Fan speed

Thread Starter

alitronic

Joined Jun 13, 2020
50
Good morning, I have found this schematics in a medical ventilator. Can you explain me how it works. Thank you very much.
1632749346587.png
 

MisterBill2

Joined Jan 23, 2018
18,179
This is an analog tachometer circuit. The Function is as follows: The input signal, an unknown waveform, is coupled into the gate of the FET amplifier transistor. The amplified signal feeds a schmidt inverter and is squared to a constant amplitude waveform. The positive portion is passed on by the diode to a voltage divider and a filter, where it is smoothed to a DC level that feeds a non-inverting op-amp set for some gain. The output is a DC level proportional to the fan speed. The circuit shown does not include any power supply nor control elements.
 

AnalogKid

Joined Aug 1, 2013
10,989
The image is too small and very fuzzy, so all of my reference designators might be wrong.

The circuit produces a variable DC output that is directly proportional to fan speed, with relatively low tach-signal ripple on it.

C816 and R817 differentiate the incoming tach signal, and produce a positive-going pseudo pulse at the Q805 gate that is a constant width for all fan speeds, with the time between pulses changing as a function of fan speed; faster speed, shorter time. The amplitude of these pulses are whatever is coming out of the fan. Because the FET threshold voltage is a constant but the tack pulse amplitude can vary with the fan's supply voltage, the resulting pulse width varies with fan voltage. For a 12 V fan it is approx. 100 us. Note - pulses from a 24 V fan will damage the FET.

Q805 is the first stage to "square-up" the signal at its gate, and inverts the signal. U804 re-inverts the signal and finishes the shaping job, producing nice clean constant-amplitude pulses. These are envelope-detected by D400 and C813. C813 is topped off during each pulse, and discharges during the time between pulses; the slower the fan, the more it discharges. This creates a signal with an average value that is directly proportional to the fan speed, but with a lot of ripple. The following five components are an additional 2-pole lowpass filter to remove much of the ripple, but also attenuate the signal. The output opamp has a gain of 2.4 to restore the signal to a useful amplitude for whatever is downstream.

ak
 

Thread Starter

alitronic

Joined Jun 13, 2020
50
This is an analog tachometer circuit. The Function is as follows: The input signal, an unknown waveform, is coupled into the gate of the FET amplifier transistor. The amplified signal feeds a schmidt inverter and is squared to a constant amplitude waveform. The positive portion is passed on by the diode to a voltage divider and a filter, where it is smoothed to a DC level that feeds a non-inverting op-amp set for some gain. The output is a DC level proportional to the fan speed. The circuit shown does not include any power supply nor control elements.
Thank you very much
 

Thread Starter

alitronic

Joined Jun 13, 2020
50
The image is too small and very fuzzy, so all of my reference designators might be wrong.

The circuit produces a variable DC output that is directly proportional to fan speed, with relatively low tach-signal ripple on it.

C816 and R817 differentiate the incoming tach signal, and produce a positive-going pseudo pulse at the Q805 gate that is a constant width for all fan speeds, with the time between pulses changing as a function of fan speed; faster speed, shorter time. The amplitude of these pulses are whatever is coming out of the fan. Because the FET threshold voltage is a constant but the tack pulse amplitude can vary with the fan's supply voltage, the resulting pulse width varies with fan voltage. For a 12 V fan it is approx. 100 us. Note - pulses from a 24 V fan will damage the FET.

Q805 is the first stage to "square-up" the signal at its gate, and inverts the signal. U804 re-inverts the signal and finishes the shaping job, producing nice clean constant-amplitude pulses. These are envelope-detected by D400 and C813. C813 is topped off during each pulse, and discharges during the time between pulses; the slower the fan, the more it discharges. This creates a signal with an average value that is directly proportional to the fan speed, but with a lot of ripple. The following five components are an additional 2-pole lowpass filter to remove much of the ripple, but also attenuate the signal. The output opamp has a gain of 2.4 to restore the signal to a useful amplitude for whatever is downstream.

ak
Thank you for your explanation
 
Top