i filed a warranty claim on my card which is less than 4 months old and had a mostly dead front fan which i discovered only because my windows nvme was acting weird so i cracked open the case to see what was going on while the machne was on and was slapped in the face by some major heat. looked at the gpu and saw only one fan was spinning. tried to spin the dead fan and it kinda did recover some spinning but truly wasn't spinning up on t's own. then i tried to install msi after burner to get the fan running but that was no help. now i have no working card and while gigabyte is honoring the warranty which is 3 years, i now have to add the cost of shipping a defective card to them on top of the purchase price of the card. i am not happy. you can bet i made some noise on my amazon review which was updated to reflect my unhappiness with a dead fan at less than 4 months.
https://www.ebay.com/itm/354640572696?hash=item529239b118:g:6Y8AAOSwrhNkC99N&amdata=enc%3AAQAIAAABAKV6hDDFeDeDeto36iKaqvtX8A3TN7Rks7rvCY4d3nCiF%2BqnAVmw4FkcXXArTpvp9e4ZwfgncOOr4bhSyQCOuqScQCnT1tFyP6WPKOAds%2Fl%2FX%2FDZOSKmYZ980zZjpwPWp0bYdrklatYDLK7IJ1IK5KiH9eyyMRFzT8LgdoAeq3yImPtSaztr1MD9Z%2B0LTZE1ip3GBZLrwMdWNVrqZgudTQfzr1i2QvGtlRwgAfIugQYqj%2F68IFvN4NoBfbRkrru9Ms20Ady9ymdeBrDFgWp%2FPOtaFeIn2fJyFU5xZGQDw7icW0ghxozndDRwpbtcVf90TLbf2jJTt6%2Fkx0JmWDBA%2BHA%3D%7Ctkp%3ABFBMksyBq7Ni
the replacement fan costs less to buy new than to ship the card back to gigabyte and though it ships from china it still arrives faster than the warranty claim process would take.
there are six very tiny screws that hold the fans and the shroud to the heatsink. the rear fan plugs into the pcb of the graphics card and has three pins. the front fan plugs into a connector coming from the rear fans leads and has two pins. The hardest part of this repair is not losing those tiny screws. the low profile faceplate doesn't even need to come off.
I wonder if i could contact an actual person if i could get them to send me replacement fans and skip the whole rigamarole of sending the card back and forth?
https://www.amazon.com/FS1250-S2053A-Graphics-Cooling-Replacement-Gigabyte/dp/B0CB2DZJPD/ref=sr_1_2?crid=2P8FOG3BF6LYB&keywords=Gigabyte+GTX1650+4GB+Graphics+Card+Cooling+Fan+FS1250-S2053A+Replace+Part&qid=1690532907&sprefix=gigabyte+gtx1650+4gb+graphics+card+cooling+fan+fs1250-s2053a+replace+part%2Caps%2C85&sr=8-2
amazon has the first item listed coming in at 15$ us with both fans and thermal paste and a plastic scraper... sweet.
Never fix something under warranty, it invalidates the warranty, and it might go wrong again in a more terminal manner yet. Send it back would be my advice learned from experience in these situations, and there is no guarantee your fix will work - what if it isn't the fan itself at fault but the electronics controlling it?
You Def make an excellent point there.
In this case though I'm pretty sure it's the fan. When I manually spin the fans with my finger the questionable fan hardly spins while the good one glides along for some time. It's night and day.
Even so, i hear you. It really is their problem to fix. I just have nothing in the interim.
Therein lies the rub.
To make sure it's the fan not the pcb you could swap the fans.
Laws vary much country to country. But in the UK the responsibility lies with the retailer to make good problems with faulty products (though many try to claim you have to contact the manufacturer. They lie, or are simply clueless).
With that in mind one tactic I've used myself is to negotiate with the retailer and buy a new one. With the understanding that the return will be refunded.
Then I'm not without the product and also the warranty restarts.
What you can't be sure of is that a faulty voltage supply could have caused the fan to fail in the first place, so even if a replacement initially worked it could get fried again. If it wasn't under warranty it would certainly be worth trying, but here it's someone else's problem - better to spend time researching cheaper shipping options I think. Many times on eBay you can get a return shipping label to print out free of charge, obviously that will vary according to the seller and this was Amazon not eBay, but it could be worth asking about.
Tiny fans are garbage unless they are those thick 4000rpm Deltas. But what am I saying cause I had a Delta fan fail on me too last month, lol, with sparks and everything.
I'm pretty sure nothing is wrong with the card. Especially if the fans run under a splitter under that shield and the other is working, and yeah do swap them around just to make sure.
They are Daisey chained and can't be swapped. The main connector is three pin the second connector is two pin. See photo. The one on the two pin connector is the failed one.
Amazon only gives 30 days I think for returns. They directed me to the manufacturer. Not sure if the retailer can be compelled to assist me in the USA.
Maybe Sunday I'll call them and ask.
Oh okay, i'm sure now they are in parralel cause the first one is 3 pin for speed sensor while the other is just powered.
