I must be as transparent as possible about my firmware, and that seems normal to me;)for temperature protection yes, I just reactivated the option of the original firmware ... even if I think that it does not really make sense to use it ...
Topic: HELP with 2 T17+ :( (Read 468 times)
Absolutely no problem! I must be as transparent as possible about my firmware, and that seems normal to me;)
for temperature protection yes, I just reactivated the option of the original firmware ... even if I think that it does not really make sense to use it ...
I must be as transparent as possible about my firmware, and that seems normal to me;)for temperature protection yes, I just reactivated the option of the original firmware ... even if I think that it does not really make sense to use it ...
Thank you for your interest, let it be clear, in my firwmare, I only modify the configuration page to display all the hidden options, and, concerning bmminer, I ONLY change the boot, it starts in fixed mode and not automatically.
EVERYTHING else remains the same,
EVERYTHING else remains the same,
Don't temperature security options exist in the stock firmware's options? I remember in the old version of S9 firmware there was an option for that, it says something like "stop mining when chip temp is over xx", have they removed all of these features in the 17 series firmware?
Thanks for being nice enough to answer all of our questions, remember we are doing this to help you improve not to bring you down.
Thank you for your interest, let it be clear, in my firwmare, I only modify the configuration page to display all the hidden options, and, concerning bmminer, I ONLY change the boot, it starts in fixed mode and not automatically.
EVERYTHING else remains the same,
Regarding pool security, in fact, the miner ping my pool, so if it is killed, or if the miner loses all internet connection, then it starts again. but contain litle bug, i resolve this problem soon and i release update
EVERYTHING else remains the same,
Regarding pool security, in fact, the miner ping my pool, so if it is killed, or if the miner loses all internet connection, then it starts again. but contain litle bug, i resolve this problem soon and i release update
Hahah, I never said it is IDEAL standard, I just said it's default, and Thierry even said that he did not modify the original firmware that much. He literally just opened the options that were already there but Bitmain locked them a bit better than on S9 machines.
However, yes, I would like for Thierry to work a bit more on features on his firmware as his fees are really competitive and firmware works like a charm usually.
I will point him again to check this thread.
Also, yeah, we do think alike, tho I never put them under 100%. I'd rather let them use more W and possibly die on me but they cool better, than to have to deal with other problems.
However, yes, I would like for Thierry to work a bit more on features on his firmware as his fees are really competitive and firmware works like a charm usually.
I will point him again to check this thread.
Also, yeah, we do think alike, tho I never put them under 100%. I'd rather let them use more W and possibly die on me but they cool better, than to have to deal with other problems.
... but I think 80 C chip temp target is default set by Bitmain.
Since when did bitmain firmware become the ideal standard?
, their firmware suck big time.Also, I never use chip temp target on my units. As you can see it's on manual 100% like on all my other units.
I saw that, and I admire, because this is how I treat my gears as well, 90-100% fixed fan speed is the only way I run these miners, fans are dirt cheap and I have a ton of them for spare.
True, but I think 80 C chip temp target is default set by Bitmain. Also, I never use chip temp target on my units. As you can see it's on manual 100% like on all my other units.
Yes, that feature where we can change temp target is nice to have, but I am not sure how much Thierry is up to adding features.
There is a lot of things he could add just looking at Vnish's firmware, and many more even.
Yes, that feature where we can change temp target is nice to have, but I am not sure how much Thierry is up to adding features.
There is a lot of things he could add just looking at Vnish's firmware, and many more even.
Agreed, also the fixed fan speed effect on temp should be modifiable, 80c as a chip temp target for auto-fan is a bit too much, the best option would be setting your desired max chip-temp on which the fans will try to achieve by spinning at faster/slower RPMs, by all means 80c on these 17 series gears is overkill.
Yeah, just ignore me for that part. I was talking totally about chip temps. PCB temps are more than fine and Thierry even has option in his firmware to stop mining if PCB is above 80C (picture below):
I actually think thierry should give us the same option as Vnish to change max chip temp by ourselves. I wouldn't like him changing it to be lower (nor higher for that matter) and not giving us an option.
