Why HPGRs did not make the list?
| LinkedIn Page 1 of 8
As of July 1, LinkedIn will no longer support the Internet Explorer 11
browser. For the best experience, please switch to one of LinkedIn’s
Why HPGRs did not make the list?
Published on December 6, 2017
Alex Doll 10 articles Following
Consultant at SAGMILLING.COM
The presentation at the CIM BC South Central Branch Meeting included one more topic that
deserves its own post. Why did High Pressure Grinding Rolls (HPGRs) not make the list of 5
technologies that BC should be using more of?
The answer is that most of BC's mining projects are large open pits treating porphyry
deposits (copper, gold and/or molybdenum). SAG mills are generally superior to HPGRs in
this application.
Summary, why SAG?
SAG mills were adopted in the 1970's and 1980's over multi-stage crushing plants because
they have a lower capital cost and they are easier to operate, especially in a Canadian winter.
They were not adopted because they were more energy efficient than multi-stage crushing:
the adopters knew they were sacrificing between 10% and 15% higher energy consumption
moving to SAG circuits.
Messaging
https://www.linkedin.com/pulse/why-hpgrs-did-make-list-alex-doll/ 5/22/2020
Why HPGRs did not make the list? | LinkedIn Page 2 of 8
One major benefit was lower capital costs;AsSAG plants were perhaps 10% to 15% cheaper
of July 1, LinkedIn will no longer support the Internet Explorer 11
because they were more compact and hadbrowser.
less equipment. The
For the best amortization
experience, pleaseof thesetolower
switch one of LinkedIn’s
capital costs mean that only the hardest orebodies would see a financial benefit from the
lower operating costs of a multi-stage crushing circuit (which HPGR is a variety of).
Canadian winter is another enemy of crushing systems (and HPGR) as evidenced by a
relatively recent example of the conversion of Gibraltar Mines. Secondary crushing,
screening, fine ore storage and rod mills were replaced with a SAG mill and recycle crusher
(the rod mills were re-purposed as ball mills). One comment from Rotzinger & Major (SAG
2011) about fine ore storage was "operations [interruption] was particularly
pronounced in the winter months when frost chunks plugging feeders was a continual
problem. Clearing frost chunks was a particularly trying task ... because of the hard
work involved in keeping the nine draw points flowing"
Energy efficiency, beware of simplistic comparisons
The most common benefit cited for HPGR is a higher energy efficiency than tumbling mills;
often this is a very simplistic "1.5 kWh/t in HPGR versus 5.0 kWh/t in SAG". The flaw in
this reasoning is that the overall energy in the circuit is more important than comparing the
energy in a single stage – the overall energy savings of an HPGR circuit should be about
10% to 15% versus a SAG mill for typical porphyry ores.
Let's pick apart the whole grinding circuit with a look at the plant-wide energy consumption
starting after the primary crusher and ending at the flotation feed:
1. A typical SABC circuit flowsheet consists of one or more SAG mills, one or more ball
mills, pebble crushers and ancillary equipment like vibrating screens and conveyors.
(Hydrocyclones and their feed pumps are the same in both circuits, so can be ignored.)
2. A typical HPGR circuit flowsheet consists of one or more secondary crushers, high
pressure grinding rolls, ball mills and ancillary equipment like vibrating screens, dust
collection and conveyors.
These flowsheets sound like they have similar equipment, but there are important differences
in scale to consider:
• typically need more secondary crushers than pebble crushers;
• conveyors in HPGR circuits are longer and do more lifting of ore;
Messaging
https://www.linkedin.com/pulse/why-hpgrs-did-make-list-alex-doll/ 5/22/2020
Why HPGRs did not make the list? | LinkedIn Page 3 of 8
• ball mills are bigger in HPGR circuitsAsthan in SAG circuits; and
of July 1, LinkedIn will no longer support the Internet Explorer 11
browser. For the best experience, please switch to one of LinkedIn’s
• HPGR circuits need a greater footprint resulting in more space heating requirements
during winter and more lighting.
The run and rise of a SAG recycle pebble crusher conveyor in a flowsheet is highly variable,
but in the simplest concept is a few metres of "lifting" and an arbitrary length of "running" on
a fraction, say 15%, of the mill's fresh feed rate. The overall energy contribution of screening
and conveying in a SABC circuit can be as little as 0.1 kWh/t.
