Minimizing Raw Material Losses at Float Glass Manufacturing Facility

[DumpmeAdrenaline]

Hi Fellow Engineers

I have been tasked with a project to minimize raw material losses during handling and processing at our float glass manufacturing facility. My current focus is on minimizing sand, dolomite and limestone losses since those are the three materials that undergo separation to the required grain size before they are transferred to their respective silos.

Through multiple inspections of the units I identified the following potential sources of sand loss: unloading of raw materials from trucks to stockpiles and from the bucket loader to hoppers, leaks from compressed air systems used for dust extraction and control , spillage at loading points where sand is conveyed between troughed belt conveyors through chutes or from chutes onto vibration screens or hoppers, losses from vibration screens, and fine and coarse sand particles collected in waste bins from vibration screens.

Quantifying intangible losses possess a challenge, any ideas on how one one can estimate losses from activities such as unloading to piles and hoppers, conveyance, and dust control?

 

[Baluncore]

What goes up, must come down.
Use several plastic trays to catch and weigh product that falls from, or blows away from, the handling equipment or stockpiles.

Distribute the trays in various directions and distances, record the wind speed and direction during each sampling period.

Make sure you have some control sites upwind, to see what else might be falling.

 

[DumpmeAdrenaline]

What goes up, must come down.
Use several plastic trays to catch and weigh product that falls from, or blows away from, the handling equipment or stockpiles.

Distribute the trays in various directions and distances, record the wind speed and direction during each sampling period.

Make sure you have some control sites upwind, to see what else might be falling.

I have attached information on the ensiling of sand, limestone, and dolomite at the facility. It is worth noting that the use of hammer and jaw crushers are no longer necessary as the raw materials are now supplied within the required specifications. Dust control systems are installed at loading or transfer points where the raw material is poured from one belt conveyor to another, hoppers to belt conveyors, screw, or bucket elevator (vice versa) and covers are placed on outdoor belt conveyors.

To provide a more comprehensive understanding, I have also included some pictures of the facility. In terms of quantifying the losses, a simple mass balance can be used: Input=Raw material discharged from scales in the silos+ loading point losses+ Screening losses (fine and coarse particles) + Dust leakage losses+ unloading of raw material into stockpiles from trucks.
The Raw material discharged can be tracked by load cells under each silo (in addition to the number of batch prepared), screening losses can be quantified by measuring the weight of waste bins collecting the fine and coarse particles.

I am not sure whether to include the last term in the mass balance or not. Sounds like a practical solution to place plastic trays or open top containers near the loading points to collect the spilled raw material and account for the loading point losses and dust leakage losses in the mass balance equation but I see one issue with this approach the consumption of raw materials in making float glass amounts to around (350 Ton/day).

Thank you Baluncore

 

[Baluncore]

I see one issue with this approach the consumption of raw materials in making float glass amounts to around (350 Ton/day).

Everything can be weighed in bulk, unless it blows away, enters the soil or water.

When you are processing 350 long/short/metric ton per day, any significant loss from the process will show up quickly. A loss of 1% would be 3.5 Ton, which I think someone should notice. With losses less than 1%, the mass of glass produced does not need to be known precisely.

The tray sampling of windblown material with controls will demonstrate your environmental credentials. If you are ever wrongly accused, that will point the finger in the direction of the real source of pollution.

There is a hidden loss of material that goes up the flue from any furnace, that includes minor components such as selenium or germanium, that may not be filtered. What minor materials are added to improve the glass quality? How much of that is lost before it reaches the product?

Analyse the glass produced to find the exact composition, including trace elements. Audit the elemental mass of inputs and outputs to identify the losses. An unimportant contaminant in a feedstock can be used to trace that stock material into the glass produced.

First identify the losses, then plug the holes.
If you cannot measure it, you cannot manage it.

 

[DumpmeAdrenaline]

Everything can be weighed in bulk, unless it blows away, enters the soil or water.

When you are processing 350 long/short/metric ton per day, any significant loss from the process will show up quickly. A loss of 1% would be 3.5 Ton, which I think someone should notice. With losses less than 1%, the mass of glass produced does not need to be known precisely.

The tray sampling of windblown material with controls will demonstrate your environmental credentials. If you are ever wrongly accused, that will point the finger in the direction of the real source of pollution.

There is a hidden loss of material that goes up the flue from any furnace, that includes minor components such as selenium or germanium, that may not be filtered. What minor materials are added to improve the glass quality? How much of that is lost before it reaches the product?

Analyse the glass produced to find the exact composition, including trace elements. Audit the elemental mass of inputs and outputs to identify the losses. An unimportant contaminant in a feedstock can be used to trace that stock material into the glass produced.

First identify the losses, then plug the holes.
If you cannot measure it, you cannot manage it.

The primary objective of this project is to identify and minimize raw material losses during handling then implement appropriate solutions based on the identified potential source of loss. For example, one identified clear area of improvement to work on are the leaks from the dust control systems as seen from the pictures posted.

Currently, there is no tracking of the raw materials (received in trucks which are sand, dolomite and limestone) during its unloading, conveying, and screening to the silos. Furthermore, the silos are equipped with high-level and low-level sensors that signal the conveyance and dust control systems to operate when the material level is below that high level so the exact amount of raw material in each silo is not known. The discharge from the scales under each silo according to the batch recipe is the point where the quantities of raw materials are determined and monitored throughout the processing of the batch (discharge from raw material scales->Batch mixing with water -> Batch charging to furnace) to form float glass.

An ECS system is utilized to remove various pollutants including dust, CO2, SOx and NOx pollutants from the flue gas generated by the furnace before releasing into the atmosphere. It is worth noting that processing losses account for approximately 20% due to the release of CO2 by Na, Mg, and Ca carbonates and free or bound water in the batch charged to the furnace.

We don't add selenium or germanium instead we add iron and carbon. I haven't experienced any contaminant issues from the chemical analysis performed daily on the glass produced or the raw materials received except for iron oxide content which influences the required optical properties in glass.

 

[Tom.G]

Many years ago I did a computerized warehouse system for sugar and corn syrup, and their mixtures. The volumes were not on your scale, but the 16 storage silos were 40 feet tall and 14 feet diameter, around 500 000 pounds of product in each as I recall. Incoming product was typically by railroad and outgoing mixtures were by tanker truck.

The liquid stuff was easy, just a pressure gauge at the bottom of the silo.

For mixing, the mix tanks were mounted on a scale, implemented as strain gauges on their mounting structure.

The granulated sugar silos were a bit more involved, we couldn't easily (and reliably) mount the silos on a scale. I don't recall the exact details, but here is the general idea:

The sugar traveled from a rail car thru an auger to a bucket elevator, which delivered the sugar to the top of the silo (like a grain elevator does).
The elevator dumped the sugar onto (or at?) a sloped platform scale (electronic of course). One problem was that sometimes some sugar would stick to the scale. Fortunately, a bucket elevator is intermittent, so to commpensate, the scale was re-zeroed between bucket loads.

The end product was dispensed thru mass flow meters to the customers truck(s). (Mass Flow Meter - based on coriolis force thru an ultra-sonically excited U-shaped quartz tube... black magic as far as I was concerned!)

Well, I hope this gives you some ideas. If your granular stuff is stored in silos, you could probably use a pressure sensor at the bottom with a Very large diaphragm to accomodate the large particle sizes.

Cheers,
Tom

p.s. Full control system design, layout, wiring, drawings, programming took about 3 years.