Topic: My review of Avalon 841 . - page 6. (Read 19331 times)

Holy moly, these are some nice new info!

which  is a twofold issue

A) a few more dollars to run at 76f vs 67f

B) a cap  of how hot  is safe?

So I will  next try an 86 f  test  which should  be max safe power cap


67f = 1366 watts
76f = 1425 watts
86f =    ?    watts





Once I get  a 86f test in.

I will attempt  to go to 90- 91 f    which is over the recommended 86f

I have ability  to adjust and I have  some areas where I can run under 86f  at all times.

If you do not have that ability  at best you spend more power per unit
At worst  you may kill unit.

That's some nice additional information for everyone, thanks Phil.
I only had one environment in which I made all the tests for the AvalonMiner.

It seems that the power consumption number is much related to the ambient temperature value.

@ yankees

well  a 60 watt power increase  when I went from 67f to 76f translates to  60 x 24 = 1440  or 1.44 kwatts at 10 cents  that is 14.4 cents a day  extra.

Here in NJ  May 1 to Oct 1 would mean 5 months  where temps will boost power  so 150 x 14.4 cents = $21.60 extra to operate it at the array

So the power boost   means it is  not unacceptable  since the other 7 months  it will drop.

I can make the gear work  at my array with buysolar.  Mostly we have  an excess of paid for gear.

The 2 year warranty has a lot of value.  and  I figure  5 warmer months = only 21.60 per machine.

I will need to test at 86f  so I  have turned on more gear in My garage to get an idea  of what the machine uses at 86f





If I test beyond 86f  it is as a service to community but for anyone running the gear in a 87f room or 31c room  it would be running gear  in a room too hot.





https://i.imgur.com/ztkRMlm.png






2663
root
cgminer --lowmem --avalon8-fan 60-100 --avalon8-voltage-level-offset 0 -o stratum+tcp://pool.ckpool.org:333 -O 16yLHLoeyuCLPMXkVpC3gyrRYvwRGwjKJr.100:x -o stratum+tcp://mmpool.org:3333 -O philipma1957.A:1234 -o stratum+tcp://us1.bitminter.com:3333 -O philipma1957_west2:x --api-allow W:127.0.0.1 --api-listen --avalon8-voltage-level-offset -2 --real-quiet
0%
0%



the above is from the processes tab  and it shows  I have -2 offset working

So

 67f = 1366 watts
 76f = 1425 watts
 86f =  yet to test.

Phil  your solar array is fan cooled  and you put in some heat shielding but if I recall  you run  2 to 4 degrees cooler then air temps.

So if you  are 90f outside temp the array is 87f
That would mean you should not run the Avalon 841  once you get to 90f .

@Haggs

I set it and I am running at -1  edit all tests below are at -2 offset.

I will post some results in an hour.


I finally got it setup in a way I am happy with for quality tests.



https://www.youtube.com/watch?v=sUaAQbRwd2Y

https://youtu.be/GZh6DSbYK5s

https://youtu.be/imdGUpVlre8

______________________________________________________________________________


cool room tests 67f   gave me

1366 watts and 13603 hash on -2 off set  or 0.1000419 watts a gh


https://youtu.be/1UU_6Zldsww

_______________________________________________________________________________ __

warmer room test 76f gave me

1425 watts and 13603 hash on -2 offset or 0.104756 watts a gh



https://youtu.be/9C1pMljvtwE

___________________________________________________________________
next test I will try to get to  86f or 30c  this is the max room temp allowed for




so a 67f room get really close to spec in terms of watts per gh
a 76f room is 4% over spec in terms of watts per gh

in both cases on -2 setting   hash  rate is 4.5 % over spec

2 year warranty is nice real nice.

Yep that is correct.

