 |
|
|
1)
Message boards :
Number crunching :
Batch 1015 Discussion/problems
(Message 70851)
Posted 12 hours ago by SolarSyonyk Post: The error message One would think that. But Windows being Windows, there are also apparently some ways it can show up that aren't "drive failure" related, as I've just learned... I've had two on a machine that, as far as I can tell, has a perfectly good drive (NVMe drive with zero reported problems, virtio block devices through to the Windows 10 VM doing the compute these days). https://www.cpdn.org/result.php?resultid=22418452 https://www.cpdn.org/result.php?resultid=22418464 I don't know what the codebase looks like, but according to: https://superuser.com/questions/1807763/inexplicable-the-system-cannot-find-the-drive-specified-how-to-solve-it and https://stackoverflow.com/questions/19843849/unexpected-the-system-cannot-find-the-drive-specified-in-batch-file that error message can occur when something in a batch file is mis-interpreted as a drive path, not what it's supposed to be. I'm sure some are the result of a failing drive, but when perfectly good hosts on modern drives are throwing it (without any subsequent errors in other tasks), it seems worth pulling the "weird corner case in a batch file" thread a bit. I'd have assumed it was purely a failing drive error message too, but... apparently not. |
|
2)
Message boards :
Number crunching :
Batch 1008, and test batches 1009 to 1014 for Windows - issues
(Message 70840)
Posted 1 day ago by SolarSyonyk Post:
It's running 9 tasks, due to disk limits, showing 6.8GB of 8 in use, and BOINC is allowed 90% of RAM. I'll just reboot the VM and up the RAM to it - it's able to have at least 12GB, the system isn't running anything else. |
|
3)
Message boards :
Number crunching :
Batch 1008, and test batches 1009 to 1014 for Windows - issues
(Message 70836)
Posted 2 days ago by SolarSyonyk Post: I have 9 tasks currently running on the system (it's a bit short on disk, I should fix that). I see 9 wah2_8.29 tasks, 9 wah2am3m2_8.29 tasks, and 9 of the wah2am3m2t_8.29 tasks. One or two of them are showing no CPU use, though. Not sure how to link PIDs to tasks, though. But looking more closely at the task, "CPU Time" reports "1d 03:51:19" on an elapsed time of 6d and change. There's no C:\Program Files\BOINC\data directory, but I've got a C:\ProgramData\BOINC directory with that sort of stuff. stderr_rm and stderr_um are both empty. stdout_mon.txt: worker: Created shared memory region key = wah2_eas25_n15e_201212_24_1008_012272746 of size 73278744 bytes (version 608) Run for 2 Years and 0 Months pShMem->PRECIS_LATITUDE 185 pShMem->PRECIS_LONGITUDE 285 pShMem->EWSPACEA 0.220000 pShMem->NSSPACEA 0.220000 pShMem->FRSTLATA 19.100000 pShMem->FRSTLONA 328.500000 pShMem->POLELATA 55.500000 pShMem->POLELONA 308.000000 pShMem->L_RUN_REGION 1 pShMem->UPLOAD_INTERVAL 0 ulTotalPhaseTimestep 276864 Starting model ID wah2_eas25_n15e_201212_24_1008_012272746 Phase 1 Launching model "C:\ProgramData\BOINC/projects/climateprediction.net/wah2am3m2_um_8.29_windows_intelx86.exe" wah2_eas25_n15e_201212_24_1008_012272746 generic_phase1_spinup_eas25_global_aabaka ic19610319_14_N96 NATclim_ancil_168months_CMIP6-HadGEM3-GC31-LL_SST_2009-01-01_2022-12-30_v2403 NATclim_ancil_168months_CMIP6-HadGEM3-GC31-LL_SIC_2009-01-01_2022-12-30_v2403 so2dms_prei_N96_1855_0000P oxi.addfa ozone_preind_N96_1879_0000Pv5 Launching model "C:\ProgramData\BOINC/projects/climateprediction.net/wah2rm3m2t_um_8.29_windows_intelx86.exe" wah2_eas25_n15e_201212_24_1008_012272746 executeModelProcess: MonID=7868, GCM_PID=9088, RCM_PID=9824 