What's new
What's new

Solid State Relays Instead of Contactor?

rpseguin

Stainless
Joined
Jun 28, 2006
Location
Napa, CA
My interlocking contactor on my 25HP RPC died this morning with a bright flash and a puff of expensive smoke.
It had been working fine.

I’m wondering if I could use ganged solid state relays (SSRs) instead of contactors.

Upsides?
Downsides?
 
What caused the bright flash and expensive smoke? What is your system voltage? What is the interlocking for, if it is to disconnect the power system from the RPC, then be aware that SSRs are not suitable for this, they need to be backed up with a mechanical disconnect.

Tom
 
My interlocking contactor on my 25HP RPC died this morning with a bright flash and a puff of expensive smoke.
It had been working fine.

I’m wondering if I could use ganged solid state relays (SSRs) instead of contactors.

Upsides?
Downsides?

You don't want to be the one trying to "gang" them. Let the factory package what you need, test it, and rate it FOR your needs. Buy what suits.

Been replacing mechanicals with Crydoms since the early '70's. Loveliest part on AC circuits is that they inherently operate right at the "zero cross".

Mind - with an INDUCTIVE load, that ain't quite the whole story!

Even so, the makers know their s**t, so have a read of their tech literature and examples / case reports / "white papers" for real-world applications.

Legally, they require a hard disconnect in series as they DO "leak" in the OFF state (not hardly - figures are in the specs..) but more importantly CAN fail, shorted, if only rarely. Failed "open" is more likely, and fail DAMNED seldom at all more likely, yet.
Even so..

'Nuther thing is that in high cyclic-rate use, they may also need heat-sinks. Seriously LARGE ones in some apps!

More-often with infrequent cycling, an Aluminium flat plate they are thermally-pasted to is plenty.

Read up.

'bout half a century of experience with them by now, and several makers besides the pioneering IRC-Crydom in the field, Europe to Asia..

I like them well-enough I am never without a stash of both AC and DC types to-hand. My next-most favorite, Mercury-Displacement contactors.

Those need more finesse. Rather a lot of it, so I'll stop short of recommending them.
 
What caused the bright flash and expensive smoke? What is your system voltage? What is the interlocking for, if it is to disconnect the power system from the RPC, then be aware that SSRs are not suitable for this, they need to be backed up with a mechanical disconnect.

Not sure what caused it.
System was working fine for some time.

Line/input voltage is 246V, 60Hz

The interlocking is for the “soft start” mechanism using a 3ph transformer that I also use as a buck transformer for my 400V German lathe
 
You don't want to be the one trying to "gang" them. Let the factory package what you need, test it, and rate it FOR your needs. Buy what suits.

Ok.
For clarity, by ganging, I mean switching different legs simultaneously, NOT trying to run the same line through multiple SSRs



Been replacing mechanicals with Crydoms since the early '70's. Loveliest part on AC circuits is that they inherently operate right at the "zero cross".

Mind - with an INDUCTIVE load, that ain't quite the whole story!

Even so, the makers know their s**t, so have a read of their tech literature and examples / case reports / "white papers" for real-world applications.

Legally, they require a hard disconnect in series as they DO "leak" in the OFF state (not hardly - figures are in the specs..) but more importantly CAN fail, shorted, if only rarely. Failed "open" is more likely, and fail DAMNED seldom at all more likely, yet.
Even so..

'Nuther thing is that in high cyclic-rate use, they may also need heat-sinks. Seriously LARGE ones in some apps!

More-often with infrequent cycling, an Aluminium flat plate they are thermally-pasted to is plenty.

Read up.

'bout half a century of experience with them by now, and several makers besides the pioneering IRC-Crydom in the field, Europe to Asia..

I like them well-enough I am never without a stash of both AC and DC types to-hand. My next-most favorite, Mercury-Displacement contactors.

Those need more finesse. Rather a lot of it, so I'll stop short of recommending them.


I’ll read up.
 
I’m wondering if I could use ganged solid state relays (SSRs) instead of contactors.

Upsides?
Downsides?

Are you referring to those small bricks that are 2.5 x 1.5 inches? Ganging 3 together on a hefty heat sink will take up some room. And will be a experiment that
will probably turn out not so great. Follow advise in post #2. When they fail they stay shorted.

I have a few of those bricks, the one in hand is rated for 10 Amps. I looked at DigiKey and they refer to them as hockey pucks.
Here is a 3 pole puck rate for 50A. You will need a hefty one for a 25Hp motor.

