Avalon ASIC users thread - page 74. | Bitcointalksearch.org

aneutronic

full member

Activity: 175

Merit: 100

Quote from: ste0024 on September 27, 2013, 07:15:18 AM

just set @ 350MHz

what is it --real-quiet

That option minimizes log output. https://github.com/ckolivas/cgminer/blob/master/README

Try adding --avalon-auto if not already there in the same option box... this lets cgminer choose the optimum freq, you may see an increase.

ste0024

sr. member

Activity: 462

Merit: 250

just set @ 350MHz

what is it --real-quiet

aneutronic

full member

Activity: 175

Merit: 100

Quote from: ste0024 on September 27, 2013, 04:50:36 AM

Quote from: jddebug on September 26, 2013, 10:09:00 PM

Quote from: ste0024 on September 26, 2013, 07:40:04 PM

Quote from: aneutronic on September 23, 2013, 06:51:48 AM

Quote from: ste0024 on September 23, 2013, 06:43:47 AM

i m sure to understand but i can read

Not Electrically Conductive:
Arctic Silver 5 was formulated to conduct heat, not electricity.

http://www.arcticsilver.com/as5.htm

good paste or not ?

Correct, as5 is not conductive, the only question I have regarding as5 is that it is capacitive unlike arctic silvers mx2 or mx4.
I don't know if that would be an issue in this application so I chose mx4 over the as5.

thermal paste change and there needed.

the thermal paste used by avalon Were dry. would no longer have anything.

Temp 2 before 69 or 70 °
Temp 2 after 58 to 62 °

Temp 3 before 60 or 62 °
Temp 3 after 50 to 52 °

the ventilos not running any more. 2500 instead of 3000rpm

think has tighten the screws that hold the radiators. on mine there was a lot of screws loosen and poor contact and poor heat disipation.

after having tighten the plate is hot

Conclusion: if you want your living avalon still a lot of time,

change the thermal paste.

for 4 modules I have used 60gr of artic silver 5. there was just enough.

I put 6 hours to do everything. disassembly and reassembly test

Now it runs at 110 instead of 108 gh - less error HW

that happiness

and all this with Firmware Version => 20130723

next step update Firmware

before

after

Did you take any pictures as you did the paste replacement?

no but you can found some picture here.

https://bitcointalksearch.org/topic/m.2687268

and

https://bitcointalksearch.org/topic/m.3216051

Nice results with the arctic silver 5! Cool

Are you running --avalon-auto or just set @ 350MHz ?

You should also notice an improvement with the firmware upgrade, highly recommended.

aneutronic

full member

Activity: 175

Merit: 100

Quote from: allinvain on September 26, 2013, 11:23:35 PM

You're probably right. The original PSU works just fine with the third module disabled, so it's likely that just a part of that module is damaged. When I get a chance I will take it apart and inspect the module. It would suck if it's the backplane that's damanged as two modules connect to one backplane. I must admit I'm not a big fan of the batch3 design. As of now the unit is hashing fine with two modules and with the original PSU in place.

I'm thinking of selling the unit. I am getting tired of babysitting these units.

If anyone wants to make me an offer for the unit (bear in mind that at the moment only two modules are officially functioning) please PM me.

The b3 backplane easily breaks apart in order to separate the modules, there are no electrical connections between the 2 modules on the same backplane.

Possible magic smoke sources on the backplane are the power connectors, 2 caps on the bottom side, or worse case circuit traces.

eestimees

hero member

Activity: 651

Merit: 500

my B1 avalon looked like this

unfortunately I was too busy replacing it and didn't take any pics after...

ste0024

sr. member

Activity: 462

Merit: 250

Quote from: jddebug on September 26, 2013, 10:09:00 PM

Quote from: ste0024 on September 26, 2013, 07:40:04 PM

Quote from: aneutronic on September 23, 2013, 06:51:48 AM

Quote from: ste0024 on September 23, 2013, 06:43:47 AM

i m sure to understand but i can read

Not Electrically Conductive:
Arctic Silver 5 was formulated to conduct heat, not electricity.

http://www.arcticsilver.com/as5.htm

good paste or not ?

Correct, as5 is not conductive, the only question I have regarding as5 is that it is capacitive unlike arctic silvers mx2 or mx4.
I don't know if that would be an issue in this application so I chose mx4 over the as5.

thermal paste change and there needed.

the thermal paste used by avalon Were dry. would no longer have anything.