stdout_rm.txt: Starting HadRM3 model for ID# wah2_eas25_n15e_201212_24_1008_012272746... Attached to shared memory segment with ID Setting run-time Fortran environment... UM environment variables in use: ASTART=dataout/region_restart.day UNIT11=dataout/xacxf.phist UM_SECTOR_SIZE=2048 UNIT02=jobs/xacxf UM_LBC_COUP=0 VN=4.5 TYPE=CRUN UNIT09=tmp/xacxf.namelists UNIT22=datain/ancil/ctldata/STASHmaster STASETS_DIR=datain/ancil/ctldata/stasets CACHE2=tmp/xacxf.cache2 UNIT08=tmp/xacxf.pipe_dummy UNIT14=tmp/xacxf.errors APSUM1=tmp/xacxf.apsum1 APSTMP1=tmp/xacxf.apstmp1 AOTRANS=tmp/xacxf.aotrans UNIT04=jobs/xacxf.stashc UNIT05=jobs/xacxf.namelists DATAM=dataout/ UNIT12=dataout/xacxf.thist UNIT10=dataout/xacxf.phist UNIT06=dataout/xacxf.out UNIT00=dataout/xacxf.err AINITIAL=dataout/region_restart.day UNIT57=jobs/spec3a_sw_3_asol2c_hadcm3 UNIT80=jobs/spec3a_lw_3_asol2c_hadcm3 SWSPECTD=jobs/spec3a_sw_3_asol2c_hadcm3 LWSPECTD=jobs/spec3a_lw_3_asol2c_hadcm3 Changing to slots dir C:\ProgramData\BOINC\slots\1 stdout_um.txt: Starting HadAM3P model for ID# wah2_eas25_n15e_201212_24_1008_012272746... Attached to shared memory segment with ID Setting run-time Fortran environment... UM environment variables in use: ASTART=datain/dumps/generic_phase1_spinup_eas25_global_aabaka UNIT15=datain/ancil/ic19610319_14_N96 SSTIN=datain/ancil/NATclim_ancil_168months_CMIP6-HadGEM3-GC31-LL_SST_2009-01-01_2022-12-30_v2403 SICEIN=datain/ancil/NATclim_ancil_168months_CMIP6-HadGEM3-GC31-LL_SIC_2009-01-01_2022-12-30_v2403 SULPEMIS=datain/ancil/so2dms_prei_N96_1855_0000P CHEMOXID=datain/ancil/oxi.addfa OZONE=datain/ancil/ozone_preind_N96_1879_0000Pv5 UM_LBC_COUP=0 UNIT11=dataout/xadae.phist UM_SECTOR_SIZE=2048 UNIT02=jobs/xadae VN=4.5 TYPE=CRUN UNIT09=tmp/xadae.namelists AINITIAL=dataout/atmos_restart.day UNIT57=jobs/spec3a_sw_3_asol2c_hadcm3 UNIT80=jobs/spec3a_lw_3_asol2c_hadcm3 SWSPECTD=jobs/spec3a_sw_3_asol2c_hadcm3 LWSPECTD=jobs/spec3a_lw_3_asol2c_hadcm3 UNIT22=datain/ancil/ctldata/STASHmaster STASETS_DIR=datain/ancil/ctldata/stasets CACHE2=tmp/xadae.cache2 UNIT08=tmp/xadae.pipe_dummy UNIT14=tmp/xadae.errors APSUM1=tmp/xadae.apsum1 APSTMP1=tmp/xadae.apstmp1 AOTRANS=tmp/xadae.aotrans UNIT04=jobs/xadae.stashc UNIT05=jobs/xadae.namelists DATAM=dataout/ UNIT12=dataout/xadae.thist UNIT10=dataout/xadae.phist UNIT06=dataout/xadae.out UNIT00=dataout/xadae.err UM_ATM_NPROCX=1 UM_ATM_NPROCY=1 UM_NPES=1 RUNID=xadae Changing to slots dir C:\ProgramData\BOINC\slots\1 Closing model... Detaching shared memory segment... I don't see anything obviously wrong in them... I'm tempted to suspend and resume that task, see if it comes back up properly. |
|
4)
Message boards :
Number crunching :
Batch 1008, and test batches 1009 to 1014 for Windows - issues
(Message 70833)
Posted 2 days ago by SolarSyonyk Post: I have a 1008 task (https://www.cpdn.org/result.php?resultid=22417298) that seems "stuck" - it's been at 6% and change for 6 days now while the rest of the tasks blow past it. Windows 10 VM on an AMD system. Is there any value to letting it continue spinning, or should I just abort it and let some other system try? |
|
5)
Message boards :
Cafe CPDN :
Off-Grid Solar/Renewable Energy Discussion
(Message 70801)
Posted 9 days ago by SolarSyonyk Post: I disagree, solid can miss the screw altogether, with stranded something will grab. "Not working at all" is a better outcome than "Being partially connected, yet not having the full wire size connected." The first is easy to troubleshoot. The second works, until it overheats the few strands that are connected and causes problems.
