Continue to Site

Welcome to EDAboard.com

Welcome to our site! EDAboard.com is an international Electronics Discussion Forum focused on EDA software, circuits, schematics, books, theory, papers, asic, pld, 8051, DSP, Network, RF, Analog Design, PCB, Service Manuals... and a whole lot more! To participate you need to register. Registration is free. Click here to register now.

circuit of fan speed control

akbarza

Full Member level 2
Joined
Dec 18, 2013
Messages
131
Helped
0
Reputation
0
Reaction score
0
Trophy points
1,296
Activity points
2,556
hi
I saw a circuit for fan speed control in:
https://www.nutsvolts.com/magazine/...nzg&utm_content=279229311&utm_source=hs_email
I attached a pic of the circuit.
my question:
1:for what R4 resistor and diode D1 are used?
2: why R1 was used?
3) what is the role of R12 between the output of D opamp and the base of the Q2 transistor?
4) according to text of the above address, U1:c creates a Schmitt-trigger. how does it do it?
thanks
Screenshot (9).png
 
1) LM324 has limited VOH (VCC-2Vbe @ no load)
so R4, R5, D1 "ensure" that Q1 can be turned all the way
off. As long as things don't get too hot in Q1 anyhow.

2) Thermistor is a resistor and all by itself would just put
12V resistor limited on the temp sense line. R1 scales,
puts midrange and "makes live".

3) R12 makes Q2 base current / U1 supply and :D output
current reasonable (better than a 40mA short circuit
current limit will...)

4) positive feedback, (R10||R11)/R9 - same as any Schmitt,
only the details vary. A nice high impedance input and a
hidden (from that input's I-V) positive feedback of lesser
authority.

These are all pretty basic and not sure giving you easy
answers is helping, or just encouraging you not to think.
There are many op amp circuit design and appllication
papers and books out there (check archive.org for e-books)
that you might find entertaining if you like this stuff.
 
  • This design powers a typical 1.2W fan @ 12V which is equivalent to 100 Ohms.
    • Thus the base R is usually chosen for 10x the load when the drop voltages are near the same to ensure rated saturation and provide a controlled base current limiting.
  • It uses a std. 10K NTC thermistor is sensed by two high Z inputs.
  • R13 is not used to U1B as a 5V unity gain amplifier.
    • Good 12V fans startup up at 5V to operate fan at low noise
    • Since the PNP collector inverts, Vin+ must be used for negative feedback.
  • U1A uses the linear gain to increase the fan voltage to full power from 33'C to 47'C per the voltage table.
  • The OpAmps then share the PNP to drive low as wired NORs = wired ANDs from de Morgan's Theorem. So U1B drives the PNP warming up then U1A takes over as it warms up 47'C at full power and a higher threshold temp is set for overtemp Fault out.
  • Since Fault Out drives a higher Z, the base resistor can be raised and still have a relatively low output impedance of 10k / hFE for noise immunity.
 
Last edited:
1698098057667.png


D1 D2 Q1 FAN basically an OR gate action

off off off off
on off on on
off on on on
on on on on

R4, R6 limit base current of Q1 so it is not over driven


Regards, Dana.
 
Positive feedback creates the hysteresis roughly by resistance ratios.

Do you know how to compute the sum of all currents on each resistor to determine the voltage hysteresis thresholds for U1C saturated states Voh, Vol?

Try it and let us know.
 
From point of view OpAmp outputs, basic NAND gate action

U1A U1B Q1 FAN
0 0 on on
0 1 on on
1 0 on on
1 1 off off
 
From point of view OpAmp outputs, basic NAND gate action

U1A U1B Q1 FAN
0 0 on on (does not exist)
0 1 on on 12 Vout
linear speed control in between
1 0 on on 5V out
1 1 off off (power off)

I understand your intent for "analogic" but only the two middle logic states exist with a linear temp-controlled fan voltage in between the two thresholds for U1B and U1A for Vfan from 5 to 12V.

- U1A is inverting linear operation between 7kohm to 4kohm while its output ranges from 11V(sat) to 10V (linear)
- then 0V (sat) when the PNP cannot pull-up any higher than 12V.
- U1B is non-inverting unity gain with Vol = "0" , "1' = 10V to 11V to drive Q1-b to regulate Fan @ 5V On for any temperature below 33'C

- When the temperature exceeds 33'C U1B goes out of regulation and saturates high and U1A takes over. (low man wins diode logic to regulate the fan.)

U1A varies between linearly inversely with temp between 7kohm to 4kohm =33'C and 47'C
for a std. 10k NTC thermistor.
 
linear speed control in between


I understand your intent for "analogic" but only the two middle logic states exist with a linear temp-controlled fan voltage in between the two thresholds for U1B and U1A for Vfan from 5 to 12V.

- U1A is inverting linear operation between 7kohm to 4kohm while its output ranges from 11V(sat) to 10V (linear)
- then 0V (sat) when the PNP cannot pull-up any higher than 12V.
- U1B is non-inverting unity gain with Vol = "0" , "1' = 10V to 11V to drive Q1-b to regulate Fan @ 5V On for any temperature below 33'C

- When the temperature exceeds 33'C U1B goes out of regulation and saturates high and U1A takes over. (low man wins diode logic to regulate the fan.)

U1A varies between linearly inversely with temp between 7kohm to 4kohm =33'C and 47'C
for a std. 10k NTC thermistor.
I should have qualified what I was posting, I fully realize this is a speed control system and
there is fdbk involved. I was responding to the "logic" conclusion posted in an earlier post
and lost my sanity in the process, eg, need to focus on poster and actual issues. Not to
create confusion with irrelevant dataum.

