Bitcoin Forum
November 20, 2017, 08:00:21 AM *
News: Latest stable version of Bitcoin Core: 0.15.1  [Torrent].
 
   Home   Help Search Donate Login Register  
Pages: « 1 2 3 4 5 6 7 8 9 10 11 12 13 [14] 15 16 17 18 19 20 21 22 23 24 25 26 27 28 »
  Print  
Author Topic: Official Avalon Technical Support Thread  (Read 48086 times)
server
Hero Member
*****
Offline Offline

Activity: 815


~\/~


View Profile
May 23, 2013, 08:39:40 AM
 #261

Any ideas why it was set to 5 min and not something shorter?

Maybe to give it some time to cool down (?)
1511164821
Hero Member
*
Offline Offline

Posts: 1511164821

View Profile Personal Message (Offline)

Ignore
1511164821
Reply with quote  #2

1511164821
Report to moderator
1511164821
Hero Member
*
Offline Offline

Posts: 1511164821

View Profile Personal Message (Offline)

Ignore
1511164821
Reply with quote  #2

1511164821
Report to moderator
1511164821
Hero Member
*
Offline Offline

Posts: 1511164821

View Profile Personal Message (Offline)

Ignore
1511164821
Reply with quote  #2

1511164821
Report to moderator
Join ICO Now Coinlancer is Disrupting the Freelance marketplace!
Advertised sites are not endorsed by the Bitcoin Forum. They may be unsafe, untrustworthy, or illegal in your jurisdiction. Advertise here.
kaerf
Hero Member
*****
Offline Offline

Activity: 624


View Profile
May 23, 2013, 08:46:21 AM
 #262

i believe if your vardiff is set high enough, then you might not submit a share every 5 minutes. this might happened due to random variance...but the higher your vardiff, the more likely you'll hit the 5 minutes mark.


i have a script that poles the cgminer API and reboots a power switch if my hash rate drops to a very low rate for a minute. i think the cgminer stats are more accurate than the last accepted share that the cronjob uses.
fpgaminer
Hero Member
*****
Offline Offline

Activity: 560



View Profile WWW
May 23, 2013, 09:57:20 AM
 #263

Quote
i believe if your vardiff is set high enough, then you might not submit a share every 5 minutes. this might happened due to random variance...but the higher your vardiff, the more likely you'll hit the 5 minutes mark.
cgminer-monitor is looking at the hardware's Diff1Shares, not shares submitted to the server, so difficulty shouldn't matter.

WinTame2012
Sr. Member
****
Offline Offline

Activity: 315


Official sponsor of Microsoft Corp.


View Profile WWW
May 23, 2013, 12:27:00 PM
 #264

Nice tuning. How did you do this?

Check the monitoring script

/usr/bin/cgminer-monitor

which is called by cron
Thanks for detailed guide.

May the WinTame Power be with you!
http://wintame.com
Sitarow
Legendary
*
Offline Offline

Activity: 1680



View Profile
May 27, 2013, 03:23:31 AM
 #265

Enermax has a really good reputation.
Can you name the exact type

ENM750EWT-S 80plus Gold

Are you running the miners at 300? If so this PSU might not be able to take such high level of load for a long time (at 300 the power consumption should be close to 710 to 720 watts...possibly less if your psu is 80+ platinum). It is also possible that your PSUs were in a "bad batch" (assuming you bought them from the same place all at once)

I got a Corsair AX860 but I've yet to install it. From what I read these are good PSUs.

I upgraded to Corsair AX860i and they are have had no problems with up time.

I have posted details about this here.

https://bitcointalk.org/index.php?topic=156323.msg1910169#msg1910169
allinvain
Legendary
*
Offline Offline

Activity: 2338



View Profile
May 27, 2013, 04:20:08 AM
 #266

Nice. I too grabbed a AX860 (not the i version as I don't need the fancy pants control feature) but have yet to put it into action - will do so when batch delivers...sometime this year hopefully  Tongue

I must also correct my statement there. At 300 the system consumes 625 to 630 watts at 120v. I finnally got around to putting a "kill-a-watt" type power meter on my unit and that what it says. Keeping in mind that the batch 1 units came with Antec 650 Watt units, I am quite amazed my unit's PSU has lasted this lost at what is pretty much almost 100% capacity.


