Solving Motor Braking Circuit Problem in Bowling Industry

[Tom.G]

OK. Double-Break contacts with NO wired in parallel. Mis-read the schematic. No need for a marked up one.

With just one relay in the pinsetter, relative timing between relays isn't an issue.

The remaining possibilities are:
1) There is a large charge retained on the Start capacitors.
2) The relay being used has overlapped contact timimg.
3) The motor is still coasting.
[5) Ghosts or Gremlins]
There just aren't any other power sources.

Can you hang a DC voltmeter on the capacitors to check for a retained charge during operation?
When a relay starts degrading, is there arcing on every operating cycle or just occassionally?

 

[H012]

Hey Tom, There are two 4PST relays. One for each assembly of the pinsetter. Table is one assembly and the sweep is the other assembly.
The circuitry for each assembly is the same. Breaks down like this: The Table has two caps, one 625 ohm resistor and a 4PST relay.
The sweep has it's own two caps., one 625 ohm resistor and a 4PST relay. The arcing is every cycle. I will check the with a DC meter tomorrow when I go to work at 3 PM Thanks for the feed back.

 

[jim hardy]

Can you hang a DC voltmeter on the capacitors to check for a retained charge during operation?

that'll be worth doing. Watch for several operating cycles because it's random where in the line cycle relay contacts open...

Here's another low-tech test
make yourself a coil about 4" diamter, 100 turns of #18 or so, whatever you have, just remember it has to carry braking current
Connect it here so braking current has to flow through it

now you have done two things
connected a tiny inductance in series
and given yourself a "poor man's surge detector"
six amps won't affect your coil
but if you are getting hundred+ amp surges from that motor, you'll feel your coil 'jump' as the wires push against one another from the sudden current rise
and that'll be a clue
maybe we can figure out how to put a better number on it.
I noticed the "jump" when applying short circuits to an inverter , but i had meters too and learned guesstimate my surge current from how hard the coil jumped.

In troubleshooting we have to use all our senses
and if your current surge is that strong we might be able to use inductance against it.
First step is investigate it further.

Thanks to TomG and Svein for diving in here - I'm over my head , but the whole is greater than the sum of its parts !

old jim
I really don't know, just one more bit of data gathering.

 

[H012]

that'll be worth doing. Watch for several operating cycles because it's random where in the line cycle relay contacts open...

Here's another low-tech test
make yourself a coil about 4" diamter, 100 turns of #18 or so, whatever you have, just remember it has to carry braking current
Connect it here so braking current has to flow through it

now you have done two things
connected a tiny inductance in series
and given yourself a "poor man's surge detector"
six amps won't affect your coil
but if you are getting hundred+ amp surges from that motor, you'll feel your coil 'jump' as the wires push against one another from the sudden current rise
and that'll be a clue
maybe we can figure out how to put a better number on it.
I noticed the "jump" when applying short circuits to an inverter , but i had meters too and learned guesstimate my surge current from how hard the coil jumped.

In troubleshooting we have to use all our senses
and if your current surge is that strong we might be able to use inductance against it.
First step is investigate it further.

Thanks to TomG and Svein for diving in here - I'm over my head , but the whole is greater than the sum of its parts !

old jim
I really don't know, just one more bit of data gathering.

Thanks Jim yea sounds easy enough. You still want me to lift the wire at TS20 first before I try the inductor? It sounds as if all you guys are
on the same page in a round about way. Can I ask a stupid question? How come put my DC meter on the caps

 

[Tom.G]

How come put my DC meter on the caps

Yea, counter-intuitive isn't it? With the motor and relay in the Stopped condition, the only place to store energy is in the caps, and they can't store AC. Ergo, if they are storing energy it must be DC.

The resistors across the caps are there to bleed off any stored charge. If there is a sustained voltage across the caps then the resistors are not doing their job. With the conditions you described, this would show up only after the relay contacts are damaged.

Another useful test is measure both the AC and DC voltage across the caps while the motor is running. Both AC & DC should be zero. If there is a sustained DC voltage, then the resistors are not doing their job. If there is an AC voltage then either the motor Start winding is not disconnecting, or the NC contacts at C1-33K to C1-31A are not opening.

@jim hardy Don't bail out on us! You are making valuable contributions.

 

[H012]

Yea, counter-intuitive isn't it? With the motor and relay in the Stopped condition, the only place to store energy is in the caps, and they can't store AC. Ergo, if they are storing energy it must be DC.

The resistors across the caps are there to bleed off any stored charge. If there is a sustained voltage across the caps then the resistors are not doing their job. With the conditions you described, this would show up only after the relay contacts are damaged.

Another useful test is measure both the AC and DC voltage across the caps while the motor is running. Both AC & DC should be zero. If there is a sustained DC voltage, then the resistors are not doing their job. If there is an AC voltage then either the motor Start winding is not disconnecting, or the NC contacts at C1-33K to C1-31A are not opening.

@jim hardy Don't bail out on us! You are making valuable contributions.

Sounds logical. I have a suspicion one of these is occurring. Will do this when I return to work tomorrow. Will let every one know. I work 3 PM-12.00 post AM Wednesday. Thanks!

 

[jim hardy]

You still want me to lift the wire at TS20 first before I try the inductor?

no, not for this test
we'll check for those voltages
and then hopefully see just how much the current shakes your coil
hopefully you can do it on a machine with open _contacts relays where you can see the arcing
if this little bit of added impedance affects arc intensity that's another clue

Don't bail out on us! You are making valuable contributions.

Thanks.. i tend to get scrambled
that's why i stick to basics

how's this for a hypothesis to check out, if those voltage tests on the caps rule out gross malfunction like sticking contacts ?
Just a hypothesis at this point...