Temp 2 before 69 or 70 °
Temp 2 after 58 to 62 °

Temp 3 before 60 or 62 °
Temp 3 after 50 to 52 °

the ventilos not running any more. 2500 instead of 3000rpm

think has tighten the screws that hold the radiators. on mine there was a lot of screws loosen and poor contact and poor heat disipation.

after having tighten the plate is hot

Conclusion: if you want your living avalon still a lot of time,

change the thermal paste.

for 4 modules I have used 60gr of artic silver 5. there was just enough.

I put 6 hours to do everything. disassembly and reassembly test

Now it runs at 110 instead of 108 gh - less error HW

that happiness

and all this with Firmware Version => 20130723

next step update Firmware

before

after

Did you take any pictures as you did the paste replacement?

no but you can found some picture here.

https://bitcointalksearch.org/topic/m.2687268

and

https://bitcointalksearch.org/topic/m.3216051

allinvain

legendary

Activity: 3080

Merit: 1080

Quote from: exahash on September 26, 2013, 11:11:27 AM

Quote from: allinvain on September 26, 2013, 10:47:35 AM

Anyone here good at troubleshooting or reviving a possibly dead hashing card? I had the "red light" issue with a batch 3 unit and while I disconnected the power supply to replace it I found out my replacement power supply was missing a second EPS 12v connect (ie CPU connector) so I plugged in a power supply from a batch 2 unit. When I tried to start it up the typical giving up the ghost smoke came out of the the third card - not a lot of smoke but it was noticeable. The unit of course refused to power on. With the third module disconnected the unit powers up just fine. It even power up just fine with it's original PSU.

Could I have killed it trying to swap the power supply or was it most likely already dead (due to the red light warning) before I tried to mess with it?

If anyone had this happen to their batch 3 unit I'm wondering if you can tell me what the exact cause was. Was it blown capacitor(s)? I looked at the exposed side of the board and no obvious burn marks or dead caps are visible. The smoke appeared to have come from the top of the backplane near where the power connector is, but from underneath.

I doubt you killed it swapping the psu, more likely it was already dead and finally burnt when you turned it back on. I found my damaged b3 units off with only the red lights on, and when I turned each back on got the smoke and some sparks. A few pages back I posted pics and people told me what the parts were. I don't know if mine were burnt at first because I didn't open them up, just toggled the psu power switch. I was able to find the burnt components and simply segregated the affected workers by bending the pins so they wouldn't plug into the bus (the top board that the cables plug into).

If you can't see anything burnt on the face of the module, try unplugging the module from the bus and looking for signs of damage on the underside of the bus or around the edge of the module where it meets the heatsink. The smoke had to come from somewhere and usually leaves residue that's fairly easy to spot. If the bus itself is damaged you'll probably have to get it repaired, not just bend some pins.

You're probably right. The original PSU works just fine with the third module disabled, so it's likely that just a part of that module is damaged. When I get a chance I will take it apart and inspect the module. It would suck if it's the backplane that's damanged as two modules connect to one backplane. I must admit I'm not a big fan of the batch3 design. As of now the unit is hashing fine with two modules and with the original PSU in place.

I'm thinking of selling the unit. I am getting tired of babysitting these units.

If anyone wants to make me an offer for the unit (bear in mind that at the moment only two modules are officially functioning) please PM me.

allinvain

legendary

Activity: 3080

Merit: 1080

Quote from: mitak64 on September 26, 2013, 11:13:41 AM

So you had 2 PSUs hooked up at the same time ?
I always wondered if that would work.
Can you post a photo of the damage ?

No, I did not have two PSU's at the same time. I never suggested that. I just had one PSU at a time. The original PSU works with the third module disabled so I doubt it was the PSU that caused the problem.

jddebug

sr. member

Activity: 446

Merit: 250

Quote from: ste0024 on September 26, 2013, 07:40:04 PM

Quote from: aneutronic on September 23, 2013, 06:51:48 AM

Quote from: ste0024 on September 23, 2013, 06:43:47 AM

i m sure to understand but i can read

Not Electrically Conductive:
Arctic Silver 5 was formulated to conduct heat, not electricity.

http://www.arcticsilver.com/as5.htm

good paste or not ?

Correct, as5 is not conductive, the only question I have regarding as5 is that it is capacitive unlike arctic silvers mx2 or mx4.
I don't know if that would be an issue in this application so I chose mx4 over the as5.

thermal paste change and there needed.

the thermal paste used by avalon Were dry. would no longer have anything.

