Control of Distributed Systems with Multiple Air

Compressor Rooms
Tim Dugan, P.E.
Compression Engineering Corporation
Keynote Speaker Sponsored by

The recording and slides of this webinar will be

made available to attendees via email later today.

PDH Certificates will be e-mailed to attendees

within 2 days
 Q&A Format

• Panelists will answer your

questions during the Q&A session
at the end of the Webinar.

• Please post your questions in the

Questions Window in your
GoToWebinar interface.
Sponsored by

• Direct all questions to

Compressed Air Best Practices®
Magazine
Handouts
 Disclaimer

All rights are reserved. The contents of this publication may not be
reproduced in whole or in part without consent of Smith Onandia
Communications LLC. Smith Onandia Communications LLC does not
assume and hereby disclaims any liability to any person for any loss or
damage caused by errors or omissions in the material contained herein,
regardless of whether such errors result from negligence, accident, or
any other cause whatsoever.

All materials presented are educational. Each system is unique and must
be evaluated on its own merits.
 Mark Your Calendars for the Best Practices 2024 EXPO & Conference!

SPEAK AT THE CONFERENCE

Track 1– Sustainability Through Track 2 – Quality, Safety and Reliability
Energy/Water Conservation Measures
As a Speaker, you will receive complimentary access to the
Conference, a $675 value!
https://cabpexpo.com/conference/speaker-submission/
Control of Distributed Systems with Multiple Air
Compressor Rooms
Introduction by
Compressed Air Best Practices® Magazine

Sponsored by
 About the Speaker

• President and Principal

Engineer, Compression
Engineering Corporation

• Over 36 years of experience

in the industry
Tim Dugan, P.E. Sponsored by
Compression Engineering Corporation
• 23 years of independent
consulting experience
OUTLINE

• Why Large Compressed Air Systems Are Distributed

• Good and Bad Reasons to Plan Distributed System
• Maintenance Challenges With Distributed Systems
• Control and Monitoring Challenges With Distributed Systems
• How to Make the Best of a Distributed System, Mechanically, Short of
Centralizing
• How to Control Screw Compressor Distributed Systems
• How to Control Centrifugal Compressor Distributed Systems
• Other Misc. Control Issues
WHY LARGE COMPRESSED AIR SYSTEMS ARE DISTRIBUTED

• Gradual Growth of Old Facility

• Different Types, Sizes, Brands
Mixed
• Example of Huge (20 Compressor)

0.64 miles
Screw Compressor System at 80-yr
Old Military Base:

0.83 miles
WHY LARGE AIR SYSTEMS ARE DISTRIBUTED - UNINTENTIONALLY

• Large Compressed Air Systems

Designed to be Segmented
• Oil and Gas Love This – They
Think it is More “Reliable”

0.25 miles
• Separate Centrifugal
Compressors, All w- Separate
Regen Dryers, All w- Diesel
Backups, One Per Process
• Example of Typical Small
Refinery Compressor System:

0.25 miles
WHY LARGE AIR SYSTEMS ARE DISTRIBUTED - INTENTIONALLY

• Process-Driven Systems
• “Process Compressors”
Behind Regulating Valves,
Feeding Most Air to Local
Process
• Mill Air & Instrument Air Fed
by What is Left Over,
Separate Piping
• Distribution X-over Lines Are
Small
• More Regulating Valves
Between Subsystems
• This Design is Intended for
“Reliability”.
• Refinery Will Not See it as
One Common System
(Though it Is!)
GOOD REASONS TO DISTRIBUTE SYSTEMS

• Very Long Distances

Between Production
Subsystems
• Piping Delta P
Calculations Needed
• Try to Centralize - Design
Piping For <0.5 psi/100ft
or < 5 psid Across
System in Worst Case X-
over Flow
• If This is Too Costly,
Separate Systems Might
be Needed
GOOD REASONS TO DISTRIBUTE SYSTEMS

• High Peak Demands at

System Extremity
• Storage Calculations
Needed
• Design Storage for <5 psi
Local Pressure Transient
in Worst Case
• Try to Centralize & Use
Storage to Feed Peak
• If This is Too Costly,
Separate Systems Might
be Needed
BAD REASONS TO DISTRIBUTE SYSTEMS

• Accounting
• Activity-Based Cost Management or
Strategic Business Unit (SBU)
Structure
• Separate P&L for Separate
Activities in Separate Buildings
• Yet There is One Common Air
System for Back-up
• Thus, Energy Into Each Meter is
Not Correlated w- Air Usage for
Each SBU
• Typically, Compressed Air Supplier /
Auditor Can’t Change This
BAD REASONS TO DISTRIBUTE SYSTEMS

