Welcome to my website


Archive
Follow up 1 - The power of the Raspberry PI
22 September 2012 22:23

This is a follow up post to the blog: The power of the Raspberry PI
So please go read that blog post first if you not already have done that.

Watch out boys, it is the new revision 2 RPI board for testing….. Don’t scratch the PCB.
Blog image

A couple of weeks ago I got the new Raspberry Pi Model B revision 2 board where they now has fixed the 1.8V power issue and done some other minor changes, so in this blog post I will test this new revision 2 board against the older Raspberry Pi Model B revision 1.
I was lucky that i have order an new board and then they announced the new board and some days later i got the new board in the mail box.

So here is what I will look at in this blog:

  •  How many mA do the new board use in idle and hardworking mode
  •  How much colder is the new board, now the 1.8V issue is fixed
  •  How hot does the board get when it is a closed case.
    This is because a number of people have asked me about this.
  •  How is it going with finding a replacement for the LDO’s

I will not in this blog go over the details about what has change on the revision 2 boards as this is already well covered on a lot other webpages, so I will limit this blog post to be the follow up items there is covered in my first blog.

Before taking readings of the new revision 2 board I will just recap the details from the testing of the revision 1 board so we have the details for both tests in this post.



Test 1
Work load: Idle mode
Board: Raspberry Pi Model B Revision 1 (Unmodified board)
in this test it is simple a revision 1 unmodified board and it is just power on with one SSH session to my pc so I can verify that the RPI is working.

Here is a photo of the test setup.
Blog image

Thermal image of the Raspberry Pi
Blog image


Power usage
Power rail
Volt
Amp
Watt
5.0V
4.960V
0.371725A
1.843756W

Heat profile (max values)
Heat location
Fluke T25 Thermal Imager
RG1 1.8V LDO
39.34C
RG2 3.3V LDO
55.78C
CPU/RAM
49.59C
LAN9512
57.56C



Test 2

Work load: Copying files over the network with SCP, omxplayer with mpeg4 movie and Stress test
Board: Raspberry Pi Model B Revision 1 (Unmodified board)

Thermal image of the Raspberry Pi
Blog image
Here you can see the RG1 is not very hot so it is not doing any work and it is the LAN9512 chip there is supply the board with the 1.8V

Power usage
Power rail
Volt
Amp
Watt
5.0V
4.901V
0.471345A
2.310061W

Heat profile (max values)
Heat location
Fluke T25 Thermal Imager
RG1 1.8V LDO
41.56C
RG2 3.3V LDO
61.16C
CPU/RAM
54.44C
LAN9512
60.28C



Test 7

Work load: Idle mode
Board: Raspberry Pi Model B Revision 2 (Unmodified board)
This test is the same as test 1 but this time it is with the new revision 2 board and as you can see it is power via the GPIO connected as I have done in all the tests, so the Polyfuse F3 do not limit the power.

Here is a photo of the test setup
Blog image

Thermal image of the Raspberry Pi
Blog image


Power usage
Power rail
Volt
Amp
Watt
5.0V
4.975V
0.371381A
1.847620W

Heat profile (max values)
Heat location
Fluke T25 Thermal Imager
RG1 1.8V LDO
42.94C
RG2 3.3V LDO
53.25C
CPU/RAM
48.72C
LAN9512
55.03C



Test 8

Work load: Copying files over the network with SCP, omxplayer with mpeg4 movie and Stress test
Board: Board: Raspberry Pi Model B Revision 2 (Unmodified board)

Thermal image of the Raspberry Pi
Blog image

Power usage
Power rail
Volt
Amp
Watt
5.0V
4.985V
0.464117A
2.313623W

Heat profile (max values)
Heat location
Fluke T25 Thermal Imager
RG1 1.8V LDO
47.66C
RG2 3.3V LDO
56.00C
CPU/RAM
53.78C
LAN9512
56.66C




Summary for test 1, 2, 7 and 8
Test description
RG1
RG2
CPU/RAM temp.
LAN9512 temp.
Total power usage
Test 1 – Raspberry Pi Model B Revision 1
Workload: Idle
39.34C
55.78C
49.59C
57.56C
1.843756W
Test 2 – Raspberry Pi Model B Revision 1
Workload: heavy stress tests
41.56C
61.16C
54.44C
60.28C
2.310061W
Test 7 – Raspberry Pi Model B Revision 2
Workload: Idle
42.94C
53.25C
48.72C
55.03C
1.847620W
Test 8 – Raspberry Pi Model B Revision 2
Workload: heavy stress tests
47.66C
56.00C
53.78C
56.66C
2.313623W

Power usage for test 1 and test 7 compared.
If we take the test 1(1.843756W) - test 7(1.847620W) then we get that the new board is using 0.003864W more than the old board, but that is so small a difference that this is below what measurements error there is with my test setup and measurement gear.

Power usage for test 2 and test 8 compared.
If we take the test 2(2.310061W) - test 8(2.313623W) then we get that the new board is using 0.003562W more than the old board.
So as you can see, there is absolute no change in how much power the two boards are using and this is making sense because fixing the LAN9512 error has just moved the 1.8V converting from one LDO in the LAN9512 chip to and external LDO on the board and because they both are LDO’s they will both be equally poor in converting the 3.3V to the 1.8V rail.