Temp 2 before 69 or 70 °
Temp 2 after 58 to 62 °

Temp 3 before 60 or 62 °
Temp 3 after 50 to 52 °

the ventilos not running any more. 2500 instead of 3000rpm

think has tighten the screws that hold the radiators. on mine there was a lot of screws loosen and poor contact and poor heat disipation.

after having tighten the plate is hot

Conclusion: if you want your living avalon still a lot of time,

change the thermal paste.

for 4 modules I have used 60gr of artic silver 5. there was just enough.

I put 6 hours to do everything. disassembly and reassembly test

Now it runs at 110 instead of 108 gh - less error HW

that happiness

and all this with Firmware Version => 20130723

next step update Firmware

before

after

Did you take any pictures as you did the paste replacement?

shmadz

legendary

Activity: 1512

Merit: 1000

@theshmadz

Quote from: xiangfu on September 25, 2013, 04:58:33 AM

The major differences is OpenWrt update to r38031 and update Cgminer to latest(Version: 3.4.3 Latest commit: d5f61e9, Sep 22 2013) you can see that at the ChangeLog

https://en.bitcoin.it/wiki/Avalon#20130923

Quote from: jermwerty on September 25, 2013, 04:37:08 AM

Thanks Xiangfu! My question is what are the major differences between ckolivas' 20130821 and this new xiangfu 20130923? I did run xiangfu's 20130723 for a bit on my Batch 3 just fine, but I've been happy with 20130821 so I am not exactly itching to change...

Seem like both Xiangfu and cklovias use OpenWRT r37008 but the new one upgrades cgminer 3.4.0 to 3.4.4?

If anyone has tested this new firmware, please report your findings.

It will have to be quite impressive for me to switch from con's 20130703

Bogart

legendary

Activity: 966

Merit: 1000

Quote from: nanobit on September 26, 2013, 05:46:03 PM

Quote from: mitak64 on September 26, 2013, 11:13:41 AM

So you had 2 PSUs hooked up at the same time ?
I always wondered if that would work.
Can you post a photo of the damage ?

Two PSU's shouldn't be an issue. I think quite a few people who've OC'd their box are using two, me included.

Two power supplies can be bad mojo if you ever happen to have one powered on and not the other. Sometimes a power supply in the off state will short +12V to GND. If this happens while the other is trying to keep things at 12V, bad things will obviously happen.

ste0024

sr. member

Activity: 462

Merit: 250

Quote from: aneutronic on September 23, 2013, 06:51:48 AM

Quote from: ste0024 on September 23, 2013, 06:43:47 AM

i m sure to understand but i can read

Not Electrically Conductive:
Arctic Silver 5 was formulated to conduct heat, not electricity.

http://www.arcticsilver.com/as5.htm

good paste or not ?

Correct, as5 is not conductive, the only question I have regarding as5 is that it is capacitive unlike arctic silvers mx2 or mx4.
I don't know if that would be an issue in this application so I chose mx4 over the as5.

thermal paste change and there needed.

the thermal paste used by avalon Were dry. would no longer have anything.

Temp 2 before 69 or 70 °
Temp 2 after 58 to 62 °

Temp 3 before 60 or 62 °
Temp 3 after 50 to 52 °

the ventilos not running any more. 2500 instead of 3000rpm

think has tighten the screws that hold the radiators. on mine there was a lot of screws loosen and poor contact and poor heat disipation.

after having tighten the plate is hot

Conclusion: if you want your living avalon still a lot of time,

change the thermal paste.

for 4 modules I have used 60gr of artic silver 5. there was just enough.

I put 6 hours to do everything. disassembly and reassembly test

Now it runs at 110 instead of 108 gh - less error HW

that happiness

and all this with Firmware Version => 20130723

next step update Firmware

before

after

nanobit

member

Activity: 77

Merit: 10

Quote from: mitak64 on September 26, 2013, 11:13:41 AM

So you had 2 PSUs hooked up at the same time ?
I always wondered if that would work.
Can you post a photo of the damage ?

