Sign in with
Sign up | Sign in
Your question

Help solve pump power issue

Last response: in Overclocking
Share
January 11, 2013 6:54:51 AM

I am having trouble testing my pump that is new. It's a Swiftech MCP655 model. This is the first time I have filled it with coolant and it appears like in the image below but with all the tubes and such connected. At this stage, only the pump is connected to an external power supply, NOT the main system as it should be at this point. The only thing I have done to the pump is added a Bitspower "Mod Kit", which required removing the molex pins and putting them back. I followed direction and made sure the wires were in the same 2 slots they were before removal (there is only two a red and black). This pump molex connector is attached to a Molex connector from the external power supply.

The problem is I get no power at all from the pump and I can't hear it running. My external power supply seems to turn on and it's fan does run. However, I noticed the fan was not very loud or spinning very fast, which it may not if no other loads are attached, it's just an old power supply being used to prime the pump. At this point I am not sure what is going on or if there is any power going through that power cable on the power supply. If there is, I do not know why the pump is not working. I don't want to have to drain the new loop and take it apart unless absolutely sure the pump is dead.

If anyone has some thoughts that has used an external power supply to fill their res and power the Swiftech 655 pump let me know what you think may be going on. Not sure what else to try, but I need to know at least if there is full power going to the cable to power the pump before looking at the pump itself.

Thanks for any help.

Here is some other info too that may be the cause but not sure:

Quote:
Did you do any work on that pump?

I did some work on mine and forgot that the 12v red wire (on pump) needs to be hooked up to the 12v yellow wire on the PSU molex. LOL
Here i was trying to run red wire (on pump) on red (on PSU) and forgot that the red wire on PSU was only 5v.



More about : solve pump power issue

January 11, 2013 10:57:17 AM

If your power supply is an ATX power supply, you have to short out the Green wire on the 24 pin power connector on the power supply to any of the black ones to simulate a power load that the motherboard typically does.

Check your power supply by adding a fan to a molex power on the power supply. Doesn't matter what type of fan, as long as it has a molex connector. If the fan runs, then you know you have good power.

If you do have power, then I would be looking toward a dead pump.
January 11, 2013 12:35:16 PM

Groveling_Wyrm said:
If your power supply is an ATX power supply, you have to short out the Green wire on the 24 pin power connector on the power supply to any of the black ones to simulate a power load that the motherboard typically does.

Check your power supply by adding a fan to a molex power on the power supply. Doesn't matter what type of fan, as long as it has a molex connector. If the fan runs, then you know you have good power.

If you do have power, then I would be looking toward a dead pump.


Actually, I am using a custom PSU cable from Performance PCs that has a 24 pin plug on one end and a toggle switch on the other. I tried a fan they worked just fine so I did some more checking.

The problem was actually the red and black wires from the pump. I had them switched on the molex connector so the red wire on the pump was going to the red on the PSU. Instead, the red on the pump needed to be on the opposite side of the molex to go to the yellow on the PSU cable. This is what was stated in my original post, but I thought they only align one way and they don't so the wires were not matching like they should.

My pump runs now, but just briefly. About 1 in 5 times when I turn on the PSU it will start for a few seconds. It gets so far filling the loops, then just stops with about 50% coolant left, then it drains back some from the tubes into the pump top res. Seems like something is stopping it from working, I can't hear any pump sounds at all and it acts like power just cuts out or something. If there is nothing pushing the fluid, this would make sense why the coolant is being pushed through the loops.

Any thoughts now on this since it may still be a power issue or something else?
Related resources
a c 324 K Overclocking
January 11, 2013 2:05:25 PM

Does the reservoir run empty? If so, you need to stop the pump, refill and restart. Your pump cannot push water with air and you can cause permanent damage by running it without water.
January 11, 2013 3:09:48 PM

rubix_1011 said:
Does the reservoir run empty? If so, you need to stop the pump, refill and restart. Your pump cannot push water with air and you can cause permanent damage by running it without water.


No, not at all as stated above, the res is about 80% full. It will run briefly, maybe 5 seconds, stops and then drains back. Could be an air lock at the pump or somewhere, but I did not notice any and have tried to check and tilted the case some. It's a HUGE case though, a 24 inch cube and HEAVY. I can't feel any vibrations or hear any noise at all coming from the pump even up close.
a c 324 K Overclocking
January 11, 2013 3:53:43 PM

Are you certain that the mod kit you added isn't causing the issue? Can you confirm the pump works by reverting to stock and testing? Can you post a pic of the setup?
January 11, 2013 5:35:57 PM

https://www.dropbox.com/s/3s4mphwca14ds72/IMG_1058_note...

I know there have been issues with the Koolace pump top mod before, but mine is Bitspower. Also spoke with tech at FrozenCPU and they said they have had no issues with any of these parts working together and do so quite often, especially the Bitspower mods. I was told to try it outside the case with a small closed loop connecting the in and out ports on the pump top mod and filling with just DI water in the res tube. It should run that way, if not take apart and see if the pump is locked or if it has any motion at all, an O ring or something may have an issue or some air may be trapped in the pump itself.

