// #eevblog — 2019-02-12
1996
1997
1998
1999
2000
2001
2002
2003
2004
2005
2006
2007
2008
2009
2010
2011
2012
2013
2014
2015
2016
00:05:27 Quit: jackbrown ([email protected]) Ping timeout: 121 seconds
00:35:42 Join: jackbrown ([email protected]) to #EEVblog
00:44:48 Quit: hsn93 ([email protected]) Quit: Leaving
00:45:06 Join: hsn93 ([email protected]) to #EEVblog
01:02:00 Quit: jackbrown ([email protected]) Ping timeout: 121 seconds
01:28:36 <hsn93> does EEPROM cells get corrupted without writing into it? (assuming voltage + environment = perfect) ?
01:31:18 <K> it shouldnt, apart from cosmic rays being force majeure
01:33:21 <hsn93> so if i validate from crc8 and it was wrong then most probably from noisy i2c with the eeprom chip?
01:33:38 <coppice> given enough years or really high temperatures EEPROM cells can lose their data. There is a quirks you'll hit on maybe 1 in 100k chips where some cells lose their data in just a few months
01:34:46 <K> is it wrong the same way every time you read it?
01:36:12 <coppice> If the EEPROM is used in an application where it gets written a lot, it can become very quirky as its approaches the point of being worn out
01:36:36 <coppice> if its inconsistent from read to read that's usually a signal quality problem
01:36:43 <hsn93> @k, no im not asking about particular eeprom, im trying to do some library that checks some data and before determining its not valid i though i might do a double check (2x eeprom) if it was not valid.
01:37:56 <hsn93> (2x eeprom reads)*
01:39:09 <hsn93> ok so i better do (2x eeprom reads). if its not valid i do (1x write + 1x read) and check if this eeprom is worn out or not?
01:39:32 <hsn93> valid = (check CRC)
01:39:57 <K> i would do it that way yea
01:40:41 <K> will you write a lot?
01:40:47 <coppice> If integrity is important with EEPROMs, a CRC check on the whole message is a good idea. because they wear out cell by cell quite a few apps put 2 copies of the message+CRC in the EEPROM, and hopefully one will show a good CRC
01:40:49 <K> i.e many times
01:41:23 <coppice> If you need to write a lot, use an FRAM instead of an EEPROM
01:41:38 <hsn93> @k, yes. but with rotating..
01:41:50 <K> ah, you have rotation, ok
01:42:24 <hsn93> does people use eeprom for logging ? XD
01:43:03 <coppice> yes. that's a common use for them, but usually only for small infrequent logs. FRAM is used quite a lot for very frequent logging
01:43:12 <K> fram looks a bit pricey
01:43:30 <coppice> not really
01:43:33 <K> hm
01:44:02 <coppice> its more expensive than EEPROM, but its not a crazy price, and when you need to log a lot EEPROM is not an option
01:44:55 <hsn93> im trying to write open source library that uses (log system).. i knew FRAM was the correct option but i allready have the hardware (it can be changed but im doing the library for fun) .. https://github.com/hsn93/Endurance_EEPROM
01:44:55 <K> i would consider writing to an sd card
01:46:53 <hsn93> im dividing eeprom into slots (some will be updated frequently = need larger slot).. data written = [small header + user data]..
01:48:31 <hsn93> im almost done with the core of it. but dont know if i should continue or its stupid project.
01:48:55 <K> hmm
01:57:09 <hsn93> i've no experience FRAM sounds compatible with eeprom.. SD card i think its more complicated and im not sure but isnt SD card = eeprom + file system?
01:57:45 <Agripa> EEPROM these days is self timed so corruption only comes from bad data or bad power.
01:57:57 <K> sd card is huge
01:58:10 <K> maybe can just write and forget
01:58:18 <K> no rotation complexity
01:58:35 <Agripa> FRAM would be more appropriate if you expect more than 100k writes over time.
01:58:50 <hsn93> @k but its done within it self right? i mean there is wearing level inside the sd card controller it self
01:58:54 <Agripa> That's per EEPROM cell. Wear levelign works fine on EEPROM.
01:58:57 Join: jackbrown ([email protected]) to #EEVblog
01:59:18 <K> hsn93, not sure about that
02:00:50 <Agripa> Another alternative in some cases is serial SRAM with battery backup.
02:01:21 <hsn93> hmm sounds good and scary at the same time :D
02:02:12 <Agripa> Some systems use SRAM which is then backed up to EEPROM or whatever when power is lost.
02:03:05 <Agripa> But I would be impressed if you could wear out an EEPROM. They support 100k+ write cycles per cell.
02:03:17 <Agripa> Some are 1M+
02:04:21 <hsn93> 1M
02:04:28 <hsn93> yeah
02:04:55 <Agripa> For critical applications, I would also use two separate non-volitile memories with separate serial links and backup power. Plus of course use a logging type file system in each one with CRC and maybe simple ECC.
