Need Fast Pulsed Current for MEMS Gas Sensor

Thread Starter

jhart2020

Joined Dec 18, 2019
1
Hello,

I am attempting to put together a circuit for a MEMS gas sensor which requires a short pulse (2 ms) of 3.0 V followed by <998 ms of no voltage. I would normally try and program this through the micro controller, but 2 ms is too short a time for me to do so. The circuit diagram for this sensor indicates using a transistor to accomplish this, but I haven't the first clue how to go about implementing that. I've posted the circuit diagram below. What would be advisable as a solution? Thanks in advance.

1576649066556.png
 

crutschow

Joined Mar 14, 2008
25,271
You could easily generate the pulse with a 555 timer IC connected as an astable oscillator.

What is the value of Vc, Rs, and RL?
 

pmd34

Joined Feb 22, 2014
507
There are also the MIC1555 and 1557 ICs as well as the 74HC221. However your comment about Micro-controllers being too slow seems a bit strange, as they typically run at >1MHz... for example the typcial ATmega ICs also used on the arduino boards run at 8MHz or greater so 1 timer pulse is 1/8th of a micro second.. you would actually have to use the divide down counters to get this clock speed slow enough.
 

TeeKay6

Joined Apr 20, 2019
572
There are also the MIC1555 and 1557 ICs as well as the 74HC221. However your comment about Micro-controllers being too slow seems a bit strange, as they typically run at >1MHz... for example the typcial ATmega ICs also used on the arduino boards run at 8MHz or greater so 1 timer pulse is 1/8th of a micro second.. you would actually have to use the divide down counters to get this clock speed slow enough.
I agree with @pmd34, any microcontroller can easily produce 2ms width pulses, at a 1 sec rate.
 

Reloadron

Joined Jan 15, 2015
5,621
A uC (micro-controller) should have no problem driving your transistor with a 1.0 Hz rate and a 2% duty cycle. Additionally a 555 timer circuit (or others) can be used as well. Just a matter of how you wish to go about it.

For reasons I don't understand all it looks like you are doing is turning the output of your sensor On and Off or On for 2 mSec and Off for 998 mSec. The value of Rs in your drawing is not a constant and will change based on the amount of gas concentration sensed. As Rs changes so goes the current from Vcc source voltage. Rs and Rl form a voltage divider circuit between Vcc and Ground where the Vout is going to be proportional to the sensed gas concentration. When a known test sample of gas is used (calibrated gas sample) Rs will provide a Vout proportional to the PPM (Parts Per Million) of your calibrated sample. Keep in mind a known calibrated standard is needed with these sensors to get any accurate results.

Ron
 

BobaMosfet

Joined Jul 1, 2009
1,194
Hello,

I am attempting to put together a circuit for a MEMS gas sensor which requires a short pulse (2 ms) of 3.0 V followed by <998 ms of no voltage. I would normally try and program this through the micro controller, but 2 ms is too short a time for me to do so. The circuit diagram for this sensor indicates using a transistor to accomplish this, but I haven't the first clue how to go about implementing that. I've posted the circuit diagram below. What would be advisable as a solution? Thanks in advance.

View attachment 194769
Stop. Make sure of your pulse length, and then do the math. How fast is your MCU, and which one are you using? Even a 16MHz MCU can easily provide a stable 1ms clock with plenty of time for housekeeping and other tasks, including more 1ms clocks for other tasks. Because you're saying you need a pulse that is two-one-thousandths of a second long, followed by the remainder of that second (nine-hundred-ninety-eight-one-thousandths of a second).

1/16000000 = 0.0000000625 seconds. or 62.5ns per clock pulse, which means 8,000-16,000 assembly language instructions can be executed, minimum, in just 1ms).
 

joeyd999

Joined Jun 6, 2011
4,425
A uC (micro-controller) should have no problem driving your transistor with a 1.0 Hz rate and a 2% duty cycle. Additionally a 555 timer circuit (or others) can be used as well. Just a matter of how you wish to go about it.

For reasons I don't understand all it looks like you are doing is turning the output of your sensor On and Off or On for 2 mSec and Off for 998 mSec. The value of Rs in your drawing is not a constant and will change based on the amount of gas concentration sensed. As Rs changes so goes the current from Vcc source voltage. Rs and Rl form a voltage divider circuit between Vcc and Ground where the Vout is going to be proportional to the sensed gas concentration. When a known test sample of gas is used (calibrated gas sample) Rs will provide a Vout proportional to the PPM (Parts Per Million) of your calibrated sample. Keep in mind a known calibrated standard is needed with these sensors to get any accurate results.

Ron
In addition, Vce,sat going to introduce an error to the output voltage.
 

Reloadron

Joined Jan 15, 2015
5,621
In addition, Vce,sat going to introduce an error to the output voltage.
Yes, should have mentioned that too. I just don't understand why sample the module's output for 2.0 mSec? Then wait 998 mSec for another sample? Yes, the transistor will have the Collector / Emitter voltage drop in there too, good point to bring up.

Ron
 
Top