Flicker Issue, Synchronous Condenser, STATCOM or SVC

[spikes2020]

I don't expect a lot of help from here, but any advice on any these items would be helpful (Synchronous Condenser, STATCOM or SVC).

I currently work at a steel mill using an arc furnace that operates 3 phase ~20,000A at 500-800V there is an 'on load' tap-changer 13.8kv 45MVA transformer. Anyway our issue is we have a flicker issue along with power factor and some harmonics. We have a 3rd harmonic filter, and a capactior bank for the power factor, but i think both of them they are slightly undersized.

The utility would like us to fix some of these issues. We are looking at a synchronous condenser, STATCOM or SVC.
Anyone else use an arc furnace? what do you use?
Has anyone looked at or used any of these?
What is the maintance on them?
Did they solve your problem?
Any advice on further reading on these devices?

Just a shot in the dark

Thanks!

Jon W

 

[Jraef]

I'm not an expert on arc furnaces, but a Synchronous Condenser, STATCON or SVC is not going to address flicker, unless the PoCo has maybe already determined that poor control of your PF is what is causing the voltage drop because you are near the limits of their T&D setup. Is that what they are saying? I for one would want that made REALLY clear, in writing, before going through the expense.

if so, and you have some large induction motor loads that run non-stop, such as compressors or blowers, then consider replacing those induction motors with synchronous and use them as synchronous condensers.

Also, if harmonics are an issue, an Active Harmonic Filter might be in order, although I'm not sure if anyone makes one big enough. They typically do what a STATCON does, plus inject mitigating harmonic corrective current.

 

[spikes2020]

I'm not an expert on arc furnaces, but a Synchronous Condenser, STATCON or SVC is not going to address flicker, unless the PoCo has maybe already determined that poor control of your PF is what is causing the voltage drop because you are near the limits of their T&D setup. Is that what they are saying? I for one would want that made REALLY clear, in writing, before going through the expense.

if so, and you have some large induction motor loads that run non-stop, such as compressors or blowers, then consider replacing those induction motors with synchronous and use them as synchronous condensers.

Also, if harmonics are an issue, an Active Harmonic Filter might be in order, although I'm not sure if anyone makes one big enough. They typically do what a STATCON does, plus inject mitigating harmonic corrective current.

Hey thanks for the fast reply!

Just imagen a 4 year old with a crayon drawing.... that is what our power waveform looks like

I don't think it is our PF that is cauing the flicker issue, i think it is more of harmonic issue... even with x2 "c-type" 3rd harmonic filters of 8MVA each we still have a significant 3rd(1%), 4th(1.2%) 5th(2%) and 7th(2%) harmonics... The filter we use is a high pass filter with a resonance frequency at 60 hz so its inpeadance looks like a "check mark" centered over the 3rd harmonic (180hz). I think this maybe some of the cause as it dose nothing for the lower frequencies (0-60hz). Isn't that is what causes the flicker? Or so i have read... around 8hz or so?

I like your idea of replacing some of the motors with synchronous ones... not sure if we would need a controller for them or what? (would help our PF a bit) We have x3 4160 motors controlled by a AllenB VFD of some sort and several large fans and other pumps.... Maybe you could expand on this a bit more, and/or ill look into it.

I think an SVC is more or less an active filter... but i like the STATCOM better as it seems more forceful and easier to upgrade later on. But i don't know which one would be better for our issue.

 

[Phil Corso]

Spikes,

To start, search WEB for EPRI papers dealing with EAF Flicker!

Regards, Phil Corso

 

[ghostbuster]

We just finished investigating a similar problem/solution and have recorded all the waveform traces etc.

Standing in the electrical room when the furnace starts up and draws an arc sounds like you are standing directly under a Nasa spaceship during taking off. 50% of the outdoor H.V. caps were blown in all the filters when we arrived.

All new caps were installed and instantly blew up and exploded when the furnace started up again.

They were on fire and there was oil everywhere.The continuous distortion injected back onto the H.V. line is extremely high--but the utility is currently not aware that they are the culprit.

Customer is unwilling to pay the big bucks to implement the new design solution until he is "pushed into it" by the utility.

There is a good IEEE paper from a Korean arc furnace group that discusses potential solutions.

Good Luck.

