View Full Version : Noisy Delta 38cfm cure...(from front page)
Just read this article....looks like it sure would quiet these fans down if you find the noise objectionable.
I've got a PEP66T, and I really don't find the noise to be too bad, especially with the side of the case on. However, having one of these fans, and being in the automotive industry, I have some of the clear silicone and electrical tape (lots, really) kicking around. In the interest of science, I'm gonna try this in the next couple of days, and see how it works.
I'm going to the in-laws for Mother's Day dinner, where I have all of my auto stuff stashed in their cellar (man, get good in-laws...they're awesome!!). I'll grab a couple of tubes of the silicone while I'm there.
I don't have any way other than by ear to test a noise differential, but will post what I find when I do this mod.
Stay tuned.....
Mr B
klosters64a
05-13-01, 11:59 AM
Thanks, Mr B! My concern is that I believe that the silicon glop is for caulking, and isn't an adhesive per se. Perhaps two wire fan covers, one on each side of the Delta, would be a sensible precaution. I want a Delta 38 bouncing around in my case--NOT!
I'm thinking that myself. I will figure out some form of a passive retention, that wont allow it to fall off. Perhaps a zip tie holding the power lead to the heatsink, to at least keep it from falling too far.
Yeah, straight below is the back of my vid card...really don't want my Delta bouncing around on that....=)
I'll arrange something, that will "hold" it in place, yet still allow the silicone to handle the vibes.
Good point. Thanks for the thought on this!
Mr B
how does separating the fan and hs reduce the noise? i don't understand that part. i thought the loud noise was the whine the fan creates because of the speed. so how does reducing the vibrations reduce the noise?
please clarify...
Door Knob
05-13-01, 12:43 PM
cel950 (May 13, 2001 12:20 p.m.):
how does separating the fan and hs reduce the noise? i don't understand that part. i thought the loud noise was the whine the fan creates because of the speed. so how does reducing the vibrations reduce the noise?
please clarify...
Yup, the vibration causes more noise. I have seen a similiar post (probably here) on how to do this with case cans using rubber gromets. Just one of many things I plan on doing when I return home to my beloved garage. Good luck Mr. B and thanks for letting us all in on your results.
OpenFriday
05-13-01, 01:10 PM
Ive got the delta in question and just made it run @7V. Proc temp went up 1C and sys temp went up 2 C but i cant ever hear it anymore. Thanks Shadow :)
This may not be for everybody but it worked for me.
sifu quoc
05-13-01, 01:33 PM
yeaup... i run it at 7 volts too.... and my 120mm at 5volts... the 2 exhaust 80mm are at 7 volts too
the loudest part now is still the 120's lil grumble....
system temp is always at room temp... system temp at about 33 under full load
pIII 700
7:00 PM, Eastern Standard Time;
Just got back from dinner at the in-laws....had Chinese take out....yummy!
Mission accomplished. Retrieved 1 (one) tube of Clear RTV Silicone (Permatex...the good s***) The silicone is setting as I type.
It mentions in the text in the article (from front page) that the silicone should set for 24 hours. So, I will be testing this tomorrow night, around 10:00 PM EST. After which, I will post my results/opinions on this hsf mod.
One note thus far...if you want to run a grill, mount it before putting the silicone on the fan as described. I did, as I could see that being a messy proposition with the silicone on the fan. (I know me...I'd drop a screw, and set the fan down on my desktop, silicone side down, while I bent down to pick up the screw.....)
Stay tuned.....
Mr B
OpenFriday
05-13-01, 06:17 PM
Cool Mr.B cant wait to see the reasults.
the silicon stuff is not meant to be used as an adhesive but it can hold stuff in place, the heat from the heatsink will make the bond stronger over time as well, I would recomend running for a while with the fan off (I can run my cCO celeron 600 at 900mhz with 1.75v and no fan all day with my lapped FOP) if the silicon goes into the screw holes of the Alpha's shroud it will have the effect of a screw and nut as it will spread out slightly, and maybe if you put some on the top side of the screw hole on the delta it will bond with the silicon below and help hold in place, I used some silicon a while ago on all my fans and it's cuts the overall noise made by the pc down a lot
Phil (May 13, 2001 06:51 p.m.):
if the silicon goes into the screw holes of the Alpha's shroud it will have the effect of a screw and nut as it will spread out slightly, and maybe if you put some on the top side of the screw hole on the delta it will bond with the silicon below and help hold in place
Due to design differences between the author's Alpha, and my PEP66T, I couldn't install screws (nor would I need to) to hold the shroud on. The shroud on the PEP66 is held in place by a quartet of screws on the sides of the shroud, not the top.
That leaves those holes open, and indeed, silicone was pushed down into those holes, which will strengthen the bond there.
In the article, the author put silicone on the fan, through the mount holes, to bond in a similar fashon as you describe. I did the same, as outlined in the article.
The electrical tape then serves not only as sides to direct the air into the heatsink, but also to keep it centered, and somewhat steady. I still will arrange some form of retension (passive) to ensure that the fan doesn't take a voyage south, onto my vid card.
