- To participate in the 911Metallurgist Forums, be sure to JOIN & LOGIN
- Use Add New Topic to ask a New Question/Discussion about Mineral Processing or Laboratory Work.
- OR Select a Topic that Interests you.
- Use Add Reply = to Reply/Participate in a Topic/Discussion (most frequent).
Using Add Reply allows you to Attach Images or PDF files and provide a more complete input. - Use Add Comment = to comment on someone else’s Reply in an already active Topic/Discussion.
Process Retention Time Estimate Evaluation (19 replies and 2 comments)
Retention time is a much used, but often understood concept. For some processes (grinding, flotation and leaching) it can be a very valuable factor, but in others (crushing, screening, and gravity separation) it means much less. Retention time is also highly dependent on the ore characteristics, and can and will vary over time. In all cases it is dependent on feed characteristics and circuit configuration. If it is expressed it usually is mentioned as the average retention time, acknowledging that some a Particles pop right through and others hang around.
The average time for concentrate will normally be much less than for tails. For processes where the tails are the smaller component it may be the revers. Although I saw one operation, phosphate, where the concentrate was depressed and the tails floated leading to the reverse. And for coal the retention time for the coal is usually greater than for the tails.
For grinding, flotation, and leaching it is determined by laboratory studies as the amount of processing time needed to achieve a certain particle size (grinding) or achieve a certain concentrate grade and recovery (flotation and leaching). It is used to determine the volume of processing equipment needed to achieve those goals. And for these processes is usually in the range of several minute to hours (leaching).
For crushing, screening, and gravity separation, as the processes move at a faster rate (seconds) it has some meaning, but is often overshadowed by sizing constraints based on particle size and shape.
Excluding stockpiles and surge bins you and not considering a complicated flotation circuit (i.e rougher scavenger and not rough- cleaner - recleaner with a regrind circuit - scavenger) you are probably looking at 30 to 60 minutes for SAG mill, 45 to 60 minutes for ball mill, and then one to 3 hours for flotation, and one to 3 hours for thickening/filtering, for a total of 3 to 6 hours which could get to 8 hours. Of which 1 to 3 hours for the concentrate and 3 to 6+ for the tails (less if you go to ponds or use high rate thickeners with filters).
The retention time during reverse flotation of phosphatic ores are suppose to be very less having calcite or dolomite as the main impurities. The acidic pH in which phosphates are depressed and allowing calcite or dolomite to be floated gives very less retention time for the particles in the Cell.
Retention time = Static volumetric capacity / volume flow. For things like screens, air beds, conveyor belts, etc., volume is based on bed depth.
Before using a radioactive material, I would try something less dangerous like fluorescence dye. This turns bright green in water and glows very bright green under ultraviolet light.
At a taconite (low grade iron ore) plant I worked at we were able to follow concentrate silica spikes through the process all the way from mine indicated silica dumped into the primary crusher to final pellet silica. This is what we came up with, pretty consistently, in a plant with relatively small in process inventory (stockpiles, bins, slurry tanks, etc.) compared to the volume of ore processed (60,000 tons of in line storage for 8000 tons per hour of plant feed).
The silica spike peaks tended to be very sharp at the crusher, and "spread out" as they went through the process.
Blasting to crusher feed took anywhere from several days to several weeks.
Three stages of crushing with the last stage in closed circuit took about 6 hours, including bin time.
The feed bins ahead of the Concentrator took 6 to 8 hours and were, surprisingly, plug flow.
The Concentrator took about 1 to 2 hours.
The slurry storage tanks at the Pellet Plant took about 5 hours. These were more of a mixed flow with a time constant.
Filtering, filter cake bins, balling, and in-durating took about 4 hours.
Dumped ore to pellets, about 24 hours.
We never had reason to identify how long it took tailings to travel from Concentrator feed to the tailings pond.
This varies from plant to plant. Some local plants have several days of storage between their crusher and concentrator, but I would expect the actual processing times would be similar.
Where do you buy this fluorescein dye of quality? The job is to get the proper metallurgical balance in real time, so also the tilings are important to measure the retention time for. Also if we close the mill circuit with derrick screen instead of cyclones, the retention tome reduce susbstancially.
Cole Parmer is one source, I'm sure there are many others.
