WHITE PAPER NW-MTBF/0610

Reliability of uninterruptible power supplies

Abstract

Index
Introduction 2
1. Single UPS without static bypass switch (SBS) 2
2. Single UPS WITH static bypass switch (SBS) 4
3. Parallel Redundant UPSS WITH Static Bypass Switch (SBS) 6

Conclusion 11

ABB Via Luserte Sud 9 CH-6572 Quartino

White_Paper_Relibility_150506.doc 1
Introduction The primary objective in the implementation of a UPS system is to improve
the reliability to the limits of technical capability, the ultimate aim being to
totally eliminate the possibility of any disturbance or downtime. In the
50s when the first static UPS systems appeared they were composed of a
rectifier, battery and inverter and where used to stabilize the output power
and to continue to supply the load for a short period of time (autonomy of
battery) in the event of a rectifier failure. The reliability of this simple UPS
chain depends dominantly on the inverter reliability. An inverter failure
meant an immediate load crash. Furthermore, the downtime (no supply to
the load) would last as long as the inverter repair. In the early 60s the
static bypass switch was introduced to enable an interruption-free load
transfer to the stand-by mains in the event of inverter failures or
overloads. The stand-by mains although far not as reliable as the UPS,
serves as a reserve power supply in case of an inverter failure to enable the
continuation of the power supply to the load while the inverter is being
repaired. This new architecture had substantially improved the overall
reliability which did not depend dominantly on the inverter reliability
anymore. The reliability of the new UPS with static bypass depended on
the quality of the mains (MTBFMAINS), the time to repair of the UPS
(MTTRUPS)and on the reliability of the static switch. Further on in this paper
(page 4) the impact of both the (MTBFMAINS) and (MTTRUPS) on the overall
UPS reliability are shown.
The everyday activities which are dependent upon computer controlled
real time information systems have grown exponentially in recent years
and the requirements for highest reliability UPS configurations have
become every day practice. Very critical loads cannot rely on a power
supply configuration of a single UPS with static bypass system; the need for
(n+1) redundant parallel UPS configurations is becoming a standard.
In this paper the reliability of the various UPS configurations is shown. The
reliability figures of the rectifier/booster, battery, inverter , static bypass
and other parts are based on the reliability figures presented in the MIL-
HDBK-217 F (Not.2, 1995).

1. Single UPS without static The reliability of a single UPS without bypass depends basically on the
bypass switch (SBS) reliability of the rectifier, battery and inverter (see electrical block-diagram
in Fig.1).
Example: In the event of an inverter fault the load would crash.

ABB Via Luserte Sud 9 CH-6572 Quartino

White_Paper_Relibility_150506.doc 2
 Electircal Block Diagram Reliability Block Diagram
(series diagram)
Mains

RECTIFIER RECT.

BATTERY BATT.

INVERTER INV.

Load

Fig. 1 Single UPS without static bypass. Electrical block diagram

and reliability diagram

Description of the variables that are used in the equations:

MTBFSU: Mean Time Between Failures of Single Unit without static bypass
UPS Failure Rate of Single Unit without static bypass switch
RECT Failure Rate Rectifier
BATT Failure Rate Battery
INV Failure Rate Inverter

Calculation of MTBF for UPS System WITHOUT bypass (MTBFSU):

MTBFUPS = 1 / UPS
UPS = RECT.+ BATT. + INV. .. (E.1)

If the figures from ABBS statistical failure-analysis RECT = 20 10-6 [h-1],

BATT = 10 10-6 [h-1], INV. = 20 10-6 [h-1] are applied in equation (E.1)the
Mean Time Between Failure (MTBFUPS) for a UPS system without static
bypass switch will be:

MTBFUPS = 20'000 [h]

ABB Via Luserte Sud 9 CH-6572 Quartino

White_Paper_Relibility_150506.doc 3
2. Single UPS WITH static bypass The reliability of a single UPS can be increased significantly by introducing a
switch (SBS) redundant mains power source by and linking it to main UPS supply source
by means of a static bypass static transfer switch.
Example: In the event of an inverter fault the load WILL NOT crash. The
load will be transferred to mains interruption-free.

Electrical Block Diagram Reliability Block Diagram

(Parallel Diagram)

Mains M
RECT.
RECTIFIER

BATTERY BATT.

INVERTER INV.

STATIC
SBS
BYPASS

Load

Fig. 2 Single UPS WITH static bypass. Electrical block diagram and
reliability diagram

Description of the variables that are used in the equations:

MTBFUPS+SBS: Mean Time Between Failures of Single Unit WITH static
bypass switch (SBS)
MTBFM: Mean Time Between Failures of MAINS
UPS+SBS: Failure Rate of Single Unit WITH static bypass switch
SBS : Failure Rate of Static Bypass Switch with control circuit
PBUS : Failure Rate of Parallel Bus (only for parallel systems)
M : Failure Rate of MAINS
SU: Repair Rate of Static Bypass Switch (SU= 1 / MTTRUPS)
M: Repair Rate of MAINS (M = 1 / MTTRM)
MTTRSBS: Mean Time To Repair of Static Bypass Switch
MTTRM: Mean Time To Repair of MAINS

ABB Via Luserte Sud 9 CH-6572 Quartino

White_Paper_Relibility_150506.doc 4
 Please note that all calculations are performed by using the following
constants:
MTBFM = 50 [h] , this figure represents a good quality mains.
MTTRUPS = 6 [h]
MTTRM = 0.1 [h]

Furthermore from ABBS statistical failure-analysis we have the following

figures for the failure rates of the static bypass switch for the power part
and the control electronic part:

SBS = 2 10-6 [h-1]

Calculation of MTBF for UPS System WITH static bypass switch

(MTBFSU+SBS):

MTBFUPS+SBS = 1 / UPS+SBS

UPS+SBS = UPS // M +SBS ... (E.2)

UPS//M= UPS M (UPS + M +UPS +M)____ =UPS M (UPS+M)=6 10-6 [h-1]

UPS M +UPS M +(UPS+M) (UPS+M)+2UPS+2M UPS M

UPS+SBS = UPS // M + SBS = 6 10-6 [h-1] + 2 10-6 [h-1] = 8 10-6 [h-1]

MTBFUPS+SBS = 125'000h

NOTE: In the above formula it can be seen that the reliability of

the UPS with static bypass switch
( MTBFUPS+SBS ) depends largely on three parameters:
reliability of the mains, the MTTR of
the UPS and the reliability of the static bypass switch. This
dependence is illustrated in Fig 3.

ABB Via Luserte Sud 9 CH-6572 Quartino

White_Paper_Relibility_150506.doc 5
 MTBFups = f (MTBFmains, MTTRups)
MTTR=6h
MTBFM=50h
600000
24 12 6 3 0.5
500000

MTBFups [h]
400000

300000

200000

100000

0
10 24 50 80 100 250 500 1000 10000
MTBFmains [h]

Fig. 3 This graph shows the dependence of the MTBFUPS+SBS on

MTBFMAINS and MTTR UPS.

3. Parallel Redundant UPSs WITH The reliability of a single UPS can be increased significantly by introducing a
Static Bypass Switch (SBS) redundant parallel configurations.

UPS 1 UPS 2 UPS (n+1)

RECTIFIER RECTIFIER RECTIFIER

BATTER BATTER BATTER

Y Y Y

INVERTER INVERTER INVERTER

STATIC STATIC STATIC

BYPASS BYPASS BYPASS

PARALLEL BUS PARALLEL BUS

ABB Via Luserte Sud 9 CH-6572 Quartino

White_Paper_Relibility_150506.doc 6
 M
RECT. RECT. RECT.

BATT. BATT. BATT.

INV. INV. INV.

PBUS

PBUS

PBUS

SBS SBS SBS

Fig. 4 Electrical and Reliability Block Diagram of (n+1) Parallel

Redundant UPS Configuration.

Calculation of MTBF for a (n+1) REDUNDANT PARALLEL UPS System

(MTBF(n+1) UPS+SBS):

We will start by the equations used for the calculation of the failure rate:

(n+1) UPS+SBS= (UPS1 //UPS2..// UPS(n+1)) + (n+1) UPBUS + (SBS1

// SBS2..// SBS(n+1)) .. (E.3)

Failure Rate: (n+1) UPS+SBS ~ (n+1) PBUS (E.4)

Reliability: MTBF(n+1) UPS+SBS = 1 / (n+1) UPS+SBS (E.5)

Availability: A (n+1) UPS+SBS = MTBF(n+1) UPS+SBS /MTBF(n+1)UPS+SBS

+ MTTR UPS (E.6)

The reliability of a (n+1) parallel redundant system depends largely on the

reliability of the failure rate of the PARALLEL BUS which is the only single
point failure.

ABB Via Luserte Sud 9 CH-6572 Quartino

White_Paper_Relibility_150506.doc 7
 The UPS parallel redundant chains, the static bypass switches and their
control electronics as well as the mains power lines are all redundant and
have therefore minor or even neglecting impact on the overall reliability.

We will perform some calculations for parallel redundant systems, by using

the following constants:
MTBFM = 50 [h] , this figure represents a good quality mains.
PBUS = 0.4 10 -6 [h-1]

Redundant Parallel Configuration Reliability (MTBF) Failure Rate ()

(1+1) redundant configuration 1250000 [h] ~ 0.8 10-6 [h-1]
(2+1) redundant configuration 830000 [h] ~ 1.2 10-6 [h-1]
(3+1) redundant configuration 625000 [h] ~ 1.6 10-6 [h-1]
(4+1) redundant configuration 500000 [h] ~ 2.0 10-6 [h-1]
(5+1) redundant configuration 420000 [h] ~ 2.4 10-6 [h-1]

