analog vs digital control


 
J

Jason H.

TVWBB Member
All,

I am a controls engineer and, once I pick up my first wsm in a month or two, I want to make my own forced air system. Most of the systems I have found online are digital (ie fan is on or off) but I was thinking of making mine analog with an analog output from my plc going to a dc motor control board going to the fan. Has anyone tried analog control and seen a performance difference between that and digital?

Also it seems to me that with a fan going in there (I am going to find a 5 cfm fan as suggested on various sites for the 18" wsm) you would get ashes blown up on the food. As there are pretty much no complaints on this that I have been able to find, can anyone explain to me why this doesn't happen?
 
The homebrew thread describes a Arduino based controller that uses PwM to modulate the fan speed beyond on/off. It works really well.

James
 
<BLOCKQUOTE class="ip-ubbcode-quote"><div class="ip-ubbcode-quote-title">quote:</div><div class="ip-ubbcode-quote-content">Originally posted by j ostheim:
The homebrew thread describes a Arduino based controller that uses PwM to modulate the fan speed beyond on/off. It works really well.

James </div></BLOCKQUOTE>

James - I was on the Arduino site and I couldn't find a description of what their modules do. Am I web challenged :-)

Curt
 
j and Curt,

Curt, for more info on pwm just google "arduino pwm" and it gives you a good description of pulse width modulation.

Basically you pick a "100% duty cycle time" which lets say is 3 seconds. This is actually what I use on several ovens where I don't have analog control of the heaters (ie its a resistive heater that either has power applied to it or not) and this does work very well. So in your controller you have something like a PID (proportional/integral/derriviative - google that for more info) control loop. The loop is "tuned" to your situation (you set your P, I and D values - google tuning its more art than science and takes some experience to get a good handle on). The PID takes in your setpoint, a reading from the thermocouple in the cooker/over and then spits out a control value from 0-100%. If the output is at 0% then the fan is off all the time. If the output is 50% then the fan is on for 1.5 seconds (50% of your 3 second 100% duty time) and off for 1.5 seconds and then repeats and of course if its at 100% then its on all the time. The on/off time is calculated for all values 0-100%. This does work very well in my experience. Although note that you still need an analog output PID loop to power these as you are looking for a control value between 0-100% duty cycle.

So the above would be a digital output for an internal analog controller. I might have to give that a try as it would be less hardware intensive. I have been there done that for this type of control loop so it wouldn't be a big deal to make it work and tune the loop.

However, note that this is still using a digital output so the fan is always either 100% on for off. Has anyone worked with something where the fan speed itself is a function of the output of the control loop?
 
My controller uses PWM to vary the blower speed, much the same was as PWM can be used to dim an LCD panel. At 50% the blower is sent a PWM signal that has voltage 50% of the time in the pulse train. This has the effect of running the blower at 1/2 speed for as long as you'd like. It works very well until you get down to very small percentages, since the blower requires a certian amount to spin at all.
 
ah ok. So you are pulling your 100% duty cycle down much further (maybe even .5-1 sec) and use that to effectively control speed rather than just having it on/off on a longer duration duty cycle. Interesting.
 
I based my controller of Bob's so we have many design features the same. I found that brushless blowers don't play tremendously well with PWM controls, because there is a digital controller in the fan that is already switching on and off power to the electromagnets. It makes a somewhat loud buzzing sound; the Arduino PWM pulses at like 500Hz.

What I, and some others, have done is to place a capacitor in parallel with the motor which smooths the PWM steps. It doesn't work like you'd think it would, providing a smooth ramp of power from 0V to 12V because the capacitor can charge really fast when the pulse is high. That said, I found a tradeoff between capacitor size and buzzing and my "smoothed" PWM ramps between 6V at 8% and 11V at 100% duty.

My fan is this 6.7CFM model and I've never really seen ash blowing up out of the coals. I do have a Big Green Egg though, so the distance from food to coal is slightly more than a WSM.
 
Disclaimer: I have an electric smoker so I have no need for fan control.

Could you let the fan blow at 100% and use a step controlled vent to direct the air flow between the smoker and the outside. This could also shut off the smoker vent completely when the duty cycle dropped to zero.

Curt
 
quick question: I did see the 6.7cfm model that Bryan posted. My question is that, of course depending on how you mount it, is there a reason that you need the "blower" type instead of a normal computer fan type? I have several of these at home from old computers and would like to use one of those if possible.
 
<BLOCKQUOTE class="ip-ubbcode-quote"><div class="ip-ubbcode-quote-title">quote:</div><div class="ip-ubbcode-quote-content">arduino pwm </div></BLOCKQUOTE>


The blower i bought was from china (it cost more to send than the blower), total about $9. It was a three wire setup. the third wire was a 0-10v control wire. that might help the noise. you can give it a pwm to adjust the voltage
 
Oh snap Anthony, I think you win. That is an excellent component right there!

Jason, I've seen some a couple custom-built controllers that use computer fans as their blower so it should work. The main reason I think blowers are chosen is because while computer fans can blow a good amount of air, that sort of fan design can build any static pressure. Fan fans assume there is a more or less unrestricted exhaust and their CFM drops off sharply if they encounter resistance. Blower-style fans can actually build up pressure. If your grill vents are almost completely closed (as mine are when I'm doing a low cook) there is a pressure inside the grill and they are less effective.

But like I said, I've seen some designs that use flat fans so I guess it does work. I just copied what the commercial controllers use because the idea seemed to make sense given the situation.
 
You must log in or register to reply here.

 

Back
Top