Aaargh! Eheim 1048 pump is LOUD

[awesterink]

So I thought I would quiet down my system by going to watercooling (using the Innovatek kit). But the vibration of the Ehaim 1048 pump makes my whole case resonate, and produces a far louder and annoying hum than I had with aircooling.

I tried rubber washers, I tried mounting the pump on a bed of matting (the stuff that you put under carpet), but no solution. I also made sure the power cable is "shielded" from direct contact with the case. But still that LOUD hum. Is it transferred through the tubing perhaps? I can FEEL the hum! Sheesh.

How can I get a quiet, non-vibrating pump going here?

Thx..

[FRAGN'STIEN]

Sounds like you may have a bad pump. My 1048 was as quiet as could be.

[Lothar5150]

Yeh, you got a bad pump

[redleader]

I don't think so. All Eheims vibrate like hell.

Try bubble wrap. The air pockets don't transmit vibration.

[DarkEdge]

Nope. Its one of two things. Either you have way to much back pressure or the pump is bad. Yes there will be some, but nothing a small hand towel, rubber washers or any other type of material can't dampen.

[Lothar5150]

Originally posted by DarkEdge
Nope. Its one of two things. Either you have way to much back pressure ........

Yep, it could be cavitation what size lines (ID) are you using?

[RedTalon19]

for a while i had my pump suspended. yes i hung it about 1/4" off the bottom of the case from my hard drive bay and it worked wonders... the only problem was it kept falling out everytime i shifted the case cauze i did a knock up job of rigging it up.

now i just use a 1/4" thick piece of neoprene and it stops virtually all vibration in the case

[bigben2k]

It sounds like the impeller might need balancing, or, as stated previously, the pump is being forced to work very hard, or even, there's cavitation (air running through).

[p0b0y]

my 1048 does that but i just suspend it with fishing line and no more damn noise!!!

[Skulemate]

Pardon me, but I think that cavitation is much different than air flowing through the pump. Cavitation is actually the formation (and eventual collapse) of vapour cavities. The collapse is associated with very high water velocities that can be very damaging to the impeller or housing.

However, it occurs when the pressure in the liquid being pumped falls below its vapour pressure. This can happen several ways:
- if the suction lift is too great
- impellor speed is too fast
- liquid temperature is too high

To aviod cavitation the pressure in the suction line must be greater than the vapour pressure of the liquid (that's where the net positive suction head comes in). Now, I seriously doubt (though I have been wrong before ) that any of us have pumps capable of causing cavitation in our small systems.

[awesterink]

Thanks for the replies guys.

I think I have been able to eliminate most of the direct vibration resulting from the pump touching the case. I bought a good size sponge, and cut out the shape of the pump, and half embedded the pump in the sponge. That seemed to reduced the hum significantly. But there is still vibration being transmitted to the case, probably both through the tubing, as well as simply through the air. If I put my case side back on (snug fit), the hum is restored in all its annoying glory because of the amplifying effect of the enclosed metal casing.

I am using 1/4" ID 3/8" OD tubing...pretty much my only option with the Innovatek kit. I bled out all the air, the Innovatek tank handles that nicely. Also put a little piece of that blue sponge that came with the Eheim into the tank to eliminate foaming.

So, assuming that I stick with the Innovatek blocks and tubing, what are my options here? Try a different pump? Same type?
Different capacity? Different brand?

Thx...

[Khledar]

Ok I gots an idea, that's some small ass tubing - could be putting a lot of strain on the pump. Since you're probably running the blocks in a series, why not try making a manifold to convert one 1/2"ID into two or three (or four!) 1/4" ID tubes and run all the devices in parallel.

Let's see
1/4" ID = pi*r^2 = 3.14*.125 ^2 =~ 0.0491

1/2" ID = 3.14*.25^2 =~ .196

.196 / .0491 = almost 4 (3.99something)

So if my math and reasoning are correct, which they probably are not, you could most efficiently run one 1/2" ID tube into a manifold, and four 1/4" tubes out. Not really sure what all the effects of this are and whatnot but someone's probably tried this so give us your insight!

GL
(AHHH! Edits all over the place except not in the math so it's probably still messed up.