So you can replace it, even if replacement is 3 pin just wire it/botch it :-) to the existing plug.
they def sell both the 3 pin and the 2 pin on ebay or the set. the 2 pin i can unplug from the daisy chain and replace for less than 10$ shipped for free from ebay seller. or pay 15 for a new set with a scraper and thermal paste, which is funny as no thermal paste is needed to replace the fans, just that it's a good idea on these cards to repaste them for better thermals though that would also require the use of plastic washers not included to get more torque on the screw but that's not something i'm interested in doing really. just fixing the fan is the goal.
I had just finished upgrading it with an m.2 wifi and an external antenna. I have an nvme for windows on the mb and an nvme to pcie adapter on the 4 lane slot in front of the graphics card fans with only the m.2 wifi antenna adapter between them. thought i was finally done. :(
sudo inxi -Fxzmc30
System:
Kernel: 6.4.6-lqx1-1-lqx arch: x86_64 bits: 64 compiler: gcc v: 13.1.1
Desktop: MATE v: 1.27.0 Distro: Artix Linux base: Arch Linux
Machine:
Type: Desktop System: HP product: HP EliteDesk 800 G3 SFF v: N/A
serial: <filter>
Mobo: HP model: 8299 v: KBC Version 06.29 serial: <filter> UEFI: HP
v: P01 Ver. 02.46 date: 03/28/2023
Battery:
Device-1: hidpp_battery_0 model: Logitech Wireless Keyboard K360 charge: 100%
(should be ignored) status: discharging
Memory:
System RAM: total: 32 GiB available: 31.1 GiB used: 7.14 GiB (23.0%)
igpu: 32 MiB
RAM Report: missing: Required tool dmidecode not installed. Check
--recommends
CPU:
Info: quad core model: Intel Core i7-7700 bits: 64 type: MT MCP
arch: Kaby Lake rev: 9 cache: L1: 256 KiB L2: 1024 KiB L3: 8 MiB
Speed (MHz): avg: 3601 min/max: 800/3601 boost: enabled cores: 1: 3601
2: 3601 3: 3601 4: 3601 5: 3601 6: 3601 7: 3601 8: 3601 bogomips: 57600
Flags: avx avx2 ht lm nx pae sse sse2 sse3 sse4_1 sse4_2 ssse3 vmx
Graphics:
Device-1: Intel HD Graphics 630 vendor: Hewlett-Packard driver: i915
v: kernel arch: Gen-9.5 bus-ID: 00:02.0
Display: x11 server: X.org v: 1.21.1.8 driver: X: loaded: intel
unloaded: fbdev,modesetting,vesa dri: i965 gpu: i915
resolution: 1920x1080~60Hz
API: OpenGL v: 4.6 Mesa 23.1.4 renderer: Mesa Intel HD Graphics 630 (KBL
GT2) direct-render: Yes
Audio:
Device-1: Intel 200 Series PCH HD Audio vendor: Hewlett-Packard
driver: snd_hda_intel v: kernel bus-ID: 00:1f.3
API: ALSA v: k6.4.6-lqx1-1-lqx status: kernel-api
Server-1: JACK v: 1.9.22 status: off
Server-2: PulseAudio v: 16.1 status: active (root, process)
Network:
Device-1: Intel Ethernet I219-LM vendor: Hewlett-Packard driver: e1000e
v: kernel port: N/A bus-ID: 00:1f.6
IF: eth0 state: up speed: 1000 Mbps duplex: full mac: <filter>
Device-2: Intel Wi-Fi 6 AX210/AX211/AX411 160MHz driver: iwlwifi v: kernel
bus-ID: 02:00.0
IF: wlan0 state: down mac: <filter>
Bluetooth:
Device-1: Intel AX210 Bluetooth driver: btusb v: 0.8 type: USB bus-ID: 1-5:2
Report: rfkill ID: hci0 rfk-id: 0 state: up address: see --recommends
Drives:
Local Storage: total: 1.95 TiB used: 553.91 GiB (27.8%)
ID-1: /dev/nvme0n1 vendor: Western Digital model: WD BLACK SN770 500GB
size: 465.76 GiB temp: 31.9 C
ID-2: /dev/nvme1n1 vendor: Fanxiang model: S500PRO 512GB size: 476.94 GiB
temp: 39.9 C
ID-3: /dev/sda vendor: Seagate model: ST1000LM035-1RK172 size: 931.51 GiB
ID-4: /dev/sdb vendor: Micron model: MT-128 size: 119.24 GiB
Partition:
ID-1: / size: 53.61 GiB used: 24.34 GiB (45.4%) fs: ext4 dev: /dev/nvme0n1p2
ID-2: /boot/efi size: 511 MiB used: 444 KiB (0.1%) fs: vfat
dev: /dev/nvme0n1p1
ID-3: /home size: 403.48 GiB used: 226.73 GiB (56.2%) fs: ext4
dev: /dev/nvme0n1p3
Swap:
ID-1: swap-1 type: partition size: 64 GiB used: 0 KiB (0.0%) dev: /dev/sda1
Sensors:
System Temperatures: cpu: 41.0 C mobo: N/A
Fan Speeds (RPM): N/A
Info:
Processes: 304 Uptime: 22h 9m Init: dinit Compilers: gcc: 13.1.1
clang: 15.0.7 Packages: 989 Shell: Sudo v: 1.9.14p3 inxi: 3.3.28
heh, speaking of thermal paste on video card, you need a good one for these modern ones, my rx5500xt worked well with this premium japanese thing that was better than arctic mx4/mx5. with generic paste it was bad, so if you have good temperatures don't replace it, maybe just blow the dust with pressure or something.
