[AusNOG] Data Suburb

Thu Jan 5 15:39:38 EST 2012

Sorry - please note that the reason the COP is higher on B at lower temps is because unit "B" it has the capability of going free air (compressorless) and operate like a car radiator (love car analogies in data centres ;)).  Note that in cold conditions the COP is much, much higher at 100% capacity than at lower levels which would contradict Tony's previous statement.  So like I said before, it's complex story subject to over-simplification.

Cheers

[b]

From: ausnog-bounces at lists.ausnog.net [mailto:ausnog-bounces at lists.ausnog.net] On Behalf Of Bevan Slattery
Sent: Thursday, 5 January 2012 11:42 AM
To: Tony de Francesco; 'Luke Smith'
Cc: ausnog at ausnog.net
Subject: Re: [AusNOG] Data Suburb

Tony,

Coefficient of Performance @ Specified Load & Ambient Temperature

Capacity

10%

25%

50%

75%

100%

A

B

A

B

A

B

A

B

A

B

Ambient Temperature °C

0

6.18

14.67

25.96

33.98

39.71

5

6.18

13.08

19.85

16.33

18.24

10

3.78

5.53

8.88

10.55

12.98

15

3.09

7.25

4.56

7.46

6.25

6.81

6.75

6.39

7.18

20

2.13

6.17

4.02

6.06

5.82

5.68

6.33

5.77

5.97

25

2.00

5.18

3.84

5.17

5.19

5.00

5.44

4.99

5.11

30

1.84

4.67

3.44

4.33

4.50

4.40

4.92

4.19

4.35

35

1.70

3.72

3.52

3.77

4.79

3.73

4.6

3.60

4.09

I think it's more appropriate to provide a fuller picture than 1 example.  Here's two (2) aircooled chillers (water coolant) and their COP's at different levels.

It's not accurate to depict that they day "N" installation is that the units are at 100%.  Obviously, you know that's not true for various reasons.  Firstly, if you have an N+1 system then depending on the number of units you have you will be operating.  In fact at 100% in some facilities will have N+2 or even N+3 for efficiency, reliability and sometime acoustical reasons.  Units have variable speed drives, multiple staged compressors which make running at lower levels as efficient as operating at day "N".  However, no one in a serious DC space (that I know of) ever expects to run outdoor cooling plant at 100% load.  The sweet spot depending on the ambient conditions in between 20-75% capacity.  But this is capacity of a chiller, not utilisation of the DC floor.  Outdoor mech is a game of swings and roundabouts at each "stage jump" which get smoothed out more and more as the facility gets bigger (assuming it's common infrastructure).

Cheers

[b]

From: ausnog-bounces at lists.ausnog.net<mailto:ausnog-bounces at lists.ausnog.net> [mailto:ausnog-bounces at lists.ausnog.net]<mailto:[mailto:ausnog-bounces at lists.ausnog.net]> On Behalf Of Tony de Francesco
Sent: Thursday, 5 January 2012 10:48 AM
To: 'Luke Smith'
Cc: ausnog at ausnog.net<mailto:ausnog at ausnog.net>
Subject: Re: [AusNOG] Data Suburb

Thanks very much for that analogy as it precisely highlights one of the reasons why the "reduced efficiency at part load" argument gets so much traction with people, i.e. people think they understand the technology and find over simplified examples to justify their position.

If you will allow me, let me educate on how an air conditioning system (and not a V8 engine) responds to part load operation:

Let's look at an air conditioning unit that is designed for 1000kW of cooling, but is delivering only 200kW of cooling:

a)     Firstly when operating at a full 1000kWs of Load:
a.     Refrigeration System Power Use:           EER of 3.5, equating to around 286kW of absorbed power
b.    Air handling System Power Use:            (relates to 69,000m³/s of supply air based on an equipment ΔT of 12°C) equating to around 60kW of fan power
TOTAL = 346kW of power use or a pPUE of 1.346

b)    The same system at 200kW of Load:
a.     Refrigeration System Power Use:           EER of 8 (as the refrigeration COP improves at part load), equating to around 25kW of absorbed power
b.    Air handling System Power Use:            (relates to 14,000m³/s of supply air based on an equipment ΔT of 12°C) equating to around 1kW of fan power (as per the fan laws).
TOTAL = 26kW of power use or a pPUE of 1.130

So you can see in this example at 20% load utilisation, the pPUE (i.e. cooling system PUE) should be around 1.130 and increase to 1.346 as the load approaches 100%.

This correlation is often not the case with electrical systems, such as static UPS and DRUPS units, which tend to have lower efficiency at part load and better efficiency at full load (as per your V8 analogy).

However, the improvement in the cooling system PUE at part load generally offsets the reduction in electrical system PUE at part load and so the overall facility PUE remains static over the load utilisation range.

Hope this helps your understanding although please feel free to email me on or off list if you would like me to explain any of this in further detail.

Kind Regards

Tony de Francesco
Technical Director
email: tonyd at pue.com.au<mailto:tonyd at pue.com.au>
mob: +61 (0) 457 701 179
.....................................................
[cid:image001.jpg at 01CCCBB7.065C55B0]
________________________________
From: ausnog-bounces at lists.ausnog.net<mailto:ausnog-bounces at lists.ausnog.net> [mailto:ausnog-bounces at lists.ausnog.net]<mailto:[mailto:ausnog-bounces at lists.ausnog.net]> On Behalf Of Luke Smith
Sent: Thursday, 5 January 2012 11:16 AM
To: ausnog at ausnog.net<mailto:ausnog at ausnog.net>
Subject: Re: [AusNOG] Data Suburb

>  This statement has permeated the industry and led to people accepting lower levels of efficiency at part load in the hope that their PUE will improve as the load increases.  There is no technical reason why PUE has to suffer at part utilization, other than people accepting a facility design that allows this.
Applying an analogy... if this were the case then you should also expect the same economy from a 6.0l V8 as you get from a 2.0l four in the same car. But the world doesn't work like that.

Unless its *very* modular and you can totally turn off the bits you are not using then you simply can't expect a half used facility to have the same efficiency as a fully utilised one half the size. Going back to the analogy modern V8s have cylinder deactivation that saves some of the overhead, but they still waste energy spinning a few extra cranks and pistons about that they don't need 99% of the time.
--
Cheers,

Luke Smith
0419 671 006
The information contained in this email and any attachments may be confidential. This email and any attachments are also subject to copyright. No part of them may be reproduced, adapted or transmitted without the written permission of the copyright owner. If you are not the intended recipient, any use, interference with, disclosure or copying of this information is unauthorised and prohibited. If you have received this email in error, please immediately advise the sender by return email and delete the message from your system. All email communications to and from NEXTDC Limited are recorded for the purposes of archival and storage.
The information contained in this email and any attachments may be confidential. This email and any attachments are also subject to copyright. No part of them may be reproduced, adapted or transmitted without the written permission of the copyright owner. If you are not the intended recipient, any use, interference with, disclosure or copying of this information is unauthorised and prohibited. If you have received this email in error, please immediately advise the sender by return email and delete the message from your system. All email communications to and from NEXTDC Limited are recorded for the purposes of archival and storage.
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.ausnog.net/pipermail/ausnog/attachments/20120105/0527620d/attachment.html>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: image001.jpg
Type: image/jpeg
Size: 24527 bytes
Desc: image001.jpg
URL: <http://lists.ausnog.net/pipermail/ausnog/attachments/20120105/0527620d/attachment.jpg>