Temperature Controlled Fans for Gas Fireplace - Equivalent Transistors and Speed Control Issues

Thread Starter

JimStokes

Joined Oct 25, 2015
38
Hi, I'm new to this forum and electronics, really, so please keep that in mind.

I found a circuit online that got me started on this project and I've modified it because I want to control three fans. My goal is to have a Thermistor (10K NTC 103) control when the fans go on and off in response to the temperature underneath my gas fireplace. The fans will blow the hot air under, up the back, and shoot out at the top front of the fireplace. I already have a 'squirrel cage' fan/blower, but it makes a lot of noise and doesn't seem very efficient.

Below is the circuit, which works. But there are a few things I'd like to resolve:

1. Note: I realize the LM1458 is a dual, but the LM741 I had on hand got fried, so I switched it for the next available 'equivalent' IC.

2. The original circuit I found was designed for a single fan using a single 2N2907. But this was overloaded when I wanted to hook up the three 12V DC 0.2A Brushless Fans. I gave up trying to find an equivalent single transistor after experimenting unsuccessfully with a TIP32C. Question 1: Any suggestions for a single equivalent transistor? (remember that I'm very new to this stuff :)))

3. I would like to be able to adjust fan speed. I've been successful by inserting a 10ohm resistor between each of the transistors and the fans. I tried using a single resistor and bringing all negative terminals together from the fans, but it didn't give me the result I wanted. I'd like to make the speed either switchable (high/medium) or with a range (like using a potentiometer), but the current configuration requires a triple switch. Question 2: Any suggestions on how I can best adjust the speed? This may be easier if there is an equivalent transistor for the three 2N2907s.

upload_2015-11-17_18-38-32.png

Thanks for your help.
Jim.
 

wayneh

Joined Sep 9, 2010
17,496
I'm wondering if that's enough fan power. 7W won't move much air.

You might look into temperature controlled PWM (pulse width modulation). This is what the computer modding guys use to cool their rigs. It'll seem a little complicated at first but it's a good way to go.

You should also learn about MOSFETs, a type of transistor that's superior for on/off switching like PWM.
 

UsagiCurry

Joined Nov 17, 2015
9
Some commercial temperature controlled fans use a TEC as a generator to directly drive the fan without an outside power source. Mostly for wood fired stoves.

I think you could use a variable voltage regulator to control the speed if you don't mind having heat loss. A lm317 or lm337 might do the trick. I could be wrong though. I have used them to make a simple welder wire feed circuit to control the motor at a constant torque under different loads for a push pull gun: so the aluminum wire wouldn't ball up in the gun if the gun feed rolls slipped.
 

#12

Joined Nov 30, 2010
18,224
I'm really miserable at flammable fuel heaters, but here's an example for one cubic foot of propane per minute:

2488 B.T.U / cubic foot of propane
2488/1.08 = CFM (of air) dT (of air)
2303 = CFM dT
For air entering at 70F and leaving at 170F, you need 23.03 CFM

That sounds to me like a very reasonable number for a small fan.
You will need less air than this because flammable fuel heaters are never 100% efficient.
You can look up whatever your fuel really is and how much you really use in a minute.

http://blog.amerigas.com/residential-propane/geeking-out-over-propane/
 

Thread Starter

JimStokes

Joined Oct 25, 2015
38
I'm wondering if that's enough fan power. 7W won't move much air.

You might look into temperature controlled PWM (pulse width modulation). This is what the computer modding guys use to cool their rigs. It'll seem a little complicated at first but it's a good way to go.

You should also learn about MOSFETs, a type of transistor that's superior for on/off switching like PWM.
Thank and I'll look into PWM and MOSFETs. The current blower provides 135cfm, but is noisy and we run it at about half speed (say 60cfm - is a guess) to keep down the noise. Each of the fans I'm proposing has a max 30cfm, which when combined will give sufficient blow. The slot under and behind the fireplace is narrow, maybe 3" by 12", and there is a natural flow of air with the heat of the fireplace. These fans will assist that flow. I'm fairly convinced these fans will provide ample push. Thanks for your help.
 

Thread Starter

JimStokes

Joined Oct 25, 2015
38
Some commercial temperature controlled fans use a TEC as a generator to directly drive the fan without an outside power source. Mostly for wood fired stoves.

I think you could use a variable voltage regulator to control the speed if you don't mind having heat loss. A lm317 or lm337 might do the trick. I could be wrong though. I have used them to make a simple welder wire feed circuit to control the motor at a constant torque under different loads for a push pull gun: so the aluminum wire wouldn't ball up in the gun if the gun feed rolls slipped.

Thanks. I looked up a TEC - LT1923 as an example - and it is expensive ($44+ at DigiKey). I'm looking for an inexpensive solution, but thanks for letting me know about TECs.