I agree, that's why I do it properly the first time, instead of constantly having to go back and check to see if I screwed up.
Fine, and entirely irrelevant. The wiring still has to handle the full rated draw, safely, for extended periods of time. The general design of wiring and circuit protection, per what I'm familiar with, is that any wire must be rated to handle the full current the upstream protection device can provide - so if you have a 20A circuit breaker, the wiring had better be able to handle 20A indefinitely. The continuous load derating factors in here too.
As clearly stated in the various charts and correction factors you refuse to consider. "12AWG Copper 90C XHHW-2" is sufficient to answer all those questions. You're just insisting on your right to not have to bother.
US fire deaths are about 3800/yr, automotive fatalities are around 38,000/yr, give or take on both numbers. So, 10x difference, a couple orders of magnitude less than your guess. And a good reason that the fire deaths are fairly low is things like the NEC, that dictates wiring standards in a fairly conservative manner, vs "Do whatever you want and hope it doesn't catch fire." |
|
6)
Message boards :
Cafe CPDN :
Off-Grid Solar/Renewable Energy Discussion
(Message 70799)
Posted 9 days ago by SolarSyonyk Post: Yeah it shouldn't and I don't know why, I just add more cable where there's a problem. If the problem is that your wiring is undersized for the job and the bulk of the wire is getting hot, this will work. If the problem, as you seem to imply, is your connections getting hot, you've got a bad crimp job, an improper connector for the job, or the wrong sized connector for your wire gauge (connectors have a rating for what wire gauges they're designed to work with, though there I go, nerding out about complex arcane specifications like "8-14 AWG" and such). You can probably cobble around it with more wire and connectors, but if you're not fixing the root problem, it's a rather cost-inefficient way of solving the issue (throwing wire wire at a connection termination problem).
What sort of "hell of a lot cheaper" prices are you seeing on it where you obtain the stuff? I know it used to be the case, but I just went poking around the site I usually order a lot of my wiring from, and 8AWG copper and 6AWG aluminum are almost the same price - 6AWG THHN-2 Al is $0.52/ft, 8AWG copper is $0.56/ft. I can't even find smaller aluminum wire anymore. Both of these are good for a 50A circuit under mostly standard conditions.
It's marginally cheaper, and if it's a "run it once and be done" sort of situation, the differences don't really matter. I tend to use the stranded stuff, though - as you note, it's a bit easier to work with. I don't mind working with solid wire, though. The house is wired up with 12AWG solid copper, and it's fine. I'm pretty sure I used stranded for the solar runs. Solid is a bit easier to work with for some terminations, though. I have a degree involving electronics, I'm not stupid. Then I have zero idea why you object to doing something like "looking up some wire ampacity charts and figuring out the proper wire gauge for your actual needs." Design it right from the start, and you don't have to go around "touching wire to see if it's too hot."
I was asking if your wire which you claim is 8A rated was rated assuming continuous or intermittent loads. Heating is by definition a continuous load, for any hard wired heaters. I would generally assume your wire was labeled for intermittent loads (which would be a bigger number for a given wire gauge), so at least with US NEC deratings, that would only be good for 6.4A continuous, and you'd need 4 of them in parallel for your 20A heater load (though 3 would probably be suitable too, just on the edge - but, again, I don't trust wire labeling like that, I do the math based on wire gauge, temperature, conduit packing, etc).