• Poor Planning
• Large Compressed Air
Systems Are Like
Uncontrolled City Sprawl
• The Old “Core” is Neglected,
Infrastructure Investment
Goes Into the Latest Fancy
Suburbs
• Like Infrastructure in
Unplanned Urban Sprawl,
Developing Countries
MAINTENANCE CHALLENGES WITH DISTRIBUTED SYSTEMS

• Plants Are Reducing Maintenance Staff

Over Time
• Many Plants Are Outsourcing
Maintenance
• Limited Time Means Crisis Management
• Further Distance Makes It Hard to
Maintain All Equipment Equally
• Lack of System Monitoring Data at the
Compressor Makes it Risky to Shut
Down Equipment
• Centralized is Much Easier for Over-
burdened & Outsourced Maintenance
• However, Technology Can Address Most
of These Concerns
CONTROL CHALLENGES WITH DISTRIBUTED SYSTEMS
Compressed Air System

• Dryers & Compressors Can Have a Compressor Room 1

“Fatal Attraction”
P1
• Compressors Behind Separate 2-stage
Oil-free
Tank
Heatless
Regenerative to Plant
Screw
Dryer/Filters Operate at Different Compressor
Dryer

Auto Drain
Pressures to Meet System Pressure Purge

• Remote Controls That Use Local

Compressor Pressure Are Difficult to
Tune & Stay Tuned
• Purging Dryers/Drains Can Keep a
Segmented Compressor Running
• VFDs & Centrifugals Have Variable
Flow, Highly Variable Dryer Delta P
CONTROL CHALLENGES WITH DISTRIBUTED SYSTEMS

• Distributed Systems Can “See-saw

Pressure.
• Automated controls Can Make it
Worse.
• If One Side’s Compressor Loades
Up, That Side Pressure Goes Up
Quickly – Other Side Not as Fast
• If Another Side’s Compressor
Unloads, That Side Pressure Goes
Down Quickly – Other Side Not as
Fast.
• Can Trigger On a Far Unit, Then
Both Sides Can “Fight”
MECHANICAL RECOMMENDATION, DISTRIBUTED SYSTEMS

• Within One Area (Several Compressed Air System

Old Compressor Room

Compressor Rooms is OK), AC1 P2

Consolidate Wet & Dry Headers Old 150hp

Modulating
Compressor
2,000
Old Heatless
Regenerative
Storage
(Est 5,000
gal)
gal. Dryer
Target
• Plumb For Parallel Dryers, Install Sequencer-
For VSD &
Fixed Speed
tank 1500 scfm to Plant A

Parallel if $ Available, Keeping AC2

150hp Newer
Load-Unload P1
Compressor
Average Flow <50%/ea.
• Allows Wet Side Control, Much
More Predictable and Tunable New Compressor Room

• Allows Independent Dryer New 300hp VSD

Compressor
P3

New Blower Purge

Redundancy 2,000
gal.
tank
Regenerative
Dryer
1500 scfm to Plant B
DISTRIBUTED SCREW COMPRESSOR CONTROL

• Use “Target Algorithm”, not “Cascade”

• But Avoid “Control Gap” (Base-load
Compressor Bigger Than Swing Typical:
Compressor Range). 5 psi/step
• That Might Mean Having Sets of Trim 20 psi for 4
Compressors
compressors
• Use Wet Side Control if Possible. Dry
Side Can Work, but Not with VFDs.
• Avoid Pressure-Flow Controllers
Unless There Are Large Peak Pressure, psig Each time pressure hits "ceiling"
Demands Regularly 116 A base-load compressor unloads
• Have Remote Control Over Load, 114 off-line
Start, and Set Point of Each 112 VFD off-line (will happen often)
Compressor 110
108 Base-load VFD speed control range
• Commission For Proper Timers & 106 Compressor
Delta Ps 104 on-line (will always be on line)
102 on-line
100 Each time pressure hits "floor"
98 A base-load compressor starts/loads
96
DISTRIBUTED CENTRIFUGAL COMPRESSOR CONTROL

• Use “Load-sharing Algorithm”, not

“Cascade”
• All Centrifugals w- Same Wet Header
Running at Same Pressure, Appx Same
% Capacity
• Run Centrifugals Behind Separate Dryers
Base-load Only
• To Start-stop a Centrifugal, Use “Flow-
based” Logic.
• Avoid Control Gap, Centrifugal Total
Turn-down > (1) Compressor’s Size
• Do Not Use a Pressure-Flow Controller
with Centrifugals. It Will Result in Blow-
off.
• Tune For Stable Start & Stop of
Compressor
DISTRIBUTED HYBRID COMPRESSOR CONTROL