Sitarow
Legendary
*
Offline Offline

Activity: 1680



View Profile
May 27, 2013, 05:31:00 AM
 #267

Nice. I too grabbed a AX860 (not the i version as I don't need the fancy pants control feature) but have yet to put it into action - will do so when batch delivers...sometime this year hopefully  Tongue

I must also correct my statement there. At 300 the system consumes 625 to 630 watts at 120v. I finnally got around to putting a "kill-a-watt" type power meter on my unit and that what it says. Keeping in mind that the batch 1 units came with Antec 650 Watt units, I am quite amazed my unit's PSU has lasted this lost at what is pretty much almost 100% capacity.



Consider posting what your voltage is at that effects the power draw in watt units.

I am below 600 watts on a hot day 28-31 Ambient temp at 120v with the 860i
Syke
Legendary
*
Offline Offline

Activity: 2436


View Profile
May 29, 2013, 06:19:32 AM
 #268

After I switched to a corsair GS800, cgminer never restarted Smiley

It seems that the original power supply (Antec Earthwatt 650) is a bit inadequate

Perhaps, but mine is holding up remarkably well. 300 MHz clock speed and current uptime has been 6 days..woohoo! Smiley

I agree the EarthWatts 650 isn't enough. Mine just died. The Avalon is just too much for it.

Buy & Hold
abracadabra
Hero Member
*****
Offline Offline

Activity: 938



View Profile
May 29, 2013, 03:12:55 PM
 #269

One of my Batch #2 Avalons is having trouble.  Every so often (~15 minutes on average) I get a long burst of these kinds of messages in the System Log (dmesg):

Code:
May 18 06:42:59 avalon1 kern.err kernel: [ 3872.320000] usb 1-1: clear tt 1 (8040) error -71
May 18 06:42:59 avalon1 kern.err kernel: [ 3872.330000] ftdi_sio ttyUSB1: failed to get modem status: -71
May 18 06:42:59 avalon1 kern.err kernel: [ 3872.340000] usb 1-1: clear tt 1 (8040) error -71
May 18 06:42:59 avalon1 kern.err kernel: [ 3872.350000] ftdi_sio ttyUSB1: failed to get modem status: -71
May 18 06:42:59 avalon1 kern.err kernel: [ 3872.350000] usb 1-1: clear tt 1 (8040) error -71
May 18 06:42:59 avalon1 kern.err kernel: [ 3872.460000] ftdi_sio ttyUSB1: failed to get modem status: -32

While that is occuring, all mining stops.  Either it eventually goes back to normal, or I see a message in the System Log saying that ttyUSB1 was disconnected, and then reconnected as ttyUSB0 (and vica-versa).  cgminer dies if that happens, and restarts.

I have another Batch #2 running smoothly.  So far, I have checked all of the connectors on the TP-LINK, PDU, and control unit.  I replaced the PSU, left the cover off, and left the PSU outside of the unit.  No joy.  All of the version numbers match my other Batch #2, so there are no differences in software/firmware.  One odd thing is that I notice mention of a usb hub on the wiki page, but there is no hub inside either of my black Avalons.  Is the hub built into the control unit?

I can only think of two other things to try.  Backup the working Avalon, and flash it to this Avalon, in case there's something weird in the software.  Or, replace the TP-LINK with a RasPi and hope that the control unit isn't at fault.
Check the usb Cable first dude.