 

[Jraef]

How old are your AB MV drives? Because if they are PF7000 with an Active Front End (AFE), they can already be run in leading PF as they are. If they are not the AFE version, they can be easily upgraded. If they are old 1557 drives however, they cannot.

 

[spikes2020]

How old are your AB MV drives? Because if they are PF7000 with an Active Front End (AFE), they can already be run in leading PF as they are. If they are not the AFE version, they can be easily upgraded. If they are old 1557 drives however, they cannot.

Um... 20-30 years old?

I think we have 1 new one that is 15 years old.

I'll go take a look at them, they each take up an entire room.

 

[jcormack]

If your Rockwell drives are that old they are not the newer gen PF7000's - even the 15 year old one is most likely a 1557, they were just beginning to provide those in 2002 methinks (we installed a 1557 in 2003, that we just upgraded to PF7000 this year). How does your tap-changer work ? is it feed forward or is it completely load reactive (switch on voltage drop or current)? Each time it changes you get a transient spike, (is it constantly changing?) you could switch to an SCR based tap changer. Changing other fixed loads to Sync motors is a good suggestion.

 

[spikes2020]

If your Rockwell drives are that old they are not the newer gen PF7000's - even the 15 year old one is most likely a 1557, they were just beginning to provide those in 2002 methinks (we installed a 1557 in 2003, that we just upgraded to PF7000 this year). How does your tap-changer work ? is it feed forward or is it completely load reactive (switch on voltage drop or current)? Each time it changes you get a transient spike, (is it constantly changing?) you could switch to an SCR based tap changer. Changing other fixed loads to Sync motors is a good suggestion.

Still haven't had time to go look at them, maybe tomorrow....

The tap changer is really cool, and I was wondering about that as well. I have the manual here, it switches on the high side 13.8kv and it makes before it breaker between each tap and uses a resistor inline. We could upgrade to vacuum bottles but it would be really expensive. We are just replacing the contacts on our EAF transformer right now and I got some pictures of the tap changer~ It can tap from 480 to 900V live at 20,000A....

Yeah we have a 2000 HP fan that runs 100% of the time that we could change out.. it would be a start anyway.

I don't think we get any spikes on our voltage from tap changing or not that we notice yet anyway. I am still trying to pull information from the PTs. Working on getting it feed to my office so i dont have to go to the substation each time...

 

[spikes2020]

Ok i had time to run over there and take a look at these drives....
Yeah all 3 look like 1557's...


Still looking for any opinions on Static Bar or SVC's

I think with all the reading that a static bar reacts faster and wih its DC capacitor bank seems more capable...

 

[Phil Corso]

Spikes,

How about providing a simple SLD?

Regards, Phil

 

[spikes2020]

Spikes,

How about providing a simple SLD?

Regards, Phil

Here is a SLD of our main facility, an example of thier SVC and STATCOM.

The STATCOM looks a lot like a VFD... also thats why i feel more comfortable with it.

We are going to add this to the 13.8kv EAF bus


Edit: the oneline was really small so i tried to fix it...

 

[Jraef]

Looks like a decent solution, but have you determined yet that low PF is the actual cause of the flicker? That looks like an expensive way to experiment.

 

[spikes2020]

Looks like a decent solution, but have you determined yet that low PF is the actual cause of the flicker? That looks like an expensive way to experiment.

Well we had a specialist come in and take a look for us. He gave us 3 solutions to the issue. The Synchronous Condenser though is way too expensive compaired to the rest.

I was just trying to see if anyone else knows about these. Has a recomendataion of one over the other or has a serious beef with one?

 

[Phil Corso]

Striker,

1) Can you correlate the flicker with a specific operation? For example, coincident with operation of the EAF's transformers primary tap-change?

2) Do you have a "flicker" monitor or instrument?

Phil

 

[jdsmith]

We really need to understand your maximum MVA load and the power factor you are running at that maximum MVA demand. Power factor is a big part of analyzing voltage drop/voltage regulation problems, and this type of flicker is just a fast acting voltage regulation problem. I dont have enough data to determine if your existing filter is undersized, but there are a couple elements that I would expect in your existing filter that are not present.

For a project of this magnitude it might be a good idea to hire a second expert and compare their recommendations with the first set of recommendations you have. Feel free to PM me if you would like more information - I know a number of folks that do this type of work regularly.