Author’s Note: Some folks think that bolts that are fully grouted
do not need bearing plates, but where that idea has been tried, the bolts
often disappear slowly up the hole as the rock and/or the grout yields
a little. Plan to use plates.)
We haven’t finished with this moisture factor yet and probably never
will understand it fully. So hear this:
At Mine C, shallow, beneath wet farm fields, we stood watching the roof
come apart in a mysterious manner—not on bedding planes or joints,
but with curved surfaces in a massive rock—and a light came on inside
my hard hat. When a large chunk fell, the newly exposed surfaces were
dark, black and wet, but in a few minutes they became light gray and dry.
Apparently, the rocks were originally in a wet expanded state, but the
ventilating air blown across the top of the mining machine was drying
it, shrinking it and causing it to fall apart…the opposite of the
more usual wetting process of spring and summer, more to be expected in
the dry season—winter. Close inspection of the roof elsewhere in
the mine detected shrinkage cracks in the rock, much as you might find
in drying concrete. File that bit of experience in your necktop.
Remember that we are still talking about defining the problem before designing
the fix. You might tell the man at the emergency room that you think you
have a broken leg, but you wouldn’t want him to wrap it in a cast
without checking first, would you?
Here’s another example: At Mine D, in desert country, MSHA had threatened
to close the mine because of a rash of significant roof falls. I was called
in to look at “ineffective resin bolts” in the early days of
resin bolting.
So we crawled around and over the top of many falls and did indeed find
some resin problems. In places, the holes were too big—so the resin
was not completely mixed. In places, resin had squirted into preexisting
separations in the roof, instead of anchoring the bolts.
During the inquisition, however, miners reported incidentally that for
days “water had been squeezing out of the roof.”
Another “Aha” moment. The roof had become abnormally wet! We
checked the weather records at the local (desert) airport and found that
the roof falls all occurred on those days when the humidity was abnormally
high—basically when the dew point of the intake air was higher than
the temperature of the rock in the mine—so leading to condensation
like that on the cold water pipes in your bathroom. It was a good fit
(see Figure 3) and, as you might expect, it was confirmed by the fact
that the falls were on the intake (fresh air) side of the mine.
The rest of the story, as told to me by the engineer a couple of years
later, was that the humidity problem was solved—by selling the mine.
Other cures have been less drastic. A gypsum mine with siltstone in the
roof confined its production efforts to the winter. Some mines use sealants,
with varying degrees of success. Some use sprayed-on urethane foam, which
helps eliminate the problem by insulating the rock surface, thus preventing
condensation. Some dehumidify locally, as in a warm, dry shop or lunch
room. On the other hand, some producers have worse problems where temperature
is too high and drying, as over diesel engines and compressor sites, so
they cool them.
A very neat idea, which I saw first at a Sahara coal mine, was to set
panels of old workings aside as tempering chambers. Intake air was coursed
through one of them before being delivered to active mining areas. The
effect was obvious, the roof broke up and fell in the chambers, but behaved
very well in the active workings. When one tempering chamber became too
badly plugged, a new one was set up.
If you take a walk around your mine, including the old workings, you will
probably see a variety of roof conditions—in dry areas, flooded areas
and in areas which are cycled wet and dry seasonally, particularly if
your roof, pillars and/or floor include some shales and siltstones. You
may want to control those factors and improve conditions before buying
your roof bolts.
Blast damage is another all-too-common ailment which should be controlled
before you design a roof-support system.
Climb up on the muck piles and see how the cracks radiating from top blast
holes go up into the roof. Look at bootlegs in the face and see how cracks
radiating from them extend about 30 hole diameters from the hole—indicating,
for example, that a 2-in. hole filled with ANFO can be expected to damage
the roof to a depth of about 5 ft.!
Observe how (again at bootlegs in the face) when two or more holes are
detonated simultaneously, cracks propagate not only from hole to hole,
but also beyond the outer holes—so a V-cut drilled too close to the
roof can send V-cracks up into the roof. And don’t forget that at
the end of the V-cut the two rows of holes are close, or touching, so
the rock there gets a double-dose of “domestic” violence.
Switch to gentler blasting near the roofline and many roof-control problems
will go away. In a new operation, it pays to seek out a good parting in
the rock, which will provide a smooth roof without much help from the
explosives. You may have to go out of your way a little to secure that
long-term benefit.
Always keep in mind the tremendous amount of energy available in a small
amount of chemical explosives. I do. Last week, a grandson built a rocket
car with rocket fuel concocted in the kitchen—ground charcoal, oxidizer
from welding class and other stuff. The fuel was packed into a tube and
taped to the car. The fuse was yarn soaked in sugar solution, then dried.
Countdown began over the phone from Mission Control in Milwaukee. “Now
I’m lighting the fuse…Whoomp! Ooooh shoot! It blew the wheels
off!” A literature search then told us that the fuel should be in
the form of a hollow tube for a controlled slow burn with moderate acceleration.
Fuel packed solid into the back of the car accelerated it like a bullet.
Now he understands. Me too.
Are your pillars too big? Some folks have trouble believing that this
could be a problem—but not a few mines have pillars so big and so
stiff that they punch into roof and/or floor, thus actually causing the
very problems they were meant to prevent. It is more likely to happen
where the roof or floor rocks are softer than the pillar rock, or where
the immediate roof or floor slab is thin, and especially where the load
on pillars is high because of a high extraction ratio and/or significantly
great mining depth.
