Phase controlled circuit

[Boncuk]

Hi,

I intend to air condition my entire house which involves a lot of room fans for equal air distribution.

The fans I can purchase here are high rpm high power types, far too loud at nominal power. Reducing fan speed with a phase contol unit would solve only a part of the problem. The ON/OFF cycles are controlled by a MASTER controller and the rpm regulation, done with normal (modified) dimmers should be left to individuals in the rooms. Setting the fan speed too low the fan won't start from a full stop and hence burn up with current flow and no rotation.

So I thought about the possibility employing a small µc which can be set to output a signal high enough to safely start the motor at any condition.

A safe start-up voltage for the fans is 130V (mains 230V).

I made up a zero-crossing-unit to be used for an interrupt in a µc, gaining the zero-crossing point without transformer (which can already cause a phase shift). The circuit output is a TTL compatible high pulse of 50 to 200µs duration every 10ms at zero crossing

My knowledge concerning µcs is very poor, not to say lousy.

What the controller should be able to do: Accept up and down key inputs to increase/decrease output pulse lengths based on the interrupt input to control the triac circuit. Have a three bit preselection speed setup (for output voltages 130,140 and 150V) to be set by two jumpers. Minumum setting via keys always to the preselected speed (voltage).

Here is the schematic of the zero-crossing circuit.

I would very much appreciate if somebody can take the time to make a source code (PIC or ATMEL) for me. I hope I can do the rest on my own.

Thanks for your attention

Regards

Hans

 

[ericgibbs]

Tag Hans,

What the controller should be able to do: Accept up and down key inputs to increase/decrease output pulse lengths based on the interrupt input to control the triac circuit. Have a three bit preselection speed setup (for output voltages 130,140 and 150V) to be set by two jumpers. Minumum setting via keys always to the preselected speed (voltage).

I would look at using one or more of the analog inputs of the MCU, by using a variable resistors it would be possible to adjust the fan's speed over the full range.

Regards to starting from a slow speed setting, consider starting the fans at full power power for say 15 seconds, then drop back to the selected speed.

The 16F877A PIC can be configured to have 8 analog inputs.

EDIT:
By connecting the zero crossing pulse into say PORTB.0 [interrupt] it should be possible to have a variable delay for a trig pulse output to a opto triac and vary the phase [power] to the fans.

 

[Boncuk]

Hi Eric,

thanks for the suggestion, but I thought about a small MCU (8pins or so). MCUs have at least one internal timing circuit which could be used to control the length of the output pulse tied to the synchronization pulse of the zero crossing unit.

I want to keep the contoller small enough to fit in a box usually used for light switches or wall outlets. VCC power will be supplied by a central power supply in the distributor cabinet.

Occupying one port pin for the /INT, two more for jumper preselction, one each for up and down keying and one for the outpulse pulse an ATTINY25/45 should be sufficient for the job. That makes a total of six I/O pins which are available with the ATTINY25/45.

I guess for a job like that the CPU might work well with its internal clock generator.

Starting the fans at 100% power (and speed) you would certainly look for the fire brigade approaching. Their sound is similar to a sirene. (Made in China. Where else in Thailand?)

Hans

P.S. Where did you learn this horrible "Tag"? Literally translated it just means day. Saying "Tag" or "Tach" is considered rude in Germany.

 

[ericgibbs]

Hi Hans.

P.S. Where did you learn this horrible "Tag"? Literally translated it just means day. Saying "Tag" or "Tach" is considered rude in Germany

I heard it a lot in Germany, 1951's at Wildenwrath, working alongside the GSO on guard duty together.
I shouldn't have used the familiar term, it wont happen again.

 

[Boncuk]

Hi Eric,

no problem, if soldiers use it. They aren't supposed to be gentlemen anyway.

Reminds me to my time in military basic training. My uniform wasn't the prescribed way, not wearing a tie, no shoe polish, and some other minor defects too. The platoon sergeant put this to my attention in a pretty rough way. I asked him: "What do you want? A dressman or a killer?"

