Potential relay noise in RPC build

Hi--been coming here from Google for a long time but hopefully I won't decrease the signal-to-noise.

I've been putting together a 10hp RPC from the 30hp integrated panel plans from this site ("/phaseconverter.pdf"). I've already tested it with the Unipoint saw I'm going to run and it works great.

The only quirk is with the potential relay. I got a $12 90-66 clone from eBay and have it wired as per the plans,

1 - one side of the coil of a 60 amp contactor for the start caps and extra L1/L2
2 - from a N-O switch, fed from L1
5 - the generated leg from the idler

As long as I hold in the N-O start switch for the motor, the relay works correctly and cuts off the start contactor after about a second. But when I release the switch, the potential relay starts making a racket, like it's opening and closing constantly.

Has anyone else had this issue? I don't mind just permanently bypassing the relay (I can hold a button in for one second) but it seems a little bit of a waste not to try to use it, and it would be a nice safety. Maybe a ferrite bead on one or more of the connections would help? If it were a DC circuit I would suggest a pull-down resistor since it seems kind of like a "floating input" problem. The relay is bolted to the panel and the panel is grounded.

Sounds to me like you have a condition where the voltage rises when it is open, the relay closes and the voltage drops, so it opens. Put a voltmeter on the relay coil or better yet, an oscilloscope.

Bill

Not knowing the circuit, is this relay intended to work with a holding contact or just voltage?

Tom

Good chance Bill (9100) has it nailed.

Thing of it is, the potential relays USUALLY have a large margin between the pull-in and the drop-out voltage, often nearly 100 volts, so there could be another issue causing the problem.

It is possible that the clone relay is (as many are) electronic instead of the original type with the magnetic coil as the voltage sensor, in which case it could have a much more narrow margin and might easily have the problem you describe. Electronic ones CAN be made to have any desired margin, so it might be defective, or maybe just not made for this type operation..

Are you starting the DPC into a load, or letting it start up and THEN connecting the load? They normally should start into no load and then have the load connected.

I've went to just using time delay relays. set the timer for about 1 or 2 seconds. That would solve your problem

https://www.amazon.com/Uxcell-a1411...078&s=gateway&sprefix=timer+re,aps,174&sr=8-9

As I have said a few times before, I don't understand why people use potential relays in this application when current sensing relays work so well. I took the coil off a standard Guardian relay and wrapped a few turns of 12ga enameled wire in its place, adjusted the spring for the range I wanted, and used it for a decade. The mill I had it on had a Dahlander type two speed motor and it coped with switching between speeds nicely.

Bill

Because they are harder to buy, and most folks do not want to be modifying relays and testing them. Some of us can do the test and get it right, most probably cannot, or do not want to.

Agreed it is a better method overeall.

When it is all said and done, the timing relay is the simplest.

Tom

Thanks for the replies. Tom, to your question re: the circuit--I don't understand, but I think "just voltage" ?

I am only starting the idler motor with no load when the behavior occurs.

I have the relay mounted kind of close to one of the contactors so maybe the magnetics are interfering? I'll test that and put a scope on it tonight.

Another fix I'm considering is to swap the N-O SPST switch with a DPDT switch, and just switch more than one of the relay inputs (probably the currently switched line and the line from L3).

I probed terminals 1 and 5 with a scope (corresponding to the contactor coil, and the generated leg). One thing I hadn't noticed, and it gave me a little tingle with the main input contactor turned off, is that 120v leaks through that contactor's coil and onto the L1 and L2 bus. Terminal 1 (the starter contactor coil) showed 120v with little spikes of an additional 30v or so (I didn't look closely). Terminal 5 was just the generated leg and uninteresting. I didn't see any patterns longer than the AC wave. I think my magnetic interference idea is probably silly but I'll still check it later.

In a few minutes of looking I haven't found a cheap DPST NO+NO switch so maybe they don't make them. I could just add another of the NO+NC switches and press both at the same time I guess.

Lets assign some letters for talking purposes. Let T1 and T2 be the wires from the RPC starter and G for the generated leg. The timer is a time delay opening upon energization (TDOE) which means that as the timer is sitting on the bench, the contacts are closed. When the coil is energized, the contacts stay closed for the timing interval, then open and stay open until the coil is released. This is connected to T1 an T2. The contact is connected to T1 (or T2) and to then to startup contactor and then hence to the other T line.

Upon energizing the RPC, the closed contact operates the startup contactor, introducing the starting caps in the circuit. After a fixed time delay, the timer times out and opens the N/C contact which shuts off the startup contactor. The time out value of the timer is short, in the 1-2 second category. By adjusting the timing value, the optimum starting conditions are arrived at.

Tom

While the current relay is the "best" method or at least a very good one, the straight timing relay as suggested by Tdegenhart is simple and effective.

One very good feature is that it limits the time the start cap is connected , in a positive way. It simply cannot be "stuck on" unless the start contactor fails short. So if you use the typical start cap of electrolytic type, you know it will not prematurely be destroyed.

Problem is that if the motor does not start in that time, motor stalls. But for an idler, started with no load / small load, that should not be an issue. An overload sensor, which is needed anyway, should take care of the problem if the idler was to stall (or get overloaded in use).

The timing relay or contactor is wired to connect the start capacitor when power is applied... normally closed. Then the start cap is disconnected (relay opens) when the time runs out. Simple, and covers most all needs of this type.