Is my pump defective

[Circaflex]

Hey guys on mobile so be kind. I'm wondering if this is normal for an mcp655 flow seems low. The pump is on but not plugged into the aquaero so the pwm is disconnected I'm assuming it runs at the lowest speed when disconnected? I've attached two videos maybe my older mcp35x was just more powerful I dunno. And yes that is a 90 degree on the outlet. Temps seem decent on my GPU, its hot in SOCAL and i max out at 45 degrees on a fresh mount, aquasuite states pump is running at 3000 rpm at 100% is this all normal?

https://onedrive.live.com/redir?resi...nt=video%2cmp4

https://onedrive.live.com/redir?resi...nt=video%2cmp4

[Church]

MCP655? Laing D5 Vario that is? It should be more then 3000 on max IIRC, except if maybe it's -B variety with no built in speed switch and running on 4th speed permanently (of 1-5 speeds of normal 655 / D5 Vario).

[m_jones_]

That does not sound right, my D5 vario on setting 2 runs at around 2800 rpm.

[Circaflex]

the pump is the mcp655-pwm from performance pcs

Ive sent PPCs a request for RMA or assistance, im worried because the loop has been up for about 3 days, money doesnt grow on trees for me, but they have very strict returns and rmas it seems hopefully i am not out of luck. but if you have any advice to throw my way please do ive refilled the loop 3 times to see if it was just air getting trapped but nothing seems to work.

[Circaflex]

So after some more research it seems the Aquaero and Swiftech mcp655 pwm are incompatible. Aquaero states Swiftech uses a different pwm standard, some users on their forums call bs but the pump works great now when connected to my mobo.

[Church]

On both AA5 & AA6? And does pwm support problems include MCP35X?

[Circaflex]

it does have problems on AA5 and AA6, there are reports of the mcp35x working fine but I cannot confirm or deny that it works. There is a fix, its a Frankenstein method, it requires adding a resistor to the pumps pwm connector, I just said screw it and hooked it up to my mobo instead.

this is a link to the fix http://www.overclock.net/t/1474470/o...#post_22637637

If you open the linked document you should have a look at point 2.1.4. It clearly states that the fan (pump in this case) must pull up the signal internally to a maximum of 5.25V, but this is exactly what the pump does not do - it has no pull up circuit. Point 2.3.1 shows how a hardware monitoring device like the aquaero has to be designed and it is exactly what we do incl. better tolerances to prevent damages due to faulty fans for example.

Well, we could have added a pull up on our side but in this case I do not want to be the person who has to reply to all the emails from customers claiming our aquaero damaged their fans because exactly that could happen then. If you go back to point 2.1.4 it states at the end that new designs should use 3.3V which means there are two different standards. Swiftechs pump requires 5V while a fan might only work with 3.3V. If we would add the pull up to 5V on our side the pump will work but the fan will be damaged...

Long story short: nothing wrong on our side. Anyway, there is a simple workaround for those who want to get it working without investing much money. You will need a 4,7kOhm resistor (a bit more would also work) some wiring and a soldering iron. Your goal will be to add 5V to the PWM signal wire of the pump. If your pump is powered directly by the PSU, then please take the 5V from the same cable where the pump is connected. If your pump is powered by the aquaero, then please take the 5V from the cable where the aquaero is connected. Remember to place the resistor into this connection.

With this little modification the pump can be controlled through the aquaero but this brings us to another problem. Normally the pump should always run at a minimum speed but due to the strange PWM implementation it will stop if you use a too low speed setting. I highly recommend that you go into the fan settings of the aquaero and set a minimum value of at least 30% for the channel where the pump is connected. Also make sure that you enable the option to hold the minimum setting so the pump will always run even when you do something wrong with the controller setup.

And last but not least: we are aware of the problem when you connect a bunch of fans like the ones from Corsair. It has been already solved with a small hardware modification and will be included into the next revision of the main controller. Devices dispatched in April 2014 should already use the newer revision.

[Martinm210]

FYI 3000rpm is what you get when the pwm cable is disconnected too. not sure about the pointing fingers at the pump but it works fine on the motherboards I tested on.

I guess all the motherboard manufacturers are crazy and doing something wrong..lol

Anyhow the real benefit of PWM is so you don't need really expensive and temper mental fan controllers to auto regulate speed.

Good old motherboard and speedfan is the better tool and free!