HPGR circuit conveyors perform more "lifting", to heights of 30 metres (100 feet) or more
on the entire fresh feed plus recycle, possibly 115% of the fresh feed rate. You are lifting
(and dropping) the whole fresh feed to an HPGR roughly three times in a typical flowsheet,
and this adds up. The potential energy wasted this way (E = m × g × h) can add up to
between 1.0 and 1.5 kWh/t, approaching the same energy consumption as the HPGR! Ref:
Doll, CMP 2015
The difference in the ball mill sizing between a SAG circuit and an HPGR circuit comes
down to two things:
• The transfer size of an SABC-A circuit is usually 2 mm in a porphyry whereas the
transfer size of an HPGR can be between 3 mm and 5 mm. Only in SABC-B
configuration would the transfer size be similar in SAG and HPGR.
Messaging
https://www.linkedin.com/pulse/why-hpgrs-did-make-list-alex-doll/ 5/22/2020
Why HPGRs did not make the list? | LinkedIn Page 4 of 8
• The "phantom cyclone effect" of a SAG mill is about twice as effective as the
As of July 1, LinkedIn will no longer support the Internet Explorer 11
"microcracking" effect of an HPGR (Ref: Doll,
browser. ForBarratt
the best&experience,
Godoy, Procemin 2010).
please switch This
to one of LinkedIn’s
will be ore-specific and difficult to assess in laboratory work, but in round numbers the
effect for a SAG product is 10% versus 5% for an HPGR. Designers usually ignore both
of these effects because they are small, difficult to measure, and reasonable to keep in
reserve as "design fat".
The "wasted" energy in a SAG mill is useful in cold climates – it heats your building so you
burn less natural gas for space heating. This isn't a direct 1:1 benefit because the humidity
that comes from SAG mills needs to be vented by the HVAC system, but it is incorrect to
describe 100% of the heat from a SAG mill as truly "wasted". Add to this the need to enclose
and heat fine ore conveyors, prevent bins from freezing, etc. and you will have a rather large
energy consumption that is typically ignored in high-level HPGR studies (Imagine the
heating requirements for enclosing Boddington's comminution circuit in a building).
Projects that benefit from HPGR
This article is dealing in generalizations, and it is worth looking at some specific cases where
an HPGR would be a better technology than SAG. In the context of copper porphyries, there
are two types of projects that would benefit from HPGR:
• underground mines where the hoisting or haulage should be the rate-limiting step (and
the mill should fit the mine, not the other way around as in an open pit);
• very competent ores that have a particular breakage profile according to particle size
(these ores are difficult to treat in a SAG mill).
Crushing equipment, including HPGRs, can generally be considered as "constant volume"
machines where the machine throughput does not vary much with ore hardness. The
circulating load can increase, reducing the fresh feed rate, but the machine itself won't see the
kind of variation in throughput one experiences in a SAG mill circuit. This HPGR
characteristic is well-suited to underground bulk-mining where the mine haulage should be
the rate-limiting step. The capital cost penalty of an HPGR circuit is also mitigated because
the milling plant tends to be a smaller component of the overall project capital cost – plus
designers might over-size a SAG mill to avoid constraining the mine, also diminishing the
capital cost benefit of SAG mill plants.
Most porphyry ores are not competent, which for our purposes means that they are easily
broken at coarse sizes. To demonstrate this, we'll use the "work index" metric across the
whole size range from coarse (low-energy impact crushing work index) to fine (ball mill
Messaging
https://www.linkedin.com/pulse/why-hpgrs-did-make-list-alex-doll/ 5/22/2020
Why HPGRs did not make the list? | LinkedIn Page 5 of 8
work index); you can do the same analysisAswith other metrics, like Hukki's Conjecture, and
of July 1, LinkedIn will no longer support the Internet Explorer 11
should get similar outcomes. browser. For the best experience, please switch to one of LinkedIn’s
The important characteristic is coarser rocks are more resistant to breakage; Boddington, for
example, has a ball mill work index of about 15 metric units and a crushing work index of
almost 30 metric units. It is important to notice that the ball mill work is not the metric that
controls this "amenability", it is the coarser sizes demonstrated by the rod mill and crushing
work indexes.
Contrast this with a copper porphyry, here is the Wi signature of the Ajax project in British
Columbia.
Here we see the coarse fractions are not competent, the crushing work index is in the range
of 5 to 10 metric units, but this ore has very high ball mill work index values, over 20 units.
The size range treated by SAG milling, from 150 mm down to 2 mm, is all "easy to break"
according this metric, and it is the ball mills that will have difficulty treating this ore (not the
SAG mill).