02:05:19 <hsn93> but i always imagine it to be 70% because that write cycle has environment conditions.
02:05:43 <Agripa> That is more of a problem with Flash which is more delicate.
02:06:50 <hsn93> i see, thanks for the info.. very helpful
02:11:08 <hsn93> eec is nice but (i only know about hamming(7,4) code and majority voting) which is data hungry.
02:11:27 <hsn93> anyway,, thanks intresting topics
02:35:44 <Ampera> I am hungr
02:39:37 <Agripa> Whatever you say Sinistar.
02:48:25 Quit: bigbadwolv ([email protected]) EOF from client
02:48:46 Join: bigbadwolv ([email protected]) to #EEVblog
02:50:44 Quit: hsn93 ([email protected]) Ping timeout: 121 seconds
02:57:41 Quit: arekm ([email protected]) Ping timeout: 121 seconds
02:57:46 Join: arekm ([email protected]) to #EEVblog
03:50:16 Join: Bluenote ([email protected]) to #EEVblog
08:19:42 Quit: jackbrown ([email protected]) Ping timeout: 121 seconds
08:52:09 Join: Guest|49777 ([email protected]) to #EEVblog
08:58:03 Quit: Guest|49777 ([email protected]) Quit: http://www.kiwiirc.com/ - A hand crafted IRC client
11:41:33 Join: Guest|66919 ([email protected]) to #EEVblog
11:42:16 Quit: Guest|66919 ([email protected]) Quit: Leaving
11:42:39 Join: Guest|86141 ([email protected]) to #EEVblog
13:16:28 Quit: Guest|86141 ([email protected]) Quit: http://www.kiwiirc.com/ - A hand crafted IRC client
14:01:48 Quit: Bluenote ([email protected]) Quit: Leaving
14:16:21 <EEVBlog> New EEVBlog Tweet: NEW EEVblog Video! EEVblog #1179 - Mailbag https://t.co/3xhWpGQKhE
14:21:35 Join: Bluenote ([email protected]) to #EEVblog
14:30:24 Quit: Bluenote ([email protected]) Quit: Leaving
14:44:32 <EEVBlog> New EEVBlog Tweet: NEW Video on EEVblog2! #1179 - Stempedia Evive Electronics Learning Kit https://t.co/UNoV94X9cn
15:29:50 <EEVBlog> New EEVBlog Tweet: NEW Video on EEVblog2! #1179 - PowerFilm Flexible Indoor Solar Cell https://t.co/sBztOcPFOw
15:31:05 Quit: ServalSpots ([email protected]) Ping timeout: 121 seconds
16:08:30 Join: hsn93 ([email protected]) to #EEVblog
16:19:15 Join: jackbrown ([email protected]) to #EEVblog
16:41:18 Quit: hsn93 ([email protected]) Quit: Leaving
16:41:36 Join: hsn93 ([email protected]) to #EEVblog
16:44:50 Join: Bluenote ([email protected]) to #EEVblog
16:53:37 Join: ServalSpots ([email protected]) to #EEVblog
16:53:38 Mode: (by ChanOP) +o ServalSpots
17:17:49 Join: niharica ([email protected]) to #EEVblog
17:17:56 <niharica> hi
17:18:22 <niharica> i was trying to find some cool material on gologicxl experiments
17:18:50 <niharica> if you could let me know of any tutorials reagarding this it would be great
17:40:04 Quit: jackbrown ([email protected]) Ping timeout: 121 seconds
17:47:30 Join: hasan93 ([email protected]) to #EEVblog
17:48:58 Quit: hsn93 ([email protected]) Ping timeout: 121 seconds
18:09:46 Quit: niharica ([email protected]) Quit: http://www.kiwiirc.com/ - A hand crafted IRC client
19:39:37 Join: hsn93 ([email protected]) to #EEVblog
19:42:26 Quit: hasan93 ([email protected]) Ping timeout: 121 seconds
20:34:37 Quit: Bluenote ([email protected]) Quit: Leaving
20:45:51 Join: jackbrown ([email protected]) to #EEVblog
20:52:09 Quit: jackbrown ([email protected]) Ping timeout: 121 seconds
20:52:59 Quit: hsn93 ([email protected]) EOF from client
21:22:44 <K> hello
21:26:15 <T3sl4> hey
22:24:25 Join: jackbrown ([email protected]) to #EEVblog
22:52:40 <EEVBlog> New EEVBlog Tweet: Youtube comment of the day https://t.co/71DAugqWxx
23:50:04 <EEVBlog> New EEVBlog Tweet: RT @rezibot: My 121GW multimeter from @eevblog arrived!! It's pretty great, and super accurate on everything I've tested it on so far. Lots…
Search this day's buffer: