Solving Motor Braking Circuit Problem in Bowling Industry

[Tom.G]

Here's one more easy attempt. Try connecting a MOV, Metal Oxide Varistor, across the motor Run winding. That would be between SM2 and SM3, or equivalently between SMP-CT and SMP-NIL.

Here is a link to one that seems adequately sized. Its wire leads are only 1 inch long though.
http://www.digikey.com/product-search/en?keywords=v20e130p

 

[jim hardy]

The NO contacts are still arcing.
Is there arcing both at NO and at NC?
If so, is there a difference between the arcs ? Intensity? I wanted to eliminate the old relays. As the contacts being the cause.
Might be one contact needs protection against overcurrent, the other against overvoltage
resistor is to limit current through NC contact
a Quencharc across the motor would try to protect the NO against overvoltage
and protect the NC against arcing during contact bounce interval,

It'd sure help to see it ourselves . Does your cellphone do short videos ?

.....................

I THINK, based on not really on anything solid,
overcurrent transient is what kills the SSR because that circuit board appears to have plenty of capacitors on it and i assume some of them are snubbers.
It'd help to know if a dead SSR shows short between 4-3, 4-7, or/and 6-5

you could risk sacrificing one on a machine that has resistor?

What you think, Tom ?

 

[Tom.G]

It'd help to know if a dead SSR shows short between 4-3, 4-7, or/and 6-5

you could risk sacrificing one on a machine that has resistor?

What you think, Tom ?

I think finding a shorted or an open Triac won't tell us what caused it. Overcurrent causes the active part to melt into an amorphous blob of Silicon with low resistance, if the current is high enough the bond wire from the blob to a package terminal becomes a fuse and opens. An overvoltage can turn the Triac on or trigger avalanche voltage breakdown and you get the same end result.

 

[H012]

Okaayy... We seem to be switching back and forth on whether it's the NO or NC contacts that are arcing. I suspect they are both arcing under different conditions/faults on different machines, but so far it is very confusing. Getting this settled is important because appropriate fixes depend on the timing and specific conditions that the arcs occur. It would be extremely useful if you could post a video of a Sweep relay for a whole sweep cycle showing arcing. Ideally using one of the old style open frame relays so we can see the contact movements and where the arcing occurs, and infer where in the operating cycle it all happens.

Here's one more easy attempt. Try connecting a MOV, Metal Oxide Varistor, across the motor Run winding. That would be between SM2 and SM3, or equivalently between SMP-CT and SMP-NIL.

Here is a link to one that seems adequately sized. Its wire leads are only 1 inch long though.
http://www.digikey.com/product-search/en?keywords=v20e130p

You guys are really trying hard. I know it's frustrating for you given the fact of my limitations. I looked up how to find the inductance of the motor. Forget that for the equations are way over my head! And yes Tom you are correct Both relays NC's fail but by nature of the fact that the sweep has 3 stops in one cycle of the machine vs 1 for the table is why I'm concentrated on the sweep. I will do the MOV simple enough. And many thanks to you guys! I forgot to add the that you are also correct that NO's on some do arc/flash why some contacts arc/flash.only adds to the confusion.

 

[H012]

It'd sure help to see it ourselves . Does your cellphone do short videos ?

.....................

I THINK, based on not really on anything solid,
overcurrent transient is what kills the SSR because that circuit board appears to have plenty of capacitors on it and i assume some of them are snubbers.
It'd help to know if a dead SSR shows short between 4-3, 4-7, or/and 6-5

you could risk sacrificing one on a machine that has resistor?

What you think, Tom ?

I will check one to see.

 

[H012]

You guys are really trying hard. I know it's frustrating for you given the fact of my limitations. I looked up how to find the inductance of the motor. Forget that for the equations are way over my head! And yes Tom you are correct Both relays NC's fail but by nature of the fact that the sweep has 3 stops in one cycle of the machine vs 1 for the table is why I'm concentrated on the sweep. I will do the MOV simple enough. And many thanks to you guys! I forgot to add the that you are also correct that NO's on some do arc/flash why some contacts arc/flash.only adds to the confusion.

Just ordered MOV's. Local electronics store didn't have them.

 

[H012]

I will check one to see.