LE: nice. i was tempted to get pcie sound card but interference is bad so usb is still for me.
Time placedJul 31, 2023 at 10:35 AM
Order number some number
Total$9.36 (1 item)
ok so for less than 10$ since it's daisy chained and we can know it's not the controller chip as both fans connect together, i'll try this and see if i can get a functioning unit.
There's a high probability it will be fine and work perfectly for many years, but fans are often speed controlled by a pulse width modulated output, say it was stuck on full power permanently or the supply was outputting an excessively high voltage due to some regulation issue, something like that could have caused the fan to overheat and fail prematurely, even if it uses the same supply as the other one it could just have been the first to burn out, that was the kind of thing I meant earlier. Wishing you good luck though! ;D
i think i will repaste my cpu. there's not much else i can do with the heat it produces, maybe replace the cpu fan? it's likely the original fan. it spins good but i don't think it's blowing like it should either. i ordered some corsair thermal paste.
This replacement cooler looks an awful lime the stock one....
and so the new fan arrived a few hours ago. very fast delivery for a fan that doesn't spin. the old fan spun more than this one. this one jerks slightly at powwer on but does nothing. the old fan can be spun a few times manually to spin a little but without power barely moves when manually spun. The so called new fan spins a little when spun manually without power but does not spin at all under power and spinnng it up manually does nothing.
You could check the part number on the fan and see if you can find a datasheet online with the input voltage spec, then you could possibly supply a suitable voltage to test it, and the original too for that matter. It might even have it written on it. If it's not plain DC that could be more difficult, and you might need to have the correct hz as well as voltage.
The new fan has a similar but different font than the one front the card.
So a 9 v battery ? Like for a smoke alarm? Costs more than I paid for the fan.
took a 9v battery from the smoke detector in the hallway. both fans spin like mad when connected to the 9v battery. when plugged into the daisey chain the old fan spins up but the new fan does not spin up. so i'm calling the new fan bad and it's probably the connector.
i sent him 3 indignant messages as he requested a video and i thought that was outrageous because what would be the point of that. it clearly doesn't work and i thought it a beginning of a run around. but now that you considered an alternate power source and i can show him how his fan clearly doesn't run even when the old jenky fan does i think i'll send him that video.
i'm assuming the 9 volt lithium battery has way more than .19 amps and that it has so much more that the old jenky fan is overpowered by the amps it puts out and runs though with some hesitation. the replacement fan runs not at all when connected to the daisy chain. as soon as it is ready ill link the videos.
And so i reversed the poles at the battery and then the first fan on the gigabyte chain didnt spin but the new gigabyte fan does. the new fan is wired backwards.
i'm half tempted to cut the wires and resplice them on the new fan. I just dont have the proper tools to do that right it would be electrical taped at best. maybe duct taped. of course i have no guarantee it will run at the proper amperage supplied by the graphics card.
https://youtu.be/UCgUHFhCcP4
Sounds more promising, some aftermarket replacement CMOS batteries are sold with terminal plugs with the wrong number of pins so you need to solder on the old connector. The fan draws 0.19A at 12V, it would use a bit less current with a 9V supply and run slower. The reason it runs fast is because PWM control switches the voltage supply on and off rapidly to limit the current and give a method of speed control, probably the first fan has an extra wire for a speed sensor output:
https://en.wikipedia.org/wiki/Pulse-width_modulation (https://en.wikipedia.org/wiki/Pulse-width_modulation)
Without that it would run at max speed when given a 12V constant current supply.
Amps = Volts / Resistance in Ohms
Volts x Amps = Watts (power) (but if the supply was off for a percentage of time the power would reduce in the same proportion)
In a motor the resistance is mostly inductance giving a resistive equivalent.
The new fan might have tighter bearings which is why it doesn't spin by hand so freely, it could free up more with use.
so is it really that the new fan is wired with the wrong polarity? the wires reversed? i think thats what im seeing
well there it is i cut the wires and reversed them so both fan ran at the same time and the new fan spins up all nice and pretty
old cpu fan is now trimmed so it fits inside the case. used a hacksaw rather than a file. less than 5 minutes. make a great ase fan that def ramps up the speed when the cpu gets toasty. the gpu runs about 83 degs Celsius and the cpu about 75-80 degs Celsius high end. I can live with that.