I would like some ideas for a self powered hotwire sensor or even using a easily available automotive mass flow sensor if viable. We currently are testing a unit that ranges from 45cfm to 110cfm through a 6" smooth metal duct.

I am working on a solar heat project where Eventually I will use a controller to set flow for best output.

Initially using this blower, Nidec Beta V TA600DC 24 VDC .82 AMP A34457-10 FAN 172MM, and a pwm circuit such as the one attached. The regulator is LM217, opamp is LM324 and the drive for the fan will be an IRFZ34N with Sr56 as D1. I would use a 11pos stepped attenuator at VR1 to get feedback from RPM out to build a flow table.

I have a brand new VW/Audi massflow sensor but the schematic in service manual doesn't explain how it works.

 

There are some ideas in this thread.

 

I recently built an anemometer using thermistors in place of the hot wire element. Thermistors have the advantage of being readily available and relatively rugged. The circuit was pretty simple, Op Amp constant current source driving the thermistors in a bridge configuration to achieve some temperature stability and an instrumentation amp on the bridge output. I can upload the circuit diagram if you are interested.

 

I recently built an anemometer using thermistors in place of the hot wire element. Thermistors have the advantage of being readily available and relatively rugged. The circuit was pretty simple, Op Amp constant current source driving the thermistors in a bridge configuration to achieve some temperature stability and an instrumentation amp on the bridge output. I can upload the circuit diagram if you are interested.

I would, While I assume the mass flow sensor is quite accurate and sensitive I have no insight into the calibration. Thanks for your help.

 

we have a lot of exhaust stacks for ovess and paint booths here, the usual method is to use a differential pressure sensor, one tube into the flow, and one tube down wind from the flow. no heat means no ignition source.
also no rebalancing the sensors for a change in air flow temprature.

 

Here it is. You can leave out U3/R9 as these are just used to switch power on/off and you probably don't need the stuff around D1/LED1 as this was just to set the bridge a bit off balance (about 0.6V). The thermistors (TR1, TR2) are surface mount components and are about 1mm across, the small size gives low thermal mass and so they are pretty responsive.

On the PCB one track is shown on the top side, just use a wire link for this and you can create it single sided. The board is designed so that the part containing the thermistors can be cut off and separated by a cable.

As for calibration, one way is to use a venturi, you measure the differential pressure across a constriction, there is plenty of stuff on the web. Or just by a cheap anemometer, depending on what flow rates and accuracy you want.