This is how many I have done since may still have 4 more to do.Next year in May I have change relays again on various chassis. This doesn't count the ones that go bad during the year. So

 

[jim hardy]

why some contacts arc/flash.only adds to the confusion.

arc is somewhat dependent on where in the line cycle the contact opens
if contact happens to open at the instant current is zero it's painless, if it happens to open at instant current is max you'll get a healthy arc

so it's random whether you'll see an arc on opening your NO's, and how big
probably random on closing your NC's too,
statistically a sinewave doesn't spend very much of its time near zero
so you'll get arcing more often than not.

Any help ?

old jim

 

[jim hardy]

This is how many I have done since may

so is this

a NO contact whose center bar moves toward camera to close
or NC contact whose center bar moves away from camera to open ?
Sure looks like melted silver showing...

 

[jim hardy]

found an article that helps understand the MOV datasheet (which i'd say is a pretty good one)

communities.leviton.com/servlet/JiveServlet/downloadBody/2374-102-1-3690/Tech%20Note%20METAL%20OXIDE%20article.pdf

Thanks Tom,G

 

[H012]

so is this

View attachment 104011

a NO contact whose center bar moves toward camera to close
or NC contact whose center bar moves away from camera to open ?
Sure looks like melted silver showing...

The ones you see on pic are NO. All 4 on top are NO. The NC are on the bottom. Very hard to see. I wish I could get a pic.It's not uncommon for me to take one of the enclosed relays apart and one of the NC's to be completely burnt off. Not just the contact surface but the whole base the contact is attached to.

found an article that helps understand the MOV datasheet (which i'd say is a pretty good one)

communities.leviton.com/servlet/JiveServlet/downloadBody/2374-102-1-3690/Tech%20Note%20METAL%20OXIDE%20article.pdf

Thanks Tom,G

Thanks Tom/ Jim I'm getting a lot of good info. I liked how you described line cycle. Me with my lack of in-depth knowledge that it could effect contacts that much. Although I do understand what line cycle I made this drawing maybe it will help to understand the sweep schematic.

 

[jim hardy]

I made this drawing maybe it will help to understand the sweep schematic.

thanks, helps me for it seems to agree with the schematic and that's what I've been working in my head.. It quiets those nagging doubts when two drawings agree.

 

[H012]

thanks, helps me for it seems to agree with the schematic and that's what I've been working in my head.. It quiets those nagging doubts when two drawings agree.

Jim on that drawing. The wire on 44 of relay
That I got noted as going to power the primary on T 1 transformer this draws 8 amps. Is this added load a problem for the contacts considering the load already on the relay.

 

[Tom.G]

Since Jim seems incommunicado at the moment, I'll give it a shot.

Jim on that drawing. The wire on 44 of relay
That I got noted as going to power the primary on T 1 transformer this draws 8 amps. Is this added load a problem for the contacts considering the load already on the relay.

According to your drawing, no, the 8Amp load of T1 does not go thru the relay at all. That current will come from C1-41 HOT to the screw terminal 14, thru the wire that connects screw terminals 14 and 44, then to T1.

 

[H012]

Since Jim seems incommunicado at the moment, I'll give it a shot.

According to your drawing, no, the 8Amp load of T1 does not go thru the relay at all. That current will come from C1-41 HOT to the screw terminal 14, thru the wire that connects screw terminals 14 and 44, then to T1.

OK thanks. Just looking at any possibilities.
I should paid more attention. Kind of goes with common sense. Hopefully MOV's will be here Monday

 

[jim hardy]

The wire on 44 of relay
That I got noted as going to power the primary on T 1 transformer this draws 8 amps. Is this added load a problem for the contacts considering the load already on the relay.

doesn't look like it to me

it just uses the relay for a hopping point, doesn't go through the contacts

what does it power ?

edit i see we crossed in the mail
and we agree !

 

[Tom.G]

I should have pointed out that the MOV I suggested (V20E130P) is for testing purposes. It was chosen for 'just enough' repetitive power capacity and a low price to see if it works. If it solves the problem you should really use the next size up in power handling. The bigger ones are chassis mount, cost around $10 each, and should last for the life of the mechanical stuff. Well worth it if they are indeed a fix.

