Overclock.net › Forums › Cooling › Water Cooling › exhausted.. days trying to figure out what is going on and its looking like temps?
New Posts  All Forums:Forum Nav:

exhausted.. days trying to figure out what is going on and its looking like temps?

post #1 of 8
Thread Starter 
So a quick back story. Have had this rig for a few years now (since the components were top of the line) and was always able to run a 4.8/4.5 overclock 24 hours prime stable @ 1.43 with temps admittedly getting into the mid 80's during the summer (during prime) but I ran that for stability, otherwise how often do you load your pc that hard?

usually high 50's during gaming which I'm fine with. Never had a blue screen or lockup (obviously after passing 24 hours of prime) I had it dialed in and no issues for years.

Fast forward I had a baby and did a lot of moving around and the pc was unhooked and sat for almost a year. I took it apart, cleaned it.. and added a new waterpump (my speed dial broke on my old one and it was stuck at like half) and I added some hose to water cool the vrms which weren't cooled before (built in vrm water blocks on my board).

After restarted the pc it was rarely posting and after days of process of elimination, naturally the last thing I checked were the temps. It was idling at 30c on one core, and high 20's on the other 3. Unfortunately the minute I figured up IBT on standard it shot into the high 80's and began throttling. this was at stock speeds! at my normal overclocked settings it wouldn't even boot properly!

I removed the waterblock from the cpu as I assumed all the moving had broken the bond from the liquid metal TIM. I took it off and sure as s---- it wasn't bonded to the chip whatsoever so I spent some time cleaning it, and re applied AS5 which is all I had on hand.

This I thought solved my issue. I am now able to boot into windows at my overclocked settings and temps at idle are roughly the same. Still, running any type of stress test immediately overloads and throttles everything.

What is going on!?

Ive flipped this thing back to front, upside down, every which way attempting to bleed it and my t line is filled to the top when the pc is running. (if I open it when its off it usually overflows)

The new pump is vibrating twice as hard as my old one (same model just with working knob so I have it on max) so I know that's good. Unsure if its possible adding the vrm cooling (admittedly uses 3/8 barbs while rest of my rig is 1/2) has overloaded my radiators or if something else is at play.

I'm sooooo stressed as ive been at this for days and I have 2 new 1080's on the way >.<<br />
Also worth noting before I reflashed the bios again, all the monitoring softwares I tried in windows were showing me the wrong speeds and voltages after about 1000 restarts and stress tests. Unsure if I corrupted the bios or this points to a problem ith the chip which is unlikely since it passes intels processor diagnostic np. Anything with the mobo that could cause the heat to be obnoxious or is this very likely some kinda loop issue?

This picture doesn't show much, ill try to take more as I fiddle when I have the heart to go back to it but I have a 360 thick XSPC rad at the top and a 240 thick XSPC rad at the front. they both have fans in push/pull on them. Chip is also delidded/naked for goodness sakes...

board warp somehow and bad contact from block even after re install? can small airbubbles still be present and really cause this bad of an issue?

due to room constraints I admittedly have my fill line at the bottom which I guess is a no. Its right before the intake on the pump which you cant see in the pic.

Anything I could be missing? I'm stumped and disheartened.



IisLfT9.jpg

rdFmEsu.jpg

DlZp83n.jpg

HMCxdRq.jpg
Edited by klepp0906 - 6/21/16 at 6:25pm
post #2 of 8
Did you drain your loop before storage? What coolant was in it?

It sounds like you are not getting any coolant flow. Check the VRM block for bugs living in the passages if it wasnt sealed while in storage. You may need to disassemble your blocks and clean them as well, they could be gunked up.
Edited by KarathKasun - 6/21/16 at 6:30pm
μRyzen
(12 items)
 
Mini Box
(4 items)
 
 
CPUMotherboardGraphicsRAM
Ryzen R5 1400 MSI B350M Gaming Pro Zotac GTX 670 4GB G.SKILL FORTIS Series 8GB (2 x 4GB) 
Hard DriveCoolingOSOS
WD Green 3tb Wraith Stealth Windows 10 Debian 8.7 
MonitorKeyboardPowerMouse
ViewSonic VX-2257-8 Chinese backlit mechanical Kingwin 850w Chinese laser optical 
CPUMotherboardGraphicsRAM
Athlon 5350 Asus AM1I-A EVGA GTX 750 Ti SC 2x4GB DDR 3 1333 
  hide details  
Reply
μRyzen
(12 items)
 
Mini Box
(4 items)
 