Retention time is very critical; it depends om the time required for a mineral(particle) to respond to what we expect in a given machine. In a comminution circuit, the retention time in a mill is large while in the cyclone(which is in closed circuit) it is fraction of that of the mill. So when ever we talk about retention time, it is better to divide the total process operations into sub-groups ; since the product from a sub-group is a feed to the next operation, it is required that we determine retention times very carefully.The problem becomes more important whenever we have circulating loads. Let us take flotation, some particles may be fast floating while some are slow floating.
The tracer technique will give an idea of the present retention times; what is that required? If you want the real retention times required to get the best grade with maximum recovery, one has to go to basic information.
We are planning to start a new concentrator plant, so we need to measure in time the mas balance to find out the recovery and grade , based on con. and tailings grade . Conventionally you must take continuous samples from the process during some hours then make the balance. It is not real on time but compensate the differences in time. So my idea to use a radioactive material or as was suggested a dye fluorescent material could work. Thanks to every body for their comments.
It should be noted that either method (dye or radioactive) will only give you the minimum retention time (e.g. when it first appears in the concentrate or tails). If you watch very carefully you may be able to detect the peak, which would correlate with the average. Quite a while latter (maybe even days) the last will clear out. To be precise you would need to run it several times. And this is only after it is built.
Installing on-line analyzers,or just doing standard sampling might give you as good information
Excellent comments, in fact on line analyzers does not give a retention time as far I know.
If you are starting a new plant, the retention times are as identified in the design parameters; essentially the flow rates of solids and water and the kinetic studies of laboratory studies. That is how mills are sized, cyclone(s) nos and sizes /flotation cell sizes are selected. When the plant starts operating, you must have a good sampling survey of each operation and optimize the process parameters to get the required grade at maximum recovery.
I do not see why tracer techniques except for curiosity. However, if you want to trace a particle, you may use tracer techniques.
The above is my understanding of the issue you raised.
Online analyzers will give you performance data based on grade and recovery, comparing this to the feed will determine how well you are doing. This information is probably of more use in balancing a plant then retention time. Retention time is good for designing, but once built is difficult to change. If the circuit is existing, retention time can usually be calculated from design parameters (of which retention time is usually one as Helena said.
I agree, however we made the design and had the parameters , however the company who won the contract for the construction has made some changes (for the worst) and this changes are not based on test), so WE WERE THINKING A SYSTEM for finding the real retention time.
I fully appreciate and understand your issue, but finding retention times will not solve your problem. You say (if I have understood well) that you are stuck with a process flow sheet and the equipment. Have a well planned sampling campaign (engage a Process Expert as an Advisor--it is worth it) around the plant and one would be able to identify the trouble spots and action to be taken. Do not worry too much with the equipment; one can always get the best from these dummy machines.
Yes, much can be done with a sub-optimal circuit, sometimes by very minor changes. In this case I would not worry about retention time, but rather circuit performance and what can be done to improve it. Add reagents in grinding can often improve surface coating, which increases the performance in the first flotation cell. Switching a cell from rougher to scavenger can increase recovery. Changing reagents can also improve performance.
I still believe you can calculate the retention time better from the drawings then any attempt to measure it in the field.
I wish there are more persons like you who appreciates that machines are not the end of the process. In my experience I found so much that could be done by manipulating process variables.
Your way of exchanging information so freely brought out many important points out into the open; thank you.
I strongly feel we should not suggest tracers for the problem being faced.
In general, tracer techniques are used to know breakage mechanisms/ kinetics of particular particulate materials etc. i.e more for fundamental studies and not to solve processing problems of the nature identified.
If the feed ore weight varied and you were able to measure both feed and product weight, you can estimate retention time using a time-adjusted correlation.
On a batch or semi-batch basis measuring the time weighted average of feed versus product weights will give an indication of the minimum retention time, but on a continuous basis the time period needed to develop it would be several weeks. It would be simpler to simple take measurements of the equipment to calculate the active volume and then determine the unit average volume feed rate.
How do you perform a Process Retention Time Estimate Evaluation to study what time needed for a mineral to travel from before the mill until to concentrate and tailings and how to estimated this time. I thought using a nuclear tracer to be measure at the feed and at the tailings. For a conventional operation saying crushing, classification, milling, flotation , including cyclones for de-watering etc, and finally thickening , can be assumed a range of 6 to 8 hrs?
I was planning to use a radioactive material on the feed and detect it on the final products:concentrate and tailings