 

[Tom.G]

Hopefully MOV's will be here Monday

And we continue with Whack-A-Mole on Tue?

 

[jim hardy]

Could one parallel MOV's ? How abrupt is their turn-on ?
Bad practice for Zeners i know, but on these I'm still a tenderfoot.

 

[Tom.G]

Could one parallel MOV's ? How abrupt is their turn-on ?

No, can't parallel 'em. Due to tolerances, one ends up doing most of the work, and fails. (They fail shorted. Then either melt their connections or burn, as in fire.) Don't know their turn-on time off hand but it is fast, will investigate, they are spec'd to clamp ESD impulses of 8us risetime.

 

[jim hardy]

No, can't parallel 'em. Due to tolerances, one ends up doing most of the work, and fails.

gotcha. Same as paralleling zeners or fuses, "Just say No" .

 

[H012]

No, can't parallel 'em. Due to tolerances, one ends up doing most of the work, and fails. (They fail shorted. Then either melt their connections or burn, as in fire.) Don't know their turn-on time off hand but it is fast, will investigate, they are spec'd to clamp ESD impulses of 8us risetime.

Here is properties of MOV see pic.

 

[Tom.G]

How abrupt is their turn-on ?

Under 1nS.

For Engineering and selection details see:
http://www.littelfuse.com/~/media/e...s_terminology_and_theory_application_note.pdf

http://www.littelfuse.com/~/media/e...ng_a_littelfuse_varistor_application_note.pdf

 

[H012]

Under 1nS.

For Engineering and selection details see:
http://www.littelfuse.com/~/media/e...s_terminology_and_theory_application_note.pdf

http://www.littelfuse.com/~/media/e...ng_a_littelfuse_varistor_application_note.pdf

The ones I ordered will work?

 

[jim hardy]

The ones I ordered will work?

I think they will based on what little we know about the motor.

Test will be to feel of it after a few sweep cycles
if it's staying nice and cool i think it'll be fine

if it gets hot we need a bigger one

TomG and i are studying those curves in datasheet
depending on what assumptions we make predicted life comes out between a few hundred cycles and forever.

When it's that close, one test is worth a thousand opinions !old jim

 

[H012]

I think they will based on what little we know about the motor.

Test will be to feel of it after a few sweep cycles
if it's staying nice and cool i think it'll be fine

if it gets hot we need a bigger one

TomG and i are studying those curves in datasheet
depending on what assumptions we make predicted life comes out between a few hundred cycles and forever.

When it's that close, one test is worth a thousand opinions !old jim

Thanks Jim/Tom. I hope I'm not taking to much time away from helping others.

 

[Tom.G]

Heck, we do this (hope you don't mind me speaking for you, Jim) to relieve the boredom of retirement.
And we're having a ball on this one. It's getting us to actuallythink again!

 

[H012]

Heck, we do this (hope you don't mind me speaking for you, Jim) to relieve the boredom of retirement.
And we're having a ball on this one. It's getting us to actuallythink again!

LOL! I can't see myself just vegitating either.

 

[jim hardy]

yes I'm learning about fault current contribution from induction motors

We have long known that power engineers have to account for it in short circuit studies, but I'm no expert in those.
and i have taken oscillograms of bus voltage in the power plant during a transfer, those big pump motors (7000 horsepower for boiler feed and 6000 horsepower for reactor coolant) act as generators because the rotor field takes many cycles to decay away.

So
after your little motor has finished arcing the NO contact, which i would think is brief ,
its terminal voltage returns to some number not far from running voltage because the rotor field is still there.
So for the next several line cycles it is an induction generator making probably almost 120 volts.
Next thing that happens is your NC closes connecting the 1000 uf capacitors right across the motor main winding which is still generating ~ 120 volts because the rotor field hasn't collapsed yet.
What happens when you throw a completely discharged 1000 uf capacitor across a generator ? You get a surge of current.
How big a surge ? sixty four dollar question, I've never done that test.
After a lot of reading I'm finding not a real concensus, but the estimates that seem best presented say the surge is 2 to 3 times starting amps, and it's called "Subtransient Fault Current" and it lasts only a fraction of a cycle
starting amps for a 1/3 hp motor ought to be in the range of 5/3 to 7/3 KVA , which at 120 volts is 14 to 19 amps and you reported measuring 22 ..
so perhaps ~60 amps flows through your NC contact at the instant they close

Your resistor will knock down that fault current. I was afraid it'd stretch out motor coastdown but you report it's fine with the resistor.
So we tend to think that'll fix the NC burnup.