If i were you i would remove p5 jumper, use USB powered hub and connect usb cable to batch one unit easiest or to linux PC to check. But first change usb cables and try other ports of internal usb hub if batch2 still has it



Batch #2 do not have a P2 jumper on the fpga controller board.  They also do not have an external USB hub. It appears the v1.5 fpga controller boards in the batch #2 avalons have a renesas usb hub chip included to attempt to account for the usb stability problem with the tp703 See: https://forum.openwrt.org/viewtopic.php?id=39956

If anyone from Avalon reads this thread, could you please provide info on whether something can be changed on the fpga board so that it won't backfeed power on the usb port?  It's been mentioned that attempting to connect the fpga board usb port to a desktop could cause problems since the fpga usb port is still powered.

What do J1 and J2 do?

Is it possible to JTAG firmware/bitstream to the fpga? Instructions?

TIA

mrb
Legendary
*
Offline Offline

Activity: 1358


View Profile WWW
May 30, 2013, 02:31:01 AM
 #270

abracadabra: shunting J1 does precisely that: it powers the USB type B port on the control unit. Remove the J1 jumper when connecting the control unit to a host computer.

As to J2, I don't know its purpose.
intron
Sr. Member
****
Offline Offline

Activity: 428


- electronics design|embedded software|verilog -


View Profile
May 30, 2013, 09:22:18 AM
 #271

Q) What is this thread for?


A) This thread is for answering TECHNICAL questions only about the Avalon ASIC Bitcoin Miner

Might be slighty off-topic, but still rather TECHNICAL:)

Is someone with an Avalon board willing to put a scope probe
on the REPORT_N/REPORT_P lines (pins 17 and 18) and observe
the timing of the signal? Interrested in:

- Up-going edge schape (it's a pulled-up pinch/release line,
   could have RC characteristics).
- Clock period and duty cycle of 1 bit period.
- Length in bits of the sent data.

Will pay a BTC 0.10 bounty for the first, clear scope screen hardcopy.

Thx in advance,

intron


bits can go here: 1intronttqV6J1PLLeQ3X5i4PxyhpE1fP
fpgaminer
Hero Member
*****
Offline Offline

Activity: 560



View Profile WWW
May 30, 2013, 11:27:06 AM
 #272

Batch #2 do not have a P2 jumper on the fpga controller board.
I have Batch #2 units, and my controller boards do still have a P5 jumper.  It's located just behind the USB type B receptacle.  As shown in this picture.

abracadabra
Hero Member
*****
Offline Offline

Activity: 938



View Profile
May 30, 2013, 01:12:39 PM
 #273

Batch #2 do not have a P2 jumper on the fpga controller board.
I have Batch #2 units, and my controller boards do still have a P5 jumper.  It's located just behind the USB type B receptacle.  As shown in this picture.

Ah!.. I seem to remember you had a "silver" case avalon.  I guessing you got the last of the <v1.5 batch#1 leftover parts.

darkip
Full Member
***
Offline Offline

Activity: 160


View Profile
May 30, 2013, 08:05:10 PM
 #274

A tip for anyone struggling with the USB modem/dropout issue:

Following discussions on IRC (Freenode #avalon) a few of us have seemed to have solved the issue by disabling the wifi part of the TP-Link.
WinTame2012
Sr. Member
****
Offline Offline

Activity: 315


Official sponsor of Microsoft Corp.


View Profile WWW
May 30, 2013, 09:04:27 PM
 #275

A tip for anyone struggling with the USB modem/dropout issue:

Following discussions on IRC (Freenode #avalon) a few of us have seemed to have solved the issue by disabling the wifi part of the TP-Link.
Thanks for letting community know. Could you give us simple step-by-step guide how to do this?

May the WinTame Power be with you!
http://wintame.com
darkip
Full Member
***
Offline Offline

Activity: 160


View Profile
May 30, 2013, 11:50:13 PM
 #276

Thanks for letting community know. Could you give us simple step-by-step guide how to do this?
I just went to the network section of the web ui and clicked the disable button for the wifi.