Mr B
Shadow рс
05-13-01, 08:21 PM
Mr B, I like the idea but he failed to mention preformance differences. I have a few Ideas of my own I'd like to work with on my FOP 32, but since you are starting yours first, I'll wait for your results.
If the adhesion properties of the silicone are in question try some Dap Aquarium Sealant it is a good clear silicone made to build custom fish tanks and seems to have a little better bond then regular silicone.
Shadow рс
05-13-01, 09:48 PM
also Mr B........I'm quiet sure you're familiar with the same goop you use on engin rebuilds.......very tolerant to heat and comes in so many pretty colors. =)
I'm betting the silicone will hold like gangbusters. I used it when I was roofing and sometimes had to pull a piece of flashing off that was sealed with it and it had surprising tensile strength. The aquarium cement seems like a great idea too. Good luck Mr B.
Thanks for the tech ideas, and the support, guys...
Had a brainstorm on a passive retention device to keep the Delta from going too far if it should detach. I'll elaborate tomorrow....
Mr B
i think the silicon will cope fine holding the delta. because the delta spins so fast, it can't have very high tolerances in lateral movement or the vibration would be tremendous. mine is almost perfectly still when i hold it.
but rather have a backup thannot, i suppose. ;-)
Breadfan
05-14-01, 11:49 AM
I guess also if you have an FOP38 or WBK38, you could put the retainer clips back on after the silicon set...that'd keep things from falling off, but would those clips transmit much vibration? I doubt they would...hmmm may have to try this very soon...
Mike
Breadfan
05-14-01, 11:52 AM
Not double posting, just thought of a few things....
I wonder why I didn't think about this...I remmeber showing a friend how loud the delta 38 was when I switched to a 90mm fan on my fop38. I was sorta dissapointed becuase when I plugged it in unattached and held it, it didn't make that much noise.
I now have 2 delta 38's becuase I got a WBK to replce the fop, and put an 80mm sunon fan on it...
This will be neat to try...maybe the other delta could be used as an extra exhaust or intake fan on my case....
Mike
Tachyon
05-14-01, 03:51 PM
I'm not sure if anyone else has noticed, but the fan shown in the article (face up) is a Pico Ace (Sanyo Denki)....not a Delta!
Shadow рс
05-14-01, 03:56 PM
ROFL......I wasn wondering if Tim was gonna move this. =)
Yeah, Shadow....frankly I didn't expect this to take off the way it did.......otherwise I would have started it here in "Cooling". Didn't expect as many replies to the thread as there have been.
Well, assembled, and installed. I took a bunch of pics w/ the webcam (sorry for the crappy pics, but I don't have a digital camera, and the webcam is better than nothing. Here goes....
I've got a half dozen or so pics to post...I ask that I get a chance to post all of them before you reply. TIA !! I'll post my results and opinions at the end.
The first pic here, is the silicone setting up, with the pencils still in place:
Here's a pic of the assembly after curing...It held!! quite well in fact,....
Reminescent of the old "Superglue" TV commercials...
I still was concerned about having some sort of restraint in place, just in case the silicone does fail. I used some 3/32" automotive wire, and some electrical terminals, and constructed this.....
Which I attached to the heatsink shroud, and fan like so...
If the fan ever did fall off, this would keep it from going too far, but doesn't support weight while mounted. Therefore doesn't affect the fact that the fan is "Rubber mounted"....
Sorry for the delay...had it all typed out and hit the damn "cancel" button by accident.
Well, how does it work??
Not as well as I had hoped for. It did somewhat quiet the roar of the Delta but not much.
On a scale of 1 to 10, with 1 being silence, and 10 being the Delta running with the case side off, I'd have to say the niose level is about an 8.5.
By itself, it really doesn't seem to make much of a difference. Used in conjunction with dropping the voltage run to the fan down to +7v, might make all of the difference in the world. I really had expected more (or less, noise that is) out of this mod.
Items used:
Alpha PEP66T, with Delta 60x60x25mm, 38 cfm "hair drier"
Permatex Clear RTV Silicone adhesive sealant
electrical tape
6" of 3/32 gauge auto electrical wire
3 electrical terminals
This is an "almost" free mod. All together, the stuff you need should run about $5, with the most expensive thing being the tube of silicone.
Every little bit of noise reduction helps, if you take umbridge to the noise level of this particular fan.
Temps don't seen to be affected at all, btw....up or down.
Again, I really hoped this might drop the noise level down more, maybe I expected too much....I wasn't expecting the level to drop to almost total silence, but I was hoping for a bit more.....
Mr B
Shadow рс
05-14-01, 10:24 PM
before you started would you have rated the noise at 10? so you dropped 1.5?
Dunno bud....looks like a very nice job with **** poor results. For the effort you put in it, is it worth the gain?
Thanks for the excellent "article" Mr B. I guess you just saved me from going through all that effort. You did a great job. Sorry the results weren't as good as we hoped for.
OpenFriday had posted in another thread the effects of reducing voltage to +7v to this fan, and that it dropped the noise level substansially.
