Sign in with
Sign up | Sign in
Your question

Want to add fan to studio xps 7100

Last response: in Systems
Share
February 1, 2013 9:51:26 AM

I have a Dell studio xps 7100. and i believe I have a bad heating issue. My room is always very hot. I've been keeping the side of the case open, and making sure gpu temp is under 70 and cpu temp is under 55. But i think i still have a heat issue.

When i'm gaming.....after a while the game starts to feel sluggish and i start getting hitches. I'm also replacing my ram to rule that out.

My cpu cooler doesn't even feel warm to the touch? but even at idle the side of gpu, opposite side where fan is, is burning hot. Also the NB cooler on this pc is scorching too. NB is at 65-70 degrees at idle. and 55 degrees at full load. I'm not sure why the temp is lower when the pc is at full load. I thought the temp sensor wrong. but its scorching hot to the touch at idle. and only a little warm at full load.

Here is a pic of inside the case same pc i got. http://www.tomshardware.com/forum/322225-28-upgrading-7... I just don't have anything in that pcie x1 slot.

there is no other spots on mobo for pluggin a fan. i would have to plug one to the power supply. I have a corsair cx600 psu. Should i put a small fan on top of that heatsink? would i even be able to fit anything there, would that help gpu too? Theres no way to screw anything in. Thats also the side of the vid card that is extremely hot. all the heat is in that area at the back of the case between the case fan and the vid card. should i jerry rig another fan right there just blasting air to the front of case? was thinking maybe this http://microcenter.com/product/311399/EVERFLOW_DC_Fan_-...

I'm wondering if this is a dumb idea or not. I appreciate any input. thanks for the help.
February 3, 2013 3:10:47 PM

Hello Rich :hello: 

First off judging from that image your case seems pretty tight I can imagine your having some heating issues. I assume your rear fan is blowing out. I think if you can take off the face plate of your case you could put a 120 mm fan in the front on the bottom blowing in right at the GPU. and use a cheap fan controller to split your fan connection and control fan speed.

I would try a Zalman Fanmate Controler with a Rosewill 120mm fan, to keep it cheap. just install the new fan, unplug the existing one, plug them both into the Zalman and plug the Zalman into the MOBO, then try to figure out the lowest fan speed to keep noise down and components cool.

Just my 2 cents do with it what you will.
m
0
l
February 3, 2013 10:54:02 PM

tks for the reply man i'll check to see if this faceplate comes off. only problem is i have no free fan port on the mobo :( 
m
0
l
Related resources
February 3, 2013 11:17:25 PM

RichAC said:
tks for the reply man i'll check to see if this faceplate comes off. only problem is i have no free fan port on the mobo :( 



I was just reading my post and realized you don't need the Zalman controller the fan comes with one. As for no power I'm sure you have a free molex plug in your case, try this.
m
0
l
February 4, 2013 1:26:56 AM

hillmanant said:
I was just reading my post and realized you don't need the Zalman controller the fan comes with one. As for no power I'm sure you have a free molex plug in your case, try this.



cool. tks man.
m
0
l
February 4, 2013 2:02:06 AM

P.S. actually opening the side of the case can produce more heat problems. The case is designed so that the fan in the back pulls air from the front of the case over and around the components inside and then out the back taking the heat with it. If you remove the side the fan pulls no air over the components it just draws it from the side and defeats the purpose.
m
0
l
February 4, 2013 3:20:03 AM

oh interesting... i thought the cpu was a a couple degrees cooler with it off...but maybe i'm wrong. i'll try that out. tks for the tip.
m
0
l
February 24, 2013 8:49:16 PM

Did this work out for you, were you able to stuff a fan in the case somewhere? Got any pics of your Fix?
m
0
l
February 24, 2013 9:37:16 PM

actually what i ended up doing is just raising the fan speed on the gpu manually as needed. so i can keep the gpu at around 55 degrees on load and it doesn't feel as hot. i usually have it between 35 and 50% fan speed on full load depending on the game. and i put it back to automatic quiet mode when i am not gaming.