I actually think thierry should give us the same option as Vnish to change max chip temp by ourselves. I wouldn't like him changing it to be lower (nor higher for that matter) and not giving us an option.
Chips' temp is a chip reported by the chip itself, while the 4 sensors on the hashboard report the PCB temp, on the T17 they can be found "near chip 9, chip 7, chip 22, chip 24.", their part number is "TMP451AIDQFR" and they look something like this.
I believe that stock firmware requires at least 2 of them to send temp-readings before mining is initialized, Vnish, on the other hand, requires at least 1 out of 4 IIRC, of course, you can get the miner to start without checking them if you would modify that piece of code, while that may be good for testing whether you really have 4 bad temp-sensors or just a loose chip/heatsink, it can set your miner on fire.
Since you mentioned 90c then it's chip temp, I assume if PCB temp is 90c then your chips would have probably melted, and as far as stock firmware, it works exactly like how you explained it, 1 chip out of all the chips across all boards goes above 90 and the miner goes to protection mode until restart, these settings are adjustable, most firmware including Vnish use 90c as a cut-off and honestly speaking even 90c is way too much for these 17 gears so perhaps thierry4wd should reduce it a bit, if it was S9, I am sure it would handle over 100 with no issue.
I believe that stock firmware requires at least 2 of them to send temp-readings before mining is initialized, Vnish, on the other hand, requires at least 1 out of 4 IIRC, of course, you can get the miner to start without checking them if you would modify that piece of code, while that may be good for testing whether you really have 4 bad temp-sensors or just a loose chip/heatsink, it can set your miner on fire.
Since you mentioned 90c then it's chip temp, I assume if PCB temp is 90c then your chips would have probably melted, and as far as stock firmware, it works exactly like how you explained it, 1 chip out of all the chips across all boards goes above 90 and the miner goes to protection mode until restart, these settings are adjustable, most firmware including Vnish use 90c as a cut-off and honestly speaking even 90c is way too much for these 17 gears so perhaps thierry4wd should reduce it a bit, if it was S9, I am sure it would handle over 100 with no issue.
I'd like to know about temp sensors too. As I've seen one of my chips reach 96 C while others were okay and neither hashboard nor whole ASIC stopped mining.
Where on default firmware, as soon as one chip reaches 90/91C, whole unit just stops mining instantly and hashrate of ASIC drops to 0.
Now, this has happened once or twice, other times when overheated, it went to 0 same as default firmware. Is this a bug or something else?
Where on default firmware, as soon as one chip reaches 90/91C, whole unit just stops mining instantly and hashrate of ASIC drops to 0.
Now, this has happened once or twice, other times when overheated, it went to 0 same as default firmware. Is this a bug or something else?
Hi, indeed, some firmware can solve some problem, concerning mine, it ignores some functionality and this content to make the miner work as simple as possible with only useful things. did you use mine?
I know it's unlikely that you are going to do it, but I want to make sure, you have not disabled all 4 temp-sensor check on your firmware right?
ok, in fact this is normal i have built in a security, if the client kill the develop pool, the miner stop hashing, but normally should restart ... i see where the problem is coming from i will correct this and update, I would be interested to have your feedback if possible 
Does this mean gyw2020 blocked the dev pool on his router? or is it just a bad connection to the pool? Make sure you have a bunch of dev pools located in different areas US/EU/ASIA, also all these pools must support Asicboost.
ok, in fact this is normal i have built in a security, if the client kill the develop pool, the miner stop hashing, but normally should restart ... i see where the problem is coming from i will correct this and update, I would be interested to have your feedback if possible
Yes but i had a problem after error: ``warn net lost´´ the miner wont start mining again something that with stock firmaware not happened now i have stock firmware and all of them working with all boards lets see for how long.
