Issue with very low current contact

[SolarPrestige]

I have an issue with a transformer PLC style controller (specifically DRMCC-T3) that doesn’t see an oil pump contactor closing. The –T3 controller energises an oil pump contactor via a digital output and gets feedback via an auxiliary contact into a digital input that the contactor has operated. Now there are two of these auxiliary contacts, one connected to the –T3 at 125 Vdc, and one connected to our station PLC (GE Mk6e) at 24 Vdc. I can see that the contactor is closing via the station PLC, but the –T3 unit has too many occasions when it can’t see the contactor has closed, and does an automatic changeover to the other oil pump. It doesn’t matter whether the –T3 starts oil pump 1 or oil pump 2, same result can occur. We get a contactor failed alarm and an oil pump failed alarm. This has been going on for years.

I suspect that the issue is related to the contactor, the –T3 input, and the 125 Vdc rating. I have measured the operating current on the –T3 input and it runs at less than 0.5 mA which is very low. I checked the contactor auxiliary contact minimum rating which turned out to be 10 mA, so clearly we were running far below its stated minimum. I changed the contactor to one with 1 mA minimum rating on the auxiliaries, and with a sealed contact unit to keep the atmosphere out (IP67) in case this was affecting the contacts. Still not low enough, but a lot better than the 10 mA. I couldn’t find auxiliary contacts that will work at 0.5 mA and at 125 Vdc.
After this, one transformer was fault free for a few months, but eventually the issue came back. The other transformer went quickly back to faulting again.

Does anyone have any suggestions on what to try next?

 

[jraef]

I had a similar issue with low energy circuits on DC traction elevator controllers that were turning on large contactors, but in my case it was because the feedback circuits were at TTL levels (5VDC) direct into their motherboard and only about 0.4mA. I tried every type of contact available to me; sealed reed contacts, gold flashed, bifurcated, waffle cross-hatched, combinations of the above, nothing worked reliably and the elevator controller mfr would not change their circuit. So what we ended up doing was putting 3 aux contacts sets (using the sealed reed contacts) in parallel on the same contactor, all three on the same circuit. When I left that project a year later, they hadn't had any more failures.


" We are all here on earth to help others; what on earth the others are here for I don't know." -- W. H. Auden

 

[itsmoked]

Lots of cures for this.

If the setup is feeding the 24V to the contact and from the contact to the PLC then the easiest solution is to put a 1W or 5W 1,200Ω resistor from the PLC side of the AUX contact to the 24Vdc return. That loads the contacts up with 20mA of oxidation busting current.

If that's not how it's wired tell us how it is.

You can also put a 1uF cap rated at 50+V to ground instead and as the points close and bounce the cap looks like a dead short and the oxidation is blasted off.

Keith Cress
kcress -

http://www.flaminsystems.com

 

[waross]

You beat me to it Keith. Load it with a resistor.

Bill
--------------------
"Why not the best?"
Jimmy Carter

 

[analogkid2digitalman]

You beat me to it Bill, had a recent design where I knew in advance I would need a load resistor.

-AK2DM

~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
"It's the questions that drive us"
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~

 

[waross]

Just do it and don't talk about it.
That part of the environmental crowd that sees any waste as bad with no consideration of actual values will be distressed by such a wanton waste of energy.
You will be wasting over 9 milli-Watts! grin


Bill
--------------------
"Why not the best?"
Jimmy Carter

 

[SolarPrestige]

Thanks for the posts. I guess the idea of three contacts in parallel is that you hope at least one is working well. It does reduce the current further by dividing it. Bit like home-made divided contacts.

Regarding the comments about a loading resistance or capacitance. The contact that I am struggling with operates at 125 Vdc, not 24 V. The manufacturer of these digital inputs has advised that the input impedance is 500k-ohm which suggests a current at 125 V of 0.25 mA, but we measured a bit more than this at about 0.4 to 0.5 mA, so maybe the impedance drops after it has closed. Would these details change your recommended resistor size and/or capacitor size?

I have also thought about converting the inputs to 24 Vdc, where low current contacts are easier to find. It's the 125 Vdc that is not easy to match.

 

[waross]

I still remember the contact section of my on Solid State Logic Control. (This went obsolete when PLCs arrived.)
The instructor recommended using higher switching voltages and an input module to match the signal to logic levels.
He claimed that tests had shown that about 100 Volts was required to pierce the contact film.
Some switch contacts had a much more aggressive scrubbing action than others. The best double break contacts had a simple, dependable action to slide the contact a couple of thousandths of an inch under spring pressure after contact was made with the fixed contacts.

It's the 125 Vdc that is not easy to match.

Buy a resistor. Let us know how it works.
You may try a variable resistor to find the resistance required and then divide by two or three.
Hey, forget the environmentalists, divide by five or more.

Bill
--------------------
"Why not the best?"
Jimmy Carter

 

[zappedagain]

Above 60V you are starting to get in the area considered dangerous, so don't run 20 mA as currently suggested. Drop it to the 1-5 mA range and you'll be in a safe range for unexpected finger-poking.

Z

 

[itsmoked]

Ahhh 125Vdc.. That's definitely a different size resistor. This is more interesting.

Definitely some crappy contacts you've got there if 125V is having troubles. General wisdom is that it takes 9V to deal with ALL oxidation levels so 12V logic thru relays is usually reliable. This is why 5Vdc doesn't work well at all and actually demands gold contacts for any kind of reliability. Since you're having trouble with 125V it's likely NOT oxidation causing you grief. You may have some other kind of contamination on your contacts, probably oil, coolant, or a dusty contaminant. If there is any possibility of that then my angle would be to clean up the air in your enclosure(s).

This limits things a bit. You can only get certain resistance values in the higher wattages. Now it becomes a find the value see-what-you-get. A value that you can get is 10kΩ. 125V/10k = 12.5mA which should wake up your 1mA contacts.

Use this:
10k 25W

It will waste a little power; 1.5W but that's not too bad.

The cap method, of course, doesn't have any steady-state power loss.
If you want to go with the capacitor method I'd try this one:
1uF 2000V







Keith Cress
kcress -

http://www.flaminsystems.com