The problem can be anticipated where those factors are present, and it
usually shows up first as crushing of the roof rock at the juncture of
roof and pillars (see Figure 4).
In class, I could demonstrate the possibility by setting a sheet of drywall
on a large number of house bricks on a flat floor, then heaping the top
with a heavy load of sand. The drywall is the roof rock; the bricks are
the rock to be mined. Now extract the bricks a few at a time. No problem
at first, but there comes a point at which the load on the remaining “pillars”
becomes so great that they punch up through the drywall “roof”.
Normally, we should avoid that situation by not extracting so much rock.
Another way to handle it is to make the pillars smaller, so that they
will not be as stiff. They are then referred to as yielding pillars, and
the idea can work under the right conditions—but the operation is
delicate, not unlike walking a tightrope over Niagara Falls.
If you tried yielding pillars on a production scale and had a major mine
failure, you would not be the first…Never forget that.
Which brings to mind an interesting medical term currently in vogue: “prudent
avoidance”.
The Problem of Residual Stress
With a bit of help we could write a book on this topic, but for present
purposes, let us look only at the highlights.
In some rocks, high stresses are locked in as you might find in a mass
of pre-stressed concrete. They were built into the system at some time,
or some times, in the history of the formation and deformation of the
rocks. We have inherited them. They are commonly called “residual
stresses”.
Some rocks are more likely to store high stresses than others because
of their makeup. Just as a steel spring stores more stress than a wooden
spring and wood stores more stress than a marshmallow—so, in general,
a hard, dense, brittle rock is more likely to be storing stress than is
a soft rock.
Note that two such rocks may occur together and be stressed differently.
For example, in a thick bed of sandstone, there may be some layers that
are cemented by silica (quartz) and some layers cemented by calcite (calcium
carbonate). Then it is not unlikely that a roof formed in the calcite-cemented
sandstone will behave quite well—whereas a roof formed in the adjacent
silica-cemented rock will be storing high stresses and will cause the
structure to fail—even though that rock would appear to be more resistant
to failure in standard lab tests. It may be a good idea to shift the roof
into the “weaker” rock. Think about it.
Recently, I heard a driller say the same thing in different words: “Over
there the roof is a lot harder (drilling), but over here it is softer,
and maybe it can ‘give’ a little, instead of breaking.”
Out of the mouths of this and other drillers shall come forth great wisdom.
Maybe roof bolts are not the answer.
A simple clue comes from the noise a rock makes when struck, by a screwdriver,
the back of a knife or anything similarly hard. A soft rock says, “thud”,
a hard brittle rock says, “chink” and intermediate rocks give
intermediate responses. If you climb over the pile of fallen rock and
it sounds like broken china, the odds are that you are dealing with a
highly stressed rock.
The stress field is usually directional, i.e., the stresses are usually
higher in one particular direction than in the others. For example, in
much of the United States the maximum stress is oriented southwest-northeast.
The implication is that rooms driven in one direction will behave worse
than the others—and that’s the way it is.
Thus in much of the United States, rooms driven either northeast or southwest
behave better than the others. Conversely, those driven northwest or southeast
are most likely to fail. Think of it like this: the rooms are canoes in
a stream. When paddled either upstream or downstream, their shape is “streamlined”
and the canoes behave fairly well, but when paddled across the current,
it is more difficult to control them. The current (force) is pushing on
a much larger area. A few years ago, when I used this word-picture to
illustrate the roof-control problem to a mine manager he said “Are
you accusing me of paddling my canoe broadside to the current all these
years?” Rest in peace, John, you got it right.
If you look at a mine map and notice that most of the roof falls are at
intersections (which are the widest spans), either they were driven too
wide or the magnitude of the horizontal stresses is low. At the other
extreme, where horizontal stresses are high, most of the roof falls will
be oriented in the same direction—perhaps elongated in the northwest-southeast
rooms. Usually, they will be in the narrowest rooms—a strong hint
that wider rooms may be more stable than narrow, which is often the case.
Sometimes we can estimate the approximate stress levels by looking at
the geological structure as follows:
- If the rock mass is badly shattered by joints and faults, the odds are fairly good that much of the ancient stress has been relieved.
- If there is one distinct set of tension joints, vertical, straight,
parallel to each other or maybe mineralized, then there is probably
one high stress, oriented parallel to those joints.
- If there is more than one set of tension joints, maybe where the rocks
have been stretched in two directions on top of a geological dome, then
we have a real problem: no horizontal stress to confine and support
the pieces of rock, which will probably fall like rubble. Dangerous.
- If the mine property covers a large area, say a square mile or two, and if it includes a variety of faults (fractures on which there has been displacement) such as normal faults, reverse faults and shears, then the stresses in the rock will likewise vary considerably in magnitude and direction.
We seem to be digressing again. But that’s OK, all of this discussion is to make the point again: We should define the roof control problem and correct it before implementing a roof bolting program. But, now it’s time to go. Next month, we’ll be back on track choosing a roof bolting system.
Jack Parker is the owner of Jack Parker and Associates. In reality, JP has downsized, and the “Associates” are Jack’s wife, Levinia, and her small dog, Dulcie.