Hi, indeed, some firmware can solve some problem, concerning mine, it ignores some functionality and this content to make the miner work as simple as possible with only useful things. did you use mine?
yes im not going to buy to bitmain more whatsminer from now on
The one that couldnt read the board was solved with custom firmware and the other one after a while drops the hasboard saying read temp error so im going to do as you say and press the heatsinks
I suspected that you would fix one with the firmware. It seems the 17 series has a lot of bad models.
The one that couldnt read the board was solved with custom firmware and the other one after a while drops the hasboard saying read temp error so im going to do as you say and press the heatsinks
Code:
2020-09-13 16:42:12 temperature.c:924:get_temp_info: read temp sensor failed: chain = 0, sensor = 0, chip = 55, reg = 1
2020-09-13 16:42:12 temperature.c:924:get_temp_info: read temp sensor failed: chain = 0, sensor = 1, chip = 40, reg = 0
2020-09-13 16:42:13 temperature.c:924:get_temp_info: read temp sensor failed: chain = 0, sensor = 2, chip = 175, reg = 0
2020-09-13 16:42:13 temperature.c:924:get_temp_info: read temp sensor failed: chain = 0, sensor = 3, chip = 160, reg = 0
2020-09-13 16:42:12 temperature.c:924:get_temp_info: read temp sensor failed: chain = 0, sensor = 1, chip = 40, reg = 0
2020-09-13 16:42:13 temperature.c:924:get_temp_info: read temp sensor failed: chain = 0, sensor = 2, chip = 175, reg = 0
2020-09-13 16:42:13 temperature.c:924:get_temp_info: read temp sensor failed: chain = 0, sensor = 3, chip = 160, reg = 0
All 4 sensors are lost, we had a whole topic discussing this issue, this is certainly isn't a temp sensor issue but rather 1 or more heatsinks lost contact to the chips or the chip itself lost contact to the board, since the chip/heatsink hasn't fallen entirely, it will work at times and stop when it gets hot, in many cases people reported success by applying some pressure on all heatsinks in an attempt to stick that heatsink/chip since you can't tell which chip has gone bad, you will have to that to all chips.
Having said that, all of these solutions including trying a custom firmware are not long-term solutions, you need to fix that board or mine with it this way until it dies completely.
As both Phil and I suggested already. Try custom firmware and see what the result is.
Thanks ill try, the other miner after a day a hashboard down again.
Code:
2020-09-13 15:36:42 thread.c:255:calc_hashrate_avg: avg rate is 59435.25 in 30 mins
2020-09-13 15:36:42 temperature.c:529:temp_statistics_show: pcb temp 38~57 chip temp 60~79
2020-09-13 16:06:43 thread.c:255:calc_hashrate_avg: avg rate is 59546.40 in 30 mins
2020-09-13 16:06:43 temperature.c:529:temp_statistics_show: pcb temp 37~56 chip temp 59~79
2020-09-13 16:36:45 thread.c:255:calc_hashrate_avg: avg rate is 59623.18 in 30 mins
2020-09-13 16:36:45 temperature.c:529:temp_statistics_show: pcb temp 37~56 chip temp 60~80
2020-09-13 16:42:11 thread.c:1372:asic_status_monitor_thread: ERROR: chain 0 get hashrate_reg_counter 6, require 44, failed times 1: oooo ooxx xxxx xxxx xxxx xxxx xxxx xxxx xxxx xxxx xxxx
2020-09-13 16:42:11 temperature.c:924:get_temp_info: read temp sensor failed: chain = 0, sensor = 0, chip = 55, reg = 0
2020-09-13 16:42:12 temperature.c:924:get_temp_info: read temp sensor failed: chain = 0, sensor = 0, chip = 55, reg = 1
2020-09-13 16:42:12 temperature.c:924:get_temp_info: read temp sensor failed: chain = 0, sensor = 1, chip = 40, reg = 0
2020-09-13 16:42:12 temperature.c:924:get_temp_info: read temp sensor failed: chain = 0, sensor = 1, chip = 40, reg = 1
2020-09-13 16:42:13 temperature.c:924:get_temp_info: read temp sensor failed: chain = 0, sensor = 2, chip = 175, reg = 0
2020-09-13 16:42:13 thread.c:1392:asic_status_monitor_thread: chain 0 can't get enough hashrate reg val for 0 times.