You claim an electronics related degree, and it's "too much complicated stuff" to look up a wire ampacity chart at the suitable temperature for your systems?
And also generally useless in designing anything more complicated than an extension cord. Wire gauge, insulation temperature rating, and the rest of the factors flow into some pretty simple charts and calculation factors based on what you want to do - if you've got more wires in a conduit carrying current, they're each derated somewhat because of total conduit heat dissipation issues, if they're being used in a high temperature environment, they derate as there's less overhead between ambient and the insulation rating temperature, etc. It doesn't take a licensed professional engineer to add a branch circuit to a house. It just requires a willingness to look up ratings in a few tables, and apply some basic percentage math beyond that. Probably only applies to non-UK temperatures, and I'd imagine all cable sold in Norway would be ok. And it's hardly dangerous, someone might get a shock or a fuse might blow, not the end of the world. And for those with the nanny state earth leakage breakers, it's very difficult to get a shock. You seem oddly resistant to the concept that electrical faults are a major cause of building fires that very much do kill people. Cracked insulation from lower than designed temperatures is a genuine hazard - and even if it doesn't arc, it's still allowing water and other things in to corrode sections of wire that are impossible to inspect, because they're in the center of segments of conduit, walls, etc. |
|
7)
Message boards :
Cafe CPDN :
Off-Grid Solar/Renewable Energy Discussion
(Message 70789)
Posted 10 days ago by SolarSyonyk Post: *shrug* Don't come into this thread if you don't want Peter and I duking it out over wiring standards? That's about all it is anymore. It's not particularly off-grid focused. I've nothing interesting to say on that front, the LFP battery bank is behaving very well. |
|
8)
Message boards :
Number crunching :
Batch 1008, and test batches 1009 to 1014 for Windows - issues
(Message 70787)
Posted 10 days ago by SolarSyonyk Post: One of my VMs (AMD system) picked up a 1012 task, on WAH 8.29. It's about 12h in, but most of my batch 1008 tasks on that machine are still churning away nicely. |
|
9)
Message boards :
Cafe CPDN :
Off-Grid Solar/Renewable Energy Discussion
(Message 70786)
Posted 10 days ago by SolarSyonyk Post: And in the case of aluminium wire coming loose, it only wastes two minutes of your time tightening it. "Ah, just tighten it up again!" is literally the path to one of the more common failures of aluminum wiring in service. Aluminum has a far higher coefficient of thermal expansion than copper, and is prone to "creep" (where it will continue to flow for some while after being tightened, so end up loose again if you over-torque the connector. I'd suggest using a torque wrench, but I'm guessing that's "too expensive" and "too complicated" for your preferences in wiring. |
|
10)
Message boards :
Cafe CPDN :
Off-Grid Solar/Renewable Energy Discussion
(Message 70785)
Posted 10 days ago by SolarSyonyk Post: GPU feeds, 12.6V, 4AWG ring. More the connections get hot. The connections shouldn't be getting that hot - usually that indicates a poor crimp job, though if you're using aluminum wire (aluminium, however you want to spell it, same material - Al, of some reasonable alloy), you've likely got some oxidation issues going on at the connection. Oxidized copper conducts. Oxidized aluminum doesn't, and a high resistance connection makes for a lot of heat, and tends to make the oxidation issues worse until something stops working - or, more often than it ought, catches fire (I've seen research that the number of "thermally faulty" outlets in a fleet of houses with aluminum wiring vs copper is about 50x higher for aluminum - and that's before you get into the long term behavior of the wire, which is "less agreeable" in terms of work hardening, thermal hardening, etc). There are reasonable places to use aluminum wire, but I'd argue (as a serious DIY sort) that aluminum wiring is best left to people doing it professionally, because it really helps to have things like explosively welded lugs on the end of large gauge aluminum wiring (they literally weld the lugs to the wire with explosives, and there's no way for it to oxidize at the joint as the metals are firmly bonded). If you try to treat it like copper wire, you'll have a genuinely bad time of it eventually, because it's not well behaved like copper is - it work hardens more easily, it's more prone to stress fractures and cracks, and you pretty much can't use it in screw terminals for the long term, because the thermal expansion and oxidation behavior will eventually make your screw terminals glow with any serious power through them. Wire is such a small part of any of my project costs that it's not worth the savings, over using something well behaved (copper). Though even there, I've learned about the problems with fine stranded cable - you can't use it in screw lugs, as it'll "flow" over time away from the screw, so you have to put a bootlace ferule on it first (or some sort of crimped terminal). Or just use regular stranded cable and skip the fine stranded for interconnects. I've got a range of good crimping tools, to include a DC wire crimper that will do up through 0000AWG or maybe a bit larger.
Wiring up high loads with "cables laying around," where you don't know the wire gauge, haven't calculated loads, etc, is the sort of "casually sloppy with high power things" I'm criticizing you for, in the sense of claiming that you don't have enough electrical background to be able to have any interesting conversations about your shortcuts. At least in your posts here, you come across as having a "Eh, whatever, it works for now..." attitude towards wiring. Your wiring should (if it's halfway competent wiring) be labeled with wire gauge, either in AWG, or in cross sectional area, and it should also be labeled with the insulation temperature rating. The lowest temperature wiring insulation in common use is 60C/140F rated, and I would be quite uncomfortable with that running as hot as you describe. As I said previously, I generally use 90C rated wire, though I also don't run it near the thermal limits as I like the headroom.
Is it actually conducting the same, with your particular terminals? "Adding more when there's a problem" isn't a replacement for a properly designed system, as by the time you notice it's hot, you've likely already gotten things hot enough to start causing thermal damage and further oxidation. A hunk of wire can heatsink a lot of heat, but it's not infinite. They're rated at 8A when in a 3 core sheath. Out of that sheath I expected them to handle 10A each. 8A continuous, or intermittent? Typically, in electrical rating for wiring, there's a different rating for "continuous" (>3 hour) loads vs intermittent loads, and at least in the US National Electric Code, an 80% derating applies for continuous loads. So 90C 12AWG is rated for 30A, but for continuous loads, it's only suitable to 24A. This handles thermal mass and inertia in wiring - running a blender for 5 minutes is going to heat up wiring less than a heater, solar backfeed, EV charger, hot tub heater, etc. This is why wire also generally isn't rated for "amps" - it's given in wire gauge and temperature rating of the insulation, and then you apply the standard math formulas (handily provided in table form in the NEC) to figure out what they can carry, or what you need for a given current - though you typically up with some standard answers, and if you end up with something radically divergent from what's expected, check your math. My solar install runs from the panels to the inverters are 12AWG for about 10A peak current, because of derating for temperature and number of wires in the conduit, among other things. But what I think I'm coming up against is the trend nowadays for things not to do what they say on the tin. What gauge is your wiring? I'm guessing 18AWG 60C or smaller? The US NEC at least doesn't go that small... But for the price I'll stick to aluminium! Given that you don't seem to know anything about the particular quirks and hazards of Al wiring, that doesn't seem wise to me. It absolutely is not the same to deal with as copper wiring. But at least there's an ocean between your place and mine. |
|
11)
Message boards :
Number crunching :
Completing a WU? Impossible. How's the situation today?