<<<<<<<<<<<<<start hypothesis>>>>>>>>>>>>>>

The initial braking current might be quite high .
The motor becomes an induction generator that gets very nearly shorted out by those 500 uf capacitors in parallel.
We think of those capacitors as 2.65 ohms of Xc, but that's at 60 hz steady state.
At the instant of NC contact closure the capacitor (if it's discharged) is very nearly a short circuit
so change your thinking from steady state 60 hz to millisecond by millisecond

Instantaneous AC is a lot like DC
if the voltage at motor terminals if high on the sinewave at instant T's NC contact closes, we have maybe 150 volts applied instantaneously to a discharged capacitor
and the only thing limiting current is the internal impedances of the motor and the capacitor (well, plus the wires)
We'd like to think current would only be as much as LRA , probably thirty amps or less
but the motor is likely capable of a lot more

i'm no expert at fault current calculations
so have searched for general info on induction motor fault current

http://ee.lamar.edu/gleb/power/Lecture%2012%20-%20Symmetrical%20faults.pdf
4/29/2010

Fault current transients in machines
It is possible to observe the three periods of fault current if the rms magnitude of the AC component current is plotted as a function of time on a semilogarithmic scale.
It is possible to determine the time constants for the three periods...

The AC current flowing in the generator during the subtransient period is called the subtransient current and is denoted by I”. This current is caused by the damper windings of synchronous machines. The time constant of the subtransient current is denoted by T” and it can be determined from the slope. This current may be 10 times the steady-state fault current
Fault transients in induction motors
An induction motor is an AC machine that has only damper windings on its rotor. Since damper windings are major source of current during the subtransient period, the induction motors in a power system should be considered during calculations of subtransient currents flowing in faults.
Since currents in a damper winding are of little importance during the transient and steady-state periods of faults, induction motors may be ignored in fault current analysis after the subtransient period

I believe your motor is capable of 100 amps of fault current for a fraction of a line cycle and that "subtransient current" may be what is wrecking your contacts..That much current will make an impressive arc like you report seeing .
Your clamp on ammeter will not report that transient, instead it'll average a few cycles and report that smaller number.
That transient is significant to the poor little relay contact though, because relay contacts don't just close and stay closed, they bounce for a millisecond or two.
Every bounce interrupts that subtransient current and makes sparks
after a few milliseconds the current settles down and the contact quits bouncing and the motor slows to a stop as intended.

Svein and TomG you guys pointed the way to this one
if it turns out this is the trouble the fix should be easy Limit subtransient current.

That 4 inch 100 turn coil i think should oppose instantaneous current rise , it's around 1.8 millihenries i think which is not quite 1/5th of the 10mh TomG measured on a similar size motor
TomG measured the motor's steady state reactance . Subtransient reactance will be several times smaller. So that coil might roughly halve subtransient current? It's hard to judge an arc, but maybe we'll be lucky and the difference will be noticeable. Again there'll be some randomness to it depending where in the line cycle the contact closes so watch several sweep cycles.
Hopefully we can play with combination of inductance and resistance to reduce contact arcing and still give desired coastdown.<<<<<<<<<<<<<<end hypotheses>>>>>>>>>>>>>>
If testing proves me 100% wrong, well, won't be first time !

sanity check :
TomG measured 10mh
which at 60 hz is 3.77 ohms, 120 volts / 3.77 ohms is 31.8 amps and HO12 reported 29

If X'' is 1/10 of X_{steady state}, then X'' is ~0.3 ohms
and our 1.8mh coil is 0.69 ohms ..
What will be rate of current rise through our 1.8 mh coil and (guess at motor's subtransient reactance of ) 1 mh X'' ?
E = L X rate of change of current
so rate of change of current = E / L
without coil 150V / 1mh =150 amps per millisecond
with coil 150V / 2.8mh = 54 amps per millisecond
we might need a core...
maybe a ferrite toroid from an old CPU power supply?
What do you guys think ?old jim

ps make sure all the turns go same way on that coil - don't use lampcord with the two wires twisted together at far end, that cancels the inductance

old jim

 

[H012]

Yea, counter-intuitive isn't it? With the motor and relay in the Stopped condition, the only place to store energy is in the caps, and they can't store AC. Ergo, if they are storing energy it must be DC.

The resistors across the caps are there to bleed off any stored charge. If there is a sustained voltage across the caps then the resistors are not doing their job. With the conditions you described, this would show up only after the relay contacts are damaged.

Another useful test is measure both the AC and DC voltage across the caps while the motor is running. Both AC & DC should be zero. If there is a sustained DC voltage, then the resistors are not doing their job. If there is an AC voltage then either the motor Start winding is not disconnecting, or the NC contacts at C1-33K to C1-31A are not opening.

@jim hardy Don't bail out on us! You are making valuable contributions.

So measure voltage of both caps at tmpZ and return [neutral] while motor is running?

 

[H012]

So measure voltage of both caps at tmpZ and return [neutral] while motor is running?

Rockwell sent me this. Must admit symbols are you guys expertise.

 

[H012]

Forgot to add that AC-15 is the line on the photo for my relay.

 

[H012]

So measure voltage of both caps at tmpZ and return [neutral] while motor is running?

Rockwell sent me this. Must admit symbols are you guys expertise.

Yea, counter-intuitive isn't it? With the motor and relay in the Stopped condition, the only place to store energy is in the caps, and they can't store AC. Ergo, if they are storing energy it must be DC.

The resistors across the caps are there to bleed off any stored charge. If there is a sustained voltage across the caps then the resistors are not doing their job. With the conditions you described, this would show up only after the relay contacts are damaged.

Another useful test is measure both the AC and DC voltage across the caps while the motor is running. Both AC & DC should be zero. If there is a sustained DC voltage, then the resistors are not doing their job. If there is an AC voltage then either the motor Start winding is not disconnecting, or the NC contacts at C1-33K to C1-31A are not opening.

@jim hardy Don't bail out on us! You are making valuable contributions.

