120VAC Control Power Conditioner 2

[LionelHutz]

I have run into a difficult situation at a site. The power is experiencing 30% to 80% THDv with a corresponding 25% voltage rise. The worst harmonics are the 59th and the 61st, with lower levels of 53rd, 55th, 65th and 67th. See the attached files showing this distortion. Anyone have a suggestion for a product that would filter this to clean up the 120VAC control power?

 

[Erro11]

Depends on the post filtered load.
If the load is not too heavy, an Uninterruptible Power Supply (UPS) may be the best because it creates its own AC output.
Otherwise, there are large power line conditioners available.
Google:
"Uninterruptible Power Supply"
"power line conditioners"

 

[racookpe1978]

You quoted a 120VAC load, but not how much was required - now, and in future upgrades.

There are a few solutions, but which might work will depend on what size the load is.

 

[LionelHutz]

The requirement is a few 100VA continuous and maybe 750VA peak. The problem is we have a 3rd party circuit board which is failing on this power. I would not trust a consumer grade conditioner on this system and I've sent a few emails to suppliers of industrial equipment and the responses weren't very positive. I doubt a typical device with a common mode choke, capacitors and varistors would fix the problem and a ferroresonant device could be overloaded by the high voltage and harmonics. Same with a UPS, even the on-line types. It's possible the distortion will cause the charging circuit to shut down or be damaged and then the battery goes dead.

 

[catserveng]

I was involved for a military project a few years ago, we had similar issues with multiple vendors control components, this company provided a power conditioner we used with great success,

http://www.synqor.com/mpc/

I also ran across these units on some diesel electric Coast Guard ice breakers after we had identified power supply issues for our controls, this is what the Coast Guard/shipyard sourced and they appeared to take care of the problems

http://www.clary.com/products/military-ups/cmn-series

Mike L.

 

[jraef]

A Ferroresonnant power conditioner, like a Sola MCR, will absorb (trap) the harmonics and essentially burn them off as heat. They have fallen out of favor of late because of their down side detractors, but this might be a case where the good outweighs the bad.

The down side will be the heat, plus the fact that ferros clip current peaks, so they have to be seriously over sized if supplying non-linear loads (which is just about everything now). You may need a 1kVA unit or maybe even 1.5 kVA depending on the nature of your loads and experience quite a heat gain if it is inside of a control cabinet. I would go for a hard-wired version and mount it on the outside if at all possible.

"Will work for (the memory of) salami"

 

[LionelHutz]

That's what the customer did Jeff, even though Sola advised that their units were not suitable for this application. If the Sola fails then we know Sola was correct with their advice.

 

[hacksaw]

what does the line voltage look like (time domain), and how is your cabling run

 

[jraef]

Most likely everyone at Sola who truly understood the CVS and MCR products is long gone now, sloughed off in one set of layoffs or another. The kids on staff now probably don't even know that the ferroresonant transformer was INVENTED by Joseph Sola...

Years ago I was did some work for Boeing Computer Services putting together an inventory management system at the assembly floor based on networked 80286 IBM PCs ("Gearboxes" if anyone remembers those behemoths). At that time the concept of "harmonics" in AC power lines caused by non-linear loads was still relatively new. BCS, being a very early adopter of PCs in mass quantities, began having problems with neutral conductors inside of walls actually starting fires. We put MCS units in front of every PC on our project in the Renton facility, the problems all went away. Like I said though, the ferros don't like high peak current gulps, they clip them, so the units had to be over sized. We ended up having to put 1kVA units on PCs with 500W power supplies.

"Will work for (the memory of) salami"

 

[racookpe1978]

Short term basis, immediate solution to the problem, is the commercial battery-backup UPS system rated for the 100 KVA with the startup surge you mentioned.

It sounds almost contradictory, but for this case, don't try to over-design something "perfect" because the ideal solution might always be "one problem away" or "one budget too expensive" or "one more part needs to be optimized" (or designed) to make the "best black box" from getting solved. For example, the Army had a problem with its boots in the desert. So, to solve it, they used the original "bad boots" in the desert for 15 years (and still are using them) while the Army/government/research labs/universities/bureaucracies tried to get new boots. They were on the shelf the whole time fro commercial fabricators. The new boots have at last been designed, but are not yet fabricated. None have been issued. "Testing" is still underway. )

 

[ScottyUK]

100kVA. Do you think that will be big enough? ;-)

 

[racookpe1978]

Nope. My error. 8<)

Motor Generator ?

 

[sibeen]

I haven't seen a harmonic issue that bad before. You might end up having to go for a bespoke arrangement. As you mentioned, an on-line UPS isn't going to the able to handle this as a stand alone unit. I'd still use a UPS but I'd place some substantial filtering at the input of the device. This would hopefully knock out most of the harmonics and let the UPS accept the input voltage. Even with that I wouldn't rely on the standard 7 to 10 minutes autonomy time that a standard UPS gives and I'd opt for a UPS that has the capacity for an extended battery to be added.

 

[controlsdude]

Constant voltage transfromers use them all the time for all the sensitive electronics inside control panels (plcs, displays, and such items)

 

[LiteYear]

Excuse me if my inexperience puts me out of line here, but is there any sense in addressing the source of the harmonics, rather than suppressing it en masse?

It looks like there could be a few noisy switch mode power supplies on the line. If upgrading them to something less noisy is not feasible, could they each be supplied with a local source of harmonic current like an appropriately sized capacitor? I suggest this it seems the THDv is particular high when the network is unloaded, so I suppose its mostly due to inductive voltage drop from the high frequency current draw of the non-linear loads. The actual current draw could be quite small, so a few well placed capacitors could work out to be a much smaller investment than a network-level power conditioner.

 

[LionelHutz]

I'm told the cause of the harmonics is a 11kV cable that resonantes with the transformer inductance at low loads. I'd think it would be fixed already if it was easy to address.

 

[iop995]

I think a good input filter (eventualy more cascaded filters) and an online UPS (metalic chassis good earth link) solve this problem. Maybe need to use good shielded cables to UPS and from UPS to sensitive load.

 

[1capybara]

bandpass filter or low-pass filter, either one

 

[djs]

The only time I have seen a ferro-resonate transformer burn out was due to high harmonics. This unit was locate on an oil drilling rig where the AC power was converted to DC power for the rig's motors using SCR converters. So I am with Sola on this, if you have a high harmonic environment, then ferro-resonate transformer is not recommended. Perhaps an online typ of UPS will work. This UPS converts the incoming power to DC to charge a battery, and the output is an inverter that converts the battery voltage to the desired AC.