Others have gone as far as disabling the interface, presumably something along the lines of:
Code:
ifconfig wlan0(?) down

Although that'd only last until reboot. A permanent settings change is probably done by clicking the disable button.
silverserpent
Member
**
Offline Offline

Activity: 83


View Profile
May 31, 2013, 06:28:07 AM
 #277

Are these temps normal? Seems like F3 is slow...

StatusMHS5s   MinerCountAsicCountFrequency   Fan1   Fan2   Fan3   Temp1   Temp2   Temp3   NMW   Firmware
Alive 74377.12   24   10               282              1440   2280   2160       22       42        44           80   20130419
drakness
Jr. Member
*
Offline Offline

Activity: 33


View Profile
May 31, 2013, 12:37:20 PM
 #278

silverserpent: That looks normal, to me.  Matches mine.
silverserpent
Member
**
Offline Offline

Activity: 83


View Profile
June 01, 2013, 02:38:49 AM
 #279

Waveform Type - Stepped Approximation to Sine Wave


is this OK for AVALON?

It all depends on the power supply. If the power supply has APFC (Active Power Factor Correction) it is highly recommended to use a UPS that outputs pure sine wave AC.

I believe that PFC power supplies should use true sine wave UPSs. The reason for this is that PFC power supplies require a much faster changeover time when the battery kicks in, and that due to the design of stepped sine wave UPSs, they were not always fast enough to satisfy the power supply. This could lead to the PSU shutting down when the battery tried to kick in. A stepped sine wave UPS wouldn't "damage" the PSU, but that it simply might not work, and that if it DID work, it might not "always" work. It all depends on the specific power supply. The UPS shutting down has happened to me when using a non sine wave UPS with a APFC power supply.

Most PSUs that you'd want to buy these days will have APFC.

The problem that I see here (and this has been discussed in another thread I believe) is that it takes a LOT of juice to keep an Avalon running so you're going to need a beefy UPS and even so most reasonably priced UPSs (1440VA) that you can attach to a 15 amp circuit are only going to be capable of keeping it alive for maybe a few minutes at full load (even less if you run it clocked at 300).

IMHO it's worth while to have a least one on a UPS so that if there is a brief blackout at least one unit is mining and you're not losing _as much_ money (in lost revenue).

Now as for what UPS to choose. Personally I recommend the APC SmartUPS series. I've been using them for a long time and they are very reliable units. The SmartUPS series output pure sine wave by the way.


I used to swear by APC too but lately I have heard a lot of complaints about APC's AVR - esp w/ high-end EQ, anybody else?

might wanna check these too : CyberPower CP1500PFCLCD  1500VA  /  900W   

this guy did some testing : Pure Sine Wave Or Not?  http://www.youtube.com/watch?feature=player_detailpage&v=4QlF5vJlXaA

this thing has smart AVR w/ pure sine wave, lcd, green tech...
I just got one, the CyberPower rep at a show said it's good to have about 50% over actual power consumption and it was good deal here : $199 (free shipping)
http://www.amazon.com/CyberPower-CP1500PFCLCD-Sinewave-Compatible-Mini-Tower/dp/B00429N19W/ref=sr_1_1?ie=UTF8&qid=1370053696&sr=8-1&keywords=cp1500pfclcd+ups+1500va+900w+pfc

please give me some feedback if your aware of anything else
solari
Member
**
Offline Offline

Activity: 82


View Profile
June 06, 2013, 02:06:46 PM
 #280

Does anyone know the precise model number of the fans the Avalon units have in them stock or a possibly good replacement for them?

NMB-MAT 7
Model 4715KL-04W-B56
Minebea-Matsushita Motor Corporation
Pages: « 1 2 3 4 5 6 7 8 9 10 11 12 13 [14] 15 16 17 18 19 20 21 22 23 24 25 26 27 28 »
  Print  
 
Jump to:  

Sponsored by , a Bitcoin-accepting VPN.
Powered by MySQL Powered by PHP Powered by SMF 1.1.19 | SMF © 2006-2009, Simple Machines Valid XHTML 1.0! Valid CSS!