(Message 70769)
Posted 11 days ago by SolarSyonyk Post: Is the problem solved or is it still almost useless to crunch? If you are constantly restarting tasks, CPDN is probably still not for you. The WUs are a good bit better behaved than they've been in the past, but given the amount of work done per WU (a CPU-week if you've got a fast CPU, more if you don't), it's not a good task for machines that are regularly interrupted. They're very much a "Start them and let them run without interruption to completion" sort of task, though they are getting better. |
|
12)
Message boards :
Cafe CPDN :
Off-Grid Solar/Renewable Energy Discussion
(Message 70768)
Posted 11 days ago by SolarSyonyk Post:
How "warm" are you running your wires if they're oxidizing on you? And are you running improperly terminated aluminum? Copper doesn't do that at any sort of sane temperatures (I typically use 90C rated wire, though occasionally have to derate for 60C rated breakers/switches/etc). (in the land of "Doing it properly," I've yet to use aluminum wire for anything, and never intend to - the slight extra cost for copper is totally worth the tradeoffs of "Isn't evil in screw terminals" like aluminum is) |
|
13)
Message boards :
Cafe CPDN :
Off-Grid Solar/Renewable Energy Discussion
(Message 70766)
Posted 12 days ago by SolarSyonyk Post: Right. I'd forgotten your electrical knowledge was far enough off in the weeds that we can't actually have a reasonable discussion about what "shortcuts" are reasonable. |
|
14)
Message boards :
Cafe CPDN :
Off-Grid Solar/Renewable Energy Discussion
(Message 70762)
Posted 13 days ago by SolarSyonyk Post: They are more important than you? Yes. Absolutely. Without question. I'd rather not kill them to save a few dollars on doing a job properly, especially when I'm saving so much by doing the work myself. I don't find "doing it properly" to be a particularly painful process, either. In the context of electrical wiring, this means NEC wire gauge or larger (I discovered that past-self was clever enough to use the full rated wire gauge for a run from my charge controller to my batteries, even though I had a smaller breaker in place at the time, meaning that when I needed to upgrade the breaker to better match my increased production, I didn't have to replace the wire run), proper junction bars as needed, etc. And I have a thermal imager to find hot spots - it's quite the fun little toy that gets used often enough to verify things are running as expected. And we go overboard with safety. Obvious safety is common sense and doesn't need legislation. The history of industrial accidents, aviation accidents, and quite a few other things disagrees. Quite a few things that matter to "not kill people" are not, in fact, "common sense" - or they'd have been done. Further, there's a difference (to my mind, at least...) between "doing something stupid that kills yourself" and "doing something stupid that kills other people." Electrical code falls into the second category - I'm familiar enough with the NEC (US electrical code) to recognize that it's exceedingly conservative, but it also does a pretty solid job of preventing the sort of penny pinching stupidity that burns down homes. How pathetic. I use two hands and it won't kick right up to my face. If it did, it would cut out before it got that far. And kicking up is damn obvious for anyone who understands the most basic law of physics, equal and opposite reactions. If you don't understand it, I don't want you to live. Imagine a world where all the stupid folk had been killed off. Man, you're an ass at times. I assume you do that on purpose. But it's far from the first time I've encountered that mentality. I typically wear a helmet and hearing protection when using a chainsaw, and as I have a perfectly good set of chainsaw chaps, I wear those often enough too. I know people who've gone through 6-8 months of rather extensive pain, suffering, and loss of employment from having a circular saw go through their leg on the job. I'd rather avoid that experience, and it's a big reason I use a compound miter saw for most of my wood cutting as opposed to a circular saw. It's the same reason I wear competent motorcycle gear when riding. It's the difference between dying and surviving in pain, or the difference between surviving in pain and walking away from a crash while inventing a string of new cursewords. Been there, done that, am damned glad I had gear on when someone turned in front of me without anything like sufficient room for me to stop or get entirely out of the way. Don't get me wrong. I fully support your right to use a chainsaw with zero protection at all. I fully support people riding motorcycles in shorts, a t-shirt, and flip flops. I just happen to think quite a few things like that are stupid. And I support my right to call people out for their overt stupidity. |
|
15)
Message boards :
Cafe CPDN :
Off-Grid Solar/Renewable Energy Discussion
(Message 70741)
Posted 15 days ago by SolarSyonyk Post: Never understood why people need to be warm to take a bath. Cold temperatures are actually very good for your health. I've no problem with it personally. I'll not fault my wife for raising the temperature up a bit with a 1 or 2 year old for bath time.
Uh... okay. Cool. I'll try not to burn down the house with my wife and kids in it, for the girls in it.