• Aside: the Ajax 43-101 that published the data in the chart above was based on an HPGR
circuit. This is actually reasonable because the owner, KGHM, had very recent costing
Messaging
https://www.linkedin.com/pulse/why-hpgrs-did-make-list-alex-doll/ 5/22/2020
Why HPGRs did not make the list? | LinkedIn Page 6 of 8
information from their Sierra Gorda process plant in Chile. If the project is a financial
As of July 1, LinkedIn will no longer support the Internet Explorer 11
success with HPGR, then it will also be a financial
browser. For thesuccess with SAG.
best experience, please switch to one of LinkedIn’s
Are there porphyry projects that have a breakage characteristic more amenable to HPGR than
to SAG milling? Yes, Cerro Verde's PKP ore type is just such an ore, as is Sierra Gorda.
British Columbia mines do not tend to publish a lot of ore characterization, so it isn't possible
to say if some of the harder BC porphyries would be better suited to HPGR (Copper
Mountain & Mount Milligan, for example). Hopefully by publishing this article, BC project
proponents will look closer at the crushing work index test when assessing grinding circuit
designs.
• Aside: Yes, the crushing work index test has "issues", but it is the only comminution test
we've got in the 100 mm size class. Some geotechnical tests operate in this size range,
Point Load Index for example, but they have many of the same flaws as the crushing
work index test. One mitigation is that projects like Ajax are not very sensitive to the
crushing Wi; if we double the crushing Wi it only results in a 5% to 10% change in
SABC specific energy consumption.
Conclusions, all technologies have their place
Both SAG milling and High Pressure Grinding Roll technologies are mature technologies
that are suitable for mining projects, but they are not completely interchangeable. Each has
its strengths and its weaknesses.
• Financial efficiency, not energy efficiency, is the goal of any design in a capitalist
economy. Use the technology that is the best use of your shareholder's money and ignore
the bleating of central planning enthusiasts who insist that energy efficiency is
paramount.
• Open pit copper porphyries, like those found in British Columbia, are generally better
treated with SAG mills. When mated with geometallurgy, it may be possible to benefit
from the throughput variability by front-loading soft material into the mine plan to
squeeze higher SAG throughput in early mine years.
• Competent craton hypogene deposits, like those found in Western Australia, are much
more likely candidates for HPGR. Deposits like Three Mile Hill (with a crushing work
index of about 50) are never going to be efficient to grind in a SAG mill.
Messaging
https://www.linkedin.com/pulse/why-hpgrs-did-make-list-alex-doll/ 5/22/2020
Why HPGRs did not make the list? | LinkedIn Page 7 of 8
• Winter conditions favour more compact, indoor plants typical of SAG circuits; desert
As of July 1, LinkedIn will no longer support the Internet Explorer 11
conditions such as southern Perú or Western
browser.Australia aren't
For the best as sensitive
experience, please to freezing
switch to one of LinkedIn’s
conditions and can favour HPGR.
• HPGR might be a better choice for underground mines as they have less variability in
throughput.
• Clays and wet fines in the ore favour SAG mills, example deposits include those with
surface laterite, oxidizes zones, and significant fault gouge.
And a final thought for project managers: EPCM companies earn more revenue designing
complicated, high capital cost projects. It is in their interests to push projects towards HPGR
to maximize their engineering fees. Don't take their word as gospel (Ref: Villanueva & Soto,
IMPC 2014), get a third party to check.
Report this
Published by
Alex Doll 10 articles Following
Consultant at SAGMILLING.COM
Published • 2yr
#SAG #HPGR #mining #comminution
Like Comment Share 107 · 14 Comments
Reactions
+95
14 Comments
Most Relevant
Add a comment…
Jorge Torrealba • 1st 1y
S
Project Manager
Thanks for sharing!
Stephan W Kirsch • 1st 2y
S
President - Mining Equipment at Metso, Executive Board Member
Alex, good article. Some good guide lines for engineers who need to make a call for their projects. Based on
some 20 years of experience with mills & HPGRs I concur with most of it. Neither SAG nor HPGR based
circuits are the one solution for all ore processing plants.
Messaging
https://www.linkedin.com/pulse/why-hpgrs-did-make-list-alex-doll/ 5/22/2020
Why HPGRs did not make the list? | LinkedIn Page 8 of 8
Load more comments
As of July 1, LinkedIn will no longer support the Internet Explorer 11
browser. For the best experience, please switch to one of LinkedIn’s
Alex Doll
Consultant at SAGMILLING.COM
Following
More from Alex Doll
More grindability metrics: Rod mill Empirical conversions, SPI™ & SGI Crushing Work Index and UCS Using home time wisely - software
updates
Alex Doll on LinkedIn Alex Doll on LinkedIn
Alex Doll on LinkedIn Alex Doll on LinkedIn
See all 10 articles
Messaging
https://www.linkedin.com/pulse/why-hpgrs-did-make-list-alex-doll/ 5/22/2020