I'm going to try this, with the silicone in place.....see what the two do together....
Yeah...**** poor results. Coulda been better, I think. Perhaps the difference in construction of the two hsf's plays a role here, as well. Had fun doing it, though!! =)
Mr B
Dropping the voltage to the fan to +7v does indeed quiet the Delta's roar. Temps went up a degree or three, though. If your temps are low to start with, this works well. If you have high temps, you be best served leaving it at 12v.
Mr B
I would say that doing this to a single fan probabally wouldn't make that much differance, if you hook up a delta and hold it be the wire whilst it is spinning so there is nothing to vibrate against it still roars, part air flow part whine, I have tyred this before on a pc with about 4 80mm fans in where each fan on it's own wasn't that loud but in the case together tit was very loud, and this cut the sound down to slightly above the noise of one of the fans alone
Just wondering if rubber grommets would have worked any better...
Sorry to hear it did work as well as you hoped Mr B
Well, here it is...
BUMP.
Had to dig back quite a few pages for this...=)
Mr B
inertia
06-29-01, 01:21 PM
It seems to me that if you take a fan that is noiseless when unattached and dangling by the power wires and that produces significant noise when actually screwed down to something that the source of the noise is the fact that the blades of the fan are not balanced. The wheels on your car have been balanced right? Well, for fans that spin in the high rpm range even a fraction of a gram difference in the weight of the fins would be enough to produce significant vibration. So, you've got to balance the thing. Perhaps you could hang it from a string from the very center at the axis and to whichever side it tips take a little bit of material off. You'd have to have just the fins though without the mount. Just a thought.
If it is noisy when it's unattached and hanging from the power wires, it's just a noisy fan.
i did the silicone trick on my gladiator (which i've sold since). it didn't lower the sound enough for me to keep it.
the question about whether the silicone would hold it or not.
with the gladiator, it has that shroud w/ holes for screws, well i just put enough silicone so that it oozed thru the hole and formed a bulb on other side of screw hole. this produced a stopper at the other end so there should be no problems w/ silicone holding fan in place.
I think that the gladiator is a particularly good candidate fo rthis because of its shroud.
i think ive just gone deaf since the sound of the delta is no longer bothering me. my case is open and its plugged into a 4 pin plug. maybe it needed some burn in to quiet down.
I just finished some benchmarking comparing the hsf setup in this thread, to the new Millenium Glaciator. You've seen the pics of the Alpha above, I put a pic of the Glaciator below...
Alpha PEP66T vs Millenium Glaciator
Tests were both run today, using my Celeron 600 @ 1008 MHz, 2.05v. Ambiant temperature at or about 78f/25c Temps measured with Asus Probe.
Test 1; PEP66T, with Delta 60x60x25 fan, 38cfm, silicone mounted to sink, "blowing" on sink
IDLE Temp 20 - 21c (idled for 5 minutes to stabilize)
LOAD Temp 28 - 33c (running Prime95 Torture Test, 30 minutes)
MB Temp 29c throughout test
FAN rpm 6490 idle, 7031 load
Test 2: Millenium Glaciator, as shipped, unmodified
IDLE Temp 21 - 23c (idled 5 min)
LOAD Temp 33 - 34c (Prime95 Torture Test, 30 min)
MB Temp 28c idle, 29c load
FAN rpm 4891 idle, 5273 load
I'm sure most of us are familiar with the "ear splitting whine" of the Delta 38cfm fan. It's loud. VERY loud. Even with the case sides on, I can hear mine running outside my computer room, and down the hall. ??? That's how the Alpha above sounds.
The Glaciator, on the other hand, cannot be heard with the case side on, over a pair of 92mm case fans (that aren't very loud, either!). This is what impressed me most about this hsf. It's lack of noise. This whole unit is a well designed piece. That's a fairly moderate to heavy overclock goin' on there, and it kept the temps under 35c, and drops down to idle temps fairly quickly after load is removed.
The Alpha was a couple of degrees celcius cooler, but the Glaciator was significantly more quiet. If you need the cooling capabilities of the Alpha/Delta, but don't want the noise, this new hsf might be just what you're looking for.
Mr B
Nice job Mr B. I use a GLADIATOR/DELTA with neoprene rubber about 1mm thk in triangle shapes(could be thicker) on the four corners and it did help. I don't get the whine with the gladiator/delta as i did when i had the pal6035/delta and i think it has to do with the height of HS and that the pal when blowing down the air is being pushed out the 4 sides of the pal6035 as with the gladiator its only pushed out 2 sides. I also use a 92mm fan on my plexiglass sidepanel facing my HS to feed it cool air all the time especially with the A/C on. What also did notice about your results Mr B is that the ALPHA's with the right fan (DELTA's) can still give these new HS's a run for their money. So don't count the ALPHA'S out anytime soon I also use sound deadening material from a CarAudio shop. Just my 2 cents.
vBulletin® v3.8.7, Copyright ©2000-2012, vBulletin Solutions, Inc.