[SOLVED] Slow start 220V single phase induction motor

[elflop]

I have a table saw which is driven by a 1.5HP single phase 220V induction motor. There are 2 capacitors in the motor's electrical connection box.
The motor drives the saw by a flat belt. The belt is 60 cm (2 foot) long and 2.5 cm (1 inch) wide.
I had to replace the original motor.
The new motor starts so quickly that the belt is immediately ejected from the flat pully. The old motor started nice and slowly and took about 5 seconds to reach maximum speed. I can't use any parts from the old motor.

I have made some experiments with an antique variac and have found that if I turn on the saw with about 115 volt it starts nice and slow. Because the variac became quite warm and smelly, I replaced it with an old movie transformer (220V -> 115V). This also works very nicely indeed but is much too large and heavy to be practical.

So now I need a lot of help. My plan is to use a timer to trigger a relay. The relay would initially connect the motor via the transformer and after about 3-5 seconds would connect the motor directly to 220V. I think I can manage the timer/relay part myself, but the transformer is my problem. I measure the current needed by the motor as at least 6A. Is it possible to use some other SIMPLE device/design instead of the transformer to regulate the initial speed of the motor?

Obviously, the less torque the better! I have googled a long time for this, and everyone says the problem with a triac is the torque - but I don't care about the torque, I just want the motor to start really nice and slowly until it gets up to speed when I can switch to direct 220V via the relay. And can I use a triac with an induction motor with 2 integrated capacitors?

I would really appreciate any help, or better still, a circuit diagram of how I can do the regulating part. Using a transformer is just too bulky (and expensive).

Thanks

 

[betwixt]

The real problem is that the electrical load presented by the motor makes a sudden change as it starts up. Even from stalled to just running there is a huge change in the current it draws. A further complication is that if you use a Triac in a conventional phase control system, it chops the power in mid cycle and creates high frequency steps in the waveform which in turn causes the capacitors to work in unintended ways.
It sounds like the original motor was less powerful than the new one and simply took longer to build up speed, possibly it was also designed to run slower as well. Changing to a motor similar to the old one is the real solution but if you want to do it electrically you could try one of two things: 1. a normal light dimmer rated at 2KW or more, you can get these for controlling floodlights or stage lighting or 2. Use a Triac with zero crossing detection and run it for say one on 10 cycles then afer a delay for 1 in 9 and so on until it runs on all cycles. This is a good electrical solution and you wouldn't need any relays but I'm not sure how stable the speed build-up would be, some experimentation would be needed.

Brian.

 

[elflop]

Thank you for this information. Option #2 is way over my head so I'll go for option #1. I've found a "**broken link removed**" that will take 1200VA for 3 mins. but stated elsewhere as 2760 VA with a heatsink, so I hope that's large enough. Like all the dimmers I can find, it says it's not suitable for "motors with starter capacitors", but I guess that's because of the torque, which I don't need if I use the relay idea. I would only use this unit for 3-5 seconds at most - just enough to get the motor spinning. Do you think this unit would work?

 

[FvM]

Using the motor with a smaller starter capacitance may be a way to reduce the initial torque. Also a resistor starter should be considered. Unfortunately, it well generate a lot of power dissipation, but it may be feasible for short time operation.

 

[RCinFLA]

There are soft starter modules but they are rather expensive.
PSR6-600-70 - ABB Control - SOFT STARTER, 6.8A, 600V, AC CTRL 100-240VAC CONTROL, 208-600VAC LOAD 1.5HP@240V, 3HP@480V

The two capacitors are run cap and start cap. You might try decreasing the start cap value.

Soft starter uses triacs like a dimmer controller with a ramp up of applied rms voltage to motor. For a 1.5 hp motor you need to have a triac capable of handling the 15-20 amp surge current during startup cycle. If you find a regular dimmer that can handle 20 amps you can use the control knob as a poor man's soft starter.

Until the motor reaches its run speed the current drawn is close to locked rotor current. Locked rotor current is a function of applied voltage.