It turns out the the hitching in mmos is very common in all mmos. and the hitching in the bf3 towards the end of long matches is also very common. The only thing that would prolly help me is an upgrade of my cpu and hdd. and thats not even a definite. PC gaming is in trouble imo. Apparenlty alot of people have it much worse then me... at least I am still able to enjoy the games. I don't believe its a heat issue anymore.

I also put the case back on the side per your advice. the cpu is one degree hotter. but I feel that the outside of the gpu and mainboard is prolly getting better air flow that way. Plus less dust getting inside.


If I do decide to try and add a fan in the future...maybe over the summer. I will keep you updated

I appreciate your advice.

Rich.
m
0
l
February 26, 2013 3:15:47 PM

If you are still interested in fixing your problem, I think I have an answer - your CPU and exhaust fan speeds are controlled by motherboard, usually Dell doesn't let users to adjust those fan speeds, so this is the question to you, how the noise when CPU is running full blast, the only cheap way is to get replacement exhaust fan with additional 4 pin Molex connector and to make this fan to run at 100%, CPU cooler is more difficult - there are 4 wires, one of them is censor wire, if this wire is no more, your CPU fan will run at 100%, another way is to change CPU cooler, not difficult task, considering your posted temperatures. I managed to Overclock Dell Inspiron 570 with AMD Phenom II x4 965 BE and it run at 60s last Summer at testing, so I am trying to say that you running hot for Winter, Summer will melt you XPS.

I just realised, your XPS 7100 has AMD processor, which one - the maximum Phenom II X4 965 BE temperature is 62C, above this is melting point.
Once again which processor you have, if it is Black Edition - it could be overclocked if you want.
m
0
l
February 26, 2013 4:12:12 PM

Kisianik said:
If you are still interested in fixing your problem, I think I have an answer - your CPU and exhaust fan speeds are controlled by motherboard, usually Dell doesn't let users to adjust those fan speeds, so this is the question to you, how the noise when CPU is running full blast, the only cheap way is to get replacement exhaust fan with additional 4 pin Molex connector and to make this fan to run at 100%, CPU cooler is more difficult - there are 4 wires, one of them is censor wire, if this wire is no more, your CPU fan will run at 100%, another way is to change CPU cooler, not difficult task, considering your posted temperatures. I managed to Overclock Dell Inspiron 570 with AMD Phenom II x4 965 BE and it run at 60s last Summer at testing, so I am trying to say that you running hot for Winter, Summer will melt you XPS.

I just realised, your XPS 7100 has AMD processor, which one - the maximum Phenom II X4 965 BE temperature is 62C, above this is melting point.
Once again which processor you have, if it is Black Edition - it could be overclocked if you want.



How do you overlock the cpu if the dell mobo doesn't let you adjust the voltages?

and I thought the problems with black editions.... is you can't get the core temps and have to rely on the mobo socket temps? which are always about 5 degrees cooler then the actual cpu.


my cpu core temp is running hot at 50 degrees on full load? what? I thought that was normal? are you sure your not confusing my gpu temps with cpu temps dude?

the fan speed at 30 degrres is about 800-900 rpms. at 50 degrees its about 2500-3000 rpms. the only time i've seen the fan go loud and jump to 4000 rpms. is when the cpu socket temp has spiked to 60 degrees. not the cores. I think it might be a a faulty sensor on the mobo i don't know. or maybe i have some dust in the socket affecting one of the cores. it only happens for a second or two then goes back to normal. The fan speed adjusts according to mobo socket temp.



I think if it gets hotter in the summer. it will only be by a couple degrees man. I've had the pc for a few years already, temps haven't changed much. But i guess i will take off the cooller if thats the case and dust out the socket and what i could do i guess is add some aftermarket thermal grease if it gets much hotter over the summer.