2020-09-13 16:42:13 temperature.c:924:get_temp_info: read temp sensor failed: chain = 0, sensor = 2, chip = 175, reg = 1
2020-09-13 16:42:13 thread.c:1372:asic_status_monitor_thread: ERROR: chain 0 get hashrate_reg_counter 6, require 44, failed times 1: oooo ooxx xxxx xxxx xxxx xxxx xxxx xxxx xxxx xxxx xxxx
2020-09-13 16:42:13 temperature.c:924:get_temp_info: read temp sensor failed: chain = 0, sensor = 3, chip = 160, reg = 0
2020-09-13 16:42:14 temperature.c:924:get_temp_info: read temp sensor failed: chain = 0, sensor = 3, chip = 160, reg = 1
2020-09-13 16:42:15 temperature.c:956:get_temp_info: ERROR: chain 0 can get NONE temp info or temp value abnormal, power it off
2020-09-13 16:42:15 thread.c:1392:asic_status_monitor_thread: chain 0 can't get enough hashrate reg val for 1 times.
2020-09-13 16:42:15 thread.c:1364:asic_status_monitor_thread: ERROR: chain 0 get hashrate_reg_counter 0, require 44, failed times 1
2020-09-13 16:42:16 frequency.c:316:get_sale_hash_rate_GH: g_sale_hash_rate = 37704
2020-09-13 17:06:46 thread.c:255:calc_hashrate_avg: avg rate is 42784.49 in 30 mins
2020-09-13 17:06:46 temperature.c:529:temp_statistics_show: pcb temp 36~55 chip temp 59~79
2020-09-13 17:36:47 thread.c:255:calc_hashrate_avg: avg rate is 39490.23 in 30 mins
2020-09-13 17:36:47 temperature.c:529:temp_statistics_show: pcb temp 36~55 chip temp 58~79
2020-09-13 18:06:48 thread.c:255:calc_hashrate_avg: avg rate is 39136.14 in 30 mins
2020-09-13 18:06:48 temperature.c:529:temp_statistics_show: pcb temp 35~55 chip temp 58~80
2020-09-13 18:36:50 thread.c:255:calc_hashrate_avg: avg rate is 39406.59 in 30 mins
2020-09-13 18:36:50 temperature.c:529:temp_statistics_show: pcb temp 35~55 chip temp 58~80
2020-09-13 19:06:51 thread.c:255:calc_hashrate_avg: avg rate is 39151.41 in 30 mins
2020-09-13 19:06:51 temperature.c:529:temp_statistics_show: pcb temp 35~55 chip temp 59~79
2020-09-13 19:36:52 thread.c:255:calc_hashrate_avg: avg rate is 39336.52 in 30 mins
2020-09-13 19:36:52 temperature.c:529:temp_statistics_show: pcb temp 35~55 chip temp 58~80
2020-09-13 15:36:42 temperature.c:529:temp_statistics_show: pcb temp 38~57 chip temp 60~79
2020-09-13 16:06:43 thread.c:255:calc_hashrate_avg: avg rate is 59546.40 in 30 mins
2020-09-13 16:06:43 temperature.c:529:temp_statistics_show: pcb temp 37~56 chip temp 59~79
2020-09-13 16:36:45 thread.c:255:calc_hashrate_avg: avg rate is 59623.18 in 30 mins
2020-09-13 16:36:45 temperature.c:529:temp_statistics_show: pcb temp 37~56 chip temp 60~80
2020-09-13 16:42:11 thread.c:1372:asic_status_monitor_thread: ERROR: chain 0 get hashrate_reg_counter 6, require 44, failed times 1: oooo ooxx xxxx xxxx xxxx xxxx xxxx xxxx xxxx xxxx xxxx
2020-09-13 16:42:11 temperature.c:924:get_temp_info: read temp sensor failed: chain = 0, sensor = 0, chip = 55, reg = 0
2020-09-13 16:42:12 temperature.c:924:get_temp_info: read temp sensor failed: chain = 0, sensor = 0, chip = 55, reg = 1
2020-09-13 16:42:12 temperature.c:924:get_temp_info: read temp sensor failed: chain = 0, sensor = 1, chip = 40, reg = 0
2020-09-13 16:42:12 temperature.c:924:get_temp_info: read temp sensor failed: chain = 0, sensor = 1, chip = 40, reg = 1
2020-09-13 16:42:13 temperature.c:924:get_temp_info: read temp sensor failed: chain = 0, sensor = 2, chip = 175, reg = 0
2020-09-13 16:42:13 thread.c:1392:asic_status_monitor_thread: chain 0 can't get enough hashrate reg val for 0 times.