Run current for a 220 vac 1.5 HP motor is between 6 and 7 amps. A triac has about 1.5 v drop across it so it will dissipate about 10 watts while motor is running. It would need a reasonable sized heat sink for that much power dissipation.

 

[jpanhalt]

How is your 220V wired? Is it like in the US, where you can split it into 2, 110V circuits with a grounded conductor? If so, why not start at 110V then use a voltage relay to switch to full 220V when the back emf reached the trigger point?

I did something like that for a single-phase to 3-phase converter.

John

 

[elflop]

@RCinFLA: Thanks for this info. Like you say, the soft starter modules are on the expensive side, especially if they wouldn't work. But I have noted them.

@jpanhalt: Our 220V are wired exactly like the US 115V, but with 230V. It is not wired like the US 220V.

@FvM: I think I should follow up on this, but I need some more help.
The capacitors are sealed in cans, but I have been able to open up the starter capacitor. It is marked: Motor Starter Capacitor", so I guess it's the right one. I don't see any polarity and it's bridged with a 15K resistor.
What size of capacitor should I experiment with - and what sort of capacitors are these? Should I also use a different resistor size?

Here is a scan of the only schematic I have for this motor (it's a bit big, but I had to make it big because the original is difficult to read)

Thank you all for your replies

---------- Post added at 22:06 ---------- Previous post was at 20:21 ----------

Since you are using 220V, then you also have half of that available for starting. Have you tried starting it on the "split" 220V?

John

John, we don't have a "split" 220V. 110V doesn't exist over here. We only have 230V, 380V and 500V.

 

[jpanhalt]

Sorry for the confusion. Right after I posted, I re-read your comment and realized I had misinterpreted it. I tried to delete my post, but apparently was too late. Sorry for the added confusion.

John

 

[RCinFLA]

If I interpret the diagram right, it looks like there is a phased winding that stays fed all the time. It looks like a centrifugal switch between V1 and V2 that just adds more capacitance during startup (adding 150 uF C1 in parallel with C2).

The resistor is just to bleed any charge on cap after startup switch opens. C1 may be a dual polarity electrolytic, meaning it can only take current for a short period of time (startup time). C2 is likely a oil filled sealed cap that can take continuous run current.

Since is goes from 300 uF plus 150 uF (450 uF) to just 300 uF for run, see what happens (startup wise) without C1 connected. If startup is too sluggish then try a lower capacitance value for C1.

 

[FvM]

A larger value (300 µF) for the permanently connected than the starter capacitor seems strange. I rather expect 30 µF.

I think, a first test can be, if the motor can possibly start with C1 disconnected.

 

[elflop]

@FvM and @RCinFLA: You are absolutely right FvM! A very bad typo on my part. The capacitor is 30µF. Sorry for the confusion. I've corrected the diagram.

And now - the great WOW-effect :-D
I disconnected C1 and applied 125V with my transformer (to be on the safe side). The motor just made a noise. Then I applied 230V directly - and the motor started nicely :-D:-D

The belt gets pushed to the edge of the pully and ovelaps the pully by a few millimeters as soon as the centrifugal switch kicks in*, but it centralizes quickly afterwards. The belt isn't ejected anymore! This is really great news, and I thank you both for this very simply (and cheap!) trick.

May I ask you a last question? Is this safe? Can I permanently remove the starter capacitor?

Thank you everyone who replied very much.

<Edit> * looking at the diagram, I don't think the centrifugal switch has any function with the starter capacitor removed, so it must just be the speed that forces the belt out a bit.

 

[RCinFLA]

In your application the motor is starting up with relatively light load. If this was a water pump or some other application that required higher starting torque it would not spin up.

I would still try to keep some starter capacitance if belt can take it. You should not let it take more then a couple of seconds to come up to full speed as this does have higher current flowing in run windings during this spin up period.

I don't really understand your description of belt being pushed to edge of pulley. A normal V-belt has cloth cord and should not stretch or expand much. If belt is good and reasonable tight on pulley maybe there can be a slight initial slippage 'squeel' but should not eject belt. Maybe you need a new belt.

