232 posts
TimePosted 22/10/2010 04:03:07
xxxx says

Separator performance

without Tromp curve what is the way to judge the separator performance by seive analysis of rejects,feed & fines. what is the normal range of residues on 212,90,63,45,32 mic for better efficiency of a separator in coarse,feed & fines & why?(for cement as well as raw meal) what is the optimum efficiency of a high efficiency separator? raj


Know the answer to this question? Join the community and register for a free guest account to post a reply.

138 posts
TimePosted 22/10/2010 13:17:18
lalbatros says

Re: Separator performance


If you have residues on {212, 90 , 63, 45, 32} µm for feed, coarse and fine, and if your product is not too fine - as compared to 32µm - , then you could evaluate a Tromp curve.
If you have less data, you have less chances to evaluate your separator, and vice-versa.

However, the main question is: what do you mean by "optimum efficiency"?

On a Tromp curve, you can observe the so-called "bypass level".
The bypass level indicates how much % of fine particles go to the coarse stream.
This measures, of course, and un-wanted behaviour or an "un-efficiency".

If the bypass is 40%, and if the production is 100 t/h, then, naïvely speaking, this means that 40 t/h of good product is returned to the mill.
This could be, naïvely, translated as an efficiency of 100/140 = 70% and would indicate that the production could be increased by 40 t/h if the separator would be 100% efficient. Actually, the impact of the bypass is lower but it increases with the fineness of the product (circulation factor). This can be understood by a model of the complete grinding circuit.

My message is: to evaluate a seperator you do need to evaluate the Tromp curve.
The least you could get from the Tromp curve is the bypass factor.
If you don't have enough information for that, you cannot evaluate the separator.



232 posts
TimePosted 23/10/2010 04:11:54
xxxx says

Re: Separator performance

my question is what is the normal range of these residues for coarse,feed & product so that i can compare for any improvement in the system raj


138 posts
TimePosted 23/10/2010 21:45:43
lalbatros says

Re: Separator performance

There is no answer to this question.
It depends on what you are producing.
For "system improvement", is it not enough to check the quality of the product and the productivity t/h ?
If your focus is on "separator improvement", then it is all about the Tromp curve.

Here is some sample data to illustrate my point:

finish product
3500 Blaines, (P32,P45,P63,P90,P212) = (78%, 90%, 97%, 99.3%, 100%) 
where P=1-R

2450 Blaines, (P32,P45,P63,P90,P212) = (53%, 67%, 80%, 90%, 99.6%) 

1200 Blaines, (P32,P45,P63,P90,P212) = (23%, 39%, 61%, 79%, 98.9%) 

Circulating load:             1.8

bypass:           10%
cut diameter:  40µm  (x50)   (x75= 26µm, x50= 40µm, x25= 63µm)
imperfection:  46%   (=(x75-x25)/2/x50)

residues reduction factors:  (= residue mill outlet / residue mill inlet)
                                        @32µm   = 40%
                                        @45µm   = 58%
                                        @63µm   = 76%
                                        @90µm   = 87%
                                        @212µm  = 100% (!)
(this is calculated from residues and circulating load)

I don't see how you could use these data and compare them to your own data.
What is your product?
What is the product fineness?
What is you circulating load?

To evaluate your separator, you need the Tromp curve, even with very few points.
The bypass is one point of the Tromp curve that can be evaluated even with few data.