 
CPUMotherboardGraphicsRAM
Ryzen R5 1400 MSI B350M Gaming Pro Zotac GTX 670 4GB G.SKILL FORTIS Series 8GB (2 x 4GB) 
Hard DriveCoolingOSOS
WD Green 3tb Wraith Stealth Windows 10 Debian 8.7 
MonitorKeyboardPowerMouse
ViewSonic VX-2257-8 Chinese backlit mechanical Kingwin 850w Chinese laser optical 
CPUMotherboardGraphicsRAM
Athlon 5350 Asus AM1I-A EVGA GTX 750 Ti SC 2x4GB DDR 3 1333 
  hide details  
Reply
post #3 of 8
Thread Starter 
I was thinking a flow issue as well. Just posted more pics. Ironically my old pump was extremely weak due to the speed pot breaking off and it being over 7 years old and I got better temps previously with that by a literal mile. Now I have this new pump in which is on full tilt and admittedly the loop is al ittle different but I'm trying to find the hangup so I can fix it. Don't want to redrain and go through all this for nothing as ive already spent a few full days on it chasing false issues. Plus ill have to buy whatever I need to fix it if anything (hopefully not)

So with what you see above, does it all seem feasible? the layout? The pump goes straight into the cpu block which goes into another rad before going to the vrms. Then back to the first rad before the pump again. The T/fill line is right after the cpu and before the radiator in the last photo. One rad on the front (240) one rad up top (360).

some tight spaces exist but certainly no creases/crinkles in the lines.

To answer your questions, the loop was not drained before storage but when I redid it I drained it (obviously) and opened the waterblock and the radiators and cleaned them with vinegar and distilled water. Also use the water as coolant with no additives or coloring as I use black tubing.

Truly at a loss but I'm at such a point I wish I had a water cooling veteran who lived down the street lol. This isn't my first loop but its certainly my most elaborate and I assumed it was all feasible and it was. No idea what caused the change/disruption but its no Bueno.

Can air cause this big of a temperature issue? I didn't use a reservoir on purpose, but I foolishly have no means of bleeding apart from the fill tube which equals a lot of rocking in what must be a 70lb pc. I don't hear water moving when I tip or turn it and I don't seem to be getting any air coming into the fill line.

as I said - I'm at a loss and now talking in circles.

No way to test flow without a meter? even inaccurately? a by-the-hip method?

it has to be something major and I know its not the pump nor the waterblock or TIM as I just redid all of those, so has to be something with the loop itself?

ugh

I truly appreciate you guys taking the time to read this though >.> I wish I was browsing and chatting atm instead of losing hair tongue.gif
post #4 of 8
Ideally the fill T should be at the top of the loop since it looks like you are running W/O a reservoir. AFAIK having a fill/bleed T at the top and a drain T at the bottom is the best practice for loops w/o a coolant reservoir. My suggestion is to put a T at the top radiator with a decent length of hose you can tuck away with a fill plug on it. When you fill the system you can pull this out and have it higher than the rest of the system with it on its side. When this tube is full (clear tube helps) you seal the system up and tuck it away again. If you can hide your fill tube and drain tubes, its best to use clear hose for it so you can see the level of coolant.

Your rads are probably full of nothing but air, unless you flip the PC upside down to fill it or have a way to bleed it.

You can flow test by hooking the pump up to a reservoir, leaving the pump return open in a container, and seeing if the pump will move water to the open return line at a respectable rate. Best to power only the pump for this.
Edited by KarathKasun - 6/21/16 at 7:24pm
μRyzen
(12 items)
 
Mini Box
(4 items)
 
 
CPUMotherboardGraphicsRAM
Ryzen R5 1400 MSI B350M Gaming Pro Zotac GTX 670 4GB G.SKILL FORTIS Series 8GB (2 x 4GB) 
Hard DriveCoolingOSOS
WD Green 3tb Wraith Stealth Windows 10 Debian 8.7 
MonitorKeyboardPowerMouse
ViewSonic VX-2257-8 Chinese backlit mechanical Kingwin 850w Chinese laser optical 
CPUMotherboardGraphicsRAM
Athlon 5350 Asus AM1I-A EVGA GTX 750 Ti SC 2x4GB DDR 3 1333 
  hide details  
Reply
μRyzen
(12 items)
 
Mini Box
(4 items)
 
 
CPUMotherboardGraphicsRAM
Ryzen R5 1400 MSI B350M Gaming Pro Zotac GTX 670 4GB G.SKILL FORTIS Series 8GB (2 x 4GB) 
Hard DriveCoolingOSOS
WD Green 3tb Wraith Stealth Windows 10 Debian 8.7 
MonitorKeyboardPowerMouse
ViewSonic VX-2257-8 Chinese backlit mechanical Kingwin 850w Chinese laser optical 
CPUMotherboardGraphicsRAM
Athlon 5350 Asus AM1I-A EVGA GTX 750 Ti SC 2x4GB DDR 3 1333 
  hide details  
Reply
post #5 of 8
Thread Starter 
Ugh. Tell me something... Does that look like too much gear for the one d5? I have had the PC on every side possible now and it's currently running on a low pump speed upside down with the t line taped to my wall straight up in the air lol.

