Syringe Pump

You might want to consider an optical encoder...

http://www.opticalencoder.com/

I'm sorry to say so, but almost nothing makes sense here.

An electrical failure can be detected by check of the motor input currents but you can have a mechanical failure and although electrical signals supply the motor it is not any more active.

For this situation the best is to check the reaction force. When the motor turns the nut and pushes the piston it is under a reaction force equal to the force the piston is pushed and which can be measured. It is a better solution since you do not need a sensor compatible with fluid you inject.

Is it a device for medicine controlled delivery? If yes I can understand the need for a 100% functional safety. It makes sens. It would be good to have a logic which in case of function stop gives a signal to ask for help and intervention.

Several years ago I was asked to analyse such a device and suggest a solution to measure the pressure I came to the force measurement which was accepted and worked fine.

I'm with Solar Eagle.
Simpler to just check that the motor is actually rotating when commanded to do so.
Del

1 micron = 39.37x10E-6 inches ... .000039 inches.

Are you sure this is the right size?

And another thing, a short travel won't build up the pressure and thus won't detect the stalled motor.
I'd suggest a motor with a nice big easilly detectable step size which is then geared down, it's then very easy to check the motor is moving.
Del

Something's not right here. He's driviing a 6 micron pump with a 6 mm shaft??????

Yeah 'cos he doesn't want it to bend under the pressure when the motor stalls
Del

The metric thread M6 has a pitch of 1mm. The motor makes 1.8°/step thus the piston makes 1/(360/1.8)= 5E-3 mm= 5 µm.

As for the pressure it depends on the fluid viscosity and on the number of steps /s.

For patients in hospitals who need a continuous supply of medicine in strictly controlled amounts there are devices which work as the one described by the OP. A programmable controller sets the rate of steps /s or /minute and also how many steps for a "supply" and of course in which intervals the medicine is given to the patient.

Any way the force measurement even if you do not accept it (as it comes from the comments) is implemented in some devices and works very fine as feed back value since the force as magnitude is a lot easier to measure as the pressure and as I mentioned the sensor is not in contact with the pumped fluid which reduces the constrains for compatibility and thus its cost.

Probably means 6 microns per step, half-step or microstep.

Pressure sensor is unlikely to work well, better to do it electronically. There are stepper motor contollers available that feature stall detection, either through current/cycle sensing or back EMF sensing. Google should pull something up.

You may well be right on the 6 microns per step, and I agree on the use of electronic stall detection.

On the other hand, the OP describes a syringe pump, in which case the important value is volume of fluid delivered. The volume delivered per step will depend greatly on the diameter of the syringe. I suspect that he/she meant microliter, not micron.

The one machine that I helped build to deliver microliters of fluid deposited the fluid onto microscope slides at atmospheric pressure, so there was virtually no pressure developed (only enough to overcome capillary tension). If his machine is at all similar, no ordinary pressure sensor will detect a stall.

Under these conditions, an overcurrent or undervoltage detector would have to be extremely sensitive, probably leading to false alarms. I agree with Lyn and Tornado: the best way is to use an optical encoder to detect motion of the stepper.

I agree with your analysis that it is probably a volumetric syringe pump. However, the medical syringe pumps are designed to alarm on an outlet obstruction, when the delivery pressure rises. This is long before the motor stalls, and I therefore suggest that the best sensor is indeed a pressure sensor, positioned on the arm which pushes the plunger.

Too bad the OP hasn't clarified anything. But then his post is so confused that it scarcely admits of clarification. (Sorry again, but that's just the way it is.)

Rigged one in the early '70s while doing research.

1. Used a pulse counter and counted # of pulses needed to move or equal 1 stepper motor displacement angle as one of my inputs.

2. Using interruptions on a series of differently angled opti-coupler devices, equate that to monitor syringe's linear displacement as my second input.

3. monitor stepper motor current drawn as the 3rd input and fed all to an AND /NAND gates the of which are used as input for alarm and other trigger functions.

Crude yet serve our purpose!