Other possiblity is your motor mount or saw blade end mounts are weak. My father had a very old table saw that the motor was attached to a hinged platform that normally sat about 45 degrees to V-belt. Just the weight of the motor against the 45 deg hinge mount is what kept the belt tight. If you have something like this you might add an adjustable spring to put more downward force on motor keeping the belt tight. If the new motor weighs less then old motor and you have this type of mount arrangement I can see that causing a problem. The good part of the motor mount was if the saw blade got wedged by a piece of wood the motor would just flip over and slacken up the the belt removing drive to the saw blade.

 

[elflop]

@RCinFLA

Thank you very much for this explanation. Now I understand why the motor usually needs a large initial torque. Obviously with a saw blade there is almost no friction at all.

My table saw sounds similar to what your father had. The motor is mounted on a hinged platform that I can adjust. You can just see the adjusting wing-nut on the bottom of the picture, in the middle. My belt is flat, it is not a V-shaped belt. I realize that it may be a rarity, hence the picture below. If the saw blade gets stuck, the belt is ejected. Since this happens rarely, I don't know how this happens because the motor mounting table is fixed by the wing nut. Like on your father's saw, the pressure of the belt is the weight of the motor - plus a wee bit more.

With the capacitor removed, the saw reaches maximum speed in about 2 seconds at the most. The old motor took longer - I would say about 5 seconds.

Do you think that I really need to use a capacitor? I'm afraid that if the saw starts any faster it will eject the belt. I'm also afraid of putting too much pressure on the belt because I have a feeling that it could wear out the saw bearing. Since the saw manufacturer has been out of business for many years, it would be difficult to find spare parts. If you think that I really should use a capacitor, what value do you suggest I start with and should I keep the 15K resistor?

 

[FvM]

Thanks to RCinFLA for a detailed explanation about different starting torque requirements, the point I coveniently left for your trial.

Do you think that I really need to use a capacitor?

The important point for a single phase motor is to get a rotary field. This is basically achieved by the 30 uF operating capacitor. If the load torque characteristic allows, you don't necessarily need a starting capacitor. The 15k resistor is simply a "bleeding" resistor, discharging the capacitor to prevent an electrical shock that may occur, if the capcitor holds it's charge a long time after unpowering the motor. The value isn't particular critical, check the respective RC time constant. The resistor isn't required to operate the motor.

Instead of requiring a certain capacitor value, I suggest an effective overload protection by a thermal circuit breaker. It protects the motor, if ist possibly doesn't spin up and draws excessive current.

 

[RCinFLA]

Flat belt arrangement.... Just like early industrial age factory belts. Like a belt sander, the alignment is critical to keep belt from walking off the pulley.

Sounds like you have a satisfactory solution just leaving off extra starting cap. If it gives you more trouble I would think about replacing pulleys and belt with V-belt type.

 

[elflop]

@RCinFLA: You seem very knowledgeable about these belts and you information has been very useful to me.

My table saw is at most 30 years old, and, except for the old motor, works perfectly. I've been very happy with it in the past and it's never let me down.
I am, however, a bit worried about the flat belt, especially if it breaks. Like you say, alignment is critical and it's all but impossible to get the belt dead center - as you can see on the picture. Once I make the electrical connections to the motor permanent, I'll try and reposition the motor to try and get the pulley centered. I doubt very much that I can convert to a V-Belt because of the top (saw blade) pulley, but I will certainly google for a possible solution.

Anyway thank you all very much for your replies, especially RCinFLA and FvM. I will definitely look into the overload protection after I get everything connected and running. This has been a much easier fix than I had expected - something that I can do without being an electronics guru! I have learned a lot from your comments, so thank you all once again. I hope this thread will help others who have the same problem with these new-age, high-torque, motors driving old-age machines.

 

[tanuki]

Take a look at this article about flat belt pulley crowning: Flat belt tracking on pulleys.

You should be able to crown the lower pulley easily with a mill file will the motor is running or take it to a machine shop and have it turned on a lathe.

Happy sawing!