I'll look at the lm317 and lm337. Thanks.
 

wayneh

Joined Sep 9, 2010
17,496
Another option is to forget about the temperature and make the fan manually controlled. A PWM circuit can give you continuous speed control using a pot. Or you could have, say, 2 or 3 settings plus off. The setting would choose a a resistor, which it turn would change the PWM setting to, say 25, 50 and 100% duty cycles. (Since 100% is just full on, you could just bypass the PWM circuit at that setting.) You can get fancy and place all this under control of the temperature sensor, but it's not really necessary.
 

Thread Starter

JimStokes

Joined Oct 25, 2015
38
Another option is to forget about the temperature and make the fan manually controlled. A PWM circuit can give you continuous speed control using a pot. Or you could have, say, 2 or 3 settings plus off. The setting would choose a a resistor, which it turn would change the PWM setting to, say 25, 50 and 100% duty cycles. (Since 100% is just full on, you could just bypass the PWM circuit at that setting.) You can get fancy and place all this under control of the temperature sensor, but it's not really necessary.
Thanks. Great idea. My fireplace is on a thermostat, so it pops on and off as needed. I'd like for the fan top come on as needed as well, so my preference is to have a sensor. It works now, just a few bugs (ie. my questions) that I'm trying to get past. Thanks again for your help.
 

wayneh

Joined Sep 9, 2010
17,496
You should post a link here to your fans. Someone that knows about such things should chime in on the appropriate method of control. I don't think all fans can use PWM control. Others are ready-made for it (with a 3rd and/or 4th wire) and need only the PWM signal so you don't need the power switching transistors external to the fan.
 

Thread Starter

JimStokes

Joined Oct 25, 2015
38
You should post a link here to your fans. Someone that knows about such things should chime in on the appropriate method of control. I don't think all fans can use PWM control. Others are ready-made for it (with a 3rd and/or 4th wire) and need only the PWM signal so you don't need the power switching transistors external to the fan.
See Below, the three fans are the last one on the list: OD8025-12H, 3300rpm, 40cfm, 33dB, 12V, .18A. I misquoted before, saying that they were 30cfm, but these are actually 40cfm.

upload_2015-11-18_16-6-20.png
 

AnalogKid

Joined Aug 1, 2013
10,987
As you are seeing from the responses, there are a lot of ways to control a fan's speed. So what is it you want? What is the control action you want to implement?

As the exhaust air or fire pan temperature increases, do you want the fans to speed up or slow down?
Is there a temperature range over which you want the fan parameters to change? For example, at 25C the fans sit at 50% of rated voltage; at 30C they start increasing speed; at 50C they are at full speed. Or something like that?

For anything under 20 watts of fan power, I strongly prefer linear fan speed control over PWM. Less electrical noise, greater reliability, no EMI, etc.

Why didn't the TIP32 circuit work? Can you post the schematic?

ak
 

Thread Starter

JimStokes

Joined Oct 25, 2015
38
As you are seeing from the responses, there are a lot of ways to control a fan's speed. So what is it you want? What is the control action you want to implement?

As the exhaust air or fire pan temperature increases, do you want the fans to speed up or slow down?
Is there a temperature range over which you want the fan parameters to change? For example, at 25C the fans sit at 50% of rated voltage; at 30C they start increasing speed; at 50C they are at full speed. Or something like that?

For anything under 20 watts of fan power, I strongly prefer linear fan speed control over PWM. Less electrical noise, greater reliability, no EMI, etc.

Why didn't the TIP32 circuit work? Can you post the schematic?

ak

Thanks for your help.

1. I want the fans to come on when the temperature gets to a certain point. That's not a problem for me. The thermistor I have provides resistance of around 10k at 25C and drops to around 5k at 45C. This part of the circuit is working fine and I intend to insert a trimpot beside R2 and R2 so that I can have some range of adjustment.

2. I want the fan speed to be set by a manual switch (eg. high, medium) or potentiometer. I'm having trouble determining where it is should go because I currently have three transistors and three parallel fans, making it necessary to either have three switches or a triple layered potentiometer. I don't like those ideas and so thought I would ask the group to help. If there is a single transistor that I could use to replace the three, then I could put in a single potentiometer or switch to control all three fans simultaneously. The fan speed switch/potentiometer is to be operated manually - I don't want the fan speed to change automatically with temperature.

3. The TIP32C just didn't respond. I don't know why. It was installed in place of the first of the three 2N2907s (the other two were removed), with the three fans coming off in parallel between the emitter and ground of the TIP32C. I thought it would respond and thought it was a good solution because it can carry a higher current (3A) than the 2N2907 (.6mA). But.... (this is where my lack of knowledge and experience shows)

Thanks for any suggestions you may have about replacing three 2N2907s for one something else and controlling fan speed.

Thanks, Jim.
 

