A little update:
As this project has sort of taken on a direction and life of its own, let me bring you all up to date.
Now that I got it to do what I started out to accomplish perfectly, that is, translate the variable voltage output of a fan controller, the new Lamptron CW611 in this case, to a proportional PWM signal, so as to control PWM pumps.
I came to the conclusion that it really isn't such a good idea to be tying pump speed to temps.
Pump speed, or flow rate, once it's high enough to create turbulent flow, doesn't have enough impact on temps at increased rates to justify the increased noise. It's better to just find the sweet spot for sufficient flow rate versus noise, then set it and forget it.
Having a nice manual PWM controller is handy for this.
Controlling temps is really all about moving air thru the rads, and having the controller vary the fan speeds is the way to do that with quiet when you want it for surfing, and performance when you need it for gaming.
So I streamlined the really sweet manual PWM controller portion and made up a PCB for it.
I also added a small piggyback PCB that allows the option of using 2 push buttons for "speed up" / "speed down" like the volume on a TV remote, as opposed to a control knob.
I did leave the option of running the interface to the CW611 board in place on the manual controller PCB, in case I want to revisit that later for some reason.
Since I decided to use the 2 new CW611's tin the stretch build, instead of the Mix 'n Match builds, to gain the really nice automatic fan control function they have, I may need to be able to run pairs of channels off of 1 temp sensor.
There's also been a lot of asking about having multiple channels run from a single sensor in the CW611 thread, so coming up with a way to do that would help others.
To that end, I came up with a working prototype, and then designed and made a PCB for that.
And of course with no good deed on my part, ever going unpunished, putting the 611's in the stretch build, means moving the FC9 controller, plus buying another one, and both of the Aerocool X-Vision displays that are already in the stretch into the Mix 'n Match builds, which originally were going to get the new CW611's. . . . . .
Means that for the displays to read RPMs from fans controlled by the FC9s, they each need a tach signal translation PCB, and while there was 1 with the FC9 in the stretch build, now I need another for the new FC9.
Fortunately, I designed that PCB the last time around, and just had to go thru the process to make it. . . .
So I made up 2, since the FC3 I use for testing stuff is like the FC9, in that the output is +12V rail referenced and needs tach signal translation to display on most displays.
Anyway, I now have PCB's ready to drill and put the parts on.
Here's some pics of the process:
First step is to print the design out onto the toner transfer paper:
Sensor expansion pcb:
PWM controller pcb
Piggy back pcb for push button controls
Then transfer the black toner layouts onto the copper clad pcb material, and apply the green sealing film:
Upper left are the tach signal translation pcb's
Then etch the unprotected copper away:
And like magic, PCB's appear:
I left the board on the top right PWM controller a bit oversized, as that one is going to Martinm210 to use for testing PWM fans and pumps, and I wanted some flexibility in how I'm going to house it.Edited by IT Diva - 6/15/13 at 9:30pm