The cgminer command in the 'More Options' field is very important, since the voltage offset function doesn't work without doing so, for some reason.

thanks  I will go back and do this

please  see if this looks good

---

Moderator's note: This post was edited by frodocooper to reduce the size of the embedded image.

did you see my comment, you can't just set the -2 in the web UI, that doesn't change the undervolt, you have to include the command line option in the more options line on the web UI.

@ yankees  yes   so far it is doing that.


 I will  test with multiple  meters and psu's


but 1469/13700 = 0.107  watts a gh in a fairly cool room  under 80f

this is  a disappointing result, but I will keep testing

I may need to upgrade firmware etc.

I want to figure how cold do I need to get  best watts per gh.

So when you are at

30/77  power is 100 extra watts


vs

25/70

correct?

Maybe a poor unit.

I opened my garage  door today  as it is 53f

So my garage dropped 10 f

from  82 f  to 72 f

power dropped to

1469 watts

13703.49  hash

Temperature©

25 / 70

so temp drop lower  power by a lot of watts on  -2 -1 0 off set

---

Moderator's note: This post was edited by frodocooper to reduce the size of the embedded image.

Wow, that power consumption is quite high phil!

room was warm  seems there is a sensitivity to heat.

I will play with it some more tonight

I am in my den and will open sliding door to get temps lower then 82f

I opened my sliding door room temps dropped to 79f

the intake on the miner reads  27c/72c

it was 31c/77c

https://i.imgur.com/Hhk6J8c.png


watts on this meter dropped to 1474  they were 1509  so the temp drop dropped watts this test is -1 offset.  using delta 2400 watt plat psu

so room temp of 77  with unit temp of 26/71 hash of 13431 watts at 1470  17 minute    -1 offset  
"   "          "        74  with unit temp of 26/71  hash of 13490 watts at  1470 21 minutes -1  offset

30 minutes into the -1 offset and   I closed sliding door.

unit temps are now going to rise

30/77  hash of 13545     watts at 1520   30 minute mark.


going to test at 0 offset

Nice to see some testing.  

15 mins is quite a short test period.

It looks like that your hash rate should most likely stabilize averaging more than the 13.2 number.

I'm surprised about the 1.5kW power consumption, since my highest measured number was 1400W (voltage offset 0, and hash rate 13.52Th/s).

  New rasp pi has arrived

It works the sd cards were good.

the oem pi was a dud.  I did a video

showing the dead on and the good one.

I have this running on -2 offset using a  delta 2400 watt plat and I am getting very high power use

1523 watts at 13229 gh

No image just a link to a screen shot

 https://i.imgur.com/TRgGOeH.png

If its the same case the the 741 it will overheat the RPi in a warm environment so be cognizant of that.

I got 54 RPis with my last 1000 order from Canaan and I had 3 bad ones out of that batch.

Serves me right to not have a second one on hand.  Still waiting for it.  My UPS guy comes from 2 to 6pm on paper  but in reality mostly from 5 to 6:30.

This does show why parts and service are really important.
MY 'free' sample came with
1 Avalon 841
1 rasp pi in a case with a micro sd card
4 usb cables
4 dongles
1 5 pin cable

A rasp pi for me is a 1 day wait from amazon  as no local store carries it.
So to anyone that wants to get the avalon  especially 4 or more  have 2 rasp pi's makes lots of sense.
While I type I will post the unboxing I did yesterday.

it will be on you tube soon.

Your video will be live at: https://youtu.be/lyHSeEOYHr0



I want to make note

a rasp pi 3 can be found on amazon
a power brick  for the rasp pi on amazon
usb cables on amazon
even the 5 pin cable is available on amazon

the usb to uart dongle is  an avalon part.  they sent four to me which means I may be able to help out with that.

I'm glad to know I'm not the only one to get a dead pi from Canaan. I got 6 from them (Don't judge me, I wanted the case)
1 DOA, 1 had a bad mem-card

Happy Testing Phil!

Well I tested the rasp pi 4 ways it is dead.  so I am waiting on amazon to send a replacement one.