Protecting AC motor on a Bandsaw from stall with Uno + IR-08H

My problem is a 1/2HP, 240VAC TEFC Cap Start/Cap Run bandsaw motor that is prone to stalling and hence burnout. I know from measuring, that it stalls instantly if revs drops below 1250 rpm (I'm on 50Hz). I figure I could use an IR-08H sensor mounted radially on the fan shroud, scanning the fins of the cooling fan, measuring rpm, and if it drops below 1250 rpm to activate a relay to cut power to the motor. The fan is 4.5" (115mm) diameter, plastic, with 12 radial fins, each ~ 0.080" (2mm) thick. The sensor could be mounted almost touching the fins if need be, though I suspect there's some minimum distance. A few questions: 1. is a 2mm thick fin wide enough to trigger the IR-08H sensor reliably (if not how wide does the reflective surface need to be?) 2. what other circuit or program elements will I need apart from the relay (something to allow the motor to start up I suppose, what else?) Can someone talk me through the logic? This is my first project with Arduino, any help appreciated - jv

I expect when the motor is straining, it's pulling more current. You might see if you can use a current sensor to identify when it's working too hard and then you won't have to mess with the fan.

For safety, I'd rather have something that presses the stop button on the saw, electrically perhaps but preferably physically. Then you don't have to modify the saw and turning it back on is just the normal power up operation.

Its the motor revs that’s critical to me, not motor current which is just a proxy. I know the motor will stall if it hits 1250rpm, I have no idea of the current at that point or even whether its the same each time it stalls. Counting fan blades and tripping a relay is positive and would seem simple for an arduino so I’d like to pursue this.
Messing with the fan is simple for me by comparison with the arduino side of things hence the questions.
Thx anyway - jv

wildbill: I expect when the motor is straining, it's pulling more current. You might see if you can use a current sensor to identify when it's working too hard and then you won't have to mess with the fan.

Optically sensing a fan is less messing than adding a current sensor to a mains supply perhaps?

You want an IR LED and IR photodiode sensor to sense the fan blades, not an obstacle avoidance sensor which is far too slow. Standard IR beam-break sensing works on microsecond timescales and will have no problem seeing the turning blades at 1500rpm. The 2mm fins will take several 100µs to pass by every 3ms or so.

You need someway (code) to prevent cutout during spin-up too.

Aha. Thanks Mark.

IR LED on one side of the fan housing and photodiode 180deg on other side
The IR LED would be high all the time.
The sensor photodiode sends a pulse when activated.
Arduino monitors pulses per second(?) (at 1250rpm there would be 20.833revs/sec*12fins = 250 pulses/sec
If it falls below that it activates a NC relay cutting motor power.

Dont know if I’ve visualised the ‘starting up’ scenario properly:
Presumably the Arduino has to ‘know’ the bandsaw has been started when saw is switched ‘on’. There’s only SPST switch on the bandsaw now, but if was replaced with double pole switch, then that could give the Arduino the startup signal.
If it takes the motor 5 sec to spin up, then put in a delay of 5000ms after startup before counting pulses/sec.
Could set a 500ms or 1000ms delay after that to loop through the ‘pulse counting’ every 1/2 sec or 1 sec which would be more than enough to protect the motor.

Then it has to reset after motor power is cut, either by the off switch at end of cut (most of time), or motor revs dropping to <1250rpm (protection circuit activated).
-If running normally the on/off power switch will be snapped off by the action of the saw completing the cut. and the Arduino would be cut off too, and it could start up again as above.

  • if protection circuit activates then the saw on/off switch will not be opened and the Arduino would still have power until it was opened by hand to reset.

Is this viable? Anything I’ve not thought of safety- or any other-wise?- jv

With all due respects, I think you are looking at this problem from the wrong end. If the motor stalls at 1250 rather than attempting to run at lower speeds then :- a) you are overloading the saw and hence causing the problem or b) your run capacitor may be faulty causing a loss of torque and permitting the motor to stall.

Looking for a protection device to overcome a frequent overload situation is not good engineering practice.

That said then measuring motor load current rather than motor speed is the correct parameter to monitor.