I've largely had no air bubbles come up or out at all. The problem with this approach (upside down) is that the outlet for the pump is pointing straight down. Doesn't this mean any potential air will just rise up into the pump as its trying to push downward?

Such a convoluted loop frown.gif. I'm really beginning to wonder if it's flow or air or something else entirely. I mean how can I be hitting 80c at stock or less than stock if I have speedstep on during IBT?

So here's where I'm at... Can't spend much of any cash... Especially when I don't know for sure what the issue is exactly. I only have the 1 T line/fitting is a drain line necessary or is a fill line adequate? Should I relocate the t from its current location to the top or is it serving the same purpose where it is now so long as I fill upside down?

Bleeding was always simple for me in the past but I had 1 Rad and 1 block (have never used a rad). If air is somehow trapped that badly that it's stopping flow id be amazed.

Would I be able to hear the water moving around the loop when I tilt the case if air was trapped or not necessarily? Or from another perspective should a fully filled/bled loop be quiet when you tip it side to side? Ie are you always shooting for 100% full of water - 0 air?

Is there anything else it can be making my non idle temps shoot up like a lightening rod to worse than air levels? What can I do going forward easiest to most difficult as a process of elimination?

I'm fried - sorry this is so long and dependent and all over the place. It's coming up on midnight and I've been at this since 4am this morning not counting the past few days ruling everything else out.

Has to be either water block not mating (unlikely), flow ultra low due to blockage, or air hiding somewhere? Or some kinda pump issue?

Thanks sincerely for the reply smile.gif.
post #6 of 8
Thread Starter 
ran all night, power cycled and checked in the morning and basically no change in fluid level. checked temps and theyre just as bad even running upside down with gravity helping the pump push.

I'm thinking someone plugged the inside of one of my tubes as a cruel joke at this point lol.

that or truly too much for the pump to push through. Probably 9 feet of tube which 8 is 1/2" and 1 is 3/8" ID. 2 rads, a mb block and a cpu block.

truly don't know. Ive noticed my multiplier changes aren't saving when I boot into windows now so I'm thinking my board may be bad and the culprit? I know if I flash they will change again but for how long, who knows?

this is a disaster tongue.gif
post #7 of 8
You haven't mentioned what your voltages are while running IBT. Your motherboard on auto may be overvolting your cpu. That would cause the high temps you are seeing. Just a thought.
post #8 of 8
Have you checked for proper flow yet? Disassemble the loop and check all of your tubes. A clog is not out of the question yet. Check EVERY piece as you disassemble.

You can also hydrolock your loop and/or pump. AFAIK, you have to tap on the rads/pump while they are low points in the loop to release hydrolocked air bubbles.

If anything, rebuild the loop, adding one part at a time and see what works and what does not.

One step at a time is the best way to debug. If you assume anything is A-OK it will come back to bite you.
μRyzen
(12 items)
 
Mini Box
(4 items)
 
 
CPUMotherboardGraphicsRAM
Ryzen R5 1400 MSI B350M Gaming Pro Zotac GTX 670 4GB G.SKILL FORTIS Series 8GB (2 x 4GB) 
Hard DriveCoolingOSOS
WD Green 3tb Wraith Stealth Windows 10 Debian 8.7 
MonitorKeyboardPowerMouse
ViewSonic VX-2257-8 Chinese backlit mechanical Kingwin 850w Chinese laser optical 
CPUMotherboardGraphicsRAM
Athlon 5350 Asus AM1I-A EVGA GTX 750 Ti SC 2x4GB DDR 3 1333 
  hide details  
Reply
μRyzen
(12 items)
 
Mini Box
(4 items)
 
 
CPUMotherboardGraphicsRAM
Ryzen R5 1400 MSI B350M Gaming Pro Zotac GTX 670 4GB G.SKILL FORTIS Series 8GB (2 x 4GB) 
Hard DriveCoolingOSOS
WD Green 3tb Wraith Stealth Windows 10 Debian 8.7 
MonitorKeyboardPowerMouse
ViewSonic VX-2257-8 Chinese backlit mechanical Kingwin 850w Chinese laser optical 
CPUMotherboardGraphicsRAM
Athlon 5350 Asus AM1I-A EVGA GTX 750 Ti SC 2x4GB DDR 3 1333 
  hide details  
Reply
New Posts  All Forums:Forum Nav:
  Return Home
  Back to Forum: Water Cooling
Overclock.net › Forums › Cooling › Water Cooling › exhausted.. days trying to figure out what is going on and its looking like temps?