• Use “Load-sharing Algorithm”, not “Cascade”

• Tune Compressors for Max Turn-down, Min
Pressure Variance
• All Centrifugals On Same Wet Header
Running at Appx Same Pressure
• Don’t Use Dry Side Pressure
• Run Centrifugals Behind Separate Dryers
Base-load Only
• If Available, Use Screw Compressors as a
“Trim Set”, Not Just One Screw Compressor
(Control Gap)
• A Pressure-Flow Controller Might Only For
Screw Compressors All Behind One PFC.
• Use Target Control for Screw Compressors
• Commission For Proper Centrifugal Load-
sharing + Trim Target
CONTROL-MONITORING ISSUES

• Only Use Amps and Pressure for Control

• Use Calculated Flow (From Amps) for “Flow-
based Control, Not Flow Meters, Unless All
Compressors Have Reliable Wet Side Meters.
• All Other Process Values in Separate
Monitoring Overlay
• Have “Operator Interface Terminal” (OIT) in
Each Compressor Room, Showing Entire
System Operationally
• Have a “Human Machine Interface” (HMI) in
Plant SCADA, The Cloud, or Dedicated PC
System, Showing All Values & “Key
Performance Indicators” (KPIs)
OTHER MISC. CONTROL ISSUES

• Use Locally Sourced & Supported Plant-standard PLCs

For Master Control Whenever Possible
• Use Compressor Ethernet or Modbus Where Possible for
Compressor Interface
• Use Relay Control for Older Compressors
• Know How to Unlock Remote Control on Compressors
• Independently Control Load-unload, Start-stop, and
Proportional Set Point (For VFD and Centrifugal)
Whenever Possible.
• Default to Local Control, at a Higher Pressure, If Network
or Master Controller Fails
• Have Predictable Algorithm. Operators Need to Know
What Should be Running In What Loads and How. Avoid
“Learning” Algorithms That Confuse Operators.
• Commission All Scenarios With Fast-sample (<=10 sec)
Data-logging
OTHER MISC. MONITORING ISSUES

• Use Proper Type and Installation of Flow

Meters. Thermal Mass After Dryer, Not Before.
• Use Reliable, Simple Transmitters That Can be
Calibrated, or Have Simple Calibrated Element
That Can be Easily Changed
• Use Transmitters That Are Rated for the
Environment and Level of Vibration.
• Allow Monitoring to Be Changed w-o Control
Program Change / Update. Separate
Software.
• Have Easy-to-use Software for Building KPIs
and Zooming Into Trends
• Allow for X-Y Performance Plots
CONCLUSIONS

• Distributed Systems Exist for Good and Bad Reasons.

• Mechanical Modifications Might be Needed Before Control
• Wet Side Control Required for Clusters of VFD & Centrifugal Compressors
• Load-sharing for Multiple VFD &/or Centrifugal Compressors is Ideal
• Flow-based Logic for Significantly Different Sized Compressors is Ideal
• Wet Side Control Preferred for Load-unload Screw Compressor Systems
• Dry Side Control Can be Used for Load-Unload w- Dryer Delta P Considered
• Be Careful with PFCs for Part of Supply Side, Distributed System
• Try to Use Separate Control and Monitoring Systems
• Commission, Using Data
 About the Speaker

• Founder and CEO, CALMS

• >30 years of active involvement

in the compressed air industry

• Founder of HPE company that is

Gorazd Bregar Sponsored by
focused on compressed air
CALMS
solutions and services

• Co-Founder of CATZ -
Compressed Air Towards Zero
coalition of independent CA
experts
NO MONITORING - NO CONTROL
A Comprehensive Guide to Control Systems
Design, Optimization, and Maintenance
• Zone Control

Gorazd Bregar March 2024

Independent Compressed Air Management System

 Certified by:

ABOUT US
• Independent Compressed Air Management System for continuous
improvements and energy savings with AI-based Auto Analyzers
• Since 2008, team of CA experts Worldwide and software developers,
• Holistic, Action driven, Independent, Employee Engagement
• Based on CSA837, support for ISO 11011, 50001, 55000 compliant
• Supported by independent experts & coalition CATZ Worldwide
• CALMS Offices: USA, Slovenia, Sweden
• 1200+ CA systems Worldwide

