Thermal Control Circuit - Comments Please

[Ed]

Hello,

After looking at some simple thermal control schematics (and not having
the parts for them) I came up with my own design with parts I had on
hand. This is just a very basic circuit, the warmer the Thermistor gets
the more power the fan gets.

Please take a look at the schematic and comment.
http://tinyurl.com/gtgy

Just want to know if the thermistor is in danger of blowing up or
anything?

Thanks,
Ed

btw, using this on a single 92mm fan inside a PC. (approx. 2.4W max)

 

[Mark Jones]

Hello,

After looking at some simple thermal control schematics (and
not having the parts for them) I came up with my own design
with parts I had on hand. This is just a very basic circuit,
the warmer the Thermistor gets the more power the fan gets.

Please take a look at the schematic and comment.
http://tinyurl.com/gtgy

Just want to know if the thermistor is in danger of blowing up
or anything?

Thanks,
Ed

btw, using this on a single 92mm fan inside a PC. (approx.
2.4W max)

Nice, simple, good design Ed. : )

The 100k drive resistor leaves not a whole lot of current to the base, is
the TIP115 a darlington device? If not, you could likely drop that base
resistor by a magnitude of 10 or so to adequately "turn it on," otherwise
the device is operating within it's linear region and wasting power as heat.

And if power is an issue, it might be possible to save a tiny bit by
reducing the 1uF and increasing the 2.2k (and chosing a higher room-temp
thermistor.)

Also it may not be much of an issue with a 470uF bypass cap already on the
motor, but usually on motors and other "heavy inductors" which are
controlled by a semiconductor, a diode is placed in parallel with the
inductance as to mitigate any possibility of "kickback" frying the
transistor.

Regards,
Mark

 

[Ed]

Nice, simple, good design Ed. : )

The 100k drive resistor leaves not a whole lot of current to the base, is
the TIP115 a darlington device? If not, you could likely drop that base
resistor by a magnitude of 10 or so to adequately "turn it on," otherwise
the device is operating within it's linear region and wasting power as heat.

And if power is an issue, it might be possible to save a tiny bit by
reducing the 1uF and increasing the 2.2k (and chosing a higher room-temp
thermistor.)

Also it may not be much of an issue with a 470uF bypass cap already on the
motor, but usually on motors and other "heavy inductors" which are
controlled by a semiconductor, a diode is placed in parallel with the
inductance as to mitigate any possibility of "kickback" frying the
transistor.

Regards,
Mark

It's a darlington from what I could find on the part #. Specs sounded
better then just a regular PNP. It's rated 2A if I am reading the data
sheet right. (?)

I used the 470 cap to make the power output smooth, otherwise it's like
a real fast on-off switch, I don't think that would be good for the fan
or PSU.

Could I just add a diode to what I have, or would the 470 cap need to
come out? Could I use both a cap and the diode? Rectifier diode OK?

I'm just a hobbyist and haven't touched an iron in probably 10 years,
couldn't even remember how to read a resistor. ;p

Thanks!
Ed

 

[FPF Info]

Why bother with the timer? Can you not just use the thermistor to control
the output voltage of a properly biased power transistor?

 

[sadolph]

Ed,

What bothers me is the AC signal to the fan-- I think it might work better
if you use the 555 to produce the duty cycle whose average is the voltage to
drive the fan, and then use lumped elements to average the duty cycle down
to DC.

How about a first stage that is a switcher like you have, (but lower power),
- take the output from the PNP and filter it to take away AC with L/C
circuit
- use a darlington emitter follower high power stage to drive the fan with
DC.

my 2 cents--steve

 

[Dan Fraser]

1. If you only change the frequency without changing the duty cycle, the
fan speed will stay the same. Duty cycle will vary a little with this
circuit but maybe not enough for your need.
2. The 470uF cap is way too big. It acts as a dead short on the TIP115
transistor and will cause it to blow eventually. Actually, all you need
is something like 0.1 uFd as a noise suppressor and nothing else.
3. The thermistor will not blow.
4. The idea of using the thermistor to control the base of a transistor
is fine if you want to generate a lot of heat in the dropping
transistor.

Dan
www.schematicsforfree.com

 

[Ed]

1. If you only change the frequency without changing the duty cycle, the
fan speed will stay the same. Duty cycle will vary a little with this
circuit but maybe not enough for your need.

Actually it works pretty well in my PC. I can't have the fans spin too
slow and I don't need full power to them either.

2. The 470uF cap is way too big. It acts as a dead short on the TIP115
transistor and will cause it to blow eventually. Actually, all you need
is something like 0.1 uFd as a noise suppressor and nothing else.

Not really worried too much about the PNP blowing with such low power,
if anything I'll put a diode coming off the PNP to the CAP and adjust
the 555 circuit to compensate for the voltage drop.

3. The thermistor will not blow.

Didn't think so. ;p

4. The idea of using the thermistor to control the base of a transistor
is fine if you want to generate a lot of heat in the dropping
transistor.

Dan
www.schematicsforfree.com

Thanks,
Ed

 

[Ed]

Ed,

What bothers me is the AC signal to the fan-- I think it might work better
if you use the 555 to produce the duty cycle whose average is the voltage to
drive the fan, and then use lumped elements to average the duty cycle down
to DC.

How about a first stage that is a switcher like you have, (but lower power),
- take the output from the PNP and filter it to take away AC with L/C
circuit
- use a darlington emitter follower high power stage to drive the fan with
DC.

my 2 cents--steve

Sounds good and all, but I would need a schematic to follow you, I'm not
much of an electrician, just a hobbyist. ;p

Thanks,
Ed