But -- that arcing at the NO contact ? How long does it persist ?
If that arc ever fails to go out before the NC contact closes, we then have a path from incoming hot through the arc then through the NC contact to that discharged 1000 uf capacitor. Incoming hot is probably capable of hundreds of amps.
Might that explain your "Gone" NC contacts , unlucky outcome of a relay contact race every now and then ?
Your 8 ohm resistor should limit fault current for that scenario too.

The MOV Varistor will help Mr NO contact extinguish the arc quickly.
and the 8 ohm resistor limits fault current from either source, subtransient from motor or direct from line through arcing NO contact

and diverse methods of protection is good design.

Guess I'm more recapping for my own clarity of thought than anything else
does this help?

Corrections welcome.

Now to figure out how big a varistor we need. ... your "touch test" will be imoprtant.

 

[H012]

yes I'm learning about fault current contribution from induction motors

We have long known that power engineers have to account for it in short circuit studies, but I'm no expert in those.
and i have taken oscillograms of bus voltage in the power plant during a transfer, those big pump motors (7000 horsepower for boiler feed and 6000 horsepower for reactor coolant) act as generators because the rotor field takes many cycles to decay away.

So
after your little motor has finished arcing the NO contact, which i would think is brief ,
its terminal voltage returns to some number not far from running voltage because the rotor field is still there.
So for the next several line cycles it is an induction generator making probably almost 120 volts.
Next thing that happens is your NC closes connecting the 1000 uf capacitors right across the motor main winding which is still generating ~ 120 volts because the rotor field hasn't collapsed yet.
What happens when you throw a completely discharged 1000 uf capacitor across a generator ? You get a surge of current.
How big a surge ? sixty four dollar question, I've never done that test.
After a lot of reading I'm finding not a real concensus, but the estimates that seem best presented say the surge is 2 to 3 times starting amps, and it's called "Subtransient Fault Current" and it lasts only a fraction of a cycle
starting amps for a 1/3 hp motor ought to be in the range of 5/3 to 7/3 KVA , which at 120 volts is 14 to 19 amps and you reported measuring 22 ..
so perhaps ~60 amps flows through your NC contact at the instant they close

Your resistor will knock down that fault current. I was afraid it'd stretch out motor coastdown but you report it's fine with the resistor.
So we tend to think that'll fix the NC burnup.

But -- that arcing at the NO contact ? How long does it persist ?
If that arc ever fails to go out before the NC contact closes, we then have a path from incoming hot through the arc then through the NC contact to that discharged 1000 uf capacitor. Incoming hot is probably capable of hundreds of amps.
Might that explain your "Gone" NC contacts , unlucky outcome of a relay contact race every now and then ?
Your 8 ohm resistor should limit fault current for that scenario too.

The MOV Varistor will help Mr NO contact extinguish the arc quickly.
and the 8 ohm resistor limits fault current from either source, subtransient from motor or direct from line through arcing NO contact

and diverse methods of protection is good design.

Guess I'm more recapping for my own clarity of thought than anything else
does this help?

Corrections welcome.

Now to figure out how big a varistor we need. ... your "touch test" will be imoprtant.

I was thinking a lot about the 1000 micro farad's that's a lot! Necessary I don't. Know. Been doing some research and far as small HP induction motors go and braking there isn't much out there, a lot of crap about. Induction motors as generators. Putting that much capacitor charge on a 1/3 HP motor seems very inefficient! In my opinion.

 

[jim hardy]

you could experiment and find out at what capacitance it no longer slows down quickly enough to park where you wnat it. Start capacitors cost around 3 to 8 bucks at my local parts house.

My guess is 1000 uf is way more than you need at 60 hz but what about when motor slows down and is generating 20 hz or 10 ? Without knowing how they chose 1000 uf i'd have to start by experiment.

...... in the spirit of ruling out what seems unlikely ...