Internal Error

If you search for "Solid State Relay vs Mechanical Relay" don't rely on the content that comes from places that sell SSR's.
 
Ok.
For clarity, by ganging, I mean switching different legs simultaneously, NOT trying to run the same line through multiple SSRs

I’ll read up.

Look at the case use of Crydom SSR's to implement, for example, the controlled "H-Bridge" for managing servo drives.

Just checked for when it was I had first designed them into office automation gear.

1973.

For something the young amateurs not yet born when first we already more expert and mature were vettng, testing, and had begun putting into widespread use?

The best rejoinder is how many tens of millions of them have been quietly beavering away "JFDI" all over planet Earth, the 45 and more years SINCE, yah?

AKA "What the f**k would a plumber be expected to know about fine Irish lace?"

:D

"Quiet", electrically, no arcs, no sparks, to hassle sensitive logic circuitry, adjacent is precisely why they were first chosen, y'see. That "zero-crossing" point of switching of a sine wave.

That far back, it wasn't even CMOS living in the same chassis. Nuisance-level "TTL", rather!

NOW? Kinda nice to not have arcs and sparks in a shop where flammable vapours may now and then arise by accident, even if not in the normally planned course of operations.

:)
 
Some days I'm not really sure what Bill is trying to say... well... most days. I'm certain for one thing- he knows what HE's talking about... the ancillary stuff is what we kinda hafta sort through...

RP-

When contactors are interlocked, it's so that there's a very intimately physical relationship between one side, and two others, and the mechanism between assures that the isolated branches DO NOT meet. Most of the time, that mechanism is such that, if contacts happen to get welded on one side, that it will NOT be able to engage the OTHER, and have much worse things happen.

You could put a pair of SSRs in there, but as Bill and others noted, the failure modes available to the SSR are dead shorts, which being non-mechanical, have no way of being a 'protected interlock'.

If it was me, and I had other reasons to pursue SSR, I would proceed cautiously. If the reasons were substantial enough, I would probably replace the interlocking contactor assembly, and then place SSRs in each leg, then STAGE the activation so that the contactor is thrown, THEN the SSR is gated on... then off... so that the contactor is not subjected to energized switching...

But I would certainly find out why the failure occurred in the first place.
 
Some days I'm not really sure what Bill is trying to say... well... most days. I'm certain for one thing- he knows what HE's talking about... the ancillary stuff is what we kinda hafta sort through...

RP-

When contactors are interlocked, it's so that there's a very intimately physical relationship between one side, and two others, and the mechanism between assures that the isolated branches DO NOT meet. Most of the time, that mechanism is such that, if contacts happen to get welded on one side, that it will NOT be able to engage the OTHER, and have much worse things happen.

You could put a pair of SSRs in there, but as Bill and others noted, the failure modes available to the SSR are dead shorts, which being non-mechanical, have no way of being a 'protected interlock'.

If it was me, and I had other reasons to pursue SSR, I would proceed cautiously. If the reasons were substantial enough, I would probably replace the interlocking contactor assembly, and then place SSRs in each leg, then STAGE the activation so that the contactor is thrown, THEN the SSR is gated on... then off... so that the contactor is not subjected to energized switching...

But I would certainly find out why the failure occurred in the first place.

If the need for interlocking is to prevent a short circuit as in reversing starters, and if that is what failed, then replace the failed unit with a like unit. Forget the SSR as the contactor is capable of starting stopping the machine.

But as Dave mentioned do find out why there was a failure.

Tom
 
Some days I'm not really sure what Bill is trying to say... well... most days. I'm certain for one thing- he knows what HE's talking about...

Ah, well.. I AM about to take another "test bench" gamble. Goal is a modern-goods 10EE "DC panel" workalike.

But neither my beloved Crydoms not Mercury displacement relays, this go.

Figure if the OEM panel was good for as much as a hundred years, given how few have failed since 1941....

Any experience with Gigavac contactors out there in PM-land?

HX22 series looks good.

Only downside is being made with slave labour in a Communist country.
 
When contactors are interlocked, it's so that there's a very intimately physical relationship between one side, and two others, and the mechanism between assures that the isolated branches DO NOT meet. Most of the time, that mechanism is such that, if contacts happen to get welded on one side, that it will NOT be able to engage the OTHER, and have much worse things happen.

Yes, I understand that.