II measured voltage at caps.Idle is zero running is 35 vac and 6.57 vdc

 

[H012]

no, not for this test
we'll check for those voltages
and then hopefully see just how much the current shakes your coil
hopefully you can do it on a machine with open _contacts relays where you can see the arcing
if this little bit of added impedance affects arc intensity that's another clueThanks.. i tend to get scrambled
that's why i stick to basics

how's this for a hypothesis to check out, if those voltage tests on the caps rule out gross malfunction like sticking contacts ?
Just a hypothesis at this point...

<<<<<<<<<<<<<start hypothesis>>>>>>>>>>>>>>

The initial braking current might be quite high .
The motor becomes an induction generator that gets very nearly shorted out by those 500 uf capacitors in parallel.
We think of those capacitors as 2.65 ohms of Xc, but that's at 60 hz steady state.
At the instant of NC contact closure the capacitor (if it's discharged) is very nearly a short circuit
so change your thinking from steady state 60 hz to millisecond by millisecond

Instantaneous AC is a lot like DC
if the voltage at motor terminals if high on the sinewave at instant T's NC contact closes, we have maybe 150 volts applied instantaneously to a discharged capacitor
and the only thing limiting current is the internal impedances of the motor and the capacitor (well, plus the wires)
We'd like to think current would only be as much as LRA , probably thirty amps or less
but the motor is likely capable of a lot more

i'm no expert at fault current calculations
so have searched for general info on induction motor fault current
I believe your motor is capable of 100 amps of fault current for a fraction of a line cycle and that "subtransient current" may be what is wrecking your contacts..That much current will make an impressive arc like you report seeing .
Your clamp on ammeter will not report that transient, instead it'll average a few cycles and report that smaller number.
That transient is significant to the poor little relay contact though, because relay contacts don't just close and stay closed, they bounce for a millisecond or two.
Every bounce interrupts that subtransient current and makes sparks
after a few milliseconds the current settles down and the contact quits bouncing and the motor slows to a stop as intended.

Svein and TomG you guys pointed the way to this one
if it turns out this is the trouble the fix should be easy Limit subtransient current.

That 4 inch 100 turn coil i think should oppose instantaneous current rise , it's around 1.8 millihenries i think which is not quite 1/5th of the 10mh TomG measured on a similar size motor
TomG measured the motor's steady state reactance . Subtransient reactance will be several times smaller. So that coil might roughly halve subtransient current? It's hard to judge an arc, but maybe we'll be lucky and the difference will be noticeable. Again there'll be some randomness to it depending where in the line cycle the contact closes so watch several sweep cycles.
Hopefully we can play with combination of inductance and resistance to reduce contact arcing and still give desired coastdown.<<<<<<<<<<<<<<end hypotheses>>>>>>>>>>>>>>
If testing proves me 100% wrong, well, won't be first time !

sanity check :
TomG measured 10mh
which at 60 hz is 3.77 ohms, 120 volts / 3.77 ohms is 31.8 amps and HO12 reported 29

If X'' is 1/10 of X_{steady state}, then X'' is ~0.3 ohms
and our 1.8mh coil is 0.69 ohms ..
What will be rate of current rise through our 1.8 mh coil and (guess at motor's subtransient reactance of ) 1 mh X'' ?
E = L X rate of change of current
so rate of change of current = E / L
without coil 150V / 1mh =150 amps per millisecond
with coil 150V / 2.8mh = 54 amps per millisecond
we might need a core...
maybe a ferrite toroid from an old CPU power supply?
What do you guys think ?old jim

ps make sure all the turns go same way on that coil - don't use lampcord with the two wires twisted together at far end, that cancels the inductance

old jim

Hey Jim all I have in quinity is 18 standed

no, not for this test
we'll check for those voltages
and then hopefully see just how much the current shakes your coil
hopefully you can do it on a machine with open _contacts relays where you can see the arcing
if this little bit of added impedance affects arc intensity that's another clueThanks.. i tend to get scrambled
that's why i stick to basics

how's this for a hypothesis to check out, if those voltage tests on the caps rule out gross malfunction like sticking contacts ?
Just a hypothesis at this point...

<<<<<<<<<<<<<start hypothesis>>>>>>>>>>>>>>

The initial braking current might be quite high .
The motor becomes an induction generator that gets very nearly shorted out by those 500 uf capacitors in parallel.
We think of those capacitors as 2.65 ohms of Xc, but that's at 60 hz steady state.
At the instant of NC contact closure the capacitor (if it's discharged) is very nearly a short circuit
so change your thinking from steady state 60 hz to millisecond by millisecond

Instantaneous AC is a lot like DC
if the voltage at motor terminals if high on the sinewave at instant T's NC contact closes, we have maybe 150 volts applied instantaneously to a discharged capacitor
and the only thing limiting current is the internal impedances of the motor and the capacitor (well, plus the wires)
We'd like to think current would only be as much as LRA , probably thirty amps or less
but the motor is likely capable of a lot more

i'm no expert at fault current calculations
so have searched for general info on induction motor fault current
I believe your motor is capable of 100 amps of fault current for a fraction of a line cycle and that "subtransient current" may be what is wrecking your contacts..That much current will make an impressive arc like you report seeing .
Your clamp on ammeter will not report that transient, instead it'll average a few cycles and report that smaller number.
That transient is significant to the poor little relay contact though, because relay contacts don't just close and stay closed, they bounce for a millisecond or two.
Every bounce interrupts that subtransient current and makes sparks
after a few milliseconds the current settles down and the contact quits bouncing and the motor slows to a stop as intended.

Svein and TomG you guys pointed the way to this one
if it turns out this is the trouble the fix should be easy Limit subtransient current.

