Avalon 841 PG 3

[ccgllc]

Most of my Avalon 841`s have a "15" under the PG column to the far right (hovering shows that as "Power Good").  One has a "3" and is underperforming.

Should I just swap the power supply?

Update:  I have spare PSUs, so went ahead and swapped it out.  Problem appears to be solved.  This is also the unit that was running about 4C higher in temp prior to the PG number changing.

Of course, will continue to monitor it.

[philipma1957]

Nice find This will be helpful to some of us.

Also a good feature put in by Avalon.

I gave you a merit point for finding it and posting it.

[ccgllc]

Thanks... lets hope it holds as stable.

FWIW - That particular unit was using an Antminer APW3++ PSU.

[HagssFIN]

Yeah, the PG value is usually something like 15 if your PSUs are perfectly ok
and it can be very low if there are some psu issues.

[ccgllc]

2 hours later it seems to be doing fine.  Its one of my slower units at 13.2 with -1 voltage.

[philipma1957]

2 hours later it seems to be doing fine.  Its one of my slower units at 13.2 with -1 voltage.

still above 13.0 spec




--avalon8-voltage-level -1    are you doing this setting or

--avalon8-voltage-level-offset -1




all of these   below reduce hash a bit and temps and watts a lot



--avalon8-voltage-level -1
--avalon8-voltage-level -2
--avalon8-voltage-level -3
--avalon8-voltage-level -4

they seem to work far better

then

--avalon8-voltage-level-offset -1 or -2

[ccgllc]

Currently doing --avalon8-voltage-level-offset -1

[philipma1957]

Currently doing --avalon8-voltage-level-offset -1

this should be better


--avalon8-voltage-level -1



especially so in hot rooms.

[ccgllc]

My results seem to differ from that.  I'm running about 2C cooler and faster using -1 offset.

I'm running short test (like 10 minute ones) in order to minimize outside air temp changes impacting the results, but so far, its repeatable.

[philipma1957]

My results seem to differ from that.  I'm running about 2C cooler and faster using -1 offset.

I'm running short test (like 10 minute ones) in order to minimize outside air temp changes impacting the results, but so far, its repeatable.

which firmware?

I have march 6? i think

[ccgllc]

which firmware?

I have march 6? i think

mm:  8411803-14b0f10

May 4th firmware

[philipma1957]

which firmware?

I have march 6? i think

mm:  8411803-14b0f10

May 4th firmware

This is why I ordered a second setup.

I want to be able to easily cross compare firmware.

Based on your info the May 4th changed something. You get different results then the March 6th firmware.

If I run on completely separate rasp pi’s I can sort out what the differences are.

[frodocooper]

mm:  8411803-14b0f10

May 4th firmware

There is no MM firmware that is dated May 4, 2018. The MM firmware that you have (version 8411803-14b0f10) is dated March 6, 2018.

[ccgllc]

mm:  8411803-14b0f10

May 4th firmware

There is no MM firmware that is dated May 4, 2018. The MM firmware that you have (version 8411803-14b0f10) is dated March 6, 2018.

That is the May 4th I was referring to.

[NotFuzzyWarm]

Ah. As it says to the left of May 4 -- that is your local time...

[ccgllc]

Ah. As it says to the left of May 4 -- that is your local time...

Ops.  OK, go with the kernel version then.