Herky Jerky DDC

[ranker]

So one of my DDC-2's has been acting strange lately. After what I thought was going to be the final rebuild of my Horizon, I noticed something strange during the bleeding process. My DDC-2 would run for anywhere between 15-30 seconds and then suddenly stop for a brief second before resuming to pump again.

The only reason I knew something was up was that I noticed and air bubble shoot from the res to the tube whenever this occurred. I looked down at the pump itself and could see the impeller stop moving for a brief second whenever this occurred. What's strange is that this occurs on an irregular interval. Most of the time its between 15-30 seconds, on occasion a full minute passes.

Is the pump failing or is there something else that may be causing this?

[Rob_B]

Have you tried it on a different connecor from your PSU already?
Or measure the voltage on the connector you use now.

[Patriote]

Damn. Yet one more problem, eh ranker? lol... That sucks. Try what Rob said above and if it still does it, id open the pump to see if the impeller is touching the top.

[nealh]

I have had an "issue" with a new DDC2 w/Petra top...pretty sure its not the pump but the molex coneection ...the 2 molex connectors fit loosely and not snug and almost any movement around the wires seems to loosen this connection

I have a Corasair HX620 and I have found if I move wires around that I loosen the connection and the pump does not start.....this has resulted in 3 times my system booting, crashing...and one very warm cpu/cpu block

I pretty sure my chip is fine

So IMHO the issue for me is partial stupidity and second watch my molex connection

PLEASE check how tight the connection between the PSU molex and the DDC molex is

[Sparky]

Sounds like a loose connection to me. My DDC did something similar when I was first leak testing until I remembered my test PSU has a bad molex plug on it. So yeah check the connection.

[hecktic]

Sigh it was only a matter of time my friend... try the suggestions in this thread maybe

[ranker]

Have you tried it on a different connecor from your PSU already?
Or measure the voltage on the connector you use now.

I have had an "issue" with a new DDC2 w/Petra top...pretty sure its not the pump but the molex coneection ...the molex connector at fits loosely and not snug and almost any movement around the wires seems to loosen this connection

I have a Corasair HX620 and I have found if I move wires around that I loosen the connection and the pump does not start.....this has resulted in 3 times my system booting, crashing...and one very warm cpu/cpu block

I pretty sure my chip is fine

So IMHO the issue for me is partial stupidity and second watch my molex connection

PLEASE check how tight the connection between the PSU molex and the DDC molex is

Sounds like a loose connection to me. My DDC did something similar when I was first leak testing until I remembered my test PSU has a bad molex plug on it. So yeah check the connection.

Thanks guys. This is the type of helpful suggestions I need. I'll check that when I get home. I can see two possible points of failure as that pump is specifically connected to a miniNG. So either the miniNG's molex connector might be faulty or the 4 pin to 3 pin molex adapter (to connect the pump to the miniNG) might be faulty.

[Snyxxx]

ranker,

Besides the above observations which I have also encountered, is the mini-nG over heating? It may be switching to PWM mode from analogue.

[virtualrain]

I had a pump that started doing that after 5 min. and got worse... I think it was overheating. I nearly burned myself when I touched the bottom of the pump. I had to RMA it.

[nealh]

Ranker....The molex on the pump is "not bad" but with the Corsair psu molex has a "pinch mech" and if you pinch it thinking your tightening like have a few times your are really making it looser and then the pins do not contact well and the pump fails to start

I will venture to guess you need to make sure they molexes are tightly connected....maybe the molexes on the DDC2 are easily loosened from thre female counterpart....hopefully this is not an issue of "impotence"

[ranker]

Well, you guys are awesome. It was narrowed down to a power problem after all. I haven't done extensive testing to narrow it down, but after disconnecting it from the miniNG, removing the 4 pin to 3 pin adapter from the pump, and connecting it straight to the same molex that was formerly connected to the miniNG, the pump's been working fine.

Now, I just need to isolate the problem to find out what the exact problem is. Here are my possible scenarios:

1) Bad 4 pin to 3 pin adapter
2) miniNG is defective
e) miniNG is going into PWM mode due to:
a) overheating
b) jumpers not set correctly
c) too much power draw

I'll have to grab another 4 pin to 3 pin sometime tonight to test that out. There is a possibility that I have a busted miniNG as I did rule out the "bad molex" scenario by using the molex that was connected to the miniNG and putting onto the pump directly.