Table 1 MTBF and Failure Rate of (n+1)-Redundant Configurations,

N=1,2,3,4 and 5

Conclusion:
In the single UPS chain (rectifier, battery and inverter) the reliability of the
UPS largely depends on the reliability of the inverter.
By introducing the static bypass switch i.e. a reserve mains power supply,
the reliability will increase by a factor of 6 if the mains MTBF is 50h (good
quality) and the MTTR of UPS is 6h. This reliability level is unfortunately not
sufficient, because it still depends substantially on the reliability of the raw
mains and on the quality of the after sales organization (reaction time,
traveling time, repair time etc.)
Modern critical loads have far higher reliability requirements and cannot
rely on the mains quality and on longer repair times.
To overcome the dependability from the raw mains, (n+1)-redundant
parallel UPS-configurations are suggested. The disadvantage of traditional
stand-alone (n+1)-redundant configuration is the relatively long repair time
of the UPS (typically 612h). By implementing modular, hot-swappable,
(n+1) redundant parallel systems the critical load will be completely mains-
independent; a faulty UPS-module may be replaced without the need to
transfer the rest of the UPS modules to raw mains (hot-swap).
Furthermore the replacement of the modules takes at most 0.5h, which
dramatically decreases the time to repair in comparison to traditional
parallel systems.

ABB Via Luserte Sud 9 CH-6572 Quartino

White_Paper_Relibility_150506.doc 8
 Below you will find an example where various UPS configurations are
compared and it is shown what impact the choice of the right
system/configuration might have on the reliability and availability:

Example:
Comparison of the Availability of a Traditional (1+1)-Redundant UPS
Configuration and a Modular (4+1)-Redundant UPS Configuration

Fig. 5 shows a block diagram of two redundant UPS configurations. The

system on the left represents a (1+1)-redundant configuration with
traditional stand-alone UPSs, whereas the system on the right side
represents a (4+1)-redundant configuration with modular hot swappable-
modular UPSs.

(4+1)-Redundant Configuration
(1+1)-Redundant Configuration (modular, hot-swappable)
(traditional stand-alone)

UPS 1 UPS 2 UPS 3 UPS 4 UPS 5

UPS 1 UPS 2 40kVA 40kVA 40kVA 40kVA 40kVA
120kVA 120kVA

Availability
Availability

LOAD
LOAD 160kVA
120kVA

Fig. 5 Block diagram of two redundant UPS configurations

AVAILABILITY (A) is an important parameter when evaluating the reliability

of UPS-configurations. A is defined as follows:

MTBFUPS
A = --------------------------------- . (E.7)
MTBFUPS + MTTR UPS
ABB Via Luserte Sud 9 CH-6572 Quartino
White_Paper_Relibility_150506.doc 9
 From equation (E.7) we see that the Availability of a UPS depends on the:

MTBF UPS = Mean Time Between Failures of a UPS configuration and

MTTR UPS = Mean Time To Repair of the UPS.

Table 2 shows the Availability Comparison of the UPS configurations in

Fig.5. Note that the MTBF UPS figures for the UPS configurations are taken
from Table 1.
Two cases are considered:

Case 1:
both UPS-configurations have the same Meant Time To Repair:
MTTRUPS= 6 [h]

Case 2:
The traditional stand-alone UPS configurations has:
MTTRUPS = 6 [h], whereas
Modular UPS configuration with hot-swappable modules has:
MTTRUPS = 0.5 [h]

Redundant-Configuration (1+1) Redundant-Configuration (4+1)

Case 1
MTBF 1250000 [h] 500000 [h]
MTTR 6 [h] 6 [h]
Availability 0,9999952 (5 nines) 0,9999888 (4 nines)
Case 2
MTBF 1250000 [h] 500000 [h]
MTTR 6 [h] 0.5 [h]
Availability 0,9999952 (5 nines) 0,9999990 (6 nines)

Comments:
Case 1: The Availability of (1+1)-Redundant Configuration is higher than
the Availability (4+1)-Redundant Configuration if the MTTR is the
same for both configurations.
This evident due to the fact that the MTBF of a (1+1)-Redundant
Configuration is higher than the MTBF of a (4+1)-
Redundant-Configuration
Case 2: The Availability of a (1+1)-Redundant-Configuration with longer
MTTR may be lower than the Availability of a (4+1)-Redundant-
Configuration with a shorter MTTR.

ABB Via Luserte Sud 9 CH-6572 Quartino

White_Paper_Relibility_150506.doc 10
Conclusion From the above cases it is shown how important the parameter MTTR is
for reaching high availabilities. If in one of the above redundant
configurations one of the ups is faulty, there will be no redundnacy left
(low-availability regime) and we need to therefore repair/replace the
faulty part/module as quickly as possible, in order to restore the
redundancy (high-availability-regime).

With ABB Modular UPS we can reach shortest MTTR and consequently
highest availabilities even if we parallel bigger numbers of modules.

[1] A. BIROLINI Reliability Engineering, Theory and Practice

ABB Via Luserte Sud 9 CH-6572 Quartino

White_Paper_Relibility_150506.doc 11