I'm using a GE potential relay (Grainger item#6X550) to pull in the start caps for my phase convertor. Looking at the schematic, it's not clear to me if this is a single or double throw switch. If it's a double, the relay can be used to pull in a 3-pole contactor on the output side of the convertor when the generated leg's up to voltage. That seems like an easy way to protect against phase loss on that 3rd leg.
That relay, like most potential relays, is a single pole single throw switch which employs a double contact set to make and break. All it can do is disconnect your start caps when the generated leg comes up to sufficient voltage.
You may want to consider putting a pushbutton start switch in series with that relay, so that in order to start your converter, you must be holding down the pushbutton. If you leave it out, your converter will start 'automatically' BUT it can also attempt to restart whenever the generated leg voltage falls too low, causing a big voltage surge to your load machine which could damage its electronic controls. I know a seasoned machinist who lost a $1500 control board just this way.
A series pushbutton also gives you protection against automatic converter restarting in the event of a power interruption.
My plan for the RPC does have a NO pushbutton in the "cap circuit". The potential relay controls the coil on a contactor that the start caps run off of. The PB goes in between the potential relay and the coil. Hit the master start on the RPC which powers both the 2 single phases to the 3-phase distribution panel and also supplies 2 legs to the idler motor. Potential relay stays closes, hold the NC pb in, capacitor's contactor closes, when 3rd leg voltage comes up the potential relay opens, breaking the cap contactor's coil and the start caps are taken out of the loop. Release PB and walk away.
What I'm trying to prevent is single-phasing on the output side of the RPC. If something "breaks" on the RPC while I'm using it, I'd rather have the whole 3-phase panel shut down than to continue running on single phase. All the control circuits of my machinery run off of 2 base, single phase legs - loss of the 3rd leg won't shut the machines down.
So - any ideas on what's the easiest way to pull in a contactor when the generated leg is running in it's normal range? Is there a variation of a potential relay that closes as voltage rises into a range and opens when the voltage drops below the range? I know there are 3-pole contactors with phase-loss sensing, but those are pretty pricey and don't seem to turn up too often in the places I scrounge in or on ebay.
Thx - Rob
I am wondering about the starting sequence you described. It sounds to me as if you want first to hold down the NO pushbutton, then hit your RPC master start. With the design you describe, it will not matter how long you hold the pushbutton down, as the relay will automatically pull the start caps out of play. But you definitely want to have that pushbutton held down BEFORE firing things up, or you'll be starting with single phase only and it won't work.
It is VERY difficult to build your own phase loss protector. That's why the ones out there cost so much; many actually have miniature open delta transformers inside them, and I won't go into details unless you want. I basically gave up on trying to build a homemade phase loss protector which will operate in the general case of loss of ANY leg.
By far the easiest way to get cheap phase loss protection is to have a three pole common trip breaker protecting your machine, sized no larger than the nameplate amps on the machine. But that ONLY protects if the cause of the lost phase leg is a short or a sustained overload on your machine.
Basically, if you do the same thing on your converter idler motor, it too would shut down for similar problems. This is precisely why I have both a smaller run and a big start breaker for the idler motor in my integrated panel design; see
So if the SOURCE of the third leg dies, the idler stops, which hopefully tells the user to stop the load machine.
Without getting really complicated, that's about as good as it gets for reasonable $$$.
The way that I addressed the issue of loosing the manufactured phase is to put a 3 pole contactor on the "output" side of the phase converter circuitry that is energized by the manufactured phase (fed through a relay that is activated by the "run" signal from the starting circuitry). If I lose the manufactured phase, the feed to my machines goes dead. Because of the way that this is connected, as soon as I hit the STOP button, the output contactor de-energizes as well killing the feed to the machinery.
I recently got some 55 amp, 3 pole solid state relays on eBay that I may swap in in place of the contactor. A lot of 5 cost me $20 plus shipping.
The design you use is what I referred to in the last bullet point under "Limitations" in the reference above. Basically you have to split out the machine supply bus from the idler supply bus, which you have done with that contactor.
If you then wired the output from your contactor to a second distribution panel supplying power to your machines, and had a three pole common trip breaker for each machine in that panel, you would indeed have complete phase loss protection for your machines.
I probably didn't word it too well, but my question was about how to energize the coil on that "output side" contactor. I do have separate panels for 240v single phase power input and 3-phase output power distribution. The voltage on that 3rd leg won't be predictable until I get the whole RPC balanced and even then I'm guessing it can vary.