Now, I have 3 YateLoon SL led fans connected to one of the two 3-pin ports on the miniNG. Each fan is 1.8W and all three together is 5.4W. The DDC-2 is 18W so that puts us over the 20W that the miniNG is capable of. Now, I was told by mcubed's support staff that the miniNG can handle over 20W and that 23.4W should be fine as long the miniNG doesn't overheat. I touched my miniNG earlier and it was cool to touch the heatsink. I'm not sure if my miniNG is overheating and switching to PWM mode. Is there any way to tell through using fans or whatnot if its in PWM mode?

If I can't isolate the problem any further, I'd have to conclude that it's a faulty miniNG. I'm hoping that's not the problem since PPC's isn't exactly as forgiving on the RMA's as petra's is.

Thanks for all the help so far.

[Sparky]

yates don't show it much IIRC but some fans when fed with a PWM signal get some clicking motor noise to them. Have any different fans to test that?

Also maybe you could use a DMM on the output to the pump to see if the voltage is right or not, or if it is bouncing around or whatever.

[nealh]

Yea..I was right..not too often I can offer much to the pro's around here....

[Snyxxx]

I thought the mini-nG only switched to PWM after like 60°C? Not sure about that, but it was pretty high and does not seem to be your case if the sink was not burning hot.

I also remember that the DDC2 has a high start-up amperage and may do something weird.

Keep us informed since I am thinking about powering my DDC2 with the big-nG for total control rather than just monitoring.

[ranker]

I thought the mini-nG only switched to PWM after like 60°C? Not sure about that, but it was pretty high and does not seem to be your case if the sink was not burning hot.

I also remember that the DDC2 has a high start-up amperage and may do something weird.

Keep us informed since I am thinking about powering my DDC2 with the big-nG for total control rather than just monitoring.

Well, I have a 2 miniNG, 1 BigNG and 1 Sensor bus setup atm, my BigNG has my CPU loop's pump + PA120.2 fans, + all case fans on it. It's working just fine without any hiccups. My other miniNG, controlling the chipset loop's pump, has only one fan on it; the fan cooling the PA120.1

I'm wondering if the amp spike of the DDC + the 3x fans all on two inputs is causing the MiniNG to switch to PWM mode. That might be why my pump was exhibiting a semi regular on-off cycle.

However, I'll spend the rest of the time tonight trying pinpoint the problem. I appreciate all the help.

Yea..I was right..not too often I can offer much to the pro's around here....

I learn new things everyday even after participating in the watercooling world on and off for the past 4 years. That's why this community rocks.

yates don't show it much IIRC but some fans when fed with a PWM signal get some clicking motor noise to them. Have any different fans to test that?

I didn't hear any clicking noise on my Yates at all. However, all of my fans are attached to alphacool attenuators before they hit their powersources (bigNG, miniNG's) anyhow.

[Waterlogged]

yates don't show it much IIRC but some fans when fed with a PWM signal get some clicking motor noise to them. Have any different fans to test that?

SilenX will do that quite easily on a PWM signal.

Tried swapping a Super Flower SF-609 to a Noise Isolator from FCPU in one of my cases to find that out first hand. Needless to say, I swapped back.

[hecktic]

dam that sucks well let us know what it is.

[nealh]

Ranker from what I cant tell you are not using a Corsair PSU correct

I think I may have found an issue with mine...the 4pin modular molex connectors have either 2 4 pins or 3 4pins on them(at least with a 620W..modular)

I had forgotten that the one with 2 4pins may have been bad..I was using it on my recent TJ07 rebuild

My pump starting acting up as well..this modualr molex loosen more with the pump molex but most important I think it was a bad modular cable as the pump on 2 occasions would not start even with perfect contact from a power on....

so my poor e6600 has been heated up quite a bit on 5 occasions now...

So my point is watch out on the modular cables ...at least in my case especially with the modular cable with 2 4 pin molexes from 620W corsair...I am throwing ti away after contacting Corsair for a replacement, if possible

[ranker]

I'm using a PCPC 750W silencer PSU in my rig. Like I pointed out, it was just the 4 pin to 3 pin adapter that caused the problem. After switching it out for another 4 to 3 pin adapter, the problem was solved. So it was definitely a power problem like you suggested, it just wasn't the PSU.

[nealh]

oops forgot you solved the issue..oh well...

maybe this will help someone else