Hans

 

[Gayan Soyza]

Just an idea

*You wait until uC receive an external zero cross interrupt.
*When it received an ZX interrupt turn on a Timer (Timer length < 9mS).
*Wait until timer overflows & after fire the triac.
*Wait until next ZX occurs.

You will be lucky your TTL ZX detector generates high pulses for both edges.So you don't need to alter the external interrupt edge bit in the code (Interrupt on Rising edge will be fine)

Varying Timer value is the one you need to control the fan speed.You can add two switches to increment/decrement the Timers length.

8 bit timer will be fine for smoother phase control.

 

[Gayan Soyza]

Or I would go with Erics suggestion to control over full range.

Connects an Preset to an AD input an sample the AD value & adjust the Timers length according to that AD value.

 

[Boncuk]

Hi Gayan,

using a pot to set the desired speed is something which can't be done in Thailand.

I don't know of any other people in the world who can destroy a mechanical control element easier than Thais can do.

I've seen a lot of TV-sets with missing knobs for channel selection. The channel selector is tougher than the knob, hence the plastic material of the knob can't withstand the power applied when attempting to turn it beyond the mechanical stop.

Even the stereo amplifier in my stepdaughter's room doesn't have any more knob and the balance pot has completely surrendered.

I'm thinking of a two button control box with vandalism prove stainless steel buttons as they are used in ATM machines. (were plastic some time ago, but all changed)

The controller must be presetable for a defined voltage output at the triac to guarantee safe operation of the fan when starting from stillstand.

Stillstand will be necessary for outside temperatures lower than inside the house and during maintenance periods.

Using two jumpers there will be the choice of three different starting voltages for the motor.

Hans

 

[rjvh]

I don't know of any other people in the world who can destroy a mechanical control element easier than Thais can do.

Cambodians

 

[Boncuk]

Cambodians

to complete the list: all Asians I guess.

The 30 year war in Europe was a piece of cake compared to the never ending war of Asians and technical devices.

 

[hafsat]

need a circuit diagram that can detect a phase sequence failure in a three phase supply

 

[rjvh]

Phase sequense relay is comercialy availible for about $60

Have a look on the MICRO website
its a Malaysian company that makes alkind of probing protection relays

Robert-Jan

 

[MrAl]

Hi,

Hans, i have to ask, did you already try driving the fans with a triac output
at some sub cycle pulse width? Doing this sometimes makes noise too,
so im not sure if that would solve your problem or not, but i thought i
would ask first.

 

[Boncuk]

Hi,

Hans, i have to ask, did you already try driving the fans with a triac output
at some sub cycle pulse width? Doing this sometimes makes noise too,
so im not sure if that would solve your problem or not, but i thought i
would ask first.

Hi MrAl,

sorry for the late reply. I assumed nobody was interested in the threat anymore.

I don't know exactly what you mean saying "at some sub cycle pulse width". If you understand firing a triac randomly there is a lot of noise and can also kill a motor. The triac works alright for a couple of cycles and then it heats up and misfires in rapid sequence making the motor change rotation at full speed, which in turn overheats the motor rapidly.

I tested that behaviour with a 5KW motor and a big radial fan connected to it. The change of rotation almost broke the shaft with the heavy 'fly wheel'.

Hans

 

[rjvh]

need a circuit diagram that can detect a phase sequence failure in a three phase supply

Just came across the brochure

Robert-Jan

 

[MrAl]

Hi MrAl,

sorry for the late reply. I assumed nobody was interested in the threat anymore.

I don't know exactly what you mean saying "at some sub cycle pulse width". If you understand firing a triac randomly there is a lot of noise and can also kill a motor. The triac works alright for a couple of cycles and then it heats up and misfires in rapid sequence making the motor change rotation at full speed, which in turn overheats the motor rapidly.

I tested that behaviour with a 5KW motor and a big radial fan connected to it. The change of rotation almost broke the shaft with the heavy 'fly wheel'.

Hans

Hi Hans,

Maybe i misunderstood the problem. I thought you wanted to switch
the triac on for parts of the cycle, not for the full cycle, in order
to try to regulate speed. If you want to switch it on for the full
cycle than that shouldnt be a problem. I surely didnt mean to
turn the triac on 'randomly' and i dont know why you would think that.