You could put a pair of SSRs in there, but as Bill and others noted, the failure modes available to the SSR are dead shorts, which being non-mechanical, have no way of being a 'protected interlock'.

If it was me, and I had other reasons to pursue SSR, I would proceed cautiously. If the reasons were substantial enough, I would probably replace the interlocking contactor assembly, and then place SSRs in each leg, then STAGE the activation so that the contactor is thrown, THEN the SSR is gated on... then off... so that the contactor is not subjected to energized switching...

But I would certainly find out why the failure occurred in the first place.

Yeah, I’m going to pull the contactor out this weekend and inspect it, look for where it is damaged.
It has 115V AC coils, so I’ll energize each side and see how each side works.

I am using the interlocking contactor to reverse the flow through my 30KVA transformer, as a soft start mechanism for spinning up my 25HP motor.
I use a 3 position switch: (start/off/run) which is used yo energize the different sides of the contactor.
Once spun up, I switch from start through off and over to run.

My notion is to use 6 SSRs, 3 switched on for starting, switch those off and then 3 for running.
As to the control circuit: Keep my simple 3 position switch? PIC chip with timing? Potential relay?
 
If the need for interlocking is to prevent a short circuit as in reversing starters, and if that is what failed, then replace the failed unit with a like unit. Forget the SSR as the contactor is capable of starting stopping the machine.

But as Dave mentioned do find out why there was a failure.

Tom

Yeah, I’m going to pull the contactor out this weekend and inspect it, look for where it is damaged.
It has 115V AC coils, so I’ll energize each side and see how each side works.
I’ll try to figure out how this happened.

I may be able to find another similar contactor.
 
My notion is to use 6 SSRs, 3 switched on for starting, switch those off and then 3 for running.

I would not do this for your application.
As noted above SSRs never turn fully off and will always leak some power when in the off state. It's not much but enough to become a problem to worry about.
Added stuff could address that but makes a mess of things.
Mechanical relays easier. Nothing is getting through that open set of contact points.
The first "servo" type drive I tried to build as a youth had one transistor on the plus and another on the minus side. Lots of smokey lessons learned there around null.
I like using SSRs for life and reliability and most certainly use a lot of them but would never back them against another set.

You do not need 6 as you can buy 3phase SSRs but zero cross switching becomes a problem with such if you want it. A delay is needed which can be done with a handful of caps and resistors.

When anything from a car to a generator fails the important thing is to understand why before you go changing the parts or design.
Only when you feel you understand the original use and design trade offs better that the guy who built it try to do better.
Bob
(Just notes from the peanut galley and 0 cents worth so maybe all wrong and then forget you read this)
 
Many years ago I looked at vacuum contactors for use as a size 5 starter. For motor applications they have many disadvantages. Current chopping, short mechanical life, expensive, maintenance, difficult to mechanically interlock. I am sure since then that some or many of those problems have been solved. Vacuum contactors are very good on DC and very high voltage AC because there is no arc to extinguish. Other than solving a problem that doesn't exist, using them in an application that was solved by simple methods 80 years ago wouldn't make sense.

Tom
 
Yes, I understand that.

Yeah, I’m going to pull the contactor out this weekend and inspect it, look for where it is damaged.
It has 115V AC coils, so I’ll energize each side and see how each side works.

I am using the interlocking contactor to reverse the flow through my 30KVA transformer, as a soft start mechanism for spinning up my 25HP motor.
I use a 3 position switch: (start/off/run) which is used yo energize the different sides of the contactor.
Once spun up, I switch from start through off and over to run.

My notion is to use 6 SSRs, 3 switched on for starting, switch those off and then 3 for running.
As to the control circuit: Keep my simple 3 position switch? PIC chip with timing? Potential relay?

Could you post sketch of your circuit?

Tom
 
Yes, I understand that.
Yeah, I’m going to pull the contactor out this weekend and inspect it, look for where it is damaged.
It has 115V AC coils, so I’ll energize each side and see how each side works.

I am using the interlocking contactor to reverse the flow through my 30KVA transformer, as a soft start mechanism for spinning up my 25HP motor.
I use a 3 position switch: (start/off/run) which is used yo energize the different sides of the contactor.
Once spun up, I switch from start through off and over to run.

My notion is to use 6 SSRs, 3 switched on for starting, switch those off and then 3 for running.
As to the control circuit: Keep my simple 3 position switch? PIC chip with timing? Potential relay?

SSR's are done in arrays often enough, but the wrong goods for this tasking as 'raw' devices.

