[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: [TCML] Correction: How To Turn A Vacuum Cleaner Motor Into A Synchronous Motor



 I made a mistake in my tests.  Test (1) below using 4 segments shorted
is correct, but tests (2) and (3) are not correct.  When I went to 3 or 2
shorted segments, the segments were no longer exactly opposed on
the commutator.  The failure to self-start in (2) and (3) were apparently
due to this condition.  I re-did test (2) using the diode with 3 segments
properly opposed on opposite sides of the commutator.  Then the 
motor self-started from any position, and the hunting gradually died
down and the phase stabilized.  The motor only held sync lock
over a narrow voltage range from around 19V to 21V when ramped
up with a variac.  But if the voltage was switched on, then it was
necessary to apply 19V to obtain sync-lock.  If 18V was applied, 
the motor would never sync lock, and if 20V was applied, the motor 
would never lock.  I didn't try 2 shorted segments on each side,
properly opposed.  

In another test, I tried 3 segments on each side *not* properly
opposed but without a diode.  This did not give good sync lock.  

In another test, I tried 3 segments on each side properly opposed
but without a diode.  The motor self started from any position if
ramped up using a variac, but if the voltage was switched in,
then it didn't start from some positions.  There was less hunting
than with the diode however (sync-lock was more stable without
the diode).  

The 6A diode blew out in one case also, so I had to replace it.  
This was during a test with 3 segments on each side, not properly
opposed.  

It would seem that further tests of various universal motor types/sizes
may be useful.  

Cheers,
John

 


 

 

-----Original Message-----
From: Futuret <futuret@xxxxxxx>
To: tesla <tesla@xxxxxxxxxx>
Sent: Tue, Dec 28, 2010 4:48 pm
Subject: Re: [TCML] How To Turn A Vacuum Cleaner Motor Into A Synchronous Motor


 Hi,



I did more experiments in modifying the same 120VAC universal motor.


1)  I added a 6A, 1000V rectifier diode across the 2 opposed sections of 

shorted commutator segments.  There were 4 adjacent segments shorted

together on one side of the commutator, and 4 adjacent segments shorted

on the opposing side of the commutator.  These opposing areas were then

connected by the diode.  I used a single diode and didn't worry about

weight balance because the diode is lightweight, and the speed (3600rpm)

is not that fast.  I didn't use any glue to hold the diode in place.  The

addition of the diode seemed to cause the motor to self start from any

spot (?), but sync operation was obtained only from 20V to 22V, and the

sync lock was not very stable.  There was constant hunting (rocking

back and forth of the visual pattern while sync-locked).  The presence of

the diode caused arcing at the brushes.



2)  I removed two shorts so that only 3 segments were shorted on

opposing sides of the commutator.  The diode was left in place.

Now the motor didn't self start from one particular spot.  There was

still arcing at the brushes.  Sync lock was obtained between 21V and

29 V.  Hunting occurred at the lower end of that voltage range, but

as the voltage was increased, the hunting died down, and the

sync lock was stable (no rocking of the pattern).  



3)  I removed two more shorts so that only 2 segments were shorted

on opposing sides of the commutator.  The diode was left in place.

Now the motor didn't self start from two spots.  There was still

arcing at the brushes.  Sync lock was only obtained between

16V and 18V.  However the motor did not hunt as in (1) above,

it was more stable, but didn't give much torque while in sync.


It seems to me from the tests above (and below), that adding

the diode improves self starting, but weakens the sync-lock

and can promote hunting.  I expected this, because the addition

of the diode causes the motor to be shorted across the opposing

segments during 1/2 of the A/C cycle, and I found that completely

shorting across the opposing segments kills sync-lock completely.

I assume that shorting for 1/2 the A/C cycle "kills" the sync-lock

only partially, and allows for some sync-locking to occur.  I didn't 

try connecting the diode to only one segment on each side of

the commutator, but my guess is this would work poorly since

the operation improved when I went from 2 segments to 3 segments.

Also Stefan's motor improved when he used more shorted segments.

Using 2 segments shorted on each side gave me bad hunting, which

Stefan also saw until he shorted more segments.  It is interesting that

sync lock occurred around 20 volts or so which happens to be 1/2

the voltage that Stefan and Clive used.  This makes sense because

they were using 240V motors, but I'm using a 120V motor.  


None of my tests gave sync-lock with self-starting in all starting

positions.


I did not try using a heavier rotor on the motor, to simulate use in

an SRSG.  


In any case the sync lock was much better without the diode, although

the self-starting was worse.  It's possible that larger more powerful

universal motors such as Stefan's washing machine motor are more

suitable for this modification.  It's also possible that 240V universal

motors are better suited for this modification since Clive had good

results using his 240 volt vacuum cleaner motor.  

Cheers,

John



-----Original Message-----

From: Futuret <futuret@xxxxxxx>

To: tesla <tesla@xxxxxxxxxx>

Sent: Fri, Dec 17, 2010 8:11 pm

Subject: Re: [TCML] How To Turn A Vacuum Cleaner Motor Into A Synchronous Motor


Hi,


I did some experiments modifying a 120VAC universal motor (series wound)

from a vacuum cleaner.  The motor current and HP ratings are unknown.  

The commutator has 22 segments in total.  

First I shorted 4 adjacent segments, and connected these to 4 more adjacent

segments on the opposite side of the commutator.  So the 4 segments on 

one side were shorted to the other 4 segments on the other side of the 

commutator

of the armature.  I didn't use any diodes.  With this setup, it was impossible 

to 

obtain synchronous operation.  I forget if the motor was always self-starting 

with this arrangement.  


Next I left the 4 segments shorted on opposite sides of the commutator, but

I cut out the short between the 2 groups of 4 segments.  So 4 segments were

still shorted together on each side, but were no longer connected to the

segments on the opposite side.  Again no diodes were used.  This arrangement

gave synchronous operation between 24V at 4.5A, and 42V at 10.5A.  

A reasonable amount of phase shift was obtained as the voltage was varied

between 24V and 42 volts.  It was much easier to pull the motor out of sync

by loading it at the lower voltage than at the higher voltage as would be

expected.  The sync-torque was good overall, but it didn't seem that 

different than an induction sync-modified motor.  I don't know if using 

diodes helps the torque.  The motor self-started at times, but not at other 

times, depending on the armature rotational position at start-up.  The higher 

the voltage, the more likely the motor was to self-start.  At 40 volts, the 

motor always self-started.  If the voltage was over 35 volts, any sudden 

loading of the motor would cause the motor to speed up and lose sync.  

Sudden loads would not tend to occur however in normal SRSG operation.  


I confirmed sync operation, and phase shifting by observing a black line 

drawn on a cardboard disc, attached to the motor shaft, under ballast-type

fluorescent lighting.  I'm experienced at doing this and I have no trouble

telling if a motor is in sync or not.  


These results show that sync operation can be obtained without using

diodes, however the diodes may perhaps improve the self-starting, and 

reduce the current draw some?  I didn't try shorting fewer than 4 segments

on each side.  I have no photos or videos available for these tests.  

In no cases did I encounter the "crazy mode" of operation.  However 

there were some hunting sounds just before the voltage was raised

enough to sync-lock the motor.  In no cases did I see any arcing

at the brushes.  I assume the motor was running at 3600 RPM (not

1800 RPM) because the motor has two poles, but I had no easy way 

to verify that.  Actually, by looking at the light pattern on the disc, it

looked like a 3600 RPM pattern, not an 1800 RPM pattern.   

Cheers,

John 



 
_______________________________________________
Tesla mailing list
Tesla@xxxxxxxxxx
http://www.pupman.com/mailman/listinfo/tesla