I'm sorry, I don't agree that motor current is the right parameter, no matter industrial practice; reasons below. The way all horizontal pivoted bandsaws work, as the angle of the head to the horizontal decreases, the cutting load increases, and the mechanism for adjusting load on 4x6, 7x9 and 7x12 bandsaws is awkward and slow. If, like I do, you constantly change size and type of material and blade TPI and speed, then you cannot always predict whether particular settings will stall under load or whether, despite adequate load settings, the blade will jam (particularly in aluminium where the swarf often welds to the wall of the cut, jamming the blade). As 1400rpm induction motors slow down under load, at about 1100rpm the centrifugal switch allows the start windings to kick back in and motor current rockets to ~10x std. I've measured my motor and it will not remain running under load at less than 1250rpm, i.e. it will stall if the revs hit this point. I believe this is the design load/revs point at which it stalls so I don't believe my run capacitor is failing (though I will check thx!). I want to switch it off before the motor current soars and starts degrading the windings. In industrial practice they're willing to accept failure and replacement in exchange for fit-and-forget heat sensors in the windings. Burnt out motors are one of the most common failures on 4x6 bandsaws, I can't afford it to fail, not to mention the fire danger. Hence my reluctance to accept motor current as the trigger for action. There are 100's of 1000's of these cheap 4x6 bandsaws(~$2-400) in workshops around the world, and none have overload protection on the motor, despite the high frequency of failure. Arduino rotation sensing is a cheap and foolproof way to protect them. Once I've figured it out the other members of our group ( will be able to do it too - jv

I also have one of the famous (or infamous) 4x6 bandsaws. It was a typical not-expensive model and I've had it for around 20 years and cut all manners of materials from solid 4" diameter aluminium billets to 4" solid steel billets. Yes cuts of that size can take a very long time but still quicker than by hand. I've never had a motor failure, never had a motor jam and never had aluminium 'welding' to the blade teeth. In all that time I've used the same tooth-count blade for all materials.

The motor sound never changes whether cutting or running free so I can only assume that for some reason you are experiencing untoward motor loading.

What I have found is that cheap carbon steel blades are as much use as a chocolate fireguard. Decent (expensive) blades from the likes of are worth every penny spent. Consistent straight cuts and long life.

Edit: I've just looked at the forum you referred to and notice your motor is rated to run at 1360RPM. Single phase motors for 50Hz are normally rated at 1425RPM or so when fully loaded to their design load. I therefore suggest that the cause of your problem lies in the quality of the motor. A 0.5HP motor running on 50Hz should run at 1425. Your motor must be underpowered and not developing its rated torque until it has slowed to the point where the load current balances the drop in speed. ie windings are being 'pushed' beyond good design practice. Source a motor rated at 1425RPM and your problems might just be resolved. On a side note my 'long life' unit (built 1992) is rated at 1/3HP

I was ready to give a similar posting lst night, but the forum server died.

I also have had one of those saws for about 20 years and it was well used when I bought it. Also had a used saw at my business until it closed. All have the original motor.

Yes, I have stalled the motor until I had time to turn the power off. Never any damage to the motor.

Stalling was caused by:

  1. pinching the saw when the block of metal moved in the vice.
  2. saw blade was worn more on one side than the other and was cutting crooked.
  3. saw slipped from the alignment wheels and twisted.

I always use a tube of solid wax on the saw blade at home to lube the cut if it is a long one. Got a used tube of was years ago and still have some left. That stops the welding of aluminum bits in the cut. I never made such big cuts with the business saw.

And like the previous post, I stay away from the cheap saw blades. The teeth wear unevenly and wear quickly.


Thanks for the comments guys. My motor is an Italian Cap start/Cap run motor which from the spec sheets are designed to get to rated horsepower at slightly lower rpm than ordinary Cap start induction motors so I don't think there's anything wrong with it. That said I think I'll look closer at its performance, maybe swap it out for another and see if there's any difference. I'm trying to push the envelope to find the limits of the saw in a methodical way, so I'm purposely running nearer to stalling the motor than most people. I couldn't do without my saw in the workshop (after the lathe and mill), its always being used, hence need for reliability and stall protection.

John_vreede: My motor is an Italian Cap start/Cap run motor which from the spec sheets are designed to get to rated horsepower at slightly lower rpm than ordinary Cap start induction motors so I don't think there's anything wrong with it.

That, I surmise, is exactly what is wrong with it. It has so little magnetic torque capability (lack of copper windings) that it can only achieve the specified horsepower when the slippage between sync speed and running speed is so great that it requires excess current to achieve its quoted HP output.

Small induction motors are generally fairly inefficient - most of the current is magnetization current I think, and single-phase powered motors are worse than three-phase too.