That 4 inch 100 turn coil i think should oppose instantaneous current rise , it's around 1.8 millihenries i think which is not quite 1/5th of the 10mh TomG measured on a similar size motor
TomG measured the motor's steady state reactance . Subtransient reactance will be several times smaller. So that coil might roughly halve subtransient current? It's hard to judge an arc, but maybe we'll be lucky and the difference will be noticeable. Again there'll be some randomness to it depending where in the line cycle the contact closes so watch several sweep cycles.
Hopefully we can play with combination of inductance and resistance to reduce contact arcing and still give desired coastdown.<<<<<<<<<<<<<<end hypotheses>>>>>>>>>>>>>>
If testing proves me 100% wrong, well, won't be first time !

sanity check :
TomG measured 10mh
which at 60 hz is 3.77 ohms, 120 volts / 3.77 ohms is 31.8 amps and HO12 reported 29

If X'' is 1/10 of X_{steady state}, then X'' is ~0.3 ohms
and our 1.8mh coil is 0.69 ohms ..
What will be rate of current rise through our 1.8 mh coil and (guess at motor's subtransient reactance of ) 1 mh X'' ?
E = L X rate of change of current
so rate of change of current = E / L
without coil 150V / 1mh =150 amps per millisecond
with coil 150V / 2.8mh = 54 amps per millisecond
we might need a core...
maybe a ferrite toroid from an old CPU power supply?
What do you guys think ?old jim

ps make sure all the turns go same way on that coil - don't use lampcord with the two wires twisted together at far end, that cancels the inductance

old jim

Hey Jim all I have enough of is 18 ga. stranded

 

[jim hardy]

So measure voltage of both caps at tmpZ and return [neutral] while motor is running?

i'd say so
while running
TMP-Z to TS-12 looks like it'd get top cap
TS-12 to 115V return gets the other

while braking
TS12 to 115V return will be interesting
read it on AC that's voltage across 1000 microfarads so we'll know how much current motor makes by ohm's law
it'd be interesting to watch a few machine cycles and see if noticeable DC voltage shows up , it'd be the DC component of fault current transient
but that's sort of idle curiosity just the kind of thing i do just to get a feel for the system.

Will you be able to patch in that coil and feel for impulse?

I know it's asking for the moon
but if you solve an industry wide problem some magazine will be interested in your story.
Check out "Sherlock Ohms" at EDN... AMF i'd think will be interested too.

old jim

 

[jim hardy]

Hey Jim all I have enough of is 18 ga. stranded

that'll carry six amps runnning just fine

 

[H012]

i'd say so
while running
TMP-Z to TS-12 looks like it'd get top cap
TS-12 to 115V return gets the other

while braking
TS12 to 115V return will be interesting
read it on AC that's voltage across 1000 microfarads so we'll know how much current motor makes by ohm's law
it'd be interesting to watch a few machine cycles and see if noticeable DC voltage shows up , it'd be the DC component of fault current transient
but that's sort of idle curiosity just the kind of thing i do just to get a feel for the system.

Will you be able to patch in that coil and feel for impulse?

I know it's asking for the moon
but if you solve an industry wide problem some magazine will be interested in your story.
Check out "Sherlock Ohms" at EDN... AMF i'd think will be interested too.

old jim

I wondered why I was a little confused. I have misled you guy,not on purpose though. the schematic I gave in photo is for the table we need the sweep. only difference is a toogle switch circuit for reversing the sweep. If you still want me to continue with the test I will.

 

[jim hardy]

Sure, continue on
do i see terminal numbers are a little different on sweep than table ?
if you want to use sweeps chematic that's fine, i'll save that drawing
i do note different terminal numbers?

I will use them hereafter unless you say otherwise.

II measured voltage at caps.Idle is zero running is 35 vac and 6.57 vdc

Idle should be zero, everything is de-energized and shorted by the motor windings

running reading is curious though
35VAC across a 500 uf capacitor suggests by ohm's law 5 amps through it which had to flow through open contacts to get there
and 6.57 volts DC suggests the resistor is open

this always happens when first start troubleshooting, you get readings that don't make sense
usually their meaning comes clear later on after you've double checked enough stuff to figure out what's different between the way you imagine the circuit and the way it really is.HO12 - do you see why I'm confused by those meter readings?
Which motor were we reading - Sweep or the other one?
Are the terminals marked on the chassis ? Or do you have to find them from a drawing ?
On what terminal was the meter's + lead ?
On what terminal was the meter's - lead ?

All terminal blocks look pretty much alike.. Is that resistor unmistakeable?
Voltage between resistor ends, 1 to 2 on drawing, running __________ VAC ____________ VDC
Voltage centertap to end , 3 to 1 running ____________ VAC __________VDC
Voltage centertap to end , 2 to 1 running ____________VAC ___________VDC

@Tom.G & @Svein Any thoughts?

old jim

 

[Tom.G]

II measured voltage at caps.Idle is zero running is 35 vac and 6.57 vdc

If the hardware matches the schematic AND you are using an Analog meter:
Two definite (and maybe a third) concurrent defects:
1) Centrifugal switch in motor is not disconnecting the Start winding
2) At least one of the caps is defective
3) Maybe a bad resistor

A current measurement, while running, on the wire between C1-24T and SWSR-5 would be a big help in clarification.[/color]

 

[jim hardy]

what tom said

 

[Tom.G]

@H012 I just downloaded the (128 pg.) manual w/ the sheet that Rockwell sent you (ref Post #59). Will plow thru it over the next few days.
If anyone is interested, the download link is: http://literature.rockwellautomation.com/idc/groups/literature/documents/td/700-td552_-en-p.pdf

 

[H012]

Sure, continue on
do i see terminal numbers are a little different on sweep than table ?
if you want to use sweeps chematic that's fine, i'll save that drawing
i do note different terminal numbers?

