Flickering LED Fault cause found, I think.

[GBDamo]

Have had a recall on an office LED swap out.

Some of the panel LEDs flicker for a fraction of a second very randomly.

The panels are fed from four three phase DBs, and all the boards and all the phases have aircon(AC) supplies fed from them. Now, some of the AC units are inverted but some of them are not.

My thinking is that the compressors in the older non-inverted AC units are possibly on their last legs and as a result are using higher than design current on start up. This is causing a dip in voltage on the circuit, as LED have no latency this voltage drop is enough that the LED power down for a fraction of a second. This can be proved by measuring the inrush current and comparing this to the design spec.

Can anyone recommend a reasonably priced clamp meter that will reliably and accurately measure inrush/start up current.

My options if the AC is causing this is to recommend replacement, preferably, or possibly fit hard start kits to the offending units.

Does this diagnosis sound reasonable? any other thoughts or possibilities.

 

[Pete999]

Have had a recall on an office LED swap out.

Some of the panel LEDs flicker for a fraction of a second very randomly.

The panels are fed from four three phase DBs, and all the boards and all the phases have aircon(AC) supplies fed from them. Now, some of the AC units are inverted but some of them are not.

My thinking is that the compressors in the older non-inverted AC units are possibly on their last legs and as a result are using higher than design current on start up. This is causing a dip in voltage on the circuit, as LED have no latency this voltage drop is enough that the LED power down for a fraction of a second. This can be proved by measuring the inrush current and comparing this to the design spec.

Can anyone recommend a reasonably priced clamp meter that will reliably and accurately measure inrush/start up current.

My options if the AC is causing this is to recommend replacement, preferably, or possibly fit hard start kits to the offending units.

Does this diagnosis sound reasonable? any other thoughts or possibilities.

Don't you have clamp meter?

 

[westward10]

Isolate the suspect ac circuits and see what happens.

 

[Wilko]

Hi - without something special you probably won't capture the initial current spike. My thought would be to reparent some equipment and see if the fault changes. @westward10 has beaten me to the draw !

 

[westward10]

Reparent ??

 

[Gavin John Hyde]

Reparent ??

Reparenting is a form of psychotherapy in which the therapist actively assumes the role of a new or surrogate parental figure for the client, in order to treat psychological disturbances caused by defective, even abusive, parenting

I can only assume @Wilko has some issues that he is currently working through with the care bears...

 

[Wilko]

I was thinking to move aircon off the phase that's got the flicky flick lights, if that's possible. But of course simplest is best as per @westward10 post.

 

[Taylortwocities]

It may be stray voltages induced on the cabling.
Often, this can be cured by fitting a C-R network across one of the light fittings. Search for 'contact suppressor'
Something like this would do you:
https://www.amazon.co.uk/CONTACT-SUPPRESSOR-0-1UF-100R-Suppressors/dp/B00SYIV79S

 

[GBDamo]

T

Isolate the suspect ac circuits and see what happens.

That'd be too easy,

There are too many AC units that are non-inverted and therefore suspect. These are nicely spread across all DBs and phases.

I was hoping to add a little meat to the bones of my AC theory before carrying out open heart surgery on the DBs.

If we were to go down the "reparenting" route all the ACs would need to go onto a phase with no lighting. I'd like to be a little more convinced i'm not talking rubbish before this happens.

 

[westward10]

Can't you go to the units and switch them off.

 

[Wilko]

Can't you go to the units and switch them off.

Agree - should have local iso - pick light that flicks most often and iso the relevant aircon units. Dare I ask, have you seen this flickering your good self?

 

[GBDamo]

Agree - should have local iso - pick light that flicks most often and iso the relevant aircon units. Dare I ask, have you seen this flickering your good self?

We are only allowed on site out of hours, typically when the AC is not in demand, so no I've not personally seen the flicker although I have no reason to doubt it as we have a good relationship with the customer and it has been reported in various areas.

What I'd like to do is force start each AC unit in turn, measure the inrush current and isolate (yes they have local rotary switches) those that give notably high readings.

Then give it a few days with these off, see if this cures the issue.

Then either;-

1, get them replaced - not likely..

2, move them off of lighting phases, I've seen inside the DBs and it makes me shudder.

3, Fit hard start kits in an attempt to cut the inrush current (units well, well outside of any warranty. I've heard good things about these but have no personal experience. Easy enough to do and not too expensive, just piggy backing a cap and relay onto the run cap.

But first I'd like to see a significantly large inrush current on one or more of the ACs that I can rely on hence asking for recommendations on clamp meters with this facility.

 

[UKMeterman]

Put a power quality analyzer on the circuit. And leave it connected for 24 hours, it will tell you if you do have a power quality issue.

 

[marconi]

You might want to think about using an electronic soft start module SSC 25AL from Reichelt electronics. Needs a heat sink. Module is circa £25. It connects into the L N feed to the compressor. You could do a trial with one to check it fixes the flickering led problem.

 

[marconi]

GBDamo - good morning. How are you getting on with this problem please?