I will reform.....I will be good______to be written longhand on chalkboard 100 times......

Regards, Dana.
 
Hi,

Code:
OR:

A B Q
0 0 0
0 1 1
1 0 1
1 1 1

here

diodes here
A B Q
0 0 0
0 1 0
1 0 0
1 1 1  == AND

diodes to out
A B Q
0 0 1
0 1 1
1 0 1
1 1 0 == NAND

You get a WIRED OR by:
* using two diodes,
* connect both cathodes as the output.
* Pull DOWN the output.
* Anodes are the input.
 
From point of view OpAmp outputs, basic NAND gate action

U1A U1B Q1 FAN
0 0 on on
0 1 on on
1 0 on on
1 1 off off
hi
--- Updated ---

Positive feedback creates the hysteresis roughly by resistance ratios.

Do you know how to compute the sum of all currents on each resistor to determine the voltage hysteresis thresholds for U1C saturated states Voh, Vol?

Try it and let us know.
hi can you guide me how do i it?
--- Updated ---

1) LM324 has limited VOH (VCC-2Vbe @ no load)
so R4, R5, D1 "ensure" that Q1 can be turned all the way
off. As long as things don't get too hot in Q1 anyhow.

2) Thermistor is a resistor and all by itself would just put
12V resistor limited on the temp sense line. R1 scales,
puts midrange and "makes live".

3) R12 makes Q2 base current / U1 supply and :D output
current reasonable (better than a 40mA short circuit
current limit will...)

4) positive feedback, (R10||R11)/R9 - same as any Schmitt,
only the details vary. A nice high impedance input and a
hidden (from that input's I-V) positive feedback of lesser
authority.

These are all pretty basic and not sure giving you easy
answers is helping, or just encouraging you not to think.
There are many op amp circuit design and appllication
papers and books out there (check archive.org for e-books)
that you might find entertaining if you like this stuff.
thanks for reply
Can you explain more about the R5 role?
--- Updated ---

1) LM324 has limited VOH (VCC-2Vbe @ no load)
so R4, R5, D1 "ensure" that Q1 can be turned all the way
off. As long as things don't get too hot in Q1 anyhow.

2) Thermistor is a resistor and all by itself would just put
12V resistor limited on the temp sense line. R1 scales,
puts midrange and "makes live".

3) R12 makes Q2 base current / U1 supply and :D output
current reasonable (better than a 40mA short circuit
current limit will...)

4) positive feedback, (R10||R11)/R9 - same as any Schmitt,
only the details vary. A nice high impedance input and a
hidden (from that input's I-V) positive feedback of lesser
authority.

These are all pretty basic and not sure giving you easy
answers is helping, or just encouraging you not to think.
There are many op amp circuit design and appllication
papers and books out there (check archive.org for e-books)
that you might find entertaining if you like this stuff.
hi
you had said that:
LM324 has limited VOH (VCC-2Vbe @ no load)
so R4, R5, D1 "ensure" that Q1 can be turned all the way
off
can give more about or introduce reference or site about it?
thanks
 
Last edited:
hi can you guide me how do i it?
I'd like to see you try 1st using what you know.
Let me use the datasheet to assume the LM324 output with 150K load goes from 10.6 to 0.1 with positive feedback to each rail.

Imagine the switch represents two two states with hystereis where the thermistor voltage input crosses these thresholds.

What are the upper and lower thresholds?

1698283893979.png
 
Last edited:
I'd like to see you try 1st using what you know.
Let me use the datasheet to assume the LM324 output with 150K load goes from 10.6 to 0.1 with positive feedback to each rail.

Imagine the switch represents two two states with hystereis where the thermistor voltage input crosses these thresholds.

What are the upper and lower thresholds?

View attachment 185764
hi and thanks for the reply
I am a little familiar with electronics (analog). but I can not understand your question completely.
in the above figure, if I suppose the 150k does not sink any current, then I think this circuit is in the middle of the hysteresis loop.
if my question is time-consuming for you, plz ignore to reply
 
This design uses all basic configurations of comparators and negative feedback using Vin+ with a driver that inverts to Vin+ making it a linear negative feedback circuit.

Final analysis
  1. All together this design with a std. 10k NTC thermistor runs the fan at half the supply voltage set by R13=4k for a 12V fan at power on from U1B.
  2. U1B sets the start range for linear speed control by switching U1B hi = diode OFF in the NAND logic using dual diodes either being low to pull up the fan.
  3. It also uses negative feedback(NFB) to set the fan voltage at 50% for a low speed below this setpoint.
  4. Don't be confused by the R7 feedback to Vin+ because the collector inverts the base voltage drive making this loop NFB for threshold.
  5. This is how the dual diode OR or NAND logic is used to switch from static low fan speed below 35'C and variable to full fan speed at 55'C.
  6. The Fault logic makes no sense to me which says fault=low if the temperature is above 25'C. Of course if the fan was on below this temperature, (which it is at V+/2 ) that seems like a design fault to me. I would have disabled the fan.
1707421785099.png


If you want to play with the simulation, the slider for the Thermistor I set to 25 to 56'C
http://tinyurl.com/2y9lovn3 can be changed to any temp. range you wish.

Rev B with wider slider range and emitter current viewed.
 
Last edited:

LaTeX Commands Quick-Menu:

Similar threads

Part and Inventory Search

Welcome to EDABoard.com

Sponsor

Back
Top