Air pressure fan regulation. [project suggestion]

[hmmpic]

Air pressure fan regulation. [project suggestion]

I have 100mm 220VAC cylinder fan (150m3 air/h), the output is connected direct to a carbon filter(air is coming out here). The other side of the fan is connected to a pipe and then into the box we want to ventilate.

How to do the regulation on the fan, maybe something from 30m3 air/h to max?

If using a speed controller (maybe it was too cheap or bad) it won't work right, because the air pressure drop so much that no air coming out of the filter, can't hold the air pressure...

Is the right way here to use on/off regulation based on time? Full power for xmS, and then off for xmS?
Or make a zero crossing regulation, and then have maybe 5 fixed power step?
Maybe reducing the speed 10% will reduce the air flow much more?

I am just confused how to do the regulation the right way, and keep up the air pressure?

[bkamen]

Well,

If pressure matters, first you need an air pressure sensor.

From there, you build a PID loop.

And yes: the smart think would be to have a clock input from the line voltage that can be used as the period for the PWM output that you'd run off to a solid-state relay.

I did something similar for an oven controller. Worked nicely.
_________________
Dazed and confused? I don't think so. Just "plain lost" will do. :D

[Ttelmah]

and, the reason ypur speed controller didn't work is down to the
nature of the motor used in the fan. What is needed to genuinely
give good speed control depends on the actual type of motor used.
The odds are that your existing controller was resulting in the motor
torque being massively reduced (and it's efficiency - so it'd probably
run hot), and the speed would reduce massively rather than adjusting
properly. Genuine speed control of such AC motors, will normally need
a proper AC synthesised inverter, not a basic 'dimmer' type controller.

Question. Could you use a DC fan instead?. Much cheaper and easier
to speed control properly.

[hmmpic]

@Ttelmah
What you say is, for to do the regulation on my 100m cylinder fan (normal used in bathroom and so) i can not just do simple ZCD and adjust the voltage with a triac and a PIC?

You are correct, the dimmer i used and failed with, was a cheap one there get the fan hot and it was running real bad.

So no simple Triac solution or "AC phase control" for the fan?

[Ttelmah]

OK.
The motors used in things like fans depend on torque being generated by
'chasing' the magnetic field being modulated by the AC frequency.
To genuinely 'speed control' these, you have to actually change the
frequency they are being fed with.
Now triac and resistive dimmers will give a 'speed effect' on these. What
happens is that reducing the effective amplitude of the waveform, reduces
the torque that the motor can produce, so it stops being able to 'keep up'
with the magnetic field, and the output speed reduces, but it is then running
massively more inefficiently, drawing more current and running hot. Done
for any length of time, this can result in motor failure....
Now you can motor speed control with a triac, by skipping cycles. Here
instead of turning the triac on for varying parts of every mains cycle, you
leave it on for perhaps ten cycles, then turn off for one or two. result is
a reduction in the effective number of cycles per second, and the speed
correctly reduces. Controllers that can do this will have a 'motor control'
specification or setting.
So for speed control you need a proper motor speed controller, not
a standard 'dimmer'.

[hmmpic]

Super thanks. I will try it out:-)

[bkamen]

Everything Ttelmah said is correct... AC motors with speed controls are pretty limited in how much control over that speed they can assert while using 50/60HZ line freq.

In the design where I used a SSR -- it was a resistive and not inductive load.

The same design would let you skip cycles as well as PWM a single cycle (I should have mentioned that).

I'd also say that if you could switch to DC, that would be better in the end.
_________________
Dazed and confused? I don't think so. Just "plain lost" will do. :D

[temtronic]

Ok, I look at this another way...
100mm is about 4" on this side of the pond, Canada,eh !
there are 1,000,000s of PCs with speed controlled cooling fans
so, if possible...
use a PC cooling fan or some other DC powered fan.

as for AC powered motors/fans... Microchip had an application note about them and a specific 'motor controller' they sold, years ago (10-15 ? ) .

BTW the current design for home HVAC units is to use DC powered, PWMed motors for fan use...if 'they' use them there HAS to be a real good reason !

Jay

[Ttelmah]

I think everyone has said 'better to go DC'....
Much easier to control.

[bkamen]

And if you go with DC fans, you can get the 4wire kind (that are so popular in computer/electronics applications) that specifically have a PWM input wire. (and tach/speed return line)

Now you're not even modulating power. It's done on the fan for you.

Based on your 150m3/hr, that's 2.5m3/min -- and that's a pretty easy fan to find.

I'm looking at www.digikey.com and there's lots to choose from.

Cheers,

-Ben
_________________
Dazed and confused? I don't think so. Just "plain lost" will do. :D

[hmmpic]

Yes I go DC, but first after this AC is running smooth
---
Anyone know what to do here.

Have a nice ZCD and a Triac driver, all look right on my scope, with a lamp from 0-100% regulation.

When using a small fan (220VAC) then i can't get it working in the 1 half Quadrant.
The Fan just run so slow, same as <10% power. Must apply a delay for about 2mS after the ZCD for to get it right. After the 2mS delay all work as expected.

The ZCD is running spot on and no noise!

Triac are a BT137X-600D (logic and low Gate control)
Opto driver a MOC3023 (the diode is powered with about 10mA)

Connection is as the application notes. (page 7 fig 9.)
Link: https://www.tme.eu/Document/4c07e045ed4ac191018a957800b78b20/MOC3023M.pdf

Any clever hints?

[temtronic]

Yeah, go DC !! Honestly, controlling an AC motor is a LOT of work (time !) and some type of motors are impossible to become variable speed.
You could easily spend 2-3 man months and still not get it to work properly.
I've used the MOC3021 for decades in my 'self indicating SSRs' that I used to control relays in HVAC systems. Again..on-off... NOT phase control / speed control.

There's a HUGE reason why 'everyone' uses DC motors...EASY and CHEAP to control.

If you can search the Microchip applications and 'old products' , they sold a preprogrammed PIC for AC motor control. NOT a 'PICxxxx', some weird product ID. Wish I had time to look it up but I kinda put my dump trailer into a 3M deep crater 3 hrs ago...so it's a 'priority'. The chip has to be 20+ years old, maybe 16C71 based ? Hopefully you can find it, GREAT reading, graphs, why and how.....charts, etc.

[hmmpic]

yes totally agree! But maybe you know the feeling, small adjustment and then it fit what you want.
DC is the way to go, but for this one i need AC.

Project work as i want, i have 32 speed step in the right speed/pressure on the fan. It is a close loop where the TC & RH are controlled with the fan. The fan is not one fan, but 50 fans, therefore no DC this time, fan are at stock.

I only thinking about why Quadrant 1 & 3 close after the ZCD not working as i expect. ZCD and pulses for the MOC3023 are spot on. So maybe something with the trigger on the logic Triac, but again i don't like 220VAC mixed on my scope, therefore all are developed with 12VAC, without the fan.

[temtronic]

I had to download the datasheet ....that's a 'random' triggering opto-coupler not a zero cross one. The one's I used were zero cross, made by Motorola, 2+ decades ago.
I presume each fan has it's own 'controller' ??

sigh, I can't find the ap note/datasheet for the PIC I've been thinking of (thought it had an MTAxxxxx part number )but...
found these apnotes..
AN887, AN967, AN984.

should give you lots to read about 'how to' control AC motors.

Jay

[bkamen]