Two PSU's shouldn't be an issue. I think quite a few people who've OC'd their box are using two, me included.

mitak64

full member

Activity: 254

Merit: 100

So you had 2 PSUs hooked up at the same time ?
I always wondered if that would work.
Can you post a photo of the damage ?

exahash

sr. member

Activity: 278

Merit: 250

Quote from: allinvain on September 26, 2013, 10:47:35 AM

Anyone here good at troubleshooting or reviving a possibly dead hashing card? I had the "red light" issue with a batch 3 unit and while I disconnected the power supply to replace it I found out my replacement power supply was missing a second EPS 12v connect (ie CPU connector) so I plugged in a power supply from a batch 2 unit. When I tried to start it up the typical giving up the ghost smoke came out of the the third card - not a lot of smoke but it was noticeable. The unit of course refused to power on. With the third module disconnected the unit powers up just fine. It even power up just fine with it's original PSU.

Could I have killed it trying to swap the power supply or was it most likely already dead (due to the red light warning) before I tried to mess with it?

If anyone had this happen to their batch 3 unit I'm wondering if you can tell me what the exact cause was. Was it blown capacitor(s)? I looked at the exposed side of the board and no obvious burn marks or dead caps are visible. The smoke appeared to have come from the top of the backplane near where the power connector is, but from underneath.

I doubt you killed it swapping the psu, more likely it was already dead and finally burnt when you turned it back on. I found my damaged b3 units off with only the red lights on, and when I turned each back on got the smoke and some sparks. A few pages back I posted pics and people told me what the parts were. I don't know if mine were burnt at first because I didn't open them up, just toggled the psu power switch. I was able to find the burnt components and simply segregated the affected workers by bending the pins so they wouldn't plug into the bus (the top board that the cables plug into).

If you can't see anything burnt on the face of the module, try unplugging the module from the bus and looking for signs of damage on the underside of the bus or around the edge of the module where it meets the heatsink. The smoke had to come from somewhere and usually leaves residue that's fairly easy to spot. If the bus itself is damaged you'll probably have to get it repaired, not just bend some pins.

allinvain

legendary

Activity: 3080

Merit: 1080

Anyone here good at troubleshooting or reviving a possibly dead hashing card? I had the "red light" issue with a batch 3 unit and while I disconnected the power supply to replace it I found out my replacement power supply was missing a second EPS 12v connect (ie CPU connector) so I plugged in a power supply from a batch 2 unit. When I tried to start it up the typical giving up the ghost smoke came out of the the third card - not a lot of smoke but it was noticeable. The unit of course refused to power on. With the third module disconnected the unit powers up just fine. It even power up just fine with it's original PSU.

Could I have killed it trying to swap the power supply or was it most likely already dead (due to the red light warning) before I tried to mess with it?

If anyone had this happen to their batch 3 unit I'm wondering if you can tell me what the exact cause was. Was it blown capacitor(s)? I looked at the exposed side of the board and no obvious burn marks or dead caps are visible. The smoke appeared to have come from the top of the backplane near where the power connector is, but from underneath.

jarhed

sr. member

Activity: 672

Merit: 254

Quote from: SolarSilver on September 26, 2013, 09:51:22 AM

Quote from: jarhed on September 25, 2013, 06:42:54 PM

Accessing ava via browser now takes longer. What may be the issue?

define longer. Does it take a while to start loading?

If you ping the machine, do you see high latency or dropped packets?

You might just suffer from a DNS issue (timeouts before load)

.....longer to load any page after login.
Everything seems to work fine now.

I'd recently update firmware to 20130923 and changed default Chip Frequency (on Cgminer Configuration Page) to 300. I lower the Frequency to 285. Unit access via browser interface now load as it should.

A fresh firmware update had the Chip Frequency at somewhere below 300 (don't remember exactly), so because "Default: 300" is stated right there on the Cgminer Configuration page, and the previous firmware was at 300 CF, naturally, anything < 300 was unacceptable. Grin

tarui

sr. member

Activity: 294

Merit: 250

Quote from: tarui on September 25, 2013, 12:00:51 PM

can someone tell me whether the batch one's controller board can handle 4 pwm fans?

the delta afc121de?
my PSU is on the outside so no issue with PSU being suffocated

bump......

SolarSilver

legendary

Activity: 1112

Merit: 1000

Quote from: jarhed on September 25, 2013, 06:42:54 PM

Accessing ava via browser now takes longer. What may be the issue?

define longer. Does it take a while to start loading?

If you ping the machine, do you see high latency or dropped packets?

You might just suffer from a DNS issue (timeouts before load)

jarhed

sr. member

Activity: 672

Merit: 254

Accessing ava via browser now takes longer. What may be the issue?

Topic: Avalon ASIC users thread - page 74. (Read 438516 times)