6 Channel Dew Heater Controller

[malc-c]

Steve, no the heaters run at 12v. The regulator is used to provide the 5v that the PCB required from the 12v supply line. I was initially hoping that the PCB would provide any 5v supply (if required) but alas it didn't (I guess that as there is a 5v line on the molex connector when used in a PC there's no need to).

[kirkster501]

Hi Malc, right think I am homing in on this now.

Soooo, on a PC of course the PSU drives the 12v and a 5V directly. So to "recreate" that 5v from the 12v batteries we use you are using the 7805 to provision that by taking a "T" off the 12v input from the battery? And then you connect the 5v output of the 7805 back to the 5v "incoming" wire of the Akasa to fully power it.... ?

Is that about it?

Sorry to be a dullard !

[kirkster501]

Also meant to add that i am thinking of building this, and a USB hub, directly into my battery box...

[malc-c]

One thing I've just found out is that it's better to use single phono sockets not the ganged ones... the GND wires from the outputs (originally the fans) can not be commoned together.

Steve, yes you're on the right lines. The power connector on the board has a red wire (+5), a yellow wire (+12v) and two black wires (GND). Basicallt wired it as per the attached sketch (NOTE check the pin out of the 7805 as some seem to swap the IN and OUT pins !)

[malc-c]

Also meant to add that i am thinking of building this, and a USB hub, directly into my battery box...

I accept no responsibility if it all goes pear shaped ! - I've posed up these notes as I like tinkering and sharing my ideas.

[kirkster501]

I accept no responsibility if it all goes pear shaped ! - I've posed up these notes as I like tinkering and sharing my ideas.

Lol !!!! No worries, 2:1 in electronics from Liverpool uni in 1988... Trust me, I'm a pro !!!!

[malc-c]

Well, I butchered the Phono sockets and re-soldered the GNDs to what was left of them - pleased to say it works using a couple of 52 ohm power resistors as loads for each output. Just need to purchase some heaters now and see how well it copes in the real world

[malc-c]

you could always email akasa, they might play ball and give you some info?

I've had a reply:

Hello Malcolm,

Thank you for your email and interest in our AK-FC-08BK fan controller.

Each of the six control channels is rated to support 30W of power.

There’s no PWM function, as fan speed is manually controlled on the dial by varying power to the fan.

Can I please ask what your application is with this fan controller?

If you have any further questions, please do not hesitate to contact us again.

Best regards,

You're hunch was right Reggie, this unit is simply a 6 variable voltage controller. However that fact it will run 30 watts should still be OK.

I've sent a further e-mail outlining what I intend to use it for, so it will be interesting to hear back from them ....

[Reggie]

That's a pity as it's not going to be very efficient by the sounds of it? Are there any other parts on the board that we can't see in the pics? I think I can make out the pots and the leds on the back of the board but there seem to be more solder lugs that are obviously attached to something else on the other side of the board, and there appear to be some 'q' parts labeled on the left of each of the fan connectors.

[kirkster501]

That's a pity as it's not going to be very efficient by the sounds of it? Are there any other parts on the board that we can't see in the pics? I think I can make out the pots and the leds on the back of the board but there seem to be more solder lugs that are obviously attached to something else on the other side of the board, and there appear to be some 'q' parts labeled on the left of each of the fan connectors.

Yes, shame. It means it wont be pulsing the current right which means a 100% duty cycle.... Hmmmm......

[kirkster501]

I've had a reply:

You're hunch was right Reggie, this unit is simply a 6 variable voltage controller. However that fact it will run 30 watts should still be OK.

I've sent a further e-mail outlining what I intend to use it for, so it will be interesting to hear back from them ....

Also, Malc, sounds like they are onto you mate !!!

[Reggie]

I would be totally frank with them, explain that you appreciate that it is totally out of warranty by your intended use and absolve them of any responsibility, they might talk to you at least a little bit more about it then.

[malc-c]

Yes, shame. It means it wont be pulsing the current right which means a 100% duty cycle.... Hmmmm......