GLOBAL CONTROL - LOCAL SUPPORT

COMPRESSED AIR SYSTEM As seen by
COMPRESSOR
MANUFACTURER

SUPPLY SIDE DISTRIBUTION SIDE DEMAND SIDE

 COMPRESSED AIR SYSTEM As seen by
AUDITOR

SUPPLY SIDE
SUPPLY SIDE DISTRIBUTION DEMAND SIDE tiny (always too
small)
compressor room
 AND WHERE ARE THE SAVINGS ?
SUPPLY SIDE DISTRIBUTION SIDE DEMAND SIDE
10% 5% 20%
Compressor room equipment Piping CA End-uses in production
• Compact space • Extensive space • Large area
• Quick assessment • Lengthy assessment • Time-intensive assessment
• Straightforward evaluation • Challenging evaluation • Complex evaluation
• Familiar solutions • Known solutions • Unfamiliar solutions
• Investment-driven • Investment-related • Exploration required

The %saving might vary based on system specifics and auditor focus (supply or full system!

• My price for energy audit as compressor seller was one new compressor or dryer or...
• My price for energy audit as indepenet auditor is 8% of annual savings
Follow 3 steps for Compressed Air Energy management

REDUCE ALTERNATIVE OPTIMIZE

DEMAND SOURCES EFFICIENT USE

Leaks Inappropriate uses Supply system optimization

Artificial demand Controls

Follow steps for Energy Optimization guided by CALMS

System Finding Audit Employee Permanent Optimisations Asset

potential Supply Training &
review Distribution monitoring & Controls management
savings Demand Engagement
Waste
 CALMS AUDIT TOOLS
Based on visualisation/graphs not tables with numbers!

Graph inspect Histogram Correlation / 3D Correlation

Duration Simulation Compressed air analyzers

Auto Reporting / Templates Events before-after Permanent monitoring with Auto analyzers...
 Control Considerations
⚬ Master controls won’t fix an otherwise broken system! Incorrect compressor size, type, number, system volume
and distribution system must be adequate...
⚬ Always start with an audit/simulation to evaluate the benefits of controls and get Functional Specification
⚬ Over expectation - avg. gains 12%, but applicability rate is 20% - so potential contribution 2.4% (CAS study 2001)
⚬ Consideration Risk (Reliability vs. Efficiency) - Will 10 minutes of downtime offset 1 year of energy savings ?
⚬ Compressors performance tested & modifications/repairs have been identified
⚬ Buy where, whom, DIY / standard products or custom design?
⚬ Finish with monitoring and defining goals and KPIs!

Common Mistakes
⚬ Implementing master controls without audit/simulation and specifications
⚬ Under specifying, not covering end-user requirements, skipping over hardware standards, programming and
communication protocols especially with different brands and model options
⚬ Over specifying, making the specification overly complicated
⚬ Fail to include a IQ/OQ test / commissioning / maintenance plan
⚬ Fail to monitor and optimize setting based on demand profile changes over time
 Multiple compressor control systems modes
Pressure reduction / Savings
p/ psi

115
C4
C3
C2 P1 -max
C1

100
p-sepoint

P2 - min p-required

85 p-critical
Cascade Pressure-band control Target pressure -PID Required pressure-Zone control

65
t

$€¥, kWh, cfm

$$$$ 100
$$$ 95
$$ 90
$ 85
 Control logic models
• Manual sequence control: Manual set sequence : 1 2 3 4 5 or 1 2 2 2 2 or 1 1 1 1 1 1 and System Pressure
Setpoint : Psp, Pmin, Pmax and Pcrit
• Rotating base load compressors: equal hours
• Manual Energy mode: selecting 3 or more sequences for starting and stopping the compressors based on flow
profile (Flow range 1, 2,3)
• Auto Energy mode: selecting best compressor sequence for starting and stopping (with same priority) based on
flow (size) and specific power or Isen. Load sharing between variable control compressors
• Zone control (CALMS unique) with dynamic pressure setpoint for compressor rooms(supply side) based on
highest required zone pressure setpoint (PFC) and flow profile (low, stable flow -> small dP)
Setup Pressure setpoint for each zone with PID parameters and schedules on Pressure Flow Controller -s(PFC)

P-SP3 auto (wet)

MCS P-SP1 by user
P-SP4 auto (wet)
CALMS EMCS is PFC-1 PFC-1
sPLC control
system fully
customizable with P-SP2 by user
standard PFC-2
functionalities.
 Zone Control - most advanced strategy Zone
M3/m
cfm
kWH Total +/- M3 Monthly cost