Temperature of the LAN9512 chip
if we look at test 1(57.56C) and test 7(55.03C) then the new board is 2.53C cooler and for test 2(60.28C) and test 8(56.66C) it is 3.62C, so the fixed LAN9512 board is giving a lower temperature of around 2.53/3.62C and that is good…. But when I have fixed my first revision 1 board, I then was seeing bigger change with 3.97C at idle and 4.84C when working hard.
But it is nice to see the fixed board now is 2.53C cooler at idle and 3.62C at hard work on the LAN9512 chip.

Temperature of the CPU/RAM chip
It seems that CPU/RAM is just under 1C cooler on my new revision 2 board and because this is so small an change then it can maybe just measurement errors to count for this but it can also be that the 1.8V rail is now an small bit lower then it was on the revision 1 board and this count for less heat.
But I will say that there is no change and it is just measurement errors.

Temperature of the RG1 LDO
as you can see the temperature is now 3-6C higher on the new revision 2 boards and this is because the RG1 is now doing some work and not just sitting idle on the board as on revision 1.

Temperature of the RG2 LDO
It is an surprise to me that RG2 is 2-5C cooler on the new board as this LDO is doing the same mA load on both boards so I think the change in temperature is because the two LDO chips are not the same, on the revision board the 3.3V LDO is an SE8117T33 and on the new board it is an ON 17-33G.
So if this change is because they now have selected a better part then it is very nice.




Heat profile for board in free air VS in a closed case

I am only having a case from modmypi.com so this is the case I will use for this test.
To make sure I got some stabile temperatures in the case testing, I have then run each test step for more than one hour and this is because the Raspberry Pi and the case has more mass and there is more to heat up and it toke up to 40+ minutes before the temperature got stabile so it fits well by running the tests longer than in the other tests.
When taking the temperature of the board on the case I then just remove the top part of the case and as fast as I can I take the thermal image because I can see on the Fluke thermal image screen that as soon as it get free air in and the top is removed then I can see the temperature is dropping and it is slowly getting down to the same temperature as if the board was in free air.
Blog image
The case is placed on the black risers on my wood board so there is the same free air underneath as there has been for all the other tests with the board in the free air.



Test 9

Work load: Idle mode
Board: Raspberry Pi Model B Revision 2 (Unmodified board)

Closed case test
Blog image
Here you can see the temperature on the top side of the closed case and it is easy to see the center gets hot because the CPU/RAM is just below and also the heat is coming out via the holes in the case.

Board in the case where the top has just been removed.
Blog image
It seems that each part is getting 3-5C hotter in the case then when it is in free air. I will have guessed that it was more but it nice it not more than this in idle mode.



Test 10

Work load: Copying files over the network with SCP, omxplayer with mpeg4 movie and Stress test
Board: Board: Raspberry Pi Model B Revision 2 (Unmodified board)

Closed case test
Blog image
now when the Pi is working hard it is easier to see the heat coming through the case.

Board in the case where the top has just been removed.
Blog image
When working hard and in a case then it seems to get 5-8C hotter on each part.



Summary of the case test
Test description
RG1
RG2
CPU/RAM temp.
LAN9512 temp.
Test 7 – Raspberry Pi Model B Revision 2 (Free air)
Workload: Idle
42.94C
53.25C
48.72C
55.03C
Test 8 – Raspberry Pi Model B Revision 2 (Free air)
Workload: heavy stress tests
47.66C
53.10C
53.78C
56.66C
Test 9 – Raspberry Pi Model B Revision 2 (In case)
Workload: Idle
48.53C
56.91C
53.56C
58.97C
Test 10 – Raspberry Pi Model B Revision 2 (In case)
Workload: heavy stress tests
54.00C
64.22C
60.13C
62.66C

It is easy to see that the board is getting a bit hotter in the case then when it is in free air, but it don’t seems to be so much as I will have guessed it will have been.
So it don’t seem to be something to be any concern about and don’t give issues, but I think it still is a good idea to add some heat sinks to the board if you have some.

 

How is it going with finding a replacement for the LDO’s
For now it is not going so well because all the switch mode converters I have tested is not that efficiency at the low mA level that the Raspberry Pi is using and I have test an number of the ones from eBay and some other ones that have.
But I am in the process of getting some PCB boards made for an switch mode converter that I think will do the job but it is an very small part from TI and I have tried to make my home made board but it was not good enough so have to get this board made from an PCB manufacture and in the time of writing this is on the way to me, so I hope I will get it soon and can start test it.
I will write about how this goes in another blog post at a later time when I have the details ready.

 

So this was a follow up post to test the new board.

Tooms @ 22 September 2012 22:23 | Comment | Direct link


Arkive list
2014 - August
2014 - April
2014 - March
2014 - February
2014 - January
2013 - December
2013 - November
2013 - October
2013 - August
2013 - June
2012 - December
2012 - September
2012 - August
2012 - May
2012 - April
2011 - November
2011 - October
2011 - September
2010 - May
2010 - March
2010 - January
2009 - December
2009 - November
2009 - October
2009 - August
2009 - July
2009 - May
2009 - April
2009 - March
2009 - February
2009 - January
2008 - December
2008 - November
2008 - August
2008 - January
2007 - December
2007 - November
2007 - October
2007 - August
2007 - July
2007 - April
2007 - March
2007 - February
2007 - January
2006 - December
2006 - November
2006 - September
2006 - August
2006 - July
2006 - May
2006 - April
2006 - March
2006 - February
2006 - January
2005 - December
2005 - November
2005 - October
2005 - September
2005 - August