View attachment 103513

I will use them hereafter unless you say otherwise.Idle should be zero, everything is de-energized and shorted by the motor windings

running reading is curious though
35VAC across a 500 uf capacitor suggests by ohm's law 5 amps through it which had to flow through open contacts to get there
and 6.57 volts DC suggests the resistor is open

this always happens when first start troubleshooting, you get readings that don't make sense
usually their meaning comes clear later on after you've double checked enough stuff to figure out what's different between the way you imagine the circuit and the way it really is.HO12 - do you see why I'm confused by those meter readings?
Which motor were we reading - Sweep or the other one?
Are the terminals marked on the chassis ? Or do you have to find them from a drawing ?
On what terminal was the meter's + lead ?
On what terminal was the meter's - lead ?

All terminal blocks look pretty much alike.. Is that resistor unmistakeable?
Voltage between resistor ends, 1 to 2 on drawing, running __________ VAC ____________ VDC
Voltage centertap to end , 3 to 1 running ____________ VAC __________VDC
Voltage centertap to end , 2 to 1 running ____________VAC ___________VDC

@Tom.G & @Svein Any thoughts?

old jim

It takes me a while to do readings I'm solo working 26 machines. ok I did the I checked TS27 which is the center tap on resistor to 115 return as per the sweep schematic. VAC= 22 and VDC = 41 I checked it twice. to make sure. this with motor running. this reading only occurred milliseconds when motor started to run the as it's running quickly drops off till the motor stops.

 

[H012]

This is the resistor. There are two of these One is for the table and one is for the sweep

 

[jim hardy]

It takes me a while to do readings I'm solo working 26 machines.

I'm sorry to be such a pest
just English is so imprecise and i can only imagine how you made the measurement.
Thanks for the expansion , i'll quote it

this with motor running. this reading only occurred milliseconds when motor started to run
AHA That's what it SHOULD do , appear until CS opens then disappear. Caps are in series with start winding until motor gets up to speed
at which time CS disconnects them
so voltage during motor acceleration is proper
and voltage should drop to zero as soon as motor reaches speed and remain there while motor runs
Voltage should return when relay T connects capacitors to motor run winding to decelerate itthen as it's running quickly drops off till the motor stops.
That sounds right.
A reading across capacitorsduring deceleration would cross check your 29 amp reading of a few days ago
29 amps through 1000 uf should make 76 volts

Thanks !

Looking forward to your coil thump test.

Got any neon indicator lamps around ?

Tom - you copy ? Red parts Sound right ?

old jim

 

[Tom.G]

...quickly drops off till the motor stops.

The RC time constant is about 15 milliSeconds so the voltage should be less than 1% of initial reading within 75 milliSeconds of Start winding disconnecting[/color].
If significant voltage remains, either resistor(s) bad, or Start winding not disconnecting.

The presence of a DC voltage strongly points to at least one defective cap.

Jim: re Red: You type faster than I do, I think we said the same thing!

EDIT: phrase in Red[/color]

 

[Tom.G]

Correction to Post #73 (#73 has also been edited in red)

The presence of a unbalanced DC voltage strongly points to at least one defective cap or resistor.

 

[Rx7man]

The sizing is beyond me at this point.. but I haven't seen the suggestion of adding a cap across the relay contacts that are arcing.. since they're arcing on opening, it would suggest a high voltage spike from the disconnection of the coils, a capacitor across them (evidently with a high voltage rating) could suppress the time and intensity of an arc. Much like the way old points style ignition systems worked, if the condenser was bad, the points burnt up.

 

[jim hardy]

this reading only occurred milliseconds when motor started to run the as it's running quickly drops off till the motor stops.

Just to make sure we're hearing each other

When motor starts you get 22AC and 41 DC but only very briefly ?

then as motor reaches speed , voltage quickly drops off, i presume to near zero but would you confirm or correct me on that ?
Most troubleshooting mistakes arise from presuming what somebody meant to say.

If that's what you saw , brief voltage during start return to near zero, it gives me confidence the circuit is as i imagined it.

Voltage should return while motor coasts down in braking cycle
because capacitors get switched back in by NC contacts on S relay
Aha now relay names T_able and S_weep make sense ...

Precision in communication sure is tedious, isn't it ?

But it feels so good to arrive !

Thanks for your work i think we are progressing nicelyold jim

 

[Tom.G]

@H012 One more clarification please.

The only fact I know is that the NC arcing occurs when they open, only because I can see it.

Does "...when they open..." refer to the instant of motor start, or to the instant of motor shutoff. i.e. the NC arcing is at motor start or stop?

 

[jim hardy]

edit--- related observation, no response needed thpugh

but I haven't seen the suggestion of adding a cap across the relay contacts that are arcing.. since they're arcing on opening,

there's an open question on which contacts arc vs burn up

from post 45

is the arcing on motor start or motor stop? or both ?

Motor stopping. Reason I can see the arcing is on seven of my machines I have the old style open contact Allen Bradley relays Jim my be familiar with these.

Arcing is on motor stop
NO contacts open on motor stop
NC contacts close on motor stop
it's NC contacts that burn up

so if the arcing is on the contacts that open on stop, it's not on the contacts that are burning up

you're right, I'm focused on the NC contacts
the NO contacts might make a more impressive arc but if so it doesn't appear to be hurting them.

Thanks ! That was one more presumption on my part, that the visible arc was wrecking the contacts,, and "it ain't necessarily so..."

old jim

 

[Tom.G]

Side Comment:

I've had a few requests regarding contact numbering on the 700-series control relays used in this thread. So here goes.

First off, realize there is no such thing as "Double Throw" contacts in industrial control relays. Contact sets are NO (Normally Open) or NC (Normally Closed). If you want Double Throw you wire up a set of NO and a set of NC contacts. Additional options such as Double Break or Overlapped Contact Timing (make-before-break) are available, but they won't be discussed here.

Contacts are numbered with a typically 2-digit number, XX

1. Right most digit is the 'Form' of the contact and which contact of the set XX
   • "1" and "2" denote a NC contact set with the "1" and "2" showing which of the contacts is referred to
   • "3" and "4" denote a NO contact set with the "3" and "4" showing which of the contacts is referred to
2. The next digit(s) to the left, XX, are just a sequential numbering of the contact sets

The relay Coil(s) have an uppercase letter as the left-character. "A1" and "A2" would be the two ends of the first coil.