The issue is twofold:

Starting inrush.

In more complex circuits, the controls can command a slow-ramp, just as a VFD or DC Drive can do. That's as much to the benefit of their own "pass elements", be they SCR's or IGBT's that need to stay alive, as it is for the load being started.

Power interruption:

OK, the SSR itself cuts-off at a no-energy-flowing zero-cross. Or not. They make them as "randomized" ON and OFF also.

But a massive rotating inductor doesn't give a shit. Transformers? They ARE "inductors", yah? See Old Skewl spark ignition coils.

A rotating motor or a transformer's suddenly de-energized fields "collapsing" can still fling back kinetic energy as electricity AFTER that time-gate has passed.

With Dinosaur Current, our motors are really GOOD DC generators. We expect roughly FIVE times the Voltage excursion as the running Voltage at time of disconnect. Worse? There can be rather serious amounts of energy "under the curve", welding-grade arcs, not just teaser spikes. DC contactors have to be FAR more rugged than AC. They ARE the "zero cross".

"Four-Quadrant" DC Drives or VFD's, neither of which even USE contactors at all, survive and thrive because they ramp the power UP..and back DOWN to Zero, typically over a full two-seconds, each direction.

Neither tribe has to deal with an abrupt start OR termination of power..

Cheating, y'say? Dambetcha!

Works well, and lasts a long time, too!

Buuuuuuttt. it needs some "smarts" controlling those ramp UP/DOWN behaviour patterns.

Now.. a "mechanical" contactor, either of hard metal or Mercury, is inherently a "binary" device. Full ON. Or full OFF. No Fine Way it can "sneak up" on a damned thing.

BFBI. Brute Force and Bloody ignorance. Not as "ignorant" as first appears, either.

Arcs are going to happen? They simply plan for them. See "Arc chutes", "blowout coils", etc. All old stuff for waaay over a hundred years already.

All you really have to know is enough to select a stock contactor rated for what you will ask it to do.

Other folk have already earned their wage making sure it happens.

And by the tens of millions, world wide, if that part matters... Asea Brown-Boveri (ABB), Eaton Cutler-Hammer, Schneider Telemechanique? Old. Big. Still here.

Trust those big, Old Line, major makers. They got to BE big, old, and major, typically over a hundred years and more in THEIR game.... because their shit Just Faithfully Works.

If you want to INVENT sumthin'? Don't waste yer time on commodity contactors. Just buy them. Plenty of good ones NOS surplus.

Now .... that faster-than-light spaceship warp-drive?

That puppy.. I could use some help with... How much money d'you have spare?

:D
 
...

OK, the SSR itself cuts-off at a no-energy-flowing zero-cross.

Do they all do this? I think not. Actually I know not.
At 3 phase each crosses at a different point in time .
If not specked as such will it hill all three or sequence them? That leads you to a needed delay for the three times two before you hit the back side.
Yes it's voltage zero cross and not a power zero cross as the poor abused solid state relay has to make a decision based on something it can sense.
So we stuff in chokes and R/C timed networks to compensate. All of which makes the simple harder.

KISS. I do like the OP idea but so many devils in the details. It just gets so messy for long term working.
Why did it die is more my concern. I would expect 30 years out of the base setup.
I have reversing relay banks in my Blanchards that have WWII stamped all over on them. I would not even try to take the old wire off of these things.
Bob
 
I’ll post pictures and more details of the contactor after I’ve pulled it out and had a chance to inspect it.

I’m perplexed because it had been working great for over a year.

I’m going to start looking for another contactor.
 
Do they all do this?
Unless intentionally engineered otherwise, yes. The sales ballyhoo is just taking advantage of their inherent nature.

The "AC" ones are just packaged Triac-arranged bi-directional SCR bridges with built-in triggers generally run off a phototransistor, in turn coupled to an LED. ISTR the first ones I ever designed-in (signature machines) responded to a control voltage anywhere beteeen 3 V and 35 V. That feature was HANDY in its own right.

And yes, they offer "randomized" switching, too. Also DC versions as well as AC.

RTFM, and read most of all "representative circuits", white papers, and case studies, and.. well.. do yer homework. Properly applied, they are reliable goods.

Designer gets instant-gratification lazy? Not so reliable.

Lot of "lazy" in the wurld. Probably outnumbers stoopid?

Why should power electronics be immune if Boeing aircraft are not?
 








 
Back
Top