I'm just weary of taking off the cooler and checking the socket for dust and applying new grease....becase the cpu fan screws in. And i'm wondering do i have to remove the whole motherboard out of the case to put the cpu fan back on? In the past i've always used cpu fans that just snapped into place.
m
0
l
February 26, 2013 4:31:58 PM

First of all, I measure temps with OCCT, I like its interface, also it gives me all info I could wish, including motherboard temperature (maybe it is NB temperature), which is usually higher than CPU temperature. Also, OCCT shows mobo socket temp and 4 cores temp!

In order to overclock Dell with AMD chip you have to have Black Edition CPU, if you don't, there is nothing I can do, so which one you have? There is a trick to do so, very simple actually. I was able to overclock Phenom II x4 965 BE (officially not even supported processor for Inspiron 570, but I was the first to install it, I think in Universe, who did it and documented this) from stock 3.4 GHz to 4 GHz on same, low voltage I think max VCore is 1.39 V, idle 1.36 V - very stable, checked with Prime95 and IntelBurn. And I was second in Universe to overclock it. I actually created complete guide at Dell Forum http://en.community.dell.com/support-forums/desktop/f/3... it is not complete yet as I am trying to polish it.

Regarding your CPU temps, are you talking Celsius? Full load at 50C is OK, but how it rises during Summer time? Again I tested mine with Prime95 and IntelBurn last summer and the highest temp I think was 63.5C, which is a bit above AMD max of 62C, but it was during testing, and normal apps and even games never exceed OCCT light testing temps.
m
0
l
February 26, 2013 5:06:08 PM

Kisianik said:
First of all, I measure temps with OCCT, I like its interface, also it gives me all info I could wish, including motherboard temperature (maybe it is NB temperature), which is usually higher than CPU temperature. Also, OCCT shows mobo socket temp and 4 cores temp!

In order to overclock Dell with AMD chip you have to have Black Edition CPU, if you don't, there is nothing I can do, so which one you have? There is a trick to do so, very simple actually. I was able to overclock Phenom II x4 965 BE (officially not even supported processor for Inspiron 570, but I was the first to install it, I think in Universe, who did it and documented this) from stock 3.4 GHz to 4 GHz on same, low voltage I think max VCore is 1.39 V, idle 1.36 V - very stable, checked with Prime95 and IntelBurn. And I was second in Universe to overclock it. I actually created complete guide at Dell Forum http://en.community.dell.com/support-forums/desktop/f/3... it is not complete yet as I am trying to polish it.

Regarding your CPU temps, are you talking Celsius? Full load at 50C is OK, but how it rises during Summer time? Again I tested mine with Prime95 and IntelBurn last summer and the highest temp I think was 63.5C, which is a bit above AMD max of 62C, but it was during testing, and normal apps and even games never exceed OCCT light testing temps.


the only thing that concerns me is the cpu socket temp spikes. doens't matter what program i use. argus. overdrive. cpu-z etc.... it has nothing to do with montoring program.. because i can actually hear the fan get louder which is controlled by bios. it will spike from like 45-60 for maybe one second or two and then go back down.

I don't see socket temp spikes when playing bf3....but I do see them when running prime 95, or even every 30 mins in skyrim. max cpu temp in prime 95 was 62c after a few hours. and the socket would start spiking after the first 30 mins of running prime 95, and then keep doing it every couple mins. the core temps don't change when it does this only the socket temp. but when playing a game like bf3 i get no socket spikes and the max core temp never goes above 50c.

It also doesn't seem to affect the performance when it happens in skyrim. and it doesn't happen in any other game not even bf3 which actually runs hotter and is more demanding then skyrim. but i can recreate the issue in prime 95 when it starts to happen after the first 30 mins. What could be causing this?

is this a problem i should worry about? why does this happen and is there anything i can do to fix it? is it a faulty socket sensor? a problem with the voltage regulator? dust? is there anything that can be done with it....or should i just live with it until the mobo dies. which may not even happen.