In the specification sheet for a relay you will find both Current rating, "I" (uppercase letter), and Voltage rating, "U" or "V". Where "V" is the American style, and "U" being used most other places.

The above was gathered from sources referencing the International standard IEC 60947-1 and IEC 60947-4-1

 

[Rx7man]

It's not a cheap solution... but a 3 phase VFD and motors would be a 'pretty' solution to this, and would be capable of operating faster... Depending on the lifespan of your current relays, the cost, downtime, etc and if you can get another solution from here, it may be worth it in the end.

 

[jim hardy]

i'd like to see what he feels on the 100 turn coil

if indeed it's a current pulse from the motor that's wrecking the contacts
some sort of snubber seems in order
a couple dozen mh of inductance should tame the inrush
or Svein's few ohm resistor might work really well too

honestly were i designing that machine i'd have sized the contacts for LRA and never given a thought to transient inrush

trial and error
i'll see if i have an old toroid to send him

 

[H012]

Just to make sure we're hearing each other

When motor starts you get 22AC and 41 DC but only very briefly ?

then as motor reaches speed , voltage quickly drops off, i presume to near zero but would you confirm or correct me on that ?
Most troubleshooting mistakes arise from presuming what somebody meant to say.

If that's what you saw , brief voltage during start return to near zero, it gives me confidence the circuit is as i imagined it.

Voltage should return while motor coasts down in braking cycle
because capacitors get switched back in by NC contacts on S relay
Aha now relay names T_able and S_weep make sense ...

Precision in communication sure is tedious, isn't it ?

But it feels so good to arrive !

Thanks for your work i think we are progressing nicelyold jim

That is what's happening Jim

 

[jim hardy]

That is what's happening Jim

Thanks !
I always struggle with communicating clearly,,,
both in transmitting and receiving
Thank you for all your effort and your patience with my "Twenty Questions"
.. but i have to ask them because I'm quite capable of going off on tangents.

i'm feeling more confidence in my understanding of the circuit now.

Can you see how sparking at the NO contacts compares to sparking at the NC contacts?

On motor start
there shouldn't be much sparking, just contact bounce on the NO's as they conduct motor starting current, maybe thirty amps
the NC's open then but they aren't carrying current so no sparks on them

but on motor stop
there'll be a spark at NO's as they open interrupting motor running current of probably six amps
AND some spark at the NC's as they close and bounce with whatever transient current the motor is pushing through the capacitors...
... if your coil 'jumps' it means that transient current is substantial
dont know yet how to measure it
but maybe we'll just try to suppress it...

This is interesting
and i hope you don't yet regret asking us !

old jim

 

[H012]

Thanks !
I always struggle with communicating clearly,,,
both in transmitting and receiving
Thank you for all your effort and your patience with my "Twenty Questions"
.. but i have to ask them because I'm quite capable of going off on tangents.

i'm feeling more confidence in my understanding of the circuit now.

Can you see how sparking at the NO contacts compares to sparking at the NC contacts?

On motor start
there shouldn't be much sparking, just contact bounce on the NO's as they conduct motor starting current, maybe thirty amps
the NC's open then but they aren't carrying current so no sparks on them

but on motor stop
there'll be a spark at NO's as they open interrupting motor running current of probably six amps
AND some spark at the NC's as they close and bounce with whatever transient current the motor is pushing through the capacitors...
... if your coil 'jumps' it means that transient current is substantial
dont know yet how to measure it
but maybe we'll just try to suppress it...

This is interesting
and i hope you don't yet regret asking us !

old jim

Jim I tried that inductor 100 turns 18 ga. I can't tell if it's jumping with the machine shaking when it's operating. It is a bad place I have to put it the way the wire duct is.

 

[jim hardy]

Jim I tried that inductor 100 turns 18 ga. I can't tell if it's jumping with the machine shaking when it's operating. It is a bad place I have to put it the way the wire duct is.

okay thanks for the try..

back later

 

[H012]

okay thanks for the try..

back later

Hey Jim with the volts seeming ok. and current from what I understand is ok. Where is this leading us? In you guys opinion? It seems a paradox.

 

[jim hardy]

I still think our trouble stems from closing a switch(NC contacts) between a power source(motor) and discharged capacitors(those 500uf's)

the surge current is less than i thought else it would have shaken your coil
but it flows through contacts that are in the process of closing and aren't fully made

http://pdfserv.maximintegrated.com/en/an/AN764.pdf
Switch Bounce
...
As a switch changes state, its contacts travel through several open-close cycles called "contact bounce"
before coming to rest in the final state. In some cases, this rapid contact bounce produces a changing
resistance as the moving contact wipes across the stationary contact.

here's a 'scope trace(from a different source) of contact bounce on closure - each bounce makes an arc
typical bouncing time is about a millisecond
contact separation on opening is much cleaner
jh

Lamp and capacitive loads represent the worst case. At the instant of switch closure, both of these loads
subject the switch to a short circuit. A lamp's cold resistance is close to 0Ω, and a discharged capacitor
is a short circuit at the instant voltage is applied! This state change can produce inrush currents up to
100 times the steady-state value. To compound the problem, switch contacts are bouncing during this
short-circuit interval. High current levels and bouncing contacts produce severe arcing across the switch
contacts that causes contact erosion. What's worse is that repeated cycles of switch operation with
severe arcing can create a short circuit by welding (fusing) the contacts together

think of your motor as a generator and those capacitors as its load. Capacitive inrush is i think our trouble.