What's the easiest way to wire in that output contactor? Use one with a 240 volt coil and grab the generated leg plus one of the other 2 phases, knowing that's always going to be about 240v? If I lose 1 of the single phase legs, the idler's gonna die so I lose the 3rd leg or if the RPC just "breaks" I'll lose the 3rd leg. Either way, I'm assuming 120v isn't enough to hold a 240v coil. Does that work or is there a btter way?
With many types of relay coils, 120 volts IS enough to ‘hold’ a 240 volt coil, but not enough to actually engage it. To put it technically, 120 volts may be above the dropout voltage of the coil but below its pickup voltage.
There are three voltages you talk about when discussing potential relays:
The PICKUP voltage is the lowest voltage which fires or engages the contacts from the resting position; it is lower than the HOLDING voltage.
The HOLDING voltage is the highest voltage you can continually apply to the relay coil without overheating it.
The DROPOUT voltage is that voltage below which an engaged contact returns to the resting position; it is always lower than the PICKUP voltage.
You can take an ordinary contactor and make it into a functional potential relay by wiring an appropriately sized resistor (sized properly for resistance AND power dissipation) in series with the contactor coil. Or you can use a potentiometer. Either way, you can tailor the coil to have the pickup and dropout voltages you desire to do a given task.
In order to do the resistor calculations, you need to know the coil current or its impedance or equivalent series resistance (ESR), then by Ohm’s law you can calculate the resistance and power dissipation for the resistor to be placed in series with the contactor coil. Often this involves considerable trial and error, but you can always get where you want to be.
So, how might this actually look in a shop? Let’s say you did the work above to find the appropriate resistor to wire in series with the ‘output’ contactor coil. You wire it in series, and energize the contactor coil with, say, L2 and L3 (generated leg). In addition, you place some kind of toggle switch in series with the coil energizing circuit, so that you have a way to manually shut off the distribution panel even if there is no phase loss on any of the three phase legs.
I think you have it! If that distribution toggle switch is ON, but L3 fails, the panel is dead; ONLY if BOTH the toggle switch is ON, AND the generated leg is ‘good’, do your three phase machines get their power.
I put a 3 pole relay on the output of my RPC to protect against the 3rd leg dropout. I had a good relay, but it had a 120 VAC coil in it. I simply ran a neutral wire to the control box and connected it to one end of the relay coil and the 3rd leg to the other. It's close enough to 120 to lock in the relay and drops out if the 3rd leg dies.
I also wired in a NO switch in this circuit that has to be pressed to energize the relay and 3 phase output after the RPC is started. I'm using a 7hp motor for the converter running a 5hp air compressor. On my original setup, it would get dark in the garage if I forgot to turn off the air compressor before starting the RPC. When I started automating things, I added this relay and switch to protect against 3rd leg loss and starting the RPC with other machinery on.
One note, my conerter isn't balance yet and every once in a while under a hard start, the relay will chatter due to the 3rd leg voltage or current dropping low. I expect that with a little balancing, it will not happen.
The voltage between the generated third leg and the neutral ideally is 208 volts, which is actually closer to 240 than 120. However if your setup works OK without the relay coil getting too hot, there's no need to change anything.
I did the potentiometer in the start contactor coil circuit before and it didn't work good for me. In the junk box I had a goodly sized wirewound pot to use for this. The problem I had was the contactor sometimes sticking in. I think the problem was caused by the coil voltage not dropping to zero fast enough and "snapping" the contacts out rather than just enough to drop out. I figure a potetial relay is designed to compensate for this. I switched to a 2 second timer setup. Those start caps smoke if left in too long.
One factor which can account for what happened to you is either worn pitted contacts or a contactor rated too low for the current it must handle. It is more difficult for a contact set to 'break' than to 'make', and if the contactor does not drop out rather sharply you can get some welding of the contacts which causes a freeze. This is especially true with the starting capacitor circuit due to the extremely high current transients occurring during engagement of the starting capacitors.
You are right that use of a real potential relay to shunt out the start caps is highly preferable. That's what they're designed to do. But it is possible to use an ordinary contactor to drop out all phase legs as I described above to protect against either power loss or loss of the generated leg.
I've picked up a new 3-phase monitor on ebay for under $10 shipped. It senses missing leg (among other things) and will allow me to cleanly control a 3-pole contactor on the output side of the RPC. A simple toggle in series between with the monitor and the contactor coil will allow me to kill output to my 3-phase distribution panel and leave the idler running.
Another nice thing about the monitor is that it's a double throw switch. I can wire in an indicator light so that, if something happens in the RPC and the monitor kills the output when I'm not actually running anything, I'll know that by looking at the RPC control panel.