The only thing annoying thing is just hearing the fan speed up for one or two seconds. which doesn't happen in any game except skyrim and only for one or two second every 30 mins.

I just find it very strange.
m
0
l
February 26, 2013 5:12:26 PM

Except reapplying good thermal paste - nothing, unless you want to change CPU cooler.
And I guess you don't have BE CPU, so no overclocking here.
m
0
l
February 26, 2013 5:14:46 PM

Kisianik said:
Except reapplying good thermal paste - nothing, unless you want to change CPU cooler.
And I guess you don't have BE CPU, so no overclocking here.


No i don't have a BE cpu. once i mentioned it on the dell forums... a day later it was sold out on newegg. lmao. I kid you not.

But after researching alot of games.... apparently especially for the mmo games it would only make a slight diff. I've talked to people in games with the 965 BE, the phenom II's are not very optimized for gaming even with overclocked cores.


apparenlty gaming in general is a bubble about to burst. All this new hardware is useless. I dont' know if its that we are reaching a wall with just our human limitations in understanding. Or if it is corruption and money holding us back. I tend to think its all about money, because believe it or not the only game right now that benches better then Intel on new age amd systems.... is COD. cooncidence? lol.


The fact is....games are running worse and worse on computers.....and are costing more and more money not less. Because the hardware is becoming more and more geared for parallel processing. And the programmers are still programming for older technology and prefer single layer programming. Even in this day and age.

So mutli cored processors and Dx11 graphic cards....are still only theoretical performance and not being use to their full potential. And they run alot of these games even worse then older hardware because the software is just not optmized for the new hardware.. I think this is all gonna come to a head soon. Not everybody is rich and pc companies are taking a big hits. Not just the gaming companies...but the whole industry in general is dying. Especially with androids and tablet pcs to do every day pc tasks, and consoles almost as powerful to run games.



But not to get sidetracked....have you ever heard of such a thing though with the cpu socket though?

would new grease or a new cpu cooler even make a diff? If the mobo socket sensor is detecting temp spikes, but the cpu core temps are fine?

only thing i can think of is dust in the socket? is that possibly affecting anything? or maybe just a bad sensor.

my worst fear is the voltage regulator is going bad. i've upgraded bios and bought a new power supply didn't make a diff.

I might of got a bad virus impossible to get out of my bios or hdd. and the only reason my pc is even still workable....is because it is a dell that doesn't let you overclock anything. so nothing got badly damaged. yet.... but over time this may change if the voltage regulator totally starts to fail.

All i can say is thank god i got a dell. cause if I built my own pc. These russians would of bricked my pc by now lol. Tk god dell also notified me of the firmware update for my hdd. Cause the hdd was dying all across the country. All they can do is spy on my like a homo sexually attracted to me. I know i know,, it sounds like a sci fi movie...lol.

But as long as i can play my games. And as long as i know any performance issues are not just me...but a problem for most people around the world, theres nothing i can do but hope times change things.
m
0
l
February 26, 2013 6:53:55 PM

For sake of argument I am running Prime95 now, will run short OCCT as well and will post results, and a bit more talking about games, Battlefiled 3 in particular.
m
0
l
February 27, 2013 12:05:35 AM

Let me say a few words about gaming. I tested on Battlefield 3, prior to overclock I had about 60 FPS and after overclock, FPS increased to about 78, not bad I say. I have stable overclock, I could increase even higher, but I am tight on money and have no time to change motherboard if something happen.

I just finished running Prime95 and IntelBurn (this one is much worse than Prime95), the worst CPU Core temp was 50.5C or so, CPU socket was actually lower at 48C and motherboard at 71C. Just to compare, with Prime95 those numbers are 47.5C, 47C, 62C - Prime95 is weak test.

I posted pictures at http://www.tomshardware.com/forum/page-261868_29_1300.h... AMD Overclocking Club thread.