What's the "make" current rating of that relay contact?

http://www.aeroelectric.com/Reference_Docs/Tyco/Tyco-Relay_Contact_Life.pdf
Material loss is due primarily to splattering of the molten and boiling
metal as contacts bounce on make. Such loss can be significant over
the course of tens of thousands of operations, and the only practical
way to minimize it is by arc suppression. Arc suppression quickly
quenches the arc, thereby holding contact temperatures lower.

so i'd like to limit the inrush to those capacitors

do you think we could experiment with some impedance where we showed it, and see if it slows the coastdown too much ?

Inductance has the appeal of elegance, but i do suffer from idealism
two or three ohms of resistance might do the trick
as Svein and Tom.G have suggested
@Tom.G thoughts?

 

[H012]

I still think our trouble stems from closing a switch(NC contacts) between a power source(motor) and discharged capacitors(those 500uf's)

the surge current is less than i thought else it would have shaken your coil
but it flows through contacts that are in the process of closing and aren't fully made
think of your motor as a generator and those capacitors as its load. Capacitive inrush is i think our trouble.

What's the "make" current rating of that relay contact?so i'd like to limit the inrush to those capacitors

do you think we could experiment with some impedance where we showed it, and see if it slows the coastdown too much ?

Inductance has the appeal of elegance, but i do suffer from idealism
two or three ohms of resistance might do the trick
as Svein and Tom.G have suggested
@Tom.G thoughts?

If you look at post #59 JPG go down to line AC-15 This gives the Make/Break of contacts. I will go down to the local to local electronics store
and pick up some resistors. I think Svein suggested 8 ohm 50 watt I will see what lower resistance ones they have. I loved Radio shack back
in the day. I could get anything for electronics. Thanks guys!

 

[jim hardy]

Paralleling two 8's makes a 4, three a 2&2/3, etc
There's some number of ohms that'll slow the coastdown enough to upset the mechanical timing
when you try one measure the voltage across it , initial when NC contacts close.While you're there, pick up a motor run capacitor, maybe maybe 10 uf, not critical should be cheap i bought a 10 yesterday for $3.

We never established unambiguously at which contacts you see the arcing on motor stop - middle NC's or outside NO's ?
If you can see arcing on NC's when they close , we at least have a visible indication of any improvement we make.

Might be worth trying to limit rate of voltage rise across motor in interval between NC's open and NC's close, also during contact bounce time

that'd be your motor run cap in red... resistor sees 1/4 amp continuous so should be a couple watts to stay cool

i tend to study things to death
you are where you can see what's going on
so don't hesitate to apprise us of anything you notice !
...

just noticed
Aha 1.5 ohms main winding ? We should be able to add at least that much which could halve your surge current
question is what's max we can add before motor coasts too far? We have to stay well below that.
....
Figuring Inductance is a little tougher, we'll have to estimate how many volts it has to hold back for how many microseconds.

If we say it only has to hold off 150 volts for a millisecond while contacts stop bouncing
and limit current rise to 30 amps while that millisecond elapses
then we have di/dt = 30 amps per millisecond = 30ka/second
at 150 volts
e = L X di/dt
so
L = 150 / 30,000 = 5 millihenries
we were perhaps ⅓ of the way there with your 100 turn coil
and 30 amps wouldn't have shook the coil very hard

did you notice any change in sparks ?

<<<<<<<<<<end chasing too many 'thought rabbits' at once..>>>>>>>> for now

 

[jim hardy]

If you look at post #59 JPG go down to line AC-15 This gives the Make/Break of contacts.

Thanks - they tested it at 10X rated current for 6050 cycles, ten cycles per minute is how i read that.

How long does it take you to accumulate six thousand cycles? Do they have counters?
Sure wish we could measure that initial burst of current to capacitors

Ahh we'll get there !

old jim

 

[jim hardy]

here's a formula for inductance of a coil with no iron core
http://home.earthlink.net/~jimlux/hv/wheeler.htm

Multi layer air core solenoid
L (uH) = 0.8 * a^2 * n^2 / (6*a + 9*b + 10*c )

where
a = average radius of windings
b = length of the coil
c = difference between the outer and inner radii of the coil.
all dimensions in inches

It states that it is accurate to 1% when the terms in the denominator are
about equal. This is also an equation by Wheeler. It applies as long as the
coil has a rectangular cross section..

cross checked at
http://66pacific.com/calculators/coil_calc.aspxi get ~1.45 mh for 100 turns on 4'' diameter if i typed right
200 turns would be 5.8 mh
a core would be nice
the toroid from an old computer or inverter type microwave oven seems right flavor , we're blocking high frequency transients
i'll check my junkpile

 

[H012]

Paralleling two 8's makes a 4, three a 2&2/3, etc
There's some number of ohms that'll slow the coastdown enough to upset the mechanical timing
when you try one measure the voltage across it , initial when NC contacts close.While you're there, pick up a motor run capacitor, maybe maybe 10 uf, not critical should be cheap i bought a 10 yesterday for $3.

We never established unambiguously at which contacts you see the arcing on motor stop - middle NC's or outside NO's ?
If you can see arcing on NC's when they close , we at least have a visible indication of any improvement we make.

Might be worth trying to limit rate of voltage rise across motor in interval between NC's open and NC's close, also during contact bounce time
View attachment 103603
that'd be your motor run cap in red... resistor sees 1/4 amp continuous so should be a couple watts to stay cool

i tend to study things to death
you are where you can see what's going on
so don't hesitate to apprise us of anything you notice !
...

just noticed
Aha 1.5 ohms main winding ? We should be able to add at least that much which could halve your surge current
question is what's max we can add before motor coasts too far? We have to stay well below that.
....
Figuring Inductance is a little tougher, we'll have to estimate how many volts it has to hold back for how many microseconds.

If we say it only has to hold off 150 volts for a millisecond while contacts stop bouncing
and limit current rise to 30 amps while that millisecond elapses
then we have di/dt = 30 amps per millisecond = 30ka/second
at 150 volts
e = L X di/dt
so
L = 150 / 30,000 = 5 millihenries
we were perhaps ⅓ of the way there with your 100 turn coil
and 30 amps wouldn't have shook the coil very hard

did you notice any change in sparks ?