The top of the res tube is on, but there is no fitting on it (see photo above, but remove the fitting at the top) so it is open at the top.

It is a 655-B model and they said maybe it got stuck in a low power mode or one of the settings is not working and to try changing it to something else and back again.

Otherwise, not sure what else to try. I had to totally drain the res and some of the loops which took awhile, so I am going to test it outside the case just to see if anything powers. If so, then it's not the pump, otherwise, I'm pretty sure it is.

Also, someone said that some PSUs don't run full power unless there is a load. This may be what is happening with my Enermax 500 watt PSU, I don't know about this either. If so, that would make sense because sometimes I get power when I first turn it on, then it quits after about 5-10 seconds like either no power or very little.
a c 324 K Overclocking
January 11, 2013 6:42:27 PM

Can you confirm your I/O ports are correct? The orientation of your res and pump in the image shows it connected to the side that is traditionally the D5 intake, but if your pump top allows for alternative flow ports I would think it should work.

If you jumper your PSU using your ATX plug, it should power up and run. It's possible you have a PSU that has a safety mechanism that kicks the powersupply off when it thinks it isn't being properly powered on (jumpering a PSU isn't exactly the very best thing you can do to it, but it isn't that 'bad' for short periods of time). This could be the reason you get intermittent power on and when it does, it is very brief.
a b ) Power supply
a b K Overclocking
January 11, 2013 7:26:27 PM

HiTekJeff said:

The problem was actually the red and black wires from the pump. I had them switched on the molex connector so the red wire on the pump was going to the red on the PSU. Instead, the red on the pump needed to be on the opposite side of the molex to go to the yellow on the PSU cable. This is what was stated in my original post, but I thought they only align one way and they don't so the wires were not matching like they should.

My pump runs now, but just briefly. About 1 in 5 times when I turn on the PSU it will start for a few seconds. It gets so far filling the loops, then just stops with about 50% coolant left, then it drains back some from the tubes into the pump top res. Seems like something is stopping it from working, I can't hear any pump sounds at all and it acts like power just cuts out or something. If there is nothing pushing the fluid, this would make sense why the coolant is being pushed through the loops.


Uh, wait can you go into more detail? You switched the negative/ground and positive leads around? yeah that could cause sporadic behavior.
January 12, 2013 7:42:02 PM

mouse24 said:
Uh, wait can you go into more detail? You switched the negative/ground and positive leads around? yeah that could cause sporadic behavior.


No, there are 2 wires on the pump, a red wire and a black wire. What I had done was placed them in ports 1 and 2 on the Molex and they should be in 3 and 4. The reason is if they are in 1 and 2 it will line up red to red which should NOT be because that is just a 5V on the PSU cable. It should be red on the pump to yellow on the 12V PSU cable which it is now that is correct.

Hope that helps explain the difference.
January 12, 2013 7:43:29 PM

Well, I finally got the pump out of the computer and have been testing. I made a small closed loop on the in and out ports and ran just pure DI water for testing. What I found is exactly what I thought the problem was, the impeller simply stops spinning after about 2-5 seconds from initial power on. With the clear Bitspower pump top, you can see it start to spin for about 2 seconds, then slow down until coming to a near stop. It will spin every now and then, but typically nothing after that initial burst. While the impeller does move by hand and is not locked, it seems like something is causing it to quit.

My next step is to try the pump with the original pump casing and see if it works that way. I know this Bitspower pump top works because I have seen this combo in youtube videos and in the "Dream Machine 2012" Maximum PC build. Koolace as stated prior did have some issues with their pump tops causing trouble, but I find it unlikely this to be the case otherwise others would've known it by now just like with Koolace until they fixed it.
a b ) Power supply
a c 76 K Overclocking
January 13, 2013 6:56:23 AM

Inspect the pump housing and see if the wires are in their right places - or It may just lead me to believe you've damaged your pump while connected to the wrong lines...maybe?
January 13, 2013 2:15:42 PM

Lutfij said:
Inspect the pump housing and see if the wires are in their right places - or It may just lead me to believe you've damaged your pump while connected to the wrong lines...maybe?


Yes, they are in the right places, that's not it. I have never heard of this being the case, in fact I found several threads over at Swiftech where others have had the pump wires switched. If anything, it would be under powered, not over powered with the wires going red to red (5V on the PSU instead of 12V). This was the case with at least 3 others I read on that site where they too had the wires like that and all they got was a dead or underpowered pump. After connecting back like I did they had no other issues.

Tried the Swiftech housing and it was a no go too. Seems like just a defective pump since I have ruled out about everything else. Sending back for replacement.
a b ) Power supply
a c 76 K Overclocking
January 13, 2013 3:10:34 PM

Quote:
Sending back for replacement.
[:lutfij:3]
!