Take CPU cooler out, clean it and place back with Arctic Silver 5, use at least alcohol to clean parts, but for best results Arctic Silver has 2 step cleaning solution, I recommend for old dried paste.
m
0
l
February 27, 2013 2:49:50 PM

Kisianik said:
Let me say a few words about gaming. I tested on Battlefield 3, prior to overclock I had about 60 FPS and after overclock, FPS increased to about 78, not bad I say. I have stable overclock, I could increase even higher, but I am tight on money and have no time to change motherboard if something happen.

I just finished running Prime95 and IntelBurn (this one is much worse than Prime95), the worst CPU Core temp was 50.5C or so, CPU socket was actually lower at 48C and motherboard at 71C. Just to compare, with Prime95 those numbers are 47.5C, 47C, 62C - Prime95 is weak test.

I posted pictures at http://www.tomshardware.com/forum/page-261868_29_1300.h... AMD Overclocking Club thread.

Take CPU cooler out, clean it and place back with Arctic Silver 5, use at least alcohol to clean parts, but for best results Arctic Silver has 2 step cleaning solution, I recommend for old dried paste.


how long did you let prime 95 run? I think you need to let it run for at least an hour or two.

whats your cpu at idle?

i max at 50c when playing a 64 man man in bf3. you think this is too high?

cpu socket temp is always about 5 degrees cooler then core temps.

what i'm wondering about is what could make the socket temp spike up and down dramatically in prime 95?
m
0
l
February 27, 2013 3:10:19 PM

You obviously did not read the whole link

I stated that I run Prime95 for 3 hours, look at timing on picture below, you see the proof



I also run IntelBurn at almost maximum memory for 5 runs, this test is more severe than Prime95



Idle temps are OCCT MIN column

How do you measure your temps. Are you sure that it is CPU socket temps are spiking, not motherboard (I think NB).

I think I have 45C to 47C on Battlefield 3 - I always run OCCT on the background to see my maximum values after the game.

Once again - what processor you have? We can not compare our numbers without knowing the other CPU. Tell me your CPU, I already know that you are using Dell CPU cooler (am I right) - knowing the wattage of CPU is important to evaluate temperatures.
m
0
l
February 27, 2013 3:46:34 PM

I just run OCCT for 30 minutes and check results - for me it is motherboard temperatures which spikes up and down, not CPU. So the question remains - how do you know that it is CPU socket spikes?
m
0
l
March 6, 2013 3:20:16 PM

Kisianik said:
I just run OCCT for 30 minutes and check results - for me it is motherboard temperatures which spikes up and down, not CPU. So the question remains - how do you know that it is CPU socket spikes?


lol NB temp changes with usage. apparently that is normal, I wouldn't consider that "spiking" during continuous use.

the reason i believe the cpu socket is spiking. Not only because ALL programs report it as so....but because the cpu fan would not get faster from the mobo temp. the case fan might possibly. According to you cpu fan is controlled by socket readings.

thats the first confirmation. As i've said the fan speed spikes...when sensor programs show the socket temp spike. No need to go beyond that really.


But just to humor you. I have three temp sensors on my dell mainboard. tmpin1 i believe is socket. it is read as so by all programs iv'e used. and is always around 5 degrees cooler then cpu core temp readins. and changes and concides with the cores change. so this also confirms, besides the fan speed changing with it, that it is the socket temp.

I believe tmpin2 is southbridge. which is always extremely cool. around hdd temps. 30 degrees.

tmpin3 i believe is the NB. which idles at 65-70 degrees.(most motherboards trehshold is 100 degrees) It is always the hottest idle on a board. But strangely on full load drops to 50-55 degrees. So like i said you can say NB spikes, but not really spiking. it goes up or down according to usage.

At first i thought this might be an error in the NB temp readings. I was afraid the temps might be reversed! But when touching the NB heatsink. At idle it is burning the finger. and On full load it becomes only slightly warm to the touch. Confirming for me the readings.