True... like I said at the beginning, if it turns out not to be suitable I've only wasted £25... and I'm sure I could find another use.... Uhmmm I have six 12v LED strips (four red and two white)... could always use it as a dimmer

[Reggie]

you could always make up a small mcu and transistor/fet board that does pwm, it's not like there isn't plenty of room in the box.

[malc-c]

Reggie, I might actually do that... I've got quite a few PICs around the place and as you say it should be fairly easy to do.

One question... what sort of frequency do the commercial units run at. This was the sort of pulse rate I developed for my PID stat

[Reggie]

no idea, I've been told that mosfets are the way to go and give yourself a duty cycle range, as long as it's fast switching it'll be more efficient.

[kirkster501]

no idea, I've been told that mosfets are the way to go and give yourself a duty cycle range, as long as it's fast switching it'll be more efficient.

There are some threads on here where people have made them from kits for a few bob. What about this?

Velleman - Velleman Kits

Steve

[malc-c]

Steve, I think that's what a lot of guys use, plus the combi-drill controller.

I've got a bit of code for an 8ch servo driver that I built many years ago. I could easily add some A-D for the use of pots and see how well that works, but as explained in the original post, I really wanted to take something that needed little conversion rather than build from components

[Psychobilly]

so have you totally dicsounted the fan controller?

The MFA Como drill controller is "noisy" they admit it in the literature which says to keep the load leads short...

The commercial dew controller I bought uses very low frquency for the PWM Cycle times in the order of a couple of seconds.. the Thermal time constants involved are very long so theres no heed for the high frequency switching which is needed for Motor control applications...

Peter...

[malc-c]

Peter,

No not really. I think the efficiency thing with PWM is really of concern for those who are running on battery power-packs. As the setup is observatory based with a 13.8v regulated power supply I'm not that bothered about it being constant DC. The only concern is that where with PWM it's easier to maintain a steady temperature, with DC this will be very much a suck it and see exercise

[Reggie]

Personally, the PWM is about efficiency but not just because it's battery packs, electricity is expensive these days, so efficiency is king in my eyes. My other reason for pwm would be to control the rate at which the telescope heats up, I intend to do this automatically, this isn't for heating efficiency as such but more to keep the telescopes at sustained temperatures for better viewing/imaging. I think we spend so much money on equipment and neglect this single area.

If it was me and i didn't want digital control of the heaters, then I would use 3x556 timers and then 6xmosfets, you should be able to reroute the pots to control the pwm rate on the timers, swap out the blue leds for red ones and incorporate those into the 556 circuit to indicate the rate they're switching at.

[Psychobilly]

Even low fequency PWM is Efficient compared to a linear design...

Peter...

[malc-c]

If it was me and i didn't want digital control of the heaters, then I would use 3x556 timers and then 6xmosfets, you should be able to reroute the pots to control the pwm rate on the timers, swap out the blue leds for red ones and incorporate those into the 556 circuit to indicate the rate they're switching at.

OK, I give in

I'll sort out on order from RS and knock something up on stripboard...

[michael.h.f.wilkinson]

Personally, the PWM is about efficiency but not just because it's battery packs, electricity is expensive these days, so efficiency is king in my eyes. My other reason for pwm would be to control the rate at which the telescope heats up, I intend to do this automatically, this isn't for heating efficiency as such but more to keep the telescopes at sustained temperatures for better viewing/imaging. I think we spend so much money on equipment and neglect this single area.

If it was me and i didn't want digital control of the heaters, then I would use 3x556 timers and then 6xmosfets, you should be able to reroute the pots to control the pwm rate on the timers, swap out the blue leds for red ones and incorporate those into the 556 circuit to indicate the rate they're switching at.

I agree. They must be using PWM or else they would be dissipating way too much heat, and would need serious heat sinks. PWM is the most efficient way to regulate voltages. Adding a capacitor can remove any high frequency components on the output (but only if you need that; in the case of resistive load: NOT).

I think they misunderstood the question slightly, and their answer is ambiguous to say the least.

Neat project, BTW, might well try the same.

[Reggie]

I suppose a non-tethered pic/atmega chip would do the same job as the 555 timers, just got to read the pots on the analog inputs and switch the fets. It would certainly keep component count down and be less of a routing headache?