Can you think of an experiment that would rule out possibility of that 'unlucky outcome of a relay contact race" ? You're a mechanical, so you are familiar with electric welding and arc voltage, freeze frame your thinking single step it through ...

Got a spare motor and relay you could put on the bench ?

..........

Well your varistors will be here soon and that's a good experiment, and good next step. Use all your senses, observe arc intensity, listen to motor, and feel the varistor's warmth.
I think arc intensity should vary randomly depending where in the line cycle contact opens.
Look for nuances like - arc is big at NO when it's small at NC ; arcs same size all the time, ; one white one blue,, you never know what little thing will turn into an AHA ! moment .
I like to have my long-shots proven wrong because then i quit worrying about them.

old jim

 

[H012]

you could experiment and find out at what capacitance it no longer slows down quickly enough to park where you wnat it. Start capacitors cost around 3 to 8 bucks at my local parts house.

My guess is 1000 uf is way more than you need at 60 hz but what about when motor slows down and is generating 20 hz or 10 ? Without knowing how they chose 1000 uf i'd have to start by experiment.

...... in the spirit of ruling out what seems unlikely ...

Can you think of an experiment that would rule out possibility of that 'unlucky outcome of a relay contact race" ? You're a mechanical, so you are familiar with electric welding and arc voltage, freeze frame your thinking single step it through ...

Got a spare motor and relay you could put on the bench ?

..........

Well your varistors will be here soon and that's a good experiment, and good next step. Use all your senses, observe arc intensity, listen to motor, and feel the varistor's warmth.
I think arc intensity should vary randomly depending where in the line cycle contact opens.
Look for nuances like - arc is big at NO when it's small at NC ; arcs same size all the time, ; one white one blue,, you never know what little thing will turn into an AHA ! moment .
I like to have my long-shots proven wrong because then i quit worrying about them.

old jim

True. As this motor calls for a. 250uf I'll try 2 of those.

 

[Tom.G]

Putting that much capacitor charge on a 1/3 HP motor seems very inefficient!

They are using the caps as a load to brake the motor, not as a charge source.
BTW, the 1000u caps with the 10mH motor inductance have a resonant frequency of 50Hz, suspiciouly close to Line frequency. Don't know exactly why, but the designer probably had a reason.

 

[Tom.G]

True. As this motor calls for a. 250uf I'll try 2 of those.

But two 250uF in series yield 125uF, not what the motor wants. The present circuit has the caps in series during Start and in parallel for Brakeing. The motor may not start under load with only 125uF.

 

[H012]

But two 250uF in series yield 125uF, not what the motor wants. The present circuit has the caps in series during Start and in parallel for Brakeing. The motor may not start under load with only 125uF.

I understand. It probably won't start. Motor plate does say 250 uF.

 

[H012]

About 6 yrs. Ago I took one of these motors with 5 20 uF run caps to just see what this motor would produce as a generator. The best it will do is 40 volts at 1500 rpm if I tried to push the motor any higher it experienced locked rotor killing the 3 HP gas motor. Just wanted to share.

 

[jim hardy]

this is how we learn !

have at it.

 

[H012]

this is how we learn !

have at it.

Yes I did it more to add more proof to tell others not to believe everything on the internet.

 

[Tom.G]

If you want to check the need for motor braking versus coasting you could just disconnect the NC contacts. Just be prepared to kill power if something goes wrong.

Still hoping for some video of the situation. If that's not possible detailed observation and notes will have to do. Of course some 'scope photos of both voltage and current at both the NC and NO contacts is the Gold standard. Probably not practical in your present environment though.

I'm beginning to feel like we are trying to shoot fish in a barrel... blindfolded and we don't know where the barrel is!

 

[jim hardy]

About 6 yrs. Ago I took one of these motors with 5 20 uF run caps to just see what this motor would produce as a generator. The best it will do is 40 volts at 1500 rpm if I tried to push the motor any higher it experienced locked rotor killing the 3 HP gas motor. Just wanted to share.

i seem to remember a similar experiment described someplace
i wonder what went on there ? A clamp around ammeter on motor lead would be interesting... and voltage across cap

 

[H012]

If you want to check the need for motor braking versus coasting you could just disconnect the NC contacts. Just be prepared to kill power if something goes wrong.