AnalogKid

Joined Aug 1, 2013
10,987
"Well, there's your problem..."

In your schematic in post #1 you show the fans connected from the collectors to GND. That is how the TIP32 should be connected, not from emitter to GND as you just stated. Also, note that the three pins are in a different order between the TIP and 2N parts.

As discussed, the simplest way to slow down BLDC fans is with fixed series resistors, but that method has limited adjustability and all of the fan power goes through the adjuster, whether it is a pot rated for the current or a set of switch contacts. Another approach is to grow what is essentially a low dropout regulator, where only the control current is going through the pot. Since you've cleared up the control requirements, this sounds like a comparator (for on-off control) (which you already have) driving a regulator (for adjustable ON speed). You already have a dual opamp and a power transistor, so this approach takes nothing but a few more passive components and one diode. Does this sound like something you want to try?

ak
 

Thread Starter

JimStokes

Joined Oct 25, 2015
38
"Well, there's your problem..."

In your schematic in post #1 you show the fans connected from the collectors to GND. That is how the TIP32 should be connected, not from emitter to GND as you just stated. Also, note that the three pins are in a different order between the TIP and 2N parts.

As discussed, the simplest way to slow down BLDC fans is with fixed series resistors, but that method has limited adjustability and all of the fan power goes through the adjuster, whether it is a pot rated for the current or a set of switch contacts. Another approach is to grow what is essentially a low dropout regulator, where only the control current is going through the pot. Since you've cleared up the control requirements, this sounds like a comparator (for on-off control) (which you already have) driving a regulator (for adjustable ON speed). You already have a dual opamp and a power transistor, so this approach takes nothing but a few more passive components and one diode. Does this sound like something you want to try?

ak
Thanks. I'll try the TIP32C again - with collector to fans to ground.

What you are describing is of interest to me and, yes, I'd like to try it.

Thanks, Jim.
 

Thread Starter

JimStokes

Joined Oct 25, 2015
38
I got the TIP32C to work. Thanks for correcting my dumb mistake.

Now, I can run the three fans in parallel off of the collector and it works fine. Now just for installing a switch for manually adjusting speed. I tried a 10ohm resistor between the collector and the fans and it worked, but got really hot fast. I also think that the 10ohm resistor actually slows the speed too much. I look forward to hearing your suggestion AnalogKid.

Thanks, Jim
 

AnalogKid

Joined Aug 1, 2013
10,987
Note that this is a first-order approximation, and the exact performance will be a bit different.

Each fan has a rated voltage and current. With Ohm's Law this gives an equivalent resistance. Divided by 3 equals the equivalent resistance of three identical fans in parallel. A resistor of the same value in series with the group of fans will cut the fan voltage approximately in half. The power in the resistor is Vsquared/R, where V is approx. 6 V. This gives the power dissipated in the resistor. You want to use a resistor rated for at least twice as much. In reality this will drop the fan speed by more than 50%, but it is a starting point. Using similar arithmetic you can select resistors for other fractional fan speeds. Most BLDC fans do not like to run at less than 50% or their rated voltage, and many that will run at 50% won't start at the low voltage.

Also, there should be a resistor in series with the base of the TIP32 to limit the base current to something the opamp can handle. With your circuit the opamp is basically running into a short circuit to the + battery, and its internal current limiting is preventing it from dying.

ak
 

wayneh

Joined Sep 9, 2010
17,496
Great to hear you're making good progress. Speed control might be accomplished with an LM317 to control the voltage supplied to the TIP 32 and fans, to reduce it from 12V down to 6V or so. You can control the minimum voltage by using a proper arrangement of a pot and an external resistor. The fans take ~1A at full power, right?
 

Thread Starter

JimStokes

Joined Oct 25, 2015
38
Great to hear you're making good progress. Speed control might be accomplished with an LM317 to control the voltage supplied to the TIP 32 and fans, to reduce it from 12V down to 6V or so. You can control the minimum voltage by using a proper arrangement of a pot and an external resistor. The fans take ~1A at full power, right?
According to the specs, the fans take 0.2A each. I am assuming 0.6A together.
  • Rating: 12VDC
  • Current: 0.2 Amps
  • Power: 2.4 W
  • Bearing: Sleeve
  • Speed: 2700 RPM
  • Noise: 29.8 dBA
  • Max Air: 37 CFM
  • Max Pressure: 3.24 mmH2O
  • Temperature: -10 to +70 C
  • Hours: 20000-30000
 

AnalogKid

Joined Aug 1, 2013
10,987
So one point on the fan-equivalent-resistance curve is 60 ohms at 12 V. If you have a variable DC supply and a VOM, connect the fan to 6.0 V and measure the current. This will give a 2nd point, and a linear approximation between them will work well for determining the details of the design.

ak
 
Top