I am trying to set up an optical trigger to measure fan speed on an oscilloscope for an automotive radiator fan. I have a QRB1134 Photo transistor see attached data sheet. I have found two references to a circuit that can be used with this. Both seam to be the same but both do not call out the capacitance for the polarized capacitor. See attached figures. I happen to have a 35V 100uF polarized capacitor. I was wondering if that would work or what size I should order.

Also one of the diagrams showe's 5 volts for power. I would like to know if they might run on 3.3V or 13.8V

 

It is an excellent idea to examine the signal with an oscilloscope. Try it first without the capacitor. Then you can observe the effect of a capacitor. Try 100nF to 1μF.

If you wish to deviate from the 5V supply as designed then modify the value of the resistors in proportion to the voltage change. In other words, double the resistance if you double the supply voltage.

 

My worry would be the dirt that is going to get blown over the sensor.
Magnets (two to keep it balanced) and hall sensor?

 

It is an excellent idea to examine the signal with an oscilloscope. Try it first without the capacitor. Then you can observe the effect of a capacitor. Try 100nF to 1μF.

If you wish to deviate from the 5V supply as designed then modify the value of the resistors in proportion to the voltage change. In other words, double the resistance if you double the supply voltage.

Thanks I might be able to give that a try tonight. Can I substitute regular ceramic capacitors for the polarized? I have an assortment of those.

 

My worry would be the dirt that is going to get blown over the sensor.
Magnets (two to keep it balanced) and hall sensor?

Good point but that wont be an issue here as it will not be installed on a car. This will only be used to characterize the fan as in getting speed vs PWM input frequency and also to balance the fan.

 

Thanks I might be able to give that a try tonight. Can I substitute regular ceramic capacitors for the polarized? I have an assortment of those.

Yes, ceramic capacitors will be ok.

 

Results are in. The circuit worked best without any capacitor. I did find that I needed to AC couple the signal.

The trigger installation is shown in the first picture.

The DC coupled wave form is in the second picture.

The AC coupled wave form is shown in the third picture.

The fourth picture shows the fan running.

The capacitors I tried are shown in the last picture. Using the capacitors attenuated the peak such that the leading and trailing edges of the pulse were nearly equal and it will be difficult to get an accurate trigger point required for balancing.

Tomorrow I will put an accelerometer on and try to balance.

Thanks to all who contributed.

 

I have never used a capacitor across a photo-transistor sensor, not sure what it does for you. I do recommend a wider white stripe to give a wider pulse. That is a cheap way to improve operation and reliability.

 

I have never used a capacitor across a photo-transistor sensor, not sure what it does for you. I do recommend a wider white stripe to give a wider pulse. That is a cheap way to improve operation and reliability.

I have no idea what the cap is for. It was just in the only two wiring diagrams I could find. As far as reliability this needs to only run for about 10 minutes for me to balance the fan. Thanks

 

OK, so you will use the pulse to trigger a strobe flash to observe the heavy spot. It still makes sense to use a wider stripe.

 

OK, so you will use the pulse to trigger a strobe flash to observe the heavy spot. It still makes sense to use a wider stripe.

I will be using an accelerator. Pulse will be used for phasing

 

Using an accelerometer? Then there will be some additional electronics. And still you will need a strobe to see where the heavy spot is. That is old technology and so there should be a lot of circuits published.

 

Using an accelerometer? Then there will be some additional electronics. And still you will need a strobe to see where the heavy spot is. That is old technology and so there should be a lot of circuits published.

Absolutely I have a PCB 480D06 power unit. No need for a strobe, I've been doing it for 25 years this way on helicopter dynamic systems. High spot is found by taking the phase angle from the trigger to the peak in the accelerometer. Using an ossiliscope I will need to convert time to phase. So I will have a little math to do. But at the moment I'm having difficulty getting the accelerator to work. All used stuff I bought on e-bay so who knows. Not exactly the best accelerator for a 20 to 40 Hz measurement. Its a 10KHz accelerometer but the specs say good from 0 to 10KHz.

 

I have watched the balancing process in a balance shop, that was interesting, and I have designed the interface to scale the pulsing to feed the balance computer. That was interesting because the encoder was 3600 PPR while the machine required 100PPR, both with phase quadrature. The mechanical guy selected the encoder before I got involved. The high pulse count was for a timing check, the thing was an engine test stand. And I was the recent hire, replacing an engineer who got fired for cheating on his hours. So no line-up conversation about anything. Keeping track of phase quadrature while dividing by 36 is a challenge with discreet logic ICs, by the way.