Faster solutions with Byblock
It seems that the past 30 years were spent in malpractice in operating
wire rod blocks Each time the rolls are machined the diameters are changed
and the original set-up conditions no longer apply. A further development
of the 'Byblock' roll pass calculation program permits rapid calculation of
correct roll gap to solve the problem.
A RAPID SOLUTION TO ROLL GAP SETTING AFTER GRINDING (*)
By Annibale Izzo, Linebow (**)
There are precise theoretical premises about roll pass design calculations
for wire rod blocks. Those premises were discussed in a paper1 by the author
which examined a wire rod block in its basic (i.e. factory set) configuration.
But in actual rolling practice basic configurations only apply for a few hours.
Roll wear soon occurs, cobbles may happen, and rolls must be changed.
The chances are that after a roll change at least one pair of rolls will have
a different diameter than the original. This situation was discussed in a
second paper2 by the author which examined how to set the correct roll gap at
a block stand with a modified roll diameter.
The problem
The problem is how to set the actual roll gaps within a wire rod block after
changing roll diameters at many stands (or at all of the stands).
As stated in the author's 'linebow' website3, "Both on the designer's desk and
in the assembly shop everything works. The problems arise on the mill floor,
where sometimes, at the first stoppage for roll changing, the rollerman has
to 'take arms against a sea of troubles'. Modifying the diameter of even a
single pair of rolls alters the delicate balance of bar speeds at each stand:
if a stand 'pushes' too much a loop will occur, with immediate danger of a cobble;
if a stand 'pulls' too much, the roll grooves will tend to be underfilled, and
the bar will be overstretched. And you know what is the common practice on the
mill floor? Grind down all the roll diameters to the amount necessary to repair
the most damaged pair and set the mill with the nominal roll gaps."
If we think this is the way to restore the basic situation (that with rolls
having nominal diameters), we better think again.
The trouble is that there are no suitable criteria available. This lack of
certainty is probably the cause of an average of two cobbles a month occurring
in each block, with an estimated loss of 20 to 30 thousand US dollars a year.
But there is more: because traditional practice produces out-of-tolerance front
ends and tail ends, these are currently scrapped. What if a new gap setting
practice eliminated those defects? In terms of mill yield, correct gap setting
could result in saving many metres of wire rod per coil or, at a conservative
estimate, some 100,000 US dollars a year.
The solution
Well, we did publish the proper criteria in the paper2 presented in 1998. But,
as remarked by Herbert Rothe Consulting Engineers, 'those criteria provide a
solution to the technical problem without accounting for a number of psychological
aspects. In fact, they assume that in a critical situation the shift foreman
disappears from the shop floor and starts the proper computer program in the
tranquillity of his office. Meanwhile the mill crew run around shouting.
And you can bet that - as stated by a notorious law of nature - if something
can go wrong, it will'.
What Mr. Rothe means is that we must give rollermen working under pressure a fast
and error-proof way to determine gap settings after changing roll diameters.
In seconds, not in minutes. We managed solving this problem, as explained
extensively on the website.
In a nutshell, we start from the 'Sun' (our Byblock program) to create a number
of 'planets'. Each planet (created in-house and then delivered to the user) refers
to a particular block AND to a single finished rod diameter. The only input
variables to be entered are roll diameters. From each planet the user can generate
an infinite number of 'satellites', i.e. block configurations with different roll
diameters.
A suitable satellite can easily be obtained in a few seconds, even on the shop floor,
by using a portable computer. A satellite has the form of an ASCII file containing
the roll pass schedule with modified roll diameters and new roll gaps. An example
can be examined on the linebow website illustrating how metal flow is left constant
and fixed rotational speed ratios are respected.
The conclusions
From exploiting the new satellites (you can freely download one from the website,
with a choice of metric or Imperial units) we can infer two 'laws':
1. Changing N roll diameters leads to changing more than N roll gaps
2. Changing all roll diameters leads to a non-predictable gap distribution
The two laws are visualised in Tables 1 & 2, taken from the website. (DIA and GAP
stand for nominal values of roll diameter and roll gap, Dia and Gap in lower case
stand for the actual values. Variations are indicated by a "<" sign.)
Stand DIA Dia GAP Gap
#21 OV 210 210 1.00 0.33 <
#22 RD 210 210 1.00 1.00
#23 OV 210 210 1.00 0.60 <
#24 RD 210 200 < 1.00 1.48 <
#25 OV 170 170 1.00 0.52 <
#26 RD 170 170 1.00 1.00
#27 OV 170 170 0.50 0.19 <
#28 RD 170 170 0.50 0.50
#29 OV 170 170 0.50 0.24 <
#30 RD 170 160 < 0.50 0.50
Table 1 - Two stands with modified roll diameters
Stand DIA Dia GAP Gap
#21 OV 210 207 < 1.00 1.04 <
#22 RD 210 207 < 1.00 1.15 <
#23 OV 210 207 < 1.00 0.97 <
#24 RD 210 207 < 1.00 1.11 <
#25 OV 170 167 < 1.00 0.97 <
#26 RD 170 167 < 1.00 1.14 <
#27 OV 170 167 < 0.50 0.47 <
#28 RD 170 167 < 0.50 0.61 <
#29 OV 170 167 < 0.50 0.48 <
#30 RD 170 167 < 0.50 0.50
Table 2 - All stands with modified roll diameters
Table 2 demonstrates the author's theorem about wire rod block malpractice.
The theorem states that setting all gaps to the nominal value after reducing
all roll diameters by the same amount can only result in uncontrolled pushes
and pulls within the block.
Interesting enough, the theorem has a corollary: the last pair of stands of
the block must have the same roll diameters. This can be seen from the last
two stands of Table 1, modified to different roll diameters: if DIA30 diminishes,
then the motor speed must be increased (to maintain constant flow); this way
speed29 rises, and the oval area must be decreased (again to maintain constant
flow); to decrease the oval area, we can only reduce GAP29 (0.50 to 0.24); and
this leads to an 'empty' round30. Checkmate.
References
1. A. Izzo, "Roll pass design software for wire rod blocks", Steel Times, January
1997 Vol 225 No 1 p 14
2. A. Izzo, "Software solutions to shop floor problems in wire rod blocks", Steel
Times International, March 1998 Vol 22 No 2 p 31
3. "LINEBOW Roll Pass Design Software", website URL http://www.passdesign.com
(*) Published by Steel Times, February 2000 (p54)
(**) Via Garda 2, I-10015 Ivrea, Italy. Phone +39 0125 616 768,
fax +39 02 700 500 073, e-mail info@passdesign.com