2020-09-13 16:42:13 temperature.c:924:get_temp_info: read temp sensor failed: chain = 0, sensor = 2, chip = 175, reg = 1
2020-09-13 16:42:13 thread.c:1372:asic_status_monitor_thread: ERROR: chain 0 get hashrate_reg_counter 6, require 44, failed times 1: oooo ooxx xxxx xxxx xxxx xxxx xxxx xxxx xxxx xxxx xxxx
2020-09-13 16:42:13 temperature.c:924:get_temp_info: read temp sensor failed: chain = 0, sensor = 3, chip = 160, reg = 0
2020-09-13 16:42:14 temperature.c:924:get_temp_info: read temp sensor failed: chain = 0, sensor = 3, chip = 160, reg = 1
2020-09-13 16:42:15 temperature.c:956:get_temp_info: ERROR: chain 0 can get NONE temp info or temp value abnormal, power it off
2020-09-13 16:42:15 thread.c:1392:asic_status_monitor_thread: chain 0 can't get enough hashrate reg val for 1 times.
2020-09-13 16:42:15 thread.c:1364:asic_status_monitor_thread: ERROR: chain 0 get hashrate_reg_counter 0, require 44, failed times 1
2020-09-13 16:42:16 frequency.c:316:get_sale_hash_rate_GH: g_sale_hash_rate = 37704
2020-09-13 17:06:46 thread.c:255:calc_hashrate_avg: avg rate is 42784.49 in 30 mins
2020-09-13 17:06:46 temperature.c:529:temp_statistics_show: pcb temp 36~55 chip temp 59~79
2020-09-13 17:36:47 thread.c:255:calc_hashrate_avg: avg rate is 39490.23 in 30 mins
2020-09-13 17:36:47 temperature.c:529:temp_statistics_show: pcb temp 36~55 chip temp 58~79
2020-09-13 18:06:48 thread.c:255:calc_hashrate_avg: avg rate is 39136.14 in 30 mins
2020-09-13 18:06:48 temperature.c:529:temp_statistics_show: pcb temp 35~55 chip temp 58~80
2020-09-13 18:36:50 thread.c:255:calc_hashrate_avg: avg rate is 39406.59 in 30 mins
2020-09-13 18:36:50 temperature.c:529:temp_statistics_show: pcb temp 35~55 chip temp 58~80
2020-09-13 19:06:51 thread.c:255:calc_hashrate_avg: avg rate is 39151.41 in 30 mins
2020-09-13 19:06:51 temperature.c:529:temp_statistics_show: pcb temp 35~55 chip temp 59~79
2020-09-13 19:36:52 thread.c:255:calc_hashrate_avg: avg rate is 39336.52 in 30 mins
2020-09-13 19:36:52 temperature.c:529:temp_statistics_show: pcb temp 35~55 chip temp 58~80
Jump to: