Problems with Voltage Sensing Relay

[kitestrings]

Well, we successfully programmed our Classic's (2- 250's), set up follow-me, set up the local network access, enabled, registered and connected the web access via mymidnite.com.  Pretty good for an old dog who doesn't know his Modbus from his baud rate.  You've nearly idiot-proofed things .

So, I'm having one pesky issue.  We incorporated an Omron voltage sensing relay in our rectifier box that is intended as a fail-safe to protect the CC's.  It is based on layout by Rob Becker's many of you may have seen.  There's more detail on the link below.  Briefly, when I energize the relay (24VDC) there is a delay, but then I start seeing a series of pulses flashing the relay & alarm LEDs (about 5-8 seconds apart).  They build in intensity until the relay actuates, at which time the alarm goes steady-state, and past a user-set delay (T), the relay drops into the over-voltage state.

Is this pulse coming from the Classic's?  It happens day or night (I'd thought it might be sweeping of the PV array, form our MX60).  Any ideas on what this is/how to remedy?  Happy New Year.  Thanks for any advise, ~ks

http://www.fieldlines.com/index.php/topic,148038.0.html?PHPSESSID=711367f1920ade2e5e77e3df10beb8ea

[kitestrings]

I'd meant to mention one other thing.  In looking thru some of the FAQs at Omron, I found some discussion related to SS relay malfunctions where a possible cause is attributed to:

"An inverter-controlled power supply was used (including high-frequency noise)"

They go on to say,

"As protection against external noise, it is recommended that an overvoltage protection element (e.g., a varistor) be externally connected to the power supply line."

Perhaps related?

[kitestrings]

Could this pulse be related to the arc-fault protection?

[Halfcrazy]

Not sure I am following the issue. Are you saying the DC line on the input side of the classic is getting pulses on it?

[zoneblue]

Reading between the lines, this is a wind system without a clipper. Your using the vc relay to limit high Voc? Is that right?

[kitestrings]

Ryan,

Yes.  I had it originally on the two of the phase leads on the AC side.  I thought perhaps a problem was that the frequency was too low.  The Omron is rated for 40-500HZ input, and the axial operates largely under 30 HZ (about 12 HZ at cut-in IIRC.  So, I moved it to the DC input, since it is rated for both AC and DC with selectable ranges of 20-200, 30-300, or 60-600V AC/DC.

I'm not sure of the origin, but it seems to be getting this pulse on the input which grows in intensity and trips out the relay.  I thought tomorrow I'd try isolating one or both CC, and/or the inverter to see where it is coming from.

I tried the relay with just a nominal 120VAC input, (and 24VDC supply) and it seems to work fine.

Zoneblue,

It does not have a MS Clipper, but it does have a "homemade" clipper.  The relay is just a fail-safe protecting the CC's, so if a $40 relay fail we don't smoke a more expensive controller (at least that's the theory).

Rob Becker's has reportedly used this effectively to protect his Aurora grid-tied inverters in wind applications:
http://www.solacity.com/Docs/Wind%20Turbine%20Overvoltage%20Protection.pdf


~kitestrings

[dgd]

Just an idea..
Have you tried inserting a diode between the Omron and the Classic so that the Omron is not trying to draw current from the Classic? Is the alarm a high voltage detection alarm or a reverse current alarm?
dgd

[dgd]

It does not have a MS Clipper, but it does have a "homemade" clipper.  The relay is just a fail-safe protecting the CC's, so if a $40 relay fail we don't smoke a more expensive controller (at least that's the theory).
~kitestrings

You could also use the PV input high on AUX1 to detect an overvoltage condition and either enable ssr to clamp turbine AC lines via the clipper resistors or stop turbine.
dgd

[kitestrings]

dhd,

I haven't tried the diode, but I sure could.  I did try a varistor that I had kicking around, but it didn't seem to change anything.

You could also use the PV input high on AUX1 to detect an overvoltage condition and either enable ssr to clamp turbine AC lines via the clipper resistors or stop turbine.

There's two problems that I see with this:

First, we're using the Aux2 to divert to our water pre-heat diversion load, and Aux1 to clip/short the windings, or maybe its the other way around(?).  I'm not home right now).  Although, we have two controllers.  Do you know if it's possible to have use of 2- Aux1's and 2- Aux2's if they are networked on a single system like this?

Perhaps more importantly, the scheme was intended to protect the CC's, independent of the CC's.  What if the failure starts with the Classic?

Appreciate your suggestions,  ~kS

[kitestrings]

Well, I finally got the Omron to work as intended this afternoon.  I may let others better explain why this works, but for anyone attempting it, here is what my experience has been. This is the unit that we're using (K8AB-VS3, 24 VDC):

http://www.ia.omron.com/data_pdf/data_sheet/k8ab-vs_ds_csm384.pdf

In the overvoltage operation mode, DIP switch #3 is ON, Operating Mode - Overvoltage; switch #4 is OFF, Relay Drive Method  - Normally Closed.  With the input voltage into terminal #2, the range should be 30-300 V AC/DC.  I'd set the set voltage, SV, to about 76%.  The threshold should then be (300 x .76= ) 228 VDC.  With the hysteresis set at 10% the reset should be 228 - (228 x .10 =) 205 V.  My input voltage today was about 36 VDC (no wind).

