• We recently switched our forum platform. If you experience any issues please email support@crystalfontz.com

HELP-Complex Control Circuit using 633 as center


New member
Ive been planning my new cooling setup for quite a while now and most of the parts have begun to arrive so now im turning my attention to the control circuit to run it all, the hub of which is the 633. Although i say hub i dont actually want any software involved if at all possible so i've been trying to think of various circuits that could control power on and off for the various things attached to the system. Most of it is outlined over in a post in the electronics forum at bit-tech.net so i wont go over it all again, theres also a couple of rough disgrams i threw together in the post.

Ive read through the forums and havent seen anything quite this complex, might be that this is gonna be one of those occassions when i will just have to try and see but maybe someone can help out.
Looking for additional LCD resources? Check out our LCD blog for the latest developments in LCD technology.

CF Mark

Can you program in a langage?

If you can, you could make things a little simpler... and much cooler ;)

You could use the FAN outs on the 633 to control the peltiers and the pump.

With a bit of electronics knowledge (switching MOSFETs) you could even control the power through the peltier (0% to 100%) using the FAN outputs PWM instead of just on/off.

Then you would just need to write a small app to control the 633.

If you got tricky, you could also possibly use a pulse outputting liquid flow meter, input the signal to a 633 fan RPM header, convert the reading, and show litres/min on the LCD.


New member
I can handle http and php, i tried C++ a while back, since then the phrase 'my head cant handle code' comes to mind :). I'm good with electronics and hardware but software kills me, http and php is about my limit. But if its the best way to do it then so be it. I read through the pdf for the 633 several times looking at the ATX Power switch functionality and wondered about redoing keypad_power_on to do more than just pulsing GPO2 low for 1 second, i.e making it use the FAN outs in sequence to control things BEFORE pulsing GPO2, and i presume this is what your talking about.

Problem is how do i get the 12v needed for the FAN out headers without pulsing GPO2 and bringing the PSU and comp to life, or can the FAN outs work with the 5vATX feed the LCD uses? Of course i could use the 12v from the AC/DC converter to power the 12v line on the LCD aswell after ive tripped the AC relay, had'nt though of that before.

Ok assuming the FAN outs work with the 5vATX feed through the LCD would need to be something like this as a recoded keypad_power_on sequence:
>keypad_power_on(Altered) Press which would show 'Power on Starting up' message

>Then FAN1 goes Live to control the AC relay for pump and fans, theres too many fans to draw from the LCD so i will just have to pick a beefy AC relay with a link to the pump and also to an AC/DC Converter for the fans and also an out to the 12v on the LCD to power its 12v, i think i would need to open JP7 and replace the standard 12v from molex with the 12v from the AC/DC ? All depends on whether the fan outs are 5v or 12v before before comp on. After the AC/DC theres a voltage regulator tied to fan2 so that the fans can still be controlled by the PWM, using the 12v from the ac/dc and then the output from fan2 as the regulator, FAN2 would go live after this whole routine im typing out has finished so that all fans are on 12v at startup being fed from the ac/dc converter and the voltage regulators default pin. Could use the dallas temp sensors aswell and like you say if i rig the right flow sensor to FAN3 then i have all the flow and temp fallover circuits in place would just need code to back it up.

>Then a wait delay of 3 minutes before FAN4 goes live and kicks in the peltier switch relay, id rather not have the pelts or the pelts PSU tied through the LCD or any other component, the PSU has a nice LCD of its own for voltage and amperage display and front controls, really all i want to do is replace its on off switch and tie it through the LCD, a simple relay should do after ive checked the voltage to the switch and figured the right one to use

>Then a delay of 10 seconds to allow the peltiers to cool a little before pulsing GPO2 low for 1 second and powering on the comp, LCD message 'Startup Complete', Then Fan2 would go live according to whatever readings from the temp sensors.

keypad_power_off would be pretty much the same as it is except waiting for power on sense to go low before shutting off the FAN outs.

So is that all impossible or is that about right ? Like i say im no C coder but i'll force myself to learn if it's actually do-able, i wanted to use the dallas t-sensors as a backup through LCDC and its plugin but if i can use them straight throguh the LCD with a little programming then your right that is FAR better.

The other thing i was thinking is if its not possible to reprogram that command directly then could i use a PIC or IC to control the commands sent to the LCD, triggered by the GPO2 low pulse from keypad_power_on and then following roughly the same sort of setup. Have absolutely 0 knowledge pf PICs but its a thought.

Just wish my m8 wasnt at university right now, never a coder round when you need one :).
Last edited:


New member
Just thought i'd make a bump post with a few updates. The PIC method seems to be the way to go and thankyou very much for the suggestion of the pulse sensing flow meter. I had a word with Ron Wlock creator of the ESDS shutdown circuit which i am now planning to intgrate and found i could use the same PFO Meter in his circuit. So now not only will the PFO meter be outputting stats to the LCD via a Fan RPM header, it will also be serving the practical purpose of the no flow shutdown circuit. Great suggestion!:cool:

Here is the Bit-Tech forum post where the new planned PIC circuit is and all the revisions.
USB LCD Displays - Graphic and Character LCDs with a Keypad