Make relays hold position during UPS Switch-over.

[JohnLoves2Ski]

I have a UPS providing 120Vac to a transformer which converts it to 24VAC. This 24Vac circuit powers several Idec relays (50-80mA coil current each).
I expected them to stay in position when power goes out since it's on UPS, but I failed to recognize that the UPS I am using has a .004 second switchover time.
Is there a simple circuit method of holding this 24V AC voltage for a fraction of a second, so it will ride through a minor power interruption while the UPS switches?

I do HVAC PLC work primarily, and my electronics knowledge level is low.
I would really like to find an inexpensive solution that doesn't require me to purchase (7) new dual conversion UPS's @ $550 each... Can you help?
Thank in advance!
John

 

[IRstuff]

You could just hang some big capacitors on the 24 VDC line. Much depends on how much droop the relays can tolerate before breaking contact. Assuming 2V maximum droop and 0.5A total current for 10 ms results in 2500 uF of capacitance.

TTFN
faq731-376

Need help writing a question or understanding a reply? forum1529


Of course I can. I can do anything. I can do absolutely anything. I'm an expert!
There is a homework forum hosted by engineering.com:

http://www.engineering.com/AskForum/aff/32.aspx

 

[JohnLoves2Ski]

I wish it were that easy, but the circuits are all AC Voltage, not DC.

 

[catserveng]

Capacitors at each relay coil get used on a lot of equipment I work on, however our best results come from using buffer power supply modules, usually from either Phoenix Contact or Allen Bradley. Google "24VDC buffer module" and you will get a number of possible options.

MikeL.

 

[xnuke]

Find out the release time of the relay model you have - it's usually around 10-50 ms for small electromechanical ice cube relays. This might be worth testing to see if they open. If you aren't using solid state relays (which tend to have faster release times), they might be able to ride through the 4 ms switchover, especially since the transformer magnetic circuit will delay loss of voltage a little bit.

Another alternative is to use a constant voltage transformer, if you can find one for the voltages you require. It should keep the AC voltage up for a brief period while your UPS switches.

xnuke
"Live and act within the limit of your knowledge and keep expanding it to the limit of your life." Ayn Rand, Atlas Shrugged.
Please see FAQ731-376 for tips on how to make the best use of Eng-Tips.

 

[JohnLoves2Ski]

Again, it's all AC circuits. Don't think capacitors will help.
CV Transformers seem to be around the same cost as a double conversion UPS.

 

[waross]

Looks like the wrong choice of UPS.
Some possible choices:
1> Change out the UPS. A true UPS charges a battery and inverts off the battery all the time. In the event of a supply failure the battery simply stops charging. There is no changeover or changeover time and no need for a delay to determine that incoming power has been lost.
2> Change out the relays for relays that will ride through a 4 milli second interruption.
3> Put in a DC power supply, change out the relays for DC relays and hang some capacitors on the power supply.


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

 

[cranky108]

Or change out your relays for DC coils, and use a bridge rectifier/ capacitor to rectify the AC to DC at each relay.

Or change out the relays to ones that have a small dropout time delay.

 

[JohnLoves2Ski]

Re-wiring the equipment to use DC relays seemed daunting, but a lack of other low cost solutions made me re-evaluate.
I found a simpler way to swap 4 of the 5 relays to DC, install a DC power supply feeding the two circuits that operate these relays. The 5th relay can cycle without causing issues to controlled equipment.
The DC Power supply has a fat little capacitor on it that may solve my issue directly, if not I can add another one on the terminal strip.
Parts are on order, I will prove the concept in a few days.
Thanks everyone!

 

[IRstuff]

A UPS that continuously charges and inverts from battery to supply line power would make the both the charger and inverted run continuously, chewing up reliability hours. That means that there is a substantially higher probability that a failure will occur in the UPS during normal operations. That possibility has to be seriously weighed against switchover time issues. Note that few things will actually drop out with a 4-ms switchover, which is very short, compared to a typical UPS, and it's longer than most system ridethrough times. APC has a couple of white papers worth glancing at:

http://www.apcmedia.com/salestools/SADE-5TNM3Y/SADE-5TNM3Y_R7_EN.pdf?sdirect=true

http://www.apcmedia.com/salestools/JSII-5YQSBR/JSII-5YQSBR_R1_EN.pdf?sdirect=true

The second paper briefly discusses the myth of needing zero switchover time.

TTFN
faq731-376

Need help writing a question or understanding a reply? forum1529


Of course I can. I can do anything. I can do absolutely anything. I'm an expert!
There is a homework forum hosted by engineering.com:

http://www.engineering.com/AskForum/aff/32.aspx

 

[JohnLoves2Ski]

@IRstuff, APC's white papers usually relate to PC power supplies and switchover times. That's not what we have here. It is imperative that the relays ride through a brief power interruption. It's the difference between seamless operation of a laboratory HVAC system, and having to spend 45 minutes manually resetting equipment all over the building. Minor interruptions are common in this area, especially through the winter.
I think I've found a way to modify the safety circuits to use DC relays and add capacitance at the same time. Will be able to test it in a few days.

 

[VEBill]

UPS systems provide an interesting and informative case study in terms of reliability and failure.

Quoting myself from several years ago: "They're like having an unreliable guard dog that is high-maintenance, sleeps through a fraction of the burglaries, makes a huge mess once in a while, and every once in a long while eats one of the children."

It may seem counter-intuitive, but one should be extremely cautious about having a UPS included in any system where reliability is important.

 

[melspuds]

If this site is a laboratory, most likely they have a UPS already installed for computers and test equipment. If there is a UPS with solidstate transfer with no line voltage dip. If it is not 120 volts UPS just change transformer. Do not forget to install fuse(s) on the primary of the transformer.
Good luck,
Dave

 

[JohnLoves2Ski]

@Melspuds - We asked early on to be on the main UPS, they deemed it less expensive to have us provide small point of use units. I've used these many times before with great success, but never to hold safety relays. <oops>

@VE1BLL - I've never had a UPS eat one of the kids... You should consider procuring a better quality UPS, and changing the batteries more frequently.

 

[Skogsgurra]

A practical remark. Not to be overlooked.

If the guard dog isn't fed properly and is so hungry that it starts eating kids, it will be very easy to pacify it with a sausage or two. So, take care to feed the dog properly.

Gunnar Englund

http://www.gke.org

--------------------------------------
Half full - Half empty? I don't mind. It's what in it that counts.

 

[VEBill]

"You should consider procuring a better quality UPS..."

Excellent advice.
[sup](Sorry, you kinda lobbed it in there...)[/sup]

Anyway, it's clear that the DC relay approach is the best given the circumstances. Good luck.

 

[daveross100]

Here we go:

1-Add to each relay a: diode bridge AC/DC 1A-100V to the control coil, now each of the AC relay will act as an DC relay
2-Add to the output of each bridge a capacitor that its energy will hold the relay to the time needed,
Do you know to calculate the capacitors value?

Any question contact : daveross100@gmail.com

 

[Skogsgurra]

Sorry Dave, that doesn't work at all.

An AC coils needs the low impedance when core is open (to produce enough current to pull in). The impedance then increases so current is reduced. That prevents the coil from overheating.

With a rectifier and an AC coil, you wont get that. And if you add a resistor to reduce current, it won't pull in.

Gunnar Englund

http://www.gke.org

--------------------------------------
Half full - Half empty? I don't mind. It's what in it that counts.

 

[daveross100]

What about: Add parallel - resistor + capacitor + diode in series with the relay coil

 

[itsmoked]

BTW Dave we are not allowed to post emails in any threads!

Keith Cress
kcress -

http://www.flaminsystems.com