Studying the operation sequence and schematic (pg5), we expected that after energizing, and a short delay (T1), that the relay should operate (Green LED ON) and terminals 11 & 14 to be closed; 11 & 12 should be open. 

What I found was that the relay light went out whenever the input voltage was present.  Energized, no reference voltage and the relay light was on, but as soon as a attached the input to V2 the relay went off.  Regardless of the dial settings I could not change the state.  I tried it with a 120VAC source; same result.

I further discovered that, at least with the DC input, that I had the reference voltage across V2 and the A1, the negative terminal of my 24VDC supply.  So, when I removed the "COM" connection - EUREKA - the relay actuated, the LED illuminated, and I could drop the SV slowly down to ~12% and have the alarm flash, then actuate (Red alarm LED illuminates.  Once the alarm state occurred it locked out as I had the manual reset selected.  I cycled thru a few more operations, but end result is as long as COM is not connected it works perfectly.

In our set-up terminals 11 & 14 close to hold the SS relay connecting the turbine under normal conditions.  If we exceed the SV the relay opens disconnecting the turbine, and simultaneously closes the circuit to our clipper relay (via terminals 11 & 12).  Rob's schematic shows this, you just have to note that the drawing is in the de-energized state.

Hope this maybe helps someone else.

~kitestrings

[dgd]

Ks,
So when the input voltage drops back into the acceptable range the OMRON reconnects the input and disconnects the Clipper circuit?
I assume there are delay times included in this process so that its not just chattering between Clipper on and Clipper off?
$40 where?

I was also not sure that the hysterisis % just simply reduced the detection of - or value of the trigger voltage, so the trigger could actually be 'from anywhere' in the range 10% below the defined switch voltage?
Or perhaps also upwards of the defined trigger voltage - 5% or 10% either side of it?
Do I have this wrong?

dgd

[kitestrings]

dgd,

You can have it auto reset (DIP switch #2 ON).  I have set for manual reset (#2 - OFF).  With manual rest you must turn off supply power to have it resume operation.  (I guess this somewhat negates the reset threshold, but to start with at least I want to know if it operates).

There is an operating time adjustment (T, .1-30s), but this I believe is on the detection of the overvoltage condition.  So, presumably you could ignore a momentary blip.  I have it set for .1s.  The delay, if you were using auto reset I believe would be controlled by lowering the reset (so the wind would have to die down before it resumed.

$40 where?

This was just illustrative - my water heater diversion relay is about that much - say a relay failed, or we tripped a breaker to the Classic, or a component in the Classic failed for whatever reason.  The Omron is an attempt to have some redundancy to the overvoltage protection is all.  My understanding is, beyond hyper VOC noted - the CCs have little to no tolerance for an overvoltage input.  I view this condition as quite possible if power is not being used, or diverted somewhere (or we're clipping).

The SV is set based on a percentage of the upper range (for V1 200, V2 300, V3 600).  You can select between 10% and 100%.  The Hys setting is 5-50% of the operating value.  Example: I have SV set at 76%, and I'm on V2.  So it should operate above 228V.  We could select between 30 -300V.  The Hys is set at 10%, so it would reconnect at 205V if auto reset were enabled.

I'm learning more as I go, but having played with this I think this is correct.

We finally had wind today.  Saw brief output up to about 1.2 kW!

~ks

[kitestrings]

This is older post, but I realized that I never followed up on it.  We're using the Omron voltage sensing relay for upper limit protection of the Classics (with wind).

I wonder though, from what you describe, if your relay is on the blink. It just doesn't sound right, and doesn't jive with what I've seen with the Omron relay...

Rob was right.  I bought a second relay; identical make model, and it works as intended.  The original relay, new out of the box, acted oddly from the start, but since it was new I'd just assumed that I had some other problem.  The original relay seemed to have lost the isolation between the DC negative and the "Com" terminal.

I had set them up side by side in a simple test application. Power (24VDC) to the unit, with an SV setting of 20%, and a probe on V1 (20-200VAC/DC), we would expect the thing to trip above (200 x .2 =) 40V.  I could choose a voltage above or below this on our battery bank.  The new one worked fine.  The original unit would drop out as soon as I connected the “Com” terminal.

The nice thing is that with the new unit I’m able to monitor the upper range of the 3-phase AC voltage from the turbine.  As Rob had pointed out this is better if the rectifier, or a diode(s) should fail.  In our case we have two isolated DC inputs, from two separate bridges, so there was no way to protect both controllers on the DC side.

Rob Beckers has a number of variations of this protection scheme for his grid-tied Aurora inverters.  We're using one to hopefully protect our CCs.  It seems like inexpensive insurance.

~kitestrings