I thought this to be because of more air flow in the case on full load. But i now i believe it to be from the way heat dissipates on the board when at full loads. when the gpu or cpu is taking more load and absorbing more heat from it away from NB. Prolly by design. Or maybe its lower voltage causing higher temps? I believe even with cool and quiet off....I still got the issue in prime 95.


what are your thoughts? Why would prime95 cause cpu socket spikes but not core spikes. I really don't care about OCCT, i'm sure the same thing would happen. I don't want to even use prime 95 again. I've stated the reasons why it is the socket temp not core or NB temp, as silly as your question was, agan I ask anyone.... Why would prime95 cause cpu socket spikes but not core spikes?

It doesn't matter what model phenom II i have. How diff could the temps thresholds be? According to amd it is 70 degrees is critical and dangerous and failing point. over 60 is unsafe. and why would version of phenom II affect the dell mobo? It is a suppoted cpu model by dell.

My fear is the voltage regulator on the mobo. A new psu did not solve this problem. I have a feeling i have been virused. And if it weren't for the fact i have a dell. My board would have been bricked a long long time ago. Or maybe my old dying psu could have damaged something possibly? causing this weird issue? I never got any instability. I have never crashed or blue screened ever on this pc in three years. The worst that has ever happened to me is a coupe times a program locked up on me. I simply changed the psu cause the old one got alot of dust and the fan started making noise, and i didn't want to take chances.

But I'm asking if this could be a voltage issue? And could the voltage regulator on the board be stressed? Is it eroded crease on a part of the cpu that is only getting stressed when at the highest possible load for extended period by programs like occt or prime95? Or dust causing it?

Has anyone ever heard of a similar issue? Should i just ignore it since it doesnt' happen in the games I play? Maybe I should ask in the cpu section.
m
0
l
March 6, 2013 6:32:29 PM

OK, I am not here to argue or trying to prove something, but lets look at abbreviations.

I have TMPIN0, TMPIN1 and TMPIN0 (2), also FANIN0, FANIN1 and FANIN0 (2).

I know for sure that FANIN1 is my exhaust fan since I have speed regulator there, next FANIN0 (2) is GPU fan, checked by GPU monitoring software, this leaves FANIN0 as CPU FAN, I hope we agree on this one.

Next temperatures. First use your common sense and correlate TMP and FAN, it is looks obvious that TMPIN0=FANIN0=CPU (socket), TMPIN1=FANIN1=exhaust fan, temperature provided by?, I say motherboard, I checked this with Aida64, but I will show another discussion on this, and finally TMPIN0 (2)=FANIN0 (2)=GPU fan and temperature, very simple with this one.

If you still see me wrong look here http://www.tomshardware.com/forum/14723-63-cpuid-hardwa...

So first figure out what are you measuring, next see if it is normal.
m
0
l
March 7, 2013 6:55:23 AM

Kisianik said:
OK, I am not here to argue or trying to prove something, but lets look at abbreviations.

I have TMPIN0, TMPIN1 and TMPIN0 (2), also FANIN0, FANIN1 and FANIN0 (2).

I know for sure that FANIN1 is my exhaust fan since I have speed regulator there, next FANIN0 (2) is GPU fan, checked by GPU monitoring software, this leaves FANIN0 as CPU FAN, I hope we agree on this one.

Next temperatures. First use your common sense and correlate TMP and FAN, it is looks obvious that TMPIN0=FANIN0=CPU (socket), TMPIN1=FANIN1=exhaust fan, temperature provided by?, I say motherboard, I checked this with Aida64, but I will show another discussion on this, and finally TMPIN0 (2)=FANIN0 (2)=GPU fan and temperature, very simple with this one.

If you still see me wrong look here http://www.tomshardware.com/forum/14723-63-cpuid-hardwa...

So first figure out what are you measuring, next see if it is normal.


sounds like you don't read my posts and you just have discussions with yourself. nvm dude. i'll post my question in the hardware forum. You obviously don't know, which is why you keep going in circles.
m
0
l
!