Still hoping for some video of the situation. If that's not possible detailed observation and notes will have to do. Of course some 'scope photos of both voltage and current at both the NC and NO contacts is the Gold standard. Probably not practical in your present environment though.

I'm beginning to feel like we are trying to shoot fish in a barrel... blindfolded and we don't know where the barrel is!

I did disconnect NC circuit somewhere at beginning of our posts. Anyway the NO braking causes the sweep to come down to hard on the bowling lane also the momentum of of the motor and gearbox causes the sweep to keep moving a couple of inches after hitting the lanes causing sweep and table to be out of time.

 

[H012]

At this point guys because of not being able to give you guys enough info due to a lack of scope readings ect. I can try the MOV's and resistor and leave it at that.You guys have been more than helpful.

 

[jim hardy]

At this point guys because of not being able to give you guys enough info due to a lack of scope readings ect. I can try the MOV's and resistor and leave it at that.You guys have been more than helpful.

i remember you said early on you didnt want a research project

If those (edit -) varistors reduce arcing at NO contacts then you will have addressed both of the things i worried about,
1. transient current capability of motor exceeding NC contact rating during bounce period
2. persistence of arc at NO causing contact overlap maybe one out of ten braking cycles, that should show as a really big arc on both contacts at same time but only once in a while.

svein's resistor helps with both 1 and 2, limits fault current
varistor addresses 2, arcing at NO and removes energy from motor during contact flight time which might help with 1 as well

i really like the combination
and i hope you write an article for Bowling World or whatever is your trade journal.

so - let us know what you see and if the varistors stay cool.

thanks,

old jim

 

[Tom.G]

A few of thoughts if you continue to have contact life problems.

I can't tell if the contacts on the relays you are using can be replaced without replacing the relay. If not, then a lower cost approach may be to use either front mounted or side mounted contacts to avoid replacing the whole relay. Maybe even for just the NC contacts since they seem to cause the most trouble.

Another alternative is use a larger, higher rated, relay.

Regardless of results and whatever followup you do, we would love to be kept up to date. Be sure to check in on progress & attempts. And of course any article you decide to write!

If you are using the same relay on most of the alleys, try getting a Field Engineer from either the manufacturer or the outfit from which you buy the relays. If that doesn't work try getting a Field Engineer from a different relay manufacturer, I bet they would be delighted to steal a competitors business!

 

[H012]

A few of thoughts if you continue to have contact life problems.

I can't tell if the contacts on the relays you are using can be replaced without replacing the relay. If not, then a lower cost approach may be to use either front mounted or side mounted contacts to avoid replacing the whole relay. Maybe even for just the NC contacts since they seem to cause the most trouble.

Another alternative is use a larger, higher rated, relay.

Regardless of results and whatever followup you do, we would love to be kept up to date. Be sure to check in on progress & attempts. And of course any article you decide to write!

If you are using the same relay on most of the alleys, try getting a Field Engineer from either the manufacturer or the outfit from which you buy the relays. If that doesn't work try getting a Field Engineer from a different relay manufacturer, I bet they would be delighted to steal a competitors business!

I will know more when MOV's get here. UPS got held up due to the flooding North of me. Tracking information shows Wednesday or Thursday.

 

[jim hardy]

I bet they would be delighted to steal a competitors business!

You should've seen the stir it caused when we replaced Copes Vulcan feedwater valve actuators with Fisher...

 

[Tom.G]

Yeah, when all else fails, call in the competitors!
I suspect that was a larger financial hit for someone than a couple dozen relays.

 

[H012]

Yeah, when all else fails, call in the competitors!
I suspect that was a larger financial hit for someone than a couple dozen relays.

Yea I was doing a little research on these relays. They all use basically the same contact material. I looked at the next step up in relays and size mainly and the big increase in price to changing type of relay off the table.

 

[jim hardy]

your fixes might let a SSR survive, if the owner will let you experiment

offer him co-authorship on your article ?

 

[Tom.G]

Just as an FYI, this popped up in a different thread referring to shunt wound motors. Seemed like it would also fit here.

Back EMF = (1/2) * (angular speed) * (number of turns) * (rotor radius) * (average magnetic flux) * (conductor length)