<<<<<<<<<<end chasing too many 'thought rabbits' at once..>>>>>>>> for now

Jim I have 3 of these. 2.7, 3.0 and 8.2 25% The 50 watt were outragous in price not to mention their size.

 

[H012]

Jim I have 3 of these. 2.7, 3.0 and 8.2 25% The 50 watt were outragous in price not to mention their size.

I will check I have some run caps not sure of their capacitance

 

[jim hardy]

Jim I have 3 of these. 2.7, 3.0 and 8.2 25% The 50 watt were outragous in price not to mention their size.

Since they only carry current during coastdown they can be less wattage than 6 amps continuous would require .

If you can see sparks on the NC's that's our best quick indication of improvement .

Try them all one at a time and see if coastdown gets longer, question is " does the thing still stop in right place...?."
If it tolerates 8 ohms that should help contact life considerably !
If you can read voltage across caps during coastdown before and after it'll tell us how much we changed braking current.

Then i guess we'd have to outfit a machine with one, install a new relay and run it for the usual life of a relay, inspect contacts to see if we reduced erosion.

If you can see sparking that's great, we can try the resistor- capacitor snubber across motor and see right away.
If you can't see sparking we could try that long term test on another machine.

Another "poor man's" test equipment is an el-cheapo AM pocket radio set between stations. It'll pick up any arcing.
You might get a clue to the intensity of arcing from how much static you hear at relay operate time. A bad contact will likely make static all the time it's conducting.
We use that in the power plant for surveying big motors, when they start to develop corona they get "transistor-radio-active" .

I envy you this experimenting, how far are you from Northeast Arkansas ?
https://en.wikipedia.org/wiki/Hardy,_Arkansas

 

[H012]

Since they only carry current during coastdown they can be less wattage than 6 amps continuous would require .

If you can see sparks on the NC's that's our best quick indication of improvement .

Try them all one at a time and see if coastdown gets longer, question is " does the thing still stop in right place...?."
If it tolerates 8 ohms that should help contact life considerably !
If you can read voltage across caps during coastdown before and after it'll tell us how much we changed braking current.

Then i guess we'd have to outfit a machine with one, install a new relay and run it for the usual life of a relay, inspect contacts to see if we reduced erosion.

If you can see sparking that's great, we can try the resistor- capacitor snubber across motor and see right away.
If you can't see sparking we could try that long term test on another machine.

Another "poor man's" test equipment is an el-cheapo AM pocket radio set between stations. It'll pick up any arcing.
You might get a clue to the intensity of arcing from how much static you hear at relay operate time. A bad contact will likely make static all the time it's conducting.
We use that in the power plant for surveying big motors, when they start to develop corona they get "transistor-radio-active" .

I envy you this experimenting, how far are you from Northeast Arkansas ?
https://en.wikipedia.org/wiki/Hardy,_Arkansas

I'm in New Jersey. I'm from smokey moutain region of Tenn. Me and my wife are out of here in 3yrs. max! Hate it here but work is good. Cost of living is outragous!. Ok will give resistors a shot. Got to do machine maitenance first.

 

[jim hardy]

Got to do machine maitenance first.

You're talking to a retired maintenance man...

i really hope this helps out

Fair Anne and I and I considered East Tennessee for retirement
but settled in the Ozarks

Further west we get the better we like it
when i cross that Mississippi bridge in Memphis i start to get that "far from the madding crowd" feeling.
It really kicks in about Laramie Wyoming.

 

[H012]

You're talking to a retired maintenance man...

i really hope this helps out

Fair Anne and I and I considered East Tennessee for retirement
but settled in the Ozarks

Further west we get the better we like it
when i cross that Mississippi bridge in Memphis i start to get that "far from the madding crowd" feeling.
It really kicks in about Laramie Wyoming.

That's how my wife JoAnn and I feel too. Had enough raising kids and yes the rat race
Too. Anyway got a full house here at work. I will try later to get the resistor in. Got to solder 12 inch wire on each end off resistor
To reach into wire duct.

 

[H012]

I still think our trouble stems from closing a switch(NC contacts) between a power source(motor) and discharged capacitors(those 500uf's)

the surge current is less than i thought else it would have shaken your coil
but it flows through contacts that are in the process of closing and aren't fully made
think of your motor as a generator and those capacitors as its load. Capacitive inrush is i think our trouble.

What's the "make" current rating of that relay contact?so i'd like to limit the inrush to those capacitors

do you think we could experiment with some impedance where we showed it, and see if it slows the coastdown too much ?

Inductance has the appeal of elegance, but i do suffer from idealism
two or three ohms of resistance might do the trick
as Svein and Tom.G have suggested
@Tom.G thoughts?

Hey Jim can you help with the resistor location. The C1 33 to TS 12 is the table. Could you show where using the sweep schematic?. Thanks

 

[jim hardy]

Could you show where using the sweep schematic?. Thanks

somewhere in the pink line
in place of that wire from C1-23N to TS-27
or C1-23N to whichever terminal that is on the S relay, 10 or 32

Schematic shows a wire between relay S terminals 2, 4 and 5
that could be a convenient place if you can see clearly how they accomplished their wiring,
just be careful , you don't want it on the wrong side of terminal 4 !
if there's a single wire on 5 find its other end, hopefully it's a short jumper to 4&2. ( with a single wire from one of those to C1-22J , that's how i woulda done it. )

even between junction of capacitors and centertap of resistor would be okay

schematics don't look a lot like the physical machine, do they ... Got a cellphone picture of it you could upload ?

old jim

 

[jim hardy]

gonna try to reload touchpad driver - back in a while

old jim

got it downloaded but windows won't recognize it, ... what else is new.

anyhow approaching bedtime here, looking forward to your results