Pressure Flow Controllers 1 25 150 3445 -15% 35 000

User Psp 2 22 132 4445 -20% 27000

3 32 192 4899 +5% 38 000

Zone 1 Zone 4 4 27 162 4333 +15% 26 000

K1 5 25 150 1225 -3% 35 000

6 32 192 2256 -12% 35 000

K2 Zone 2 Zone 5
Main Control System
Dynamic Psp K3
Zone 3 Zone 6 5.1 5.2
5.5 5.6
M3 kWH
5.3 5.4 5.1 5.7
K1 65 100%

K2 90 92% 5.8 5.9 5.10 5.11

Virtual meter - calculations 5.12
K3 90 45%

Tot 245
MASTER CONTROLLER AND FLOW CONTROLLER MUST BE ALIGNED

If the compressors are not set

correct the system pressure
can increase even PFC is
holding stable pressure - also
specific power gets worse -

PFC & MC must be aligned

actively - we use dynamic
pressure setpoint based on
highest FC and based on flow
profile
SAVINGS WITH FLOW EMCS CONTROL SYSTEM
Savings or avoided energy
consumption: before-after event

Based on IPMVP - International

Performance Measurement and
Verification Protocol
 USA

THANK YOU !
NORDIC
Central Europe
CALMS AIR INC.
19 West 34th Street Suite CALMS Nordic AB
CALMS AIR d.o.o.
1018 NY10001 New York Linnégatan 30
Dolenjska cesta 83
413 04 Göteborg

Any questions ?
USA
1000 Ljubljana, SI
Sweden
E: [email protected]
E: [email protected]
T: +1-646-5532175 E: [email protected]
T: +386-41-432-202

[email protected]
T: +46-31-757-15-15
www.calms.com
www.calms.com
www.calms.com
 About the Speaker

• International Sales Director,

Comate Intelligent Sensor

• >11 years experience in flow

instrument sales

Ray Fang • 8 years experience in Sponsored by

Comate Intelligent Sensor
compressed air efficiency
measurement
 Efficiency Measurement in
Monitoring and Control Systems

21th Mar,2024

Ray Fang
Overseas sales director, COMATE INTELLIGENT SENSOR
Efficiency Measurement in Monitoring & Control Systems

AIR QUALITY

DEMAND & PRODUCTION BALANCE

COST OF AIR
Efficiency Measurement in Monitoring & Control Systems

Focus point Key Spec Unit

Capacity Air Supply m3/min

FIND KEY SPEC
Quality Dew Point Ctd
ACCORDING TO
Quality Oil Content mg/m3

USER’S FOCUS POINT Cost Unit power consumption Kw/m3

Lost Air loss ratio %

Efficiency Measurement in Monitoring & Control Systems

FIND PROPER MEASUREMENT POINTS

Different
measurement points
lead to different facts
Efficiency Measurement in Monitoring & Control Systems

Pitot tube flow meter

VFM60MV TGF460
FIND THE RIGHT 1st Vortex flow meter that can
With industrial grade heavy duty
adapt to compressed air
sensors, easy for maintenance.
INSTRUMENTS applications, with multi-
variable, best small flow and
Superior long-term stability , no
0-off problem
best anti-vibration

Vortex flow meter for direct Thermal mass flow meter

output measurement
Different instruments
answer different
measurement
challenges PTF520
Special designed PTF520 can
deal with liquid water blockage
problem better than other
brands, it is also multi-variable,
and can measure bi-direction
flow
Efficiency Measurement in Monitoring & Control Systems

ANALYZE & USE OF THE DATA

Problem
DATA
Analyze
DATA
Plan
DATA
Improvement
DATA
Evaluate
Efficiency Measurement in Monitoring & Control Systems

DISTRIBUTED MONITORING SYSTEM

Flexible
Wireless
Distributed logging
Multi-level data access
Thank you!
Looking forward to further discussion!
 Control of Distributed Systems with Multiple Air
Compressor Rooms
Q&A
Please submit any questions through the Question Window on
your GoToWebinar interface, directing them to Compressed Air
Best Practices Magazine. Our panelists will do their best to Sponsored by
address your questions and will follow up with you on anything
that goes unanswered during this session.
Thank you for attending!
 Thank you for attending!

The recording and slides of this webinar will be made

available to attendees via email later today.

PDH Certificates will be e-mailed to Attendees within 2 days.

 April 2024 Webinar
Refrigerated vs Desiccant Dryers and Choosing the Right One

Don Van Ormer

APEnergy
Keynote Speaker

Thursday, April 4, 2024– 2:00 PM EST

Register for free at
www.airbestpractices.com/webinars