[Skulemate]

Umm... well, your math is fine, though sometimes we rely on calculators too much... you need exactly four 1/4" ID tubes to make one 1/2" ID tube (pi is a constant... so the only thing that changes is r^2... and since the radius is doubled from 1/4 to 1/2, the area is four times larger). However, there'd likely be a fair bit of flow resistance in the manifold, especially if you were splitting the 1/2" tube into the four 1/4" ones.

[awesterink]

Okay. So what is the likelyhood that the Eheim simply has to work too hard? I'd be willing to try a different selection of components. I have a BIX "laying around", so I could go to a 1/2"ID system with e.g. the MAZE blocks. I suppose that would be beneficial to the flow, but would it help reduce the hum/vibration? Or do I need to look for a different pump?

[DarkEdge]

Nailed it on the head when i said there was to much back pressure. For that size of tubing the 1046 would be working to hard. I ran that with my maze 1 when I first got into water cooling and it only pumps 75gph at 0 head.

Grab some bigger tubing and change the fittings as needed.

[Lothar5150]

I took a look at the Innovatek blocks they look like they have a very small channel. I would go with changing out the tubing and go with 1/2 OD. If your lines are stiff, right now it is probably from too much backpressure. I recommend getting "quick connect fittings" like mine


You can get them at US Plastics or at McMaster-Carr.
Make sure you get the plastic compression fitting inserts.

[koslov]

Lothar, what part number did you get? They all look the same to me. I am looking for some disconnects for 1/2in ID tygon.

Also, do these disconnects impair flow at all or introduce air into the system when reconnected?

[Lothar5150]

Originally posted by koslov
Lothar, what part number did you get? They all look the same to me. I am looking for some disconnects for 1/2in ID tygon.

Also, do these disconnects impair flow at all or introduce air into the system when reconnected?

At McMaster-Carr look under Pipe Fittings and Pipe, Tubing and Fittings, Hose and Fittings then Instant Tube Fittings. You will need to figure out what size thread you have on your blocks. There is a guide on the site or you can match them up at your local hardware store.

They do not restrict flow at all they and they are airtight (same fittings that swiftech uses except they are 1/2 inch)

Edit: Ehiem pumps are BPT fittings since they are German but most US manufactured blocks will be NPTF

[gmat]

awesterink: did you glue the impeller to its shaft ? If not it *will* produce lots of noise...

(note to pro/staff: IMHO this should be a FAQ)

[koslov]

Thanks Lothar. I was actually thinking of using the disconnects in the middle of the tubing, but perhaps on the WB would be better. You see, I am building an external enclosure for my WC gear and I need to be able to disconnect it from my case for transport. Do you have any ideas on which would be easier / better?

edit: I apologize for getting a little OT.

[Sanjuro]

the ones from usplastics are pretty good. look for the high flow ones that close on both ends

[awesterink]

gmat: no, I did not glue the impeller to the shaft...did not know about this...worth a try! Is that described in a thread somewhere?

Other guys: if I go to 1/2" tubing, would the small channel in my Innovatek blocks not negate that effect? I know that a smaller channel in a block results in faster flow, which is a good thing for the block. But perhaps a 1/4" channel in the block is TOO small? Which blocks have a larger channel...Swiftech? MAZE3?

Appreciate all you replies guys! I'm not giving up darned...with 60C load on my CPU, 46C case, I better do SOMETHING :-)

[gmat]

yes look at "eheim 1048 rattling explained" just below.

[awesterink]

duh. :-)

[Lothar5150]

Originally posted by awesterink
gmat: no, I did not glue the impeller to the shaft...did not know about this...worth a try! Is that described in a thread somewhere?

Other guys: if I go to 1/2" tubing, would the small channel in my Innovatek blocks not negate that effect? I know that a smaller channel in a block results in faster flow, which is a good thing for the block. But perhaps a 1/4" channel in the block is TOO small? Which blocks have a larger channel...? MAZE3?

Appreciate all you replies guys! I'm not giving up darned...with 60C load on my CPU, 46C case, I better do SOMETHING :-)

Swiftech MCW462-UH