It's smaller and lighter, especially if you're looking at derating for temperature and such. Just not as much as most people assume. The 12.8V/400Ah EG4-LL is just under 100 pounds. The Trojan SPRE 06-415s are a similar rated capacity (somewhat less in practice, but they're broadly close), and a pair of them to make a 12V system would be 236 pounds. So, slightly more than twice as much for a similar capability. |
|
16)
Message boards :
Cafe CPDN :
Off-Grid Solar/Renewable Energy Discussion
(Message 70732)
Posted 16 days ago by SolarSyonyk Post: Know where I could find Windows instructions/people? Not a clue. My experience with Windows for the past 6 years doesn't extend beyond "Installing Windows 10 in a VM to infrequently run some Windows-only software." I can't even interact with it as an operating system anymore, since Microsoft seems to specialize in "changing things, removing functionality, and adding more ads or MSN articles." It's fine, you just have to modify this registry key to get the old functionality back! If you can export the battery system SoC somewhere that the machines can access it, I'm sure it would be easy enough to write some scripts to poll for that and shut down, or have a Wake on LAN/Remote Sleep Command machine that manages it. I've just been going away from automation as I find my life easier and simpler without bothering with it. I recently removed the last bits of "smart" in my "smart thermostat" by removing its internet connection, because I don't use any of those features, and every other "clever feature" is has is about 50% too clever to be useful. My wife raising the temperature some Tuesday afternoon because one of our kids is taking a bath doesn't mean that the temperature needs to go up every Tuesday afternoon from there on out.
It really depends on the system, the local net metering agreements, and the costs. I'm grandfathered into a kWh for kWh plan until 2045 with my grid-tie house system. Newer installs (at least where I am) get paid about half of the retail rate for exports most of the year, but about 1.5x the retail rate during "peak summer hours" for a few months of a year, which encourages the sort of system I built - east-west facing panels, though one could be fine with just southwest facing panels too. Given our power rates and the costs of UL listed energy storage systems out here, it's hardly worth putting in batteries with a grid-tie system for that reason alone.
Indeed! And your power bill goes to $0, too, when the power company disconnects you from the grid for unlisted generating equipment connected to it, which is explicitly listed as an option they have in the relevant rate schedules to me. If they find unapproved generation attached, and there are no suitable disconnects, they'll just take the meter until you sort it out to their satisfaction. We've gone over this before, and I find nothing interesting about re-litigating it. You prefer to ignore the rules and build as cheaply as possible, my social group and I prefer to build better-quality-than-professionals sort of stuff for radically less money than it pays to get someone to slap it together. It's not hard to do it right, not significantly more expensive in materials, and it tends to both perform better long term and avoid some of the more energetic failure modes that seem to get YouTube views.
Bare cells, or complete packs with BMS? The stuff I'm using now (EG4-LLs, LiFePO4) looks to be around US$3.50/(12.8V*Ah), which looks like about £2.75? That's UL listed 4U packs, 400Ah (mine are configured as 48V/100Ah), full BMS, 10 year warranty. I've only had my set in for about 6 months, and they're as staggeringly boring as I'd hoped.
Sure, but unless your workload is purely register bound, it doesn't scale performance with clocks. Memory access latency tends to limit a lot of GPU workloads. In any case, a typical DVFS curve (dynamic voltage and frequency scaling) means that the last few MHz use quite a bit more power - the system has to crank the voltage up. I believe power consumption, handwavingly, scales linearly with frequency, and with the square of voltage - so that extra little bit of voltage uses a lot more watts per unit incremental compute per time. If you're power limited, underclocking/undervolting (or just reducing the power limits, which gets you the same thing on any modern bit of silicon) will improve compute per joule substantially. |
|
17)
Message boards :
Cafe CPDN :
Off-Grid Solar/Renewable Energy Discussion
(Message 70728)
Posted 16 days ago by SolarSyonyk Post: If only I could simply have the computers suspend (or just pause Boinc) when battery voltage is under x. Sounds like a very simple thing, but unless you're a programmer, impossible. I can think of half a dozen ways to do it. But I'm a programmer, and I operate in the Linux ecosystem.
Yes. I've got 5200W of panel hung for my office in various orientations. Which is an 8' x 12' shed of about 86 square feet internal floor space (8 square meters, if you prefer). The house system is 15.9kW of panel, though optimized for "long solar day" production vs "peak kWh per panel kW," given what I expected to happen with net metering agreements out here. ... plus another kW of panel on a solar trailer.
Yes, the bulk of my system was built 8 years ago. I added panels a few years ago (2019 or so?) with a small test A-frame before I launched into building the large house A-frames. For a DIY, code compliant, ground mound, grid tie solar install around here, $1/W is doable (I know someone who's done that), but batteries are just expensive.
That adds a LOT of power consumption for only marginally additional compute power. I'd be surprised if you got more than about 10-15% extra compute from the 50% bump in power limit. If you're power limited, go the other way. I run my 2080 (240W nominal power limit) at 125W quite a bit of the time, and it definitely isn't halving compute power. |
|
18)
Message boards :
Cafe CPDN :
Off-Grid Solar/Renewable Energy Discussion
(Message 70725)
Posted 16 days ago by SolarSyonyk Post: Ah, so pretty much manual then, I'll stick to that too. Yup. I poke the "suspend time" a couple times throughout the year, will manually change things as needed, but I've avoided too much automation on it. I just don't find automation helpful for a lot of this stuff, and I've seen it go badly wrong more than a few times. It's not a production system, it's a hobby system, but "the rest of my office" is a production system (I work from here), so I err on the side of battery state of charge when needed.
It's not, but it generally will give me day or two ahead guidance on total solar flux. I don't really care if it's raining or not, I care if there's a heavy cloud layer or not, and the forecast for that is at least better. Sometimes. Or sometimes I'm wrong and hook the backup power up. But I'm seriously overpaneled out here, so can make up a lot of it on "Well, I screwed that up, but 5kW of panels solves a lot."
I haven't totaled up my system... it's certainly far more, but I also have a system suited to a small house on and around my shed out here. It's also partly R&D for me on other projects - knowing my way around off grid power systems is useful for some other side projects I have. I'll probably start buying some newer more efficient CPU/GPUs instead of more solar. Take a look at the power settings on your CPUs. If I'm not mistaken, you have some 3900X chips too - I've disabled turbo on mine, because it adds a lot of power consumption for rather marginal additional compute throughput when the system is loaded up (and on hotter days, it starts throttling more) - I've seen a net loss in throughput, at higher power consumption (on a Wraith Pro cooler, that genuinely can't keep up with a loaded 3900X unless it's very cold in here). The same goes for GPUs - you can lower the power use on them, and substantially increase compute-per-joule. |
|
19)
Message boards :
Cafe CPDN :
Off-Grid Solar/Renewable Energy Discussion
(Message 70720)
Posted 16 days ago by SolarSyonyk Post: But you can have a sunny week or a rainy week, where presumably you adjust something. Do you get computers to run for half the day? Or reduce the cores used? If it's a sunny week, there's nothing to adjust. The computers just run. If it's particularly hot, I may let them sleep overnight to avoid cooling loads, though I'm planning to move some hardware into a box outside with filtered vent fans this year to help reduce the summer cooling demand. If it's rainy, I'll only run one computer at a time, or let them sleep, though I try to avoid grabbing too many short-lived tasks if I know the weather is going to be cloudy in the next week or so - I just set my machines to not grab new tasks, drain them out, and wait until I have more power.
That depends entirely on what the deadlines are. If it's a week deadline, it's probably a fairly short (4-6h task). If it's got a 90 day deadline, not working for a week doesn't matter much. Though outside the dead of winter, it's rare to get a week of "no compute at all." It's more a day or two at a time where I'll let the machines idle. But I can also swap around which machines run - a few of them share a power supply, so I try to run those machines together, and just toggle between what machines run one day to another to help spread the load out. Lots of ways to solve it.
I don't automate any of this beyond "Sometimes I have a cron job that will S3 suspend machines at a certain time of the evening." I used to have wake on LAN capability to power machines on remotely (from either powered off or to wake them from sleep), but I just walk out and turn the machines on physically now.
If you really want to do it, you can do wake on LAN with most hardware - you'll have to tweak some settings in your BIOS or NIC or something, but I had a Windows box that would wake-on-LAN just fine years back. But if I wanted to automate more, I'd just have scripts on the systems that poll the state of my battery (I have a VM that reads charge controller state and reports useful things) and act locally. I've just not found it to be worth the hassle, over doing it manually and keeping an eye on things. If I miss a couple hours of compute because I'm out, oh well. So be it. |
|
20)
Message boards :
Number crunching :
New Work Announcements 2024
(Message 70716)
Posted 16 days ago by SolarSyonyk Post: It's great there's new work, but it'd be even better if it would actually let machines requesting it HAVE some. At least as of right now, there's nothing meaningful in the unsent queue. The Windows machines